Karsten Müller, Chenzi Xu, Mohamed Lehbib, Ziliang Chen — International Finance and Macroeconomics Data Session
This paper introduces the Global Macro Database, an open-source, continuously updated dataset of macroeconomic statistics that unifies and extends existing resources, covering 46 variables across 243 countries from 1086 to 2030.
Finance Application
- This extensive, long-run dataset is invaluable for asset pricing research, particularly for modeling disaster risk and long-run risk.
- The persistent GDP losses from financial crises and climate shocks could be incorporated into models to better price equity risk premia, bond yields, and real estate, reflecting the true long-term macroeconomic state.
- For household finance, understanding these protracted impacts can inform optimal savings and investment strategies, especially for long-horizon goals like retirement.
- In insurance, the data on persistent climate-related GDP contractions can refine pricing for climate risk transfer products and improve sovereign risk assessments for exposed countries.
MacroeconomicsFinancial CrisesClimate RiskDataLong-Run RiskAsset PricingEconomic HistoryGDPTemperature ShocksDisaster RiskHousehold FinanceInsurance
Core finding, identification, data
Core Finding
- The Global Macro Database reveals that financial crises lead to statistically detectable real GDP contractions for up to 50 years, with losses potentially twice as large as previously estimated.
- Global temperature shocks also predict significant real GDP contractions for up to 30 years, particularly in emerging economies, highlighting the profound and persistent long-term economic costs of these events.
Identification Strategy
- The paper's core innovation is a systematic data harmonization and integration methodology, using a transparent ratio-splicing approach to combine 110 diverse historical and contemporary sources.
- For its applications, it employs local projections (Jordà, 2005) to estimate the long-run impact of financial crises and global temperature shocks on real GDP, treating crisis events and temperature deviations as the primary shocks.
Data
The paper introduces the Global Macro Database, which integrates data from 32 major contemporary sources (IMF, World Bank, OECD) and 78 historical datasets. It provides annual time series for 46 macroeconomic variables across 243 countries, spanning from 1086 to 2030.
Paul Schmelzing, Kenneth S. Rogoff, Barbara Rossi — Macro Public Finance
This paper re-examines the econometric properties of short-maturity real interest rates and term spreads using multi-century, high-frequency data for the UK and US.
Finance Application
- The finding of trend stationarity and short half-lives for real rates and term spreads implies that long-term bond yields and their risk premia may be more predictable and mean-reverting than current models assuming unit roots suggest, leading to new specifications for affine term structure models and improved pricing of long-duration assets.
- If real rates are mean-reverting to a long-run trend and term spreads are secularly rising, households might adjust their intertemporal consumption-saving decisions, particularly regarding long-term investments and mortgage choices.
- Insurance companies, especially those with long-duration liabilities like annuities, could significantly revise their asset-liability management (ALM) strategies, altering assumptions about future investment returns and discount rates, influencing product pricing, reserving, and hedging strategies for interest rate risk.
Interest RatesTerm StructureReal RatesTerm SpreadsStationarityUnit RootsStructural BreaksLong-Run DataMonetary PolicyAsset PricingBond MarketsRisk PremiaInflation VolatilityHistorical DataEconometrics
Core finding, identification, data
Core Finding
- The paper finds strong and consistent evidence of trend stationarity in very long-run series of short-maturity real interest rates and term spreads, with relatively fast adjustment speeds (half-lives of 1-3 years) and a paucity of structural breaks.
- This contradicts the prevailing consensus of non-stationarity and permanently lower r* post-financial crisis.
- Furthermore, it shows that term spreads have been secularly rising while inflation volatility has trended in the opposite direction, questioning influential term structure models.
Identification Strategy
- The paper's identification strategy relies on leveraging newly available, multi-century, high-frequency (quarterly), and methodologically consistent time series data for short-maturity real interest rates and term spreads in the UK and US.
- It employs rigorous econometric tests, including ADF-GLS, Zivot-Andrews (allowing for structural breaks), and Bai-Perron tests, to establish trend stationarity and assess structural breaks.
- The robustness of findings is further validated through out-of-sample forecasting comparisons, where stationary models outperform non-stationary ones.
Data
The paper utilizes multi-century quarterly data for short-maturity real interest rates and term spreads for the United Kingdom (from 1704/1718/1753) and the United States (from 1831). These series are constructed from nominal rates (e.g., prime commercial paper rates, Bank of England rates, consol yields, mortgage rates) and inflation series (wholesale price indices, CPI) from various historical sources and financial historians (e.g., Bigelow, Macaulay, Board, Dimsdale and Thomas, Warren and Pearson, BLS, Officer).
Katharina Bergant, Andrés Fernández, Ken Teoh, Martín Uribe — International Finance and Macroeconomics Data Session
This paper develops a novel, high-resolution, daily dataset of cross-border flow restrictions worldwide from 1950 to 2022 by applying large language models to semi-structured IMF documents.
Finance Application
- This granular, daily dataset of cross-border flow restrictions offers significant opportunities for finance research.
- In asset pricing, it could be used to conduct high-frequency event studies on how the *announcement* or *implementation* of specific capital controls (e.g., taxes on foreign investment, repatriation requirements) impacts local and international asset prices, currency exchange rates, and sovereign bond yields.
- Researchers could examine whether these restrictions create market segmentation, affect risk premia for different asset classes (equities, bonds, FX), or influence cross-border capital allocation decisions by institutional investors.
- For household finance, the data could shed light on how restrictions on foreign asset ownership or transfers affect household portfolio choices, wealth accumulation, and consumption smoothing, especially in emerging markets.
Capital ControlsCross-Border FlowsFinancial RegulationAsset PricingInternational FinanceTextual AnalysisMachine LearningEvent StudiesEmerging MarketsExchange RatesSovereign DebtMarket SegmentationHousehold Portfolio Choice
Core finding, identification, data
Core Finding
The paper identifies five stylized facts: current account restrictions were significant pre-Bretton Woods collapse, shifting to capital account and financial sector measures later; financial liberalization has been non-linear with periods of tightening; liberalization pace is uneven across income groups (higher income liberalizing faster); quantity-based restrictions liberalized most, while administrative controls persist in low-income countries; and outflow restrictions liberalized faster than inflow restrictions.
Identification Strategy
- The methodological innovation lies in using state-of-the-art Large Language Models (LLMs) to extract, classify, and analyze de jure restrictions from unstructured and semi-structured textual data (IMF AREAER reports).
- This allows for unprecedented granularity, including daily frequency of policy adjustments and classification across multiple dimensions (direction, type, category, flow direction), which can be leveraged for event-study type identification.
Data
The primary data source is the International Monetary Fund's (IMF) Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER) from 1950 to 2022. It also uses the IMF's Taxonomy of Capital Flow Management Measures for analyzing motivations.
Simone Lenzu, David Rivers, Joris Tielens, Shi Hu — Corporate Finance
This paper studies the interconnection between the productivity and pricing effects of financial shocks, showing how firms' pricing policies play a crucial role in mitigating the long-run impact on physical productivity.
Finance Application
- This research offers several avenues for finance applications.
- In asset pricing, the delayed and persistent impact on TFPQ, and the differential pricing strategies of firms, could be used to construct novel risk factors that explain cross-sectional stock returns or credit spreads during and after financial crises.
- For corporate finance, lenders could integrate firms' inventory management capabilities and pricing flexibility into credit risk models, as these factors are shown to mitigate the negative impact of credit shocks on long-term productivity.
- Furthermore, understanding firms' liquidity generation through inventory liquidation could inform supply chain finance, helping assess the financial resilience of trading partners.
Financial ShocksProductivityPricingCredit ConstraintsFirm DynamicsInventory ManagementTFPQTFPRAsset PricingCorporate FinanceCredit RiskInnovationNatural Experiment
Core finding, identification, data
Core Finding
- A tightening of credit conditions has a persistent, yet delayed, negative effect on firms' long-run physical productivity growth (TFPQ), but also induces firms to change their pricing policies.
- Revenue-based productivity measures (TFPR) offer biased predictions, underestimating the long-run elasticity of physical productivity to credit supply by half.
- Firms use low pricing as a source of internal financing, allowing them to avoid cutting expenditures on productivity-enhancing activities, thereby softening the impact of financial shocks.
Identification Strategy
- The study uses the 2010-2012 European sovereign debt crisis as a natural experiment.
- Firm-specific credit supply shifters are constructed based on firms' heterogeneous exposure to banks differentially impacted by the crisis, allowing for the identification of causal impacts on firm-level productivity and pricing, separate from demand-side factors.
Data
The paper utilizes a novel micro-level panel dataset for Belgian manufacturing firms, combining confidential administrative records from PRODCOM (firm/product-specific output prices and quantities), firm annual accounts (balance sheets, income statements), corporate credit registry (firm-bank credit relationships), and bank balance sheet data from supervisory records.
Benny Kleinman, Ernest Liu, Stephen J. Redding — International Trade & Investment
This paper develops a tractable approach to characterize the distributions of endogenous variables in quantitative trade models under aggregate uncertainty and incomplete markets, allowing agents to make ex ante investments in import and export capacity that determine resilience.
Finance Application
- This paper's framework, which models how agents make ex ante investments in import and export capacity to manage aggregate uncertainty and its second moments (variances and covariances), has direct applications in corporate finance and macro-finance.
- In corporate finance, it could inform optimal supply chain design and resilience strategies for multinational corporations, where firms choose supplier diversification and logistics investments to balance expected costs against the covariance risk of disruptions, akin to portfolio optimization.
- In macro-finance, the concept of a country's 'risk profile' as a determinant of welfare and trade could be used to explain cross-country differences in asset risk premia (e.g., sovereign bonds, equities), as investors demand higher compensation for exposure to countries with less diversified or more volatile productivity shocks, especially when financial markets are incomplete.
UncertaintyRisk ManagementSupply Chain ResilienceInternational FinancePortfolio ChoiceMacro-FinanceGeneral Equilibrium
Core finding, identification, data
Core Finding
- The paper finds that a country's risk profile, including the variance and covariance of costs across countries, is a key determinant of its comparative advantage, terms of trade, and welfare.
- Trade liberalization affects welfare through both expected real income and the variance of real income, with endogenous ex ante investments in import/export capacity playing a crucial mediating role.
- Changes in global uncertainty or the covariance structure of shocks have heterogeneous welfare effects across countries, depending on their ability to hedge against aggregate uncertainty through these investments.
Identification Strategy
- The core methodological innovation is a tractable approach to incorporate aggregate uncertainty into quantitative trade models with constant trade elasticity.
- This is achieved by characterizing the general equilibrium using a second-order Taylor-series expansion of the system of general equilibrium conditions around the mean values of log productivities and bilateral trade costs.
- This allows for analytical characterization of the first and second moments of endogenous variables (like wages, prices, welfare) in a multi-country setting with incomplete markets, even with general distributions of aggregate shocks.
Data
The paper uses panel data for 106 countries from 1993-2017, drawing international trade data from IMF DOTS, national accounts data (GDP, population, price indexes) from Penn World Tables (PWT) and Global Macro Database (GMD), and gross output data from WIOD. These data are used to estimate productivity and trade costs, and their variance-covariance matrix.
Paul Beaudry, Paolo Cavallino, Tim Willems — Capital Markets and the Economy
This paper develops a Finitely-Lived Agent New Keynesian (FLANK) model to explain how central banks can affect long-term real interest rates without strong effects on activity, due to offsetting asset valuation and consumption-savings effects.
Finance Application
- This research has significant implications for asset pricing, particularly for long-duration assets like long-term bonds, real estate, and growth stocks.
- If central bank actions can persistently influence long-term real rates without causing large swings in real activity, then the perceived 'natural rate' becomes less relevant, and central bank beliefs or policy targets could directly drive long-term yields, impacting term premia.
- In household finance, the 'interest income effect' suggests that households, especially those saving for or in retirement, may increase savings and demand for income-generating assets (e.g., dividend stocks, annuities) in a low-rate environment, affecting their portfolio allocation and optimal saving strategies.
- For insurance companies, this framework could inform the pricing and risk management of long-duration liabilities, such as life insurance and annuities, as the stability and level of long-term real rates become more directly influenced by monetary policy.
Monetary PolicyYield CurveLong-Term Real RatesLife-Cycle ModelAsset ValuationMarginal Propensity to ConsumeHousehold FinanceAsset PricingRetirement SavingsDuration Riskr-star
Core finding, identification, data
Core Finding
- The paper finds that highly persistent monetary policy shocks have an ambiguous, potentially near-zero, net effect on economic activity.
- This occurs because a standard asset valuation effect (higher rates lower demand) is largely offset by an 'interest income effect' where the marginal propensity to consume out of financial wealth decreases with lower rates, as households save more to compensate for reduced income flow.
- This allows central banks to influence long-term real rates persistently, making the 'natural rate of interest' (r*) less of a binding constraint for monetary policy.
Identification Strategy
- The paper's primary identification strategy is theoretical model-based, developing a Finitely-Lived Agent New Keynesian (FLANK) model that incorporates life-cycle considerations, including retirement savings motives and finite lives.
- This model allows for the derivation of an Euler equation that distinguishes between the effects of temporary versus persistent interest rate changes.
- The model's predictions are then empirically validated using a recursive VAR analysis on US data, examining impulse responses to monetary policy shocks of varying persistence (transitory vs. persistent).
Data
The paper uses quarterly U.S. data from FRED (1982Q1-2019Q4) for real consumption per capita (PCE), real wealth holdings per capita (TABSHNO), CPI (CPIAUCSL), and the ex-ante 10-year real rate (REAINTRA-TREARAT10Y). For empirical validation, it uses monetary policy shock series from Romer and Romer (2004) and Gertler and Karadi (2015), and other macroeconomic data (Federal funds rate, CPI, commodity price index, Industrial Production Index) from Ramey (2016) and Wieland and Yang (2020).
Aaditya Mattoo, Michele Ruta, Robert W. Staiger — International Trade & Investment
This paper theoretically examines how geopolitical rivalry impacts international trade cooperation and the functioning of the WTO, particularly its core principles of reciprocity and non-discrimination.
Finance Application
- This paper's insights could inform asset pricing models by introducing geopolitical risk factors that affect cross-border capital flows, foreign exchange rates, and the valuation of multinational corporations, especially those with complex global supply chains.
- For insurance, it highlights the rising demand for political risk insurance and trade credit insurance, as geopolitical tensions increase the probability of trade disruptions, tariffs, and non-MFN policies, leading to new product development or repricing strategies for insurers.
- In household finance, the impact of trade fragmentation on inflation and job security in specific sectors could influence household consumption and savings decisions, affecting demand for various financial products.
GeopoliticsInternational TradeWTOTariffsTrade AgreementsEconomic ModelsAsset PricingRisk ManagementSupply ChainsInternational Relations
Core finding, identification, data
Core Finding
- The paper finds that while geopolitical rivalry increases non-cooperative tariffs, it does not fundamentally alter the set of internationally efficient tariffs, implying a persistent role for trade cooperation unless rivalry completely overshadows own-country well-being.
- However, it argues that geopolitical rivalry creates significant tension for the WTO's traditional principles of reciprocity and most-favored-nation (MFN), necessitating adaptive policy responses to maintain multilateral trade cooperation.
Identification Strategy
- The paper develops a two-country (and later three-country) general equilibrium neoclassical trade model, augmenting government objective functions with a 'relative power' term to capture geopolitical rivalry.
- This allows for a theoretical analysis of how changes in the weight placed on relative power (parameterized by ρ and ρ*) affect Nash tariffs and the international efficiency frontier.
Data
The paper is primarily theoretical, using a neoclassical trade model. It references real-world examples like the US-China trade war and the Phase 1 Agreement for illustrative purposes, and mentions 'data from the New Industrial Policy Observatory' and 'quantitative trade model' in the context of other research.
Ezequiel Garcia-Lembergman, Natalia Ramondo, Andrés Rodríguez-Clare, Joseph S. Shapiro — International Trade & Investment
This paper develops a quantitative general equilibrium model of trade, sectoral linkages, and fossil fuel markets to analyze the impact of various carbon tax instruments on global emissions, welfare, and value chain reorganization.
Finance Application
- This model offers a robust framework for quantitative climate finance research.
- In asset pricing, it could be used to stress-test portfolios against various carbon tax regimes, modeling how different policy mixes (e.g., global vs. unilateral, extraction vs. consumption taxes) impact the valuation of fossil fuel companies, carbon-intensive sectors, and green energy firms, thus informing climate transition risk assessments.
- For household finance, the welfare and consumption effects of carbon taxes could be analyzed to understand their distributional impact on household wealth, savings, and investment decisions in energy-related assets.
- In insurance, the model's ability to quantify climate-related liabilities and opportunities could help insurers price risks for carbon-intensive industries, develop new climate-resilient products, and assess systemic risks from global carbon policy shifts.
Climate FinanceESG InvestingGeneral Equilibrium ModelsCarbon TaxesTrade PolicyEnergy MarketsSupply ChainsAsset ValuationTransition RiskWelfare Analysis
Core finding, identification, data
Core Finding
- A global carbon tax of $100 per ton CO2 would reduce global emissions by nearly half and increase global welfare by 1.4 percent.
- Unilateral optimal consumption taxes (similar to CBAMs) are more effective than production taxes in limiting leakage and yielding welfare gains, and combining consumption and extraction taxes further enhances these gains, particularly for fossil fuel exporters.
Identification Strategy
- The paper develops a quantitative general equilibrium trade model that explicitly incorporates upward-sloping supply curves for homogeneous fossil fuels.
- It estimates fossil fuel supply elasticities using mine-level cost data from Welsby et al. (2021) across 16 world regions, allowing for endogenous price adjustments and general equilibrium responses to various carbon tax instruments (extraction, production, consumption, import, and export taxes) implemented at different stages of the global value chain.
Data
The study uses global multi-region input-output tables from the World Input Output Database (WIOD) and Exiobase for 28 countries (including a rest-of-world aggregate) and 28 sectors in 2014. It also incorporates mine-level cost schedules from Welsby et al. (2021), emission intensity data from the U.S. Energy Information Administration, and country-level Social Cost of Carbon estimates from Ricke et al. (2018a).
Lorena Keller, Ricardo Santiago Pique Cebrecos — International Trade & Investment
This paper examines the long-run effects of trade liberalization on local economic activity using a historical free trade zone in the French Alps.
Finance Application
- The paper's findings on the significant and persistent boost to local economic activity and wealth from trade liberalization have direct implications for local asset pricing, particularly real estate and local business valuations.
- Researchers could investigate how local property values and the performance of small, unlisted businesses responded to the GFZ's establishment and abolition.
- For household finance, the increased real income and purchasing power could be linked to changes in household savings, investment in local businesses, or demand for financial services like mortgages and insurance, providing a historical context for understanding how trade policy affects household balance sheets and financial decisions.
Trade liberalizationLocal economic activityRegression discontinuityHistorical dataReal estateHousehold wealthRegional financeAsset pricingEconomic historyTaxation
Core finding, identification, data
Core Finding
- The study finds that communes within the Savoy Great Free Zone (GFZ), which benefited from tariff exemptions, experienced significantly higher direct tax revenues (a proxy for local wealth and economic activity) compared to neighboring non-GFZ communes.
- This effect was persistent, driven by greater land tax revenues, and associated with increased agro-industrial and manufacturing establishments, suggesting benefits from lower living costs and cheaper inputs.
Identification Strategy
- The paper employs a sharp geographic regression discontinuity (RD) design.
- The GFZ's boundary, established for political rather than economic reasons, arbitrarily partitioned the department of Haute-Savoie.
- This allows for a quasi-experimental comparison of communes just inside and outside the tariff-exempt zone, leveraging the exogenous nature of the border to identify the causal effects of trade liberalization.
Data
The authors construct a novel commune-level dataset by digitizing eight archival series from 1862 to 1914, including municipal budgets, land and personal tax records, and detailed agricultural and industrial censuses. Pre-treatment data from the 1848 Sardinian Census and 1860 Agricultural Survey are used for balance checks, alongside geographic covariates.
Olivia Bordeu — International Trade & Investment
This paper develops a quantitative spatial model to examine how metropolitan fragmentation impacts the provision of commuting infrastructure, the spatial distribution of economic activity, and aggregate welfare, using Santiago, Chile as a case study.
Finance Application
- The paper's insights on infrastructure misallocation due to fragmented local governance could be applied to municipal bond pricing and real estate investment.
- Researchers could investigate whether municipal bonds issued by 'underinvesting' municipalities in fragmented metropolitan areas carry a higher credit risk premium or have lower credit ratings.
- For real estate, the model's predictions on dispersed employment and polycentric urban patterns could inform REIT investment strategies, identifying areas where property values might be undervalued or overvalued due to suboptimal infrastructure provision and its impact on market access and economic activity.
Urban EconomicsInfrastructureLocal GovernanceSpatial ModelMunicipal BondsReal EstateCredit RiskMetropolitan FragmentationCommuting CostsWelfare
Core finding, identification, data
Core Finding
- The model predicts that municipalities underinvest in areas near their boundaries and overinvest in core locations, leading to higher cross-jurisdiction commuting costs, dispersed employment, more polycentric urban patterns, and lower aggregate welfare compared to a centralized planner.
- Centralizing investment decisions in Santiago, Chile, would increase infrastructure expenditure by 44-60%, population by 1.9-2.4%, and welfare by 1.4-1.7%.
Identification Strategy
- The paper estimates key parameters by exploiting the discontinuity in infrastructure at municipal borders in Santiago, Chile, to identify the infrastructure elasticity of travel times.
- It also uses variation in road density at the boundary between municipalities to estimate political weights for residents and workers, assuming building costs are uniform across borders.
Data
The paper uses Santiago's 2012 travel survey (Encuesta Origen Destino de Viajes) for commuting flows, a public database of real estate appraisals from Chile's Servicio de Impuestos Internos (SII) for land use and prices, and Open Street Maps data combined with official government road network data for infrastructure characteristics.
Hâle Utar, Carlos Alfon Cebreros Zurita, Luis B. Torres Ruiz — International Trade & Investment
This paper examines how the 2018/19 US-China trade war led to adjustments in global value chains (GVCs) and increased nearshoring activities to Mexico, using confidential firm-level trade data.
Finance Application
- The findings suggest that the US-China trade war induced a structural shift in global supply chains, leading to increased manufacturing and export activity in Mexico.
- This could inform asset pricing models by identifying 'nearshoring exposure' as a new factor affecting corporate profitability and stock returns for MNEs with operations in Mexico or those heavily reliant on China-US supply chains.
- It also has implications for real estate investment, particularly industrial properties in Mexico, and could influence currency markets through increased Mexican trade and foreign direct investment.
- Furthermore, it highlights supply chain risk management, which could be relevant for insurance products covering geopolitical trade disruptions.
Trade WarGlobal Value ChainsNearshoringMexicoMultinational Enterprises (MNEs)Supply ChainExportsImportsTariffsFirm-level dataNatural ExperimentAsset PricingCorporate FinanceReal EstateFX MarketsRisk Management
Core finding, identification, data
Core Finding
- The US-China trade war significantly increased Mexican firms' exports to the US, imports from Asia and the US, and overall net exports.
- This effect was primarily driven by GVC participant firms, particularly foreign multinational enterprises (MNEs), with export growth concentrated in technology-intensive industries and existing intermediate products.
- Foreign MNEs' exports to the US grew 2.5 times more than domestic GVC firms.
Identification Strategy
- The study leverages the abrupt US trade policy shift as a natural experiment, employing a generalized difference-in-differences (DD) and triple difference-in-differences (DDD) approach.
- Firm-level trade policy exposures are constructed based on pre-shock product portfolios, and the DDD model specifically compares GVC firms to non-GVC firms, controlling for firm fixed effects, baseline firm size, non-manufacturing status, and concurrent trade policy changes.
Data
The paper uses confidential longitudinal firm-level trade data from Mexico (2015–2021), monthly directories of IMMEX export platform firms from the Ministry of Economy, parent-country information from the Dun & Bradstreet Hierarchy and Connections database, and tariff changes on US imports and exports from Fajgelbaum et al. (2020, 2021).
Edward Wiles, Deivy Houeix — International Trade & Investment
This paper uses a field experiment in Senegal to study how search and trust frictions (adverse selection and moral hazard) limit international sourcing for small firms, and how social media can alleviate these barriers.
Finance Application
- This research offers insights for financial institutions operating in emerging markets or serving SMEs.
- The findings on information asymmetry and trust could inform the design of credit scoring models for small businesses, incorporating 'social commerce' data (e.g., supplier reviews, network activity) to assess creditworthiness and mitigate adverse selection.
- For supply chain finance, understanding how social networks build trust could lead to innovative platforms that facilitate trade credit or invoice financing for SMEs, especially in cross-border transactions where formal institutions are weak.
- In household finance, the role of social media in building trust could be applied to the adoption of new fintech products or micro-insurance, where peer information and community-based reputation might reduce perceived risk and foster engagement.
Supply Chain FinanceInformation FrictionsAdverse SelectionMoral HazardRelational ContractingEmerging MarketsSME FinanceFintechMicro-insuranceField Experiment
Core finding, identification, data
Core Finding
- The study finds that alleviating search frictions (connecting firms to foreign suppliers via social media) increases access to foreign goods, but only when trust frictions (adverse selection and moral hazard) are also addressed do these connections develop into lasting, profitable relationships.
- The combined alleviation of adverse selection and moral hazard has the largest impact on relationships and profits, suggesting a strong complementarity between these two types of trust frictions.
Identification Strategy
- The authors conduct a field experiment with 1,862 garment firms in Senegal, randomly assigning them to different treatment arms.
- These arms include matching firms to Turkish suppliers (search), providing information about supplier types (adverse selection), and manipulating perceived supplier incentives (moral hazard) through cross-randomization, allowing for causal identification of each friction's impact.
Data
The paper uses a baseline survey of 1,862 firms, a mystery shopping exercise to measure access to foreign goods, real-time transaction data from Senegal's largest mobile money provider (Wave Mobile Money) to track supplier payments, and a follow-up survey.
Treb Allen, Winston L. Chen, Suresh Naidu — Development of the American Economy
This paper quantifies the economic and welfare impacts of American slavery using a quantitative spatial general equilibrium model, analyzing how coercion, market power, and misallocation shaped the antebellum U.S. economy and the effects of emancipation.
Finance Application
- This paper's framework could be applied to asset pricing by analyzing how the 'coercion premium' or 'misallocation discount' was embedded in historical asset prices, such as land values or firm valuations in slave-holding regions, and how these risks were priced (e.g., political risk of emancipation).
- In household finance, the model's welfare calculations and data on slave property values could inform studies on intergenerational wealth accumulation for both slaveholders and formerly enslaved households, and how the lack of property rights for enslaved persons affected their ability to accumulate human and financial capital.
- For insurance, the explicit modeling of 'slave property values' and risks (e.g., runaways, mortality, emancipation) could be used to reconstruct or analyze the historical market for slave insurance, examining how geographic factors and coercion influenced premium pricing and risk transfer mechanisms.
Economic HistorySlaverySpatial EconomicsGeneral EquilibriumLabor MarketsHuman CapitalWelfare EconomicsAsset PricingHousehold FinanceInsuranceHistorical Data
Core finding, identification, data
Core Finding
- Emancipation generated substantial welfare gains for formerly enslaved persons (exceeding 1,200%) but reduced welfare for free persons by 1.6%.
- The overall impact on U.S.
- GDP was small, as the reduction in coerced labor supply was offset by migration to more productive Northern regions.
- Slavery led to significant misallocation of labor to lower-amenity, higher-productivity locations, driven by slaveholders' profit maximization.
Identification Strategy
- The paper 'inverts' a microfounded spatial general equilibrium model to recover underlying productivities, amenities, and coercion parameters.
- It estimates the intensive labor supply elasticity using U.S.
- Army fort data, comparing payments to free and enslaved workers for the same task.
- Extensive labor supply elasticities are estimated by comparing relative wages and slave prices to labor shares, instrumenting with geographic amenity proxies like malaria suitability and distance to the North.
Data
The study uses 100% linked census counts of free and enslaved persons (1860), novel data on free wages, slave property values, and enslaved worker rental payments. It also incorporates 1860 Census County Level Social Statistics, agricultural and manufacturing censuses, U.S. Army Fort quartermaster records, and geographic data on cotton suitability, malaria suitability, and distance to the North.
Tishara Garg — International Trade & Investment
This paper develops a novel econometric methodology using algebraic geometry to empirically study the impact of industrial policy on equilibrium selection in settings with multiple equilibria and coordination failures, applying it to industrial zones in India.
Finance Application
- The paper's methodology for empirically identifying and quantifying equilibrium selection in the presence of multiple equilibria, particularly those driven by coordination failures, offers significant potential for finance research.
- In asset pricing, this framework could be applied to model and test for multiple equilibria in phenomena like market bubbles, crashes, or liquidity traps, quantifying how policy announcements (e.g., central bank interventions, regulatory changes) shift financial markets between different equilibria.
- For household finance, the method could analyze local housing markets, where neighborhood quality and prices might exhibit multiple equilibria, and assess how urban development policies influence household wealth and mortgage decisions by triggering equilibrium shifts.
- The algebraic geometry and homotopy continuation techniques could also be adapted to identify multiple equilibria in models of financial contagion or systemic risk.
Industrial PolicyCoordination FailuresMultiple EquilibriaEquilibrium SelectionAlgebraic GeometryHomotopy ContinuationEvent StudySpilloversEconomic DevelopmentIndiaEmpirical MethodsAsset PricingHousehold FinanceSystemic Risk
Core finding, identification, data
Core Finding
- Industrial zones in India lead to a 60% increase in non-farm employment over 15 years, with significant spillovers to non-targeted sectors and municipalities.
- The methodology reveals that industrial zones increase the probability of escaping a low-industrialization equilibrium by 38%, with coordination effects explaining roughly one-third of the observed change in outcomes.
Identification Strategy
- The paper employs a three-step procedure: 1) model estimation and inversion using GMM and instrumental variables (industrial zones in non-targeted sectors/locations) to identify the structural system and unobserved fundamentals; 2) equilibrium enumeration using homotopy continuation methods from algebraic geometry to find all possible equilibria; and 3) type assignment to classify equilibria and decompose policy effects into fundamental-driven and coordination-driven changes.
- Causal inference for policy impact relies on staggered Difference-in-Differences event studies with propensity score matching.
Data
A newly constructed dataset of over 4,000 industrial zones in India (with establishment dates for 1,500), combined with Economic Census data (1990-2013) for 600,000 municipalities, Population Census data (1991-2011), DMSP-OLS annual nighttime luminosity data (1994-2013), new company registrations data (from 2001), and novel spatial data on historical road networks and natural attributes.
Jeffery A. Jenkins, Thomas R. Gray — Development of the American Economy
This paper examines county-level voting patterns on constitutional referendums that disenfranchised voters in the early 20th-century US South, revealing widespread electoral manipulation and the role of racial threat dynamics.
Finance Application
- This paper's findings on political disenfranchisement and voter manipulation could be applied to finance research by examining how such events impact local asset prices, credit markets, and wealth accumulation.
- For instance, researchers could investigate whether successful disenfranchisement referendums led to measurable changes in local property values, the cost of capital for businesses, or the availability and terms of credit for different demographic groups in affected counties.
- The methodology for detecting electoral manipulation could also be adapted to identify anomalies or undue influence in corporate governance votes or shareholder activism, especially in contexts with concentrated ownership or potential 'disenfranchisement' of minority shareholders.
political riskhistorical financewealth inequalitylocal asset pricescredit marketscorporate governancevoter manipulationracial discriminationeconomic history
Core finding, identification, data
Core Finding
- Disenfranchisement referendums were most successful in counties with larger African American populations, consistent with a 'racial threat' hypothesis.
- However, analysis of 'naïve White support' and turnout reveals significant electoral manipulation, with results in many majority-Black counties being mathematically impossible without widespread voter suppression or ballot fabrication.
- Poor Whites, particularly in diverse areas, supported these referendums, indicating successful division of potential biracial Populist coalitions by Democratic elites.
Identification Strategy
- The paper analyzes county-level vote totals for 14 constitutional referendums (1892-1916) against racial demographics (Black population share from preceding decennial censuses) and 1892 Populist presidential vote share.
- It employs a 'naïve White support' estimation, assuming minimal Black support (0.5%) and equal turnout, to identify implausible results (e.g., requiring >100% White support), thereby inferring electoral manipulation like suppression or ballot stuffing.
Data
The study uses county-level vote totals for 14 constitutional referendums across Southern and Border states between 1892 and 1916, Black population share data from the immediately preceding decennial censuses, and county-level vote shares for the Populist presidential candidate in 1892.
Allison E. Green, Kaan Cankat — Development of the American Economy
This paper investigates how municipal boundary expansions, particularly in Sun Belt cities, affected local public finance and public good provision during post-war suburbanization.
Finance Application
- This research offers insights for municipal bond pricing, as cities achieving economies of scale through annexation may present lower fiscal risk and more stable revenue streams, potentially leading to lower bond yields and better credit ratings.
- For household finance, the reduction in per capita expenditures and potential tax burden could influence household disposable income, property values, and migration patterns.
- In insurance, the finding that core services like fire protection are maintained or improved despite per capita spending declines could inform property insurance underwriting and pricing models, particularly in rapidly expanding urban areas.
Municipal FinanceAnnexationLocal GovernmentPublic GoodsEconomies of ScaleSun BeltUrban EconomicsFiscal PolicyMunicipal BondsReal EstateHousehold WealthInsurance Risk
Core finding, identification, data
Core Finding
- The average large annexation increased municipal population by over 30%, leading to a 21-22% decrease in per capita revenue and expenditures, which persisted for nine years.
- This suggests cities leveraged economies of scale in high fixed-cost sectors, maintaining core services like fire and policing without increased building fires, while increasing total debt issued by 24% to finance capital investments.
Identification Strategy
- The study employs a staggered difference-in-differences design around large annexation events.
- For each treatment city-year, a synthetic control city is constructed by matching on pre-period population and baseline 1950 demographics (share Black, share native, and share native-born) to identify causal effects.
Data
The paper digitizes annual annexation data from Municipal Yearbooks (1948-1968) and the Boundary and Annexation Survey (1970 onwards). It supplements this with annual municipal spending and revenue data from the Annual Survey of Governments (from 1952) and municipal employment data from the Census Bureau (1952-2012), alongside Decennial Censuses and Current Population Reports for population figures.
Carlos Fernando Avenancio-León, Troup Howard, William Mullins — Development of the American Economy
This paper examines how the rollout of the Food Stamp Program in the U.S. during the 1960s and 1970s induced long-lasting racial and political polarization.
Finance Application
- The paper's findings on long-run policy politicization and racial polarization could inform asset pricing models by introducing a 'policy risk premium' for sectors sensitive to social welfare debates, affecting long-term investment horizons.
- In household finance, the persistent racial political alignment could explain differential access to credit, wealth accumulation, and investment behaviors across racial groups over decades.
- For ESG investing, the framework for how political narratives shape public perception of policies can be applied to assess the long-term viability and financial impact of corporate social responsibility or environmental initiatives.
- Furthermore, the model's concept of political actors deliberately politicizing voter perceptions aligns with narrative economics in finance, suggesting how narratives can be strategically deployed to influence investor sentiment and market cycles.
Political EconomyRacial PolarizationSocial Safety NetFood Stamps ProgramLong-Run EffectsDifference-in-DifferencesVoter BehaviorPolicy PoliticizationNarrative EconomicsPolitical RiskWealth InequalityESG
Core finding, identification, data
Core Finding
- The Food Stamps Program's rollout generated significant racial political polarization that persisted for fifty years.
- White individuals exposed as adults became more likely to register and vote Republican, while Black voters became more likely to register and vote Democrat.
- This effect is not driven by age or historical experiences, and the paper links these findings to a model where parties deliberately politicize voter perceptions of public policies for strategic advantage.
Identification Strategy
- The study employs a county-level staggered rollout of the Food Stamp Program combined with an experience-exposure Difference-in-Differences (DiD) design.
- It compares lifetime voting patterns of individuals who experienced the policy expansion as adults versus those who came of age in a world where the program was established, using various fixed effects (county, birth year, race) and bias-robust estimators (Callaway and Sant'Anna).
Data
The paper uses voter microdata for 175 million Americans from L2 (as of 2020), county-level Food Stamp program rollout data, congressional speeches from the Congressional Record, Gallup Organization surveys, historical county-level voting data from ICPSR and Dave Leip's Atlas, NAACP Voter Education Project registration data, Black Elected officials data, DW-NOMINATE project data, and Bureau of Economic Analysis (BEA) data for recessions.
Myera Rashid — Development of the American Economy
This paper examines how the adoption of the typewriter in US workplaces from 1880-1920 impacted women's labor force participation, marriage, fertility, and intergenerational mobility.
Finance Application
- The paper's insights into how a technological shock altered women's labor force participation, marriage, and fertility have profound implications for household financial decisions.
- Increased female earnings stability could reduce household precautionary savings, alter demand for life insurance (as women become primary earners), and shift investment horizons.
- Changes in household formation (delayed marriage, lower fertility) could influence demand for housing and related mortgage products, impacting real estate asset pricing.
- The observed upward mobility through marriage could also lead to different patterns of intergenerational wealth transfers and investment strategies across demographic groups.
Technological ChangeLabor EconomicsWomen's Economic OutcomesHousehold FinanceIntergenerational MobilityMarriageFertilityHistorical DataShift-Share IVSocial InteractionsReal EstateInsurance
Core finding, identification, data
Core Finding
- The introduction of the typewriter significantly increased women's labor force participation, leading to lower marriage and fertility rates for White women.
- It also created an indirect crowding-in effect, drawing Black women into domestic services.
- For White women, the typewriter acted as a "meeting technology," enabling them to marry men of higher socioeconomic status and achieve upward mobility.
Identification Strategy
- The main empirical strategy uses a shift-share instrumental variable approach, interacting local industry employment shares before typewriter adoption with national shifts in typist demand across sectors (1880-1920).
- A complementary difference-in-differences approach leverages exogenous variation in the supply of female typists from early typing schools (Women's Christian Association chapters) that existed prior to the typewriter's adoption.
Data
The paper uses historical US complete count censuses (1870-1940), a novel hand-collected database of local Women's Christian Association (WCA) organizations (1875-1880), and genealogical linked data following White women from the 1920 to 1940 Census.
Paula A. Calvo, Murat Iyigun, Jeanne Lafortune — Development of the American Economy
This paper demonstrates that unilateral divorce laws, by considering separation as an alternative, significantly and heterogeneously impact women's economic welfare, particularly benefiting low-educated women.
Finance Application
- This research offers significant insights for household finance, particularly regarding how legal frameworks influence financial planning, risk management, and wealth accumulation.
- The heterogeneous welfare impacts of divorce laws could be studied in relation to household portfolio choices, debt levels, and housing market decisions.
- For instance, changes in divorce legislation could affect the demand for specific insurance products (e.g., 'divorce insurance' or enhanced life/disability coverage) or influence mortgage default rates and housing prices in affected regions.
- Insurers could also price policies differently based on the prevailing divorce regime and its impact on financial stability.
household financedivorce lawswomen's welfareeconomic outcomesseparationlabor economicslegal impactrisk managementwealth accumulationhousing marketinsurance
Core finding, identification, data
Core Finding
- Separation is a prevalent marital status in the US, and separated women generally experience worse economic outcomes than divorced women.
- The adoption of unilateral divorce (UD) laws reduced separation and increased divorce, especially among low-educated women.
- While some women lose welfare with the transition from mutual consent divorce (MCD) to UD, low-educated women whose husbands would have otherwise separated under MCD rules experience significant welfare gains (8-32% of their annual income) under UD.
- Short desertion laws under MCD can generate similar welfare gains for women.
Identification Strategy
- The paper leverages the staggered adoption of unilateral divorce (UD) laws across US states over time as a natural experiment to identify the causal impact of these laws on separation and divorce rates.
- It uses a difference-in-differences approach by comparing outcomes in states before and after UD implementation, controlling for state and year fixed effects, and state-specific time trends.
Data
The paper uses microdata from the US Census (1880-2010), Panel Study of Income Dynamics (PSID), National Longitudinal Survey of Mature Women (NLS-MW), and administrative data from US divorce certificates (1968-1995) compiled by the NBER.
Richard K. Crump, Stefano Eusepi, Emanuel Moench, Bruce Preston — Monetary Economics
This paper uses a unique panel dataset of professional forecasts to show that long-run macroeconomic expectations are heterogeneous, time-varying, and formed using multivariate unobserved component models.
Finance Application
- The insights directly challenge the common assumption of anchored or stationary long-run expectations in asset pricing models.
- Time-varying, heterogeneous long-run expectations about inflation, growth, and interest rates could drive time-varying risk premia in equity and bond markets, explaining anomalies or the pricing of long-duration assets.
- In household finance, these dynamics could explain shifts in household savings, investment, and mortgage choices, as individuals' long-run income or inflation expectations are revised based on short-run news.
- For insurance, the time-varying and multivariate nature of long-run expectations about interest rates and inflation introduces significant uncertainty for pricing and reserving long-term liabilities like annuities, impacting hedging strategies and product design.
expectationslong-run forecastsmacroeconomic variablespanel dataunobserved components modelasset pricinghousehold financerisk premiainflationinterest rateseconomic growthforecasting
Core finding, identification, data
Core Finding
- The paper establishes three new stylized facts: (1) forecasters' rankings are weakly correlated for short vs. long horizons but strongly correlated for longer horizons; (2) long-term and short-term forecast revisions co-vary, with long-run forecasts exhibiting substantial time-variability; and (3) long-term forecast revisions depend on short-run expectations of multiple macroeconomic variables.
- These findings are consistent with a multivariate unobserved trend and cycle model where forecasters have imperfect information.
Identification Strategy
- The paper's methodological innovation is to propose and empirically test a multivariate unobserved component forecasting model.
- This model posits that forecasters perceive macroeconomic data as driven by permanent 'trend' and transitory 'cycle' components, have imperfect information about the current state, and use multivariate models.
- The authors test predictions derived from this model (e.g., correlations of forecast revisions across horizons and variables) using panel data regressions.
Data
The paper utilizes a novel and unique panel dataset of individual-level professional forecasts from the Blue Chip Economic Indicators (BCEI) survey's 'Long-Range Consensus U.S. Economic Projections,' covering 16 US macroeconomic variables from 1998 to 2016, with forecast horizons ranging from one quarter to six to eleven years ahead.
Andrés Blanco, Virgiliu Midrigan, Corina Boar, Callum J. Jones — Monetary Economics
This paper develops a tractable sticky price model with endogenous price changes, revealing an "inflation accelerator" that significantly steepens the Phillips curve during high inflation periods.
Finance Application
- The time-varying slope of the Phillips curve and the inflation accelerator have significant implications for asset pricing.
- For instance, the changing trade-off between inflation and output could alter inflation risk premiums in nominal bonds, making TIPS more or less attractive depending on the inflation regime.
- Equity markets might react differently to monetary policy surprises when the Phillips curve is steep (high inflation) versus flat (low inflation), as the real effects of monetary shocks are smaller during high inflation.
- Furthermore, household finance decisions, such as optimal portfolio allocation between nominal and real assets or debt management, would be influenced by these non-linear inflation dynamics, requiring a more sophisticated understanding of inflation's impact on real wealth.
MacroeconomicsMonetary PolicyInflationPhillips CurveSticky PricesAsset PricingMacro-FinanceHousehold FinanceEconomic ModelsNon-linearityRisk Premiums
Core finding, identification, data
Core Finding
- The model features a powerful inflation accelerator, a feedback loop where an increase in the fraction of price changes increases inflation (more so at higher inflation levels), and higher inflation, in turn, increases firms' incentives to adjust prices.
- This mechanism significantly increases the slope of the Phillips curve during periods of high inflation, implying that reducing inflation is much less costly when inflation is high compared to when it is low, as reflected in a time-varying sacrifice ratio.
Identification Strategy
- The authors develop a structural sticky price model with multi-product firms that choose the *fraction* of prices to adjust, allowing for exact aggregation and tractability.
- They calibrate the model to U.S. data (mean/standard deviation of inflation, mean frequency of price changes, and the comovement of price changes with inflation) and then use log-linear perturbations or non-linear solutions to analyze impulse responses to monetary shocks.
- For time-series analysis, they back out the sequence of monetary policy shocks that reproduce observed U.S. inflation.
Data
The paper uses U.S. CPI (excluding shelter) from 1962:Q1 to 2023:Q4, monthly median fraction of price changes (excluding sales) derived from BLS price quotes (1978-2023), and NBER recession indicators. For robustness, it also incorporates real GDP growth and the federal funds rate.
Gergely Buda, Vasco M. Carvalho, Giancarlo Corsetti, Joao B. Duarte, Stephen Hansen, Afonso S. Moura, Alvaro Ortiz, Tomasa Rodrigo, Jose Rodriguez Mora, Guilherme Alves da Silva — Monetary Economics
This paper demonstrates that monetary policy shocks transmit to real economic activity (consumption, output, investment) much faster than conventionally believed, often within weeks, using novel high-frequency daily data from Spain.
Finance Application
- The rapid and heterogeneous transmission of monetary policy to real activity, especially in specific consumption categories and supply-chain positions, offers rich avenues for finance research.
- In asset pricing, this could inform models of cross-sectional stock returns, where firms in 'fast-responding' sectors (e.g., durables, luxury goods, downstream) might exhibit higher sensitivity to monetary policy surprises, leading to distinct risk premia or trading opportunities around central bank announcements.
- For household finance, the granular daily consumption data could be leveraged to study how households adjust spending on different goods in real-time in response to interest rate changes, impacting their liquidity management, credit demand, and portfolio rebalancing.
- Furthermore, the identified time aggregation bias suggests that many existing finance studies using quarterly data might misestimate the speed and channels of monetary policy transmission to financial markets and firm valuations, warranting re-evaluation with higher-frequency data.
Monetary PolicyHigh-Frequency DataConsumptionInvestmentEmploymentReal ActivityTransmission MechanismSupply ChainsSectoral HeterogeneityTime Aggregation BiasEvent StudyLocal ProjectionsAsset PricingHousehold FinanceCorporate FinanceMarket EfficiencyCredit Risk
Core finding, identification, data
Core Finding
- Monetary policy shocks have economically and statistically significant effects on aggregate real economic activity (gross output, consumption, investment) within weeks, not quarters or years.
- This fast adjustment is led by downstream sectors, particularly those producing luxuries and durables, while aggregate employment and consumer prices respond more slowly and with longer lags.
- The paper also finds that time aggregation to quarterly frequency obscures these short-lag responses, shifting significant effects to longer lags.
Identification Strategy
- Monetary policy shocks are identified using daily monetary policy surprises for the Euro Area from Jarociński and Karadi (2020).
- These shocks are constructed from high-frequency changes in financial assets around ECB policy announcements, employing sign restrictions in a Bayesian VAR to control for the 'information channel' and isolate pure monetary policy shocks.
- Local projections are then used to estimate impulse response functions.
Data
The paper utilizes novel high-frequency daily data from Spain, including consumption and investment aggregates built from BBVA bank transaction records, and daily gross output (corporate sales) and employment from the Spanish Tax Authority and Ministry for Inclusion, Social Security, and Migration. It also incorporates high-frequency financial market data (Euribor, IBEX35) and monthly consumer prices, housing prices, and various expectation and confidence indicators.
Miguel Acosta, Lydia Cox — Monetary Economics
This paper investigates whether Buy American restrictions on government procurement lead to stronger macroeconomic effects of government spending.
Finance Application
- This paper's insights could be applied to asset pricing by examining whether firms with a higher proportion of 'constrained' government contracts exhibit lower stock returns or higher credit spreads, particularly during periods of heightened BAA enforcement, due to the identified higher labor costs and lower private sector output.
- In corporate finance, researchers could study how BAA stringency influences firms' capital expenditure decisions, supply chain reshoring investments, or financing choices, linking the 'negative labor supply shock' mechanism to firm-specific labor market risk.
- The detailed contract and import data could also be used to construct novel firm-level exposure measures to government policy risk.
Macro-financeGovernment ProcurementBuy American ActFiscal MultiplierSupply ChainFirm BehaviorLabor MarketsRegional EconomicsAsset PricingCredit RiskCorporate InvestmentPolicy Risk
Core finding, identification, data
Core Finding
- Government spending constrained by Buy American restrictions has a lower cross-sectional fiscal multiplier than unconstrained spending.
- This is because constrained spending shocks act like negative labor supply shocks to the private sector, reallocating labor, raising wages, dampening private GDP, and indirectly inducing 'leakage' through increased imports in the consumption-good sector.
Identification Strategy
- The authors leverage the complexity of the Buy American Act (BAA) legislation, including loopholes, amendments (Micro-Purchase Threshold, Berry Amendment), and the WTO Government Procurement Agreement, to identify government spending that is more- and less-constrained by domestic content restrictions.
- They validate this by showing firms exhibit significantly smaller import responses to 'constrained' contracts.
- For fiscal multipliers, they use an instrumental variables approach (building on Nakamura and Steinsson 2014) exploiting variation in state military procurement associated with aggregate changes in defense spending, distinguishing between constrained and unconstrained spending.
Data
The paper uses U.S. federal defense procurement contracts (1979-2000) and the universe of procurement contracts (2001-2019) from USAspending.gov. It also incorporates firm-level import shipments from S&P Panjiva (bills of lading data, 2007 onwards), state-level GDP from the BEA, employment data from the BLS, and inflation data from Hazell et al. (2022).
Boris Hofmann, Cristina Manea, Benoit Mojon — Monetary Economics
This paper refines conventional Taylor rules to allow for a differentiated monetary policy response to demand- versus supply-driven inflation, finding that the Federal Reserve reacts significantly more forcefully to demand-driven inflation and showing the welfare implications of such a 'targeted Taylor rule' in a New Keynesian model.
Finance Application
- This research offers significant insights for asset pricing and household finance.
- Asset pricing models could incorporate the asymmetric monetary policy response to demand versus supply inflation shocks to better explain asset price dynamics (e.g., equity returns, bond yields, real estate) and inflation risk premia, as different sources of inflation may imply distinct policy paths and thus different discount rates or cash flow impacts.
- In household finance, understanding how households form inflation expectations based on the *source* of inflation, and how these expectations interact with the Fed's targeted policy, could inform models of consumption, saving, and borrowing behavior, especially for inflation-indexed products or long-term financial planning.
- The use of LLMs to extract policy signals from central bank communications also presents a novel methodological tool for predicting market reactions to policy statements.
Monetary PolicyInflationDemand ShocksSupply ShocksTaylor RuleFederal ReserveNew Keynesian ModelLLMInflation TargetingAsset PricingHousehold Finance
Core finding, identification, data
Core Finding
- The Federal Reserve has historically reacted almost fourfold stronger to demand-driven inflation than to supply-driven inflation, a 'targeted' approach consistent with its dual mandate.
- When modeled in a New Keynesian framework, this asymmetric response leads to less volatile output and different inflation dynamics compared to conventional Taylor rules, and improves welfare by better approximating optimal policy, even in the presence of measurement error.
Identification Strategy
- The paper estimates Taylor-type rules by replacing aggregate inflation with its demand- and supply-driven components, derived from recent inflation decomposition methods (Shapiro, 2024; Eickmeier and Hofmann, 2022) which are agnostic to monetary policy.
- Robustness is checked by correlating these decomposed inflation series with LLM-derived assessments of demand/supply pressures from FOMC transcripts, confirming the Fed's historical awareness of these drivers.
Data
The empirical analysis uses quarterly data from 1979Q3 to 2007Q4 (and extended periods for robustness) for the federal funds rate, year-on-year core PCE inflation (and headline inflation for robustness), and the CBO's potential GDP estimate for the output gap. It leverages inflation decomposition series from Shapiro (2024) and Eickmeier and Hofmann (2022), and analyzes FOMC transcripts (1939-2019) using Large Language Models.
Tom Nicholas, Christophe Spaenjers — Development of the American Economy
This paper investigates racial housing inequality in pre-WWII Manhattan by linking investor returns and maintenance incentives to long-term property decay and neighborhood segregation.
Finance Application
- The core insight that high cash flow (rental yields) in segmented markets can disincentivize long-term asset maintenance and capital appreciation has broad implications for asset pricing.
- For instance, it could explain why certain commercial real estate (CRE) assets or even specific equity sectors (e.g., those investing in distressed urban areas) might exhibit high dividend yields but suffer from persistent underinvestment and lower long-term capital gains.
- In household finance, this framework could be extended to study how discriminatory practices in mortgage or insurance markets affect homeowners' incentives to maintain properties, impacting their wealth accumulation and intergenerational wealth transfers.
- For insurance, the use of visual machine learning to detect property decay could be directly applied to develop dynamic risk assessment models, allowing insurers to price policies more accurately or offer incentives for maintenance based on visual cues from satellite or street-view imagery.
Housing MarketsReal EstateRacial InequalityRental YieldsProperty DecayUrban EconomicsMachine LearningComputer VisionHistorical DataAsset MaintenanceSegregationLandlord BehaviorAsset PricingHousehold FinanceInsurance Risk
Core finding, identification, data
Core Finding
- Properties housing Black residents in pre-WWII Manhattan transacted at lower values but generated disproportionately high rental yields for landlords.
- This higher profitability reduced landlords' economic incentives to maintain these properties, leading to delayed alterations and greater visual decay in Black neighborhoods by 1980, thereby perpetuating disinvestment and inequality.
Identification Strategy
- The paper constructs a novel dataset linking real estate transactions (1912-1939), federal Census records (1930, 1940) for household demographics and rents, property images (c.1940, 1980) for visual decay, and administrative data on building alterations.
- It employs hedonic models, random forests, and visual machine learning (ResNet CNN, cosine similarity) to estimate property values and rental yields, and to track property decay over time.
- Racial disparities are identified by comparing properties occupied by Black vs.
- White households, controlling for property and household characteristics, and redlining districts.
Data
The study uses real estate transactions from the Real Estate Record and Builders' Guide (1912-1939), federal Census records (1930, 1940) for household demographics and rents, property images from WPA/NYC Tax Department (c.1940) and NYC Department of Finance (1980s), PLUTO dataset for property information, NYC Buildings Information System for alteration data, and HOLC redlining maps.
Scott Fulford, Fabio Schiantarelli — Development of the American Economy
This paper constructs the first estimates of U.S. county nominal and real GDP by industrial sectors from 1870 to 2018 to analyze spatial economic convergence and divergence.
Finance Application
- The granular, long-term county GDP data and its industrial decomposition could be used to study regional equity premiums or real estate asset pricing, by linking local economic structure to asset returns and risk.
- For household finance, the findings on within-state inequality and the changing 'path to riches' could inform research on regional wealth accumulation, consumption smoothing, and mortgage default risk, especially in areas experiencing divergence.
- In insurance, the detailed industrial composition and inequality trends could be critical for regional risk assessment in property & casualty insurance (e.g., economic resilience of a region to shocks) or life/health insurance (e.g., how regional economic shifts affect long-term health outcomes and mortality rates).
Regional EconomicsEconomic HistoryInequalityGDPIndustrial CompositionSpatial DataConvergenceDivergenceReal EstateLocal Asset MarketsHousehold WealthRegional Risk
Core finding, identification, data
Core Finding
- U.S. counties converged in GDP per worker from 1870 to 1970, primarily due to falling inequality between states, but have since diverged, largely driven by increasing inequality within states.
- This divergence is linked to shifts from manufacturing to tradable services, with tradable services concentrating in richer counties and government services reducing inequality in poorer ones, alongside changes in population mobility and education convergence patterns.
Identification Strategy
- The paper's core innovation is the construction of a novel, long-term, granular dataset of county-level GDP by industrial sector.
- They use a method to allocate state-level GDP to counties based on relative wage earnings or employment, accounting for industrial composition and relative price changes.
- They also adapt the Theil index decomposition to quantify each sector's contribution to inequality, considering its share and correlation with county GDP per worker.
Data
The paper constructs the first estimates of U.S. county nominal and real GDP by 16 industrial sectors from 1870 to 2018. Data sources include individual census records, county employment records, BEA county GDP estimates (post-2001), and various historical sources for wages, output, and inputs (e.g., NBER macro history database, Kendrick (1961), Gallman (1960), Barger (1955)).
Florin O. Bilbiie, Sigurd Galaasen, Refet S. Gürkaynak, Mathis Mehlum, Krisztina Molnar — Impulse and Propagation Mechanisms
This paper investigates whether household heterogeneity amplifies aggregate demand effects in Norway using novel micro-level consumption, income, and wealth data, finding that heterogeneity is largely irrelevant for aggregate fluctuations due to strong automatic stabilizers.
Finance Application
- The finding that strong automatic stabilizers (taxes and transfers) significantly dampen income heterogeneity and its aggregate demand amplification has direct implications for household finance and asset pricing.
- In economies with robust welfare states like Norway, households might exhibit lower demand for precautionary savings in liquid assets, potentially leading to higher allocations to riskier, higher-return assets like equities or illiquid real estate.
- This reduced aggregate consumption risk, stemming from effective fiscal insurance, could also imply a lower equity premium or different pricing of income-contingent securities in such markets, as the 'macro-risk' component from household heterogeneity is mitigated.
Heterogeneous Agent Models (HANK)Marginal Propensity to Consume (MPC)Fiscal PolicyAutomatic StabilizersIncome InequalityConsumption SmoothingHousehold FinanceAsset PricingRisk SharingMicrodataWelfare StateLiquidity ConstraintsPortfolio Choice
Core finding, identification, data
Core Finding
- The main empirical finding is that, unlike labor earnings, using disposable income (which accounts for capital income, taxes, and transfers) reveals that heterogeneity does not amplify aggregate demand shocks in Norway; instead, it shows a dampening effect or near-irrelevance.
- This is primarily attributed to the strong insurance effect of the tax and transfer system, which smooths income for high-MPC individuals, a result that largely persists even under a counterfactual US tax system.
Identification Strategy
- The paper estimates individual Marginal Propensities to Consume (MPCs) using an instrumental variable approach, where unemployment serves as an exogenous income shock.
- It then calculates income betas (elasticities of individual income to aggregate income) and consumption betas (elasticities of individual consumption to aggregate consumption) for different household groups, comparing these to assess amplification effects.
- A robustness check employs the Blundell et al. (2008) method, extended by Commault (2022), to estimate MPCs out of transitory income shocks.
Data
The study utilizes a unique Norwegian dataset from 1993-2018, combining comprehensive administrative records (demographics, income, and wealth from tax records) with granular, transaction-level consumption data derived from electronic payments (debit card and online bank transfers).
Paul Beaudry, Fabrice Collard, Patrick Fève, Alain Guay, Franck Portier — Impulse and Propagation Mechanisms
This paper introduces a novel method, Dynamic Structural VAR (D-SVAR), to identify structural shocks and impulse responses in VAR models by leveraging dynamic restrictions on exogenous driving forces, allowing for formal testing of traditional identification strategies.
Finance Application
- This D-SVAR methodology could be directly applied in asset pricing to re-evaluate the identification of macro-financial shocks (e.g., monetary policy, credit supply, technology, uncertainty) that drive asset returns, volatility, or corporate investment.
- Researchers could use D-SVAR to test the validity of instruments used in proxy VARs for financial shocks or to assess the robustness of impulse responses derived from traditional Cholesky or sign restrictions on financial variables.
- For instance, one could identify 'pure' financial shocks (e.g., liquidity shocks, risk aversion shocks) without ad-hoc restrictions and then analyze their impact on equity markets, bond yields, or credit spreads, potentially uncovering new asset pricing factors or re-interpreting existing ones.
Structural VARDynamic IdentificationMacroeconomic ShocksImpulse Response FunctionsEconometricsProxy VARMonetary PolicyAsset PricingFinancial MarketsIdentification Testing
Core finding, identification, data
Core Finding
- The core theoretical finding is that if macroeconomic data is generated by a dual dynamic structure where exogenous components follow restricted autoregressive processes (e.g., linearly independent AR(1) processes), then the full set of structural shocks and impulse responses can be identified without imposing conventional identifying restrictions.
- This D-SVAR framework can then be used to formally test the validity of commonly used SVAR identification strategies, such as impact restrictions, long-run restrictions, and proxy-VAR approaches.
Identification Strategy
- The paper's identification strategy, D-SVAR, relies on imposing restrictions on the autoregressive matrix (R) governing the dynamics of unobserved exogenous driving forces, rather than on the contemporaneous loading matrix (F) or external instruments.
- For instance, assuming a diagonal R matrix with distinct diagonal elements (implying mutually orthogonal and dynamically distinct exogenous shocks) is shown to be sufficient for local identification.
- This allows for the recovery of structural shocks and impulse responses, and critically, enables formal testing of overidentifying restrictions.
Data
The paper uses both simulated data from an estimated three-equation New Keynesian model and real-world macroeconomic data. For empirical applications, it revisits Blanchard and Quah (1989) using output growth and unemployment gap, Gertler and Karadi (2015) using CPI, industrial production, one-year government bond rate, and excess bond premium, and Christiano et al. (1999) using real GDP, unemployment rate, CPI inflation, commodity price inflation, and federal funds rate.
Vasco M. Carvalho, Matias Covarrubias, Galo Nuño — Impulse and Propagation Mechanisms
This paper explores how optimal capital allocation in dynamic, multi-sector production networks can mitigate the aggregate impact of cascading shocks, revealing an efficient strategy of over-investing in upstream sectors to reduce disaster risk at the cost of lower average consumption.
Finance Application
- The finding that optimal capital allocation involves over-investing in upstream sectors to mitigate disaster risk suggests that these 'insurance' sectors might exhibit lower systematic risk and thus lower equity risk premia, especially during tail events.
- This framework could inform models of cross-sectional asset returns based on firms' positions within production networks and their exposure to network-induced disaster risk.
- The large welfare cost of business cycles, driven by lower average consumption due to risk mitigation, also has implications for household savings and portfolio choices, potentially leading to higher precautionary savings or a stronger preference for safe assets.
Production NetworksDisaster RiskGeneral EquilibriumDeep LearningCapital AllocationWelfare CostsSupply Chain RiskAsset PricingRisk ManagementMacro-Finance
Core finding, identification, data
Core Finding
- Analytically, the optimal capital allocation under uncertainty involves deliberately over-investing in upstream sectors to mitigate severe economic downturns, which reduces average consumption but significantly mitigates disaster risk.
- Quantitatively, the simulated nonlinear economy features higher mean capital levels in key upstream sectors, lower mean levels of macroeconomic aggregates, realistic aggregate volatility, and a welfare cost of business cycles one order of magnitude larger than in standard linear models, primarily due to lower average consumption rather than consumption disasters.
Identification Strategy
- The paper's methodological innovation is deploying novel deep-learning techniques, specifically extending the 'deep equilibrium nets' method, to solve a high-dimensional dynamic general equilibrium model with 74 state variables.
- This approach allows them to capture full nonlinear effects and Monte-Carlo based expectations for shock propagation, enabling the simulation of ergodic distributions and generalized impulse responses in a way traditional methods cannot.
Data
The model is calibrated to data moments for 37 sectors in the US, using data spanning 1948 to 2018, drawing primarily from BEA Tables, BEA Fixed Assets data, Input-Output data, and GDP-by-Industry data.
Zhen Huo, Jieran Wu, Minghao Li, Huan Xiong — Impulse and Propagation Mechanisms
This paper investigates how firm-level attention allocation, modeled as an information network, amplifies micro-level shocks into aggregate economic fluctuations, using EDGAR browsing data.
Finance Application
- The 'diluted in-degree index' could serve as a novel measure of institutional investor attention or firm salience, predicting cross-sectional stock returns, trading volume, or liquidity beyond traditional metrics.
- The information network structure among firms could also be mapped to investor or analyst networks, explaining contagion or information spillover effects in asset markets.
- Furthermore, the granular impact of highly-attended firms on macro aggregates suggests these firms might disproportionately affect diversified portfolio risk and returns, offering insights into portfolio construction and systemic risk.
Information NetworksGranularityFirm AttentionEDGARMacroeconomic FluctuationsPower LawsMicro ShocksAsset PricingInvestor AttentionInformation Asymmetry
Core finding, identification, data
Core Finding
- Empirically, firm attention (measured by EDGAR browsing activity) follows fat-tailed power-law distributions, weakly correlated with firm size, and browsing-weighted sales growth strongly predicts macroeconomic forecasts.
- Theoretically, granular effects emerge if a firm's 'diluted in-degree index' (effective attention received) grows sufficiently with the number of firms, and attention heterogeneity alone can drive significant aggregate fluctuations, with its interplay with firm size depending critically on their correlation.
Identification Strategy
- The paper empirically identifies firm attention using detailed firm-to-firm browsing data on EDGAR filings, constructing a 'browsing-weighted measure of sales growth' and testing its predictive power for macroeconomic forecasts.
- Theoretically, it extends the noisy business cycle framework to incorporate asymmetric attention via a directed information network, deriving conditions for granular effects based on network properties and power-law distributions of firm characteristics.
Data
The study uses detailed firm-to-firm browsing information on EDGAR filings from 2009-2016, de-anonymized using IP-info.io. It also incorporates firm-level data from Compustat (sales, TFP), macroeconomic forecasts from the Survey of Professional Forecasters (SPF), and firm linkage data from FactSet Revere Supply Chain.
Saleem Bahaj, Ricardo Reis — Monetary Economics
This paper examines how China manages its dual-currency system (mainland CNY and offshore CNH) through capital controls and distinct monetary/liquidity policies to maintain a peg, influence exchange rates, and internationalize the RMB.
Finance Application
- The paper's framework of parallel currencies, capital controls, and a 'UIP wedge' could be applied to asset pricing research to quantify the capital control premium/discount in cross-listed assets or the basis between onshore/offshore derivatives.
- The detailed analysis of liquidity policies could inform studies on liquidity risk premia in short-term bond markets, especially in emerging economies.
- Furthermore, the predictable policy responses to exchange rate deviations could be used to model and test the profitability and risk of FX carry trades involving dual-currency pairs.
International FinanceExchange RatesMonetary PolicyCapital ControlsDual CurrenciesLiquidity ManagementCentral BankingArbitrageAsset PricingEmerging MarketsFX Markets
Core finding, identification, data
Core Finding
- The paper finds that exogenous increases in offshore yuan supply depreciate the exchange rate, with the elasticity of money demand for reserves being significantly higher than for deposits.
- While monetary policy (money supply adjustments) accounts for only one-sixth of peg reversals, liquidity policies (reserve requirements, discount window, interbank market management) achieve the remaining five-sixths of the adjustment, allowing China to manage its exchange rate and current account while maintaining capital controls.
Identification Strategy
- For the causal effect of money supply on exchange rates, the authors exploit exogenous, transitory increases in offshore yuan supply caused by changes in the PBoC's bill issuance schedule (specifically, roll-offs not immediately replaced).
- For shocks to money demand, they use an instrumental variable: the deviation of the CNY-USD exchange rate from its central parity rate, which proxies for binding exchange rate bands and unfulfilled pressure on the CNY-USD rate, acting as an 'escape valve' for CNH.
Data
The study uses daily data on CNY-CNH and CNH-USD exchange rates, PBoC bills outstanding, CNH one-week interbank rates, HKMA Primary Liquidity Provider (PLP) drawings, and HKMA discount window facility usage. It also incorporates estimates of CNH deposits and reserves, monthly CNH deposits, and analyzes specific events like the 2015-16 financial crisis and the August 2023 devaluation.
Javier Bianchi, Alisdair McKay, Neil Mehrotra — Impulse and Propagation Mechanisms
This paper develops a New Keynesian model with search frictions and demand rationing in the housing sector to analyze optimal monetary policy responses to housing inflation.
Finance Application
- This research suggests that financial markets might misprice housing-related assets if they assume central banks will actively target housing inflation.
- For asset pricing, investors in REITs, mortgage-backed securities, or homebuilder stocks could refine their valuation models by factoring in that monetary policy will likely allow housing market imbalances to persist.
- In household finance, the demand rationing and search cost framework could explain heterogeneous household responses to interest rate changes in the housing market, impacting mortgage demand and homeownership rates.
- For real estate investment, understanding that housing cycles are less directly managed by monetary policy could inform long-term development and acquisition strategies, focusing more on fundamental supply-demand dynamics and search costs.
Monetary PolicyHousing InflationNew Keynesian ModelSearch FrictionsDemand RationingReal EstateAsset PricingHousehold FinanceMacroeconomicsCentral Banking
Core finding, identification, data
Core Finding
- The optimal monetary policy should largely disregard housing inflation and instead focus on stabilizing inflation in the non-housing sector.
- This is because the housing market, characterized by demand rationing and search frictions, behaves differently from the demand-determined sectors assumed in standard models, and the costs of a recession needed to counter housing overheating outweigh the benefits.
Identification Strategy
- The methodological innovation is a two-sector New Keynesian model that incorporates search frictions and demand rationing in the housing market, contrasting with the standard assumption of demand-determined output.
- This framework allows for a 'short-side rule' where equilibrium quantity is determined by the lesser of notional demand or supply, especially as search costs diminish.
- The model is calibrated to match US data on housing tenure, vacancy rates, and the size of the real estate sector.
Data
The paper utilizes US PCE inflation data from the Bureau of Economic Analysis, the 2019 American Community Survey for housing tenure and mobility, Zillow Observed Rent Index, BLS New Tenant Rent Index, and NIPA Table 5.4.5U for brokers' commissions and ownership transfer costs. Monthly CPI-shelter price index and Zillow data (2005-2024) and quarterly New Tenant Rent Index data (2005-2024) are used for calibration and simulation.
José Luis Montiel Olea, Mikkel Plagborg-Møller, Eric Qian, Christian K. Wolf — Impulse and Propagation Mechanisms
This paper formally proves the robustness of Local Projection (LP) confidence intervals to model misspecification, contrasting it with the fragility and undercoverage of Vector Autoregression (VAR) confidence intervals, especially with short lag lengths.
Finance Application
- Many asset pricing, household finance, and insurance studies rely on VAR models to estimate impulse response functions (IRFs) for understanding how macroeconomic or financial shocks propagate and affect financial variables (e.g., equity returns, bond yields, consumption, housing prices, insurance claims).
- This paper suggests that if these studies use VARs with short-to-moderate lag lengths, their IRF confidence intervals might be severely undercovering, leading to misleading conclusions about the significance and magnitude of shock effects.
- Researchers should consider using Local Projections or significantly increasing VAR lag lengths, even if it means wider confidence intervals, to ensure robust inference in the presence of model misspecification.
Local ProjectionsVector AutoregressionImpulse Response FunctionsModel MisspecificationEconometricsTime Series AnalysisConfidence IntervalsMacro-FinanceAsset PricingHousehold FinanceRisk Management
Core finding, identification, data
Core Finding
- The conventional LP confidence interval maintains correct coverage even under significant local misspecification (detectable with high probability), due to a "double robustness" property analogous to partially linear regression estimators.
- In stark contrast, conventional VAR confidence intervals with short-to-moderate lag lengths can severely undercover, even for small and hard-to-detect misspecification, unless the lag length is so large that the interval becomes as wide as the LP interval, leading to a "no free lunch" result.
Identification Strategy
- The paper's methodological innovation is a formal analytical comparison of LP and VAR inference procedures under local misspecification within a VARMA(1,∞) model.
- It derives the worst-case bias and coverage properties for VARs and provides a formal proof of the "double robustness" of LP estimators, demonstrating their asymptotic invariance to misspecification under certain conditions.
Data
The paper uses simulations based on the Smets and Wouters (2007) dynamic stochastic general equilibrium (DSGE) model to illustrate its theoretical findings in a macro-economic context, specifically analyzing impulse responses to cost-push or monetary shocks.
Peter Andre, Joel P. Flynn, George Nikolakoudis, Karthik Sastry — Impulse and Propagation Mechanisms
This paper introduces a model of 'quick-fixing' behavior in consumption-savings decisions, empirically validates it with a novel survey, and quantifies its macroeconomic implications, showing that small optimization costs lead to large behavioral deviations and significant aggregate effects.
Finance Application
- This quick-fixing behavior could explain suboptimal financial planning, such as insufficient emergency savings or inconsistent debt repayment strategies, especially for small, frequent financial events, potentially leading to persistent low wealth accumulation.
- The size-dependent MPCs and aggregate consumption responses imply that standard consumption-based asset pricing models, which often assume rational optimization, might misprice assets whose returns are sensitive to small vs. large aggregate income shocks.
- For instance, assets that perform well during periods of small, frequent positive shocks (leading to high MPCs) might be overvalued if the market assumes a lower, more rational MPC.
- Quick-fixing could also influence insurance demand and claims behavior, as households might under-insure against small, frequent risks or use insurance payouts inefficiently for small losses, impacting insurer profitability and product design.
Household FinanceConsumption-SavingsNear-RationalityMarginal Propensity to Consume (MPC)Survey DataBehavioral EconomicsIncome ShocksFinancial PlanningAsset Pricing ImplicationsInsurance Behavior
Core finding, identification, data
Core Finding
- Almost 70% of households follow one of four simple quick-fixes that fully consume or fully save out of small income shocks, but abruptly adjust their behavior for large shocks.
- This quick-fixing behavior is near-rational, with average opportunity costs of only $17 per quarter, yet it significantly alters aggregate consumption responses to income shocks of varying sizes and explains a large share of MPC heterogeneity unexplained by traditional demographics.
Identification Strategy
- The study uses a novel, large-scale survey administered to 5,000 US households, eliciting detailed within-respondent consumption policy functions in response to 14 hypothetical income shocks (seven gains and seven losses) ranging from $50 to $10,000.
- This allows for the direct measurement of how households' marginal propensities to consume (MPCs) change with shock size and sign, revealing abrupt transitions from extreme (0 or 1) to interior MPCs.
Data
The primary data source is a novel survey of 4,981 US households conducted in October and November 2023, balanced to approximate the US adult population. Additional data from the American Community Survey (ACS) 2022 and Survey of Consumer Finances (SCF) 2022 are used for demographic and wealth comparisons.
Saki Bigio, Nicolas Caramp, Dejanir Silva — Impulse and Propagation Mechanisms
This paper develops a New Keynesian model where the anticipation of future inflationary finance links public debt to inflation expectations, shaping optimal monetary policy responses to fiscal shocks.
Finance Application
- The paper's mechanism of debt-driven sticky inflation and central bank "underreaction" could be used to model the inflation risk premium embedded in the term structure of interest rates, particularly for inflation-linked securities like TIPS or inflation swaps.
- The heterogeneous beliefs between firms and households regarding future monetary accommodation could explain divergences in market-implied versus survey-based inflation expectations, impacting asset allocation strategies.
- In household finance, the "stepping on a rake" result suggests that households anticipating future inflation due to fiscal debt might adjust their portfolios towards inflation-hedging assets (e.g., real estate, commodities) or alter consumption-saving decisions, affecting long-term wealth accumulation and retirement planning.
Monetary PolicyFiscal PolicyInflation ExpectationsPublic DebtNew Keynesian ModelRegime SwitchingCentral Bank CredibilityAsset PricingTerm StructureInflation Risk PremiumHousehold FinanceSavingsInvestmentInsuranceMacro-Finance
Core finding, identification, data
Core Finding
- The paper finds that in fiscally sensitive environments, optimal monetary policy deviates from the Taylor principle, requiring persistently low real interest rates and accommodating fiscal shocks to front-load inflation and mitigate debt accumulation.
- This "sticky inflation" phenomenon, driven by expectations of debt monetization, means that aggressive rate hikes can backfire, leading to higher inflation and debt in the medium run, and that price-level targeting is not optimal.
Identification Strategy
- The theoretical identification relies on a tractable New Keynesian model with regime switches (fiscal-expansion phase vs. inflationary-finance phase) and heterogeneous beliefs (households vs. firms) to demonstrate how debt influences inflation expectations and monetary policy.
- For empirical counterfactuals, the paper uses a Kalman filter to decompose post-COVID-19 inflation into fiscal, cost-push, monetary (deviations from a Taylor rule), and bond-valuation shocks, inferring these shocks from observed paths of primary surpluses, debt-to-GDP, inflation, and nominal rates.
Data
The paper uses U.S. historical time series data from the post-COVID-19 period, including the market value of debt to GDP ratio (from Hall, Payne and Sargent, and Fred TOTRESNS), primary deficits to GDP ratio (NIPA Tables), inflation (GDP deflator from NIPA), nominal policy rates (Federal Funds Effective Rate from Fred DFF), and various measures of inflation expectations (Michigan, SOFIE, SPF, Cleveland Fed, 5-year breakeven inflation, and market-implied inflation-disaster probability from Hilscher et al. (2022)).
Javier Bianchi, Louphou Coulibaly — International Finance & Macroeconomics
This paper analyzes the optimal monetary policy response to tariffs within an open-economy New Keynesian model, finding that an expansionary stance is optimal.
Finance Application
- This paper's findings offer a novel framework for understanding how central banks might react to trade policy shocks, which is highly relevant for global macro and asset pricing.
- The prediction of an expansionary monetary policy response to tariffs, leading to higher inflation and a depreciating currency, could inform trading strategies in fixed income (e.g., inflation-linked bonds), foreign exchange markets, and equity sectors sensitive to import costs or exchange rates.
- Furthermore, the fiscal externality mechanism could be integrated into models of international portfolio choice and hedging, allowing for more nuanced predictions of capital flows and asset returns under trade policy uncertainty.
Monetary PolicyTariffsInternational TradeInflationExchange RatesNew Keynesian ModelFiscal ExternalityAsset PricingGlobal MacroFX
Core finding, identification, data
Core Finding
- Tariffs create a fiscal externality where households undervalue imported goods because they fail to internalize that higher imports generate additional tariff revenue, which, in equilibrium, raises household income.
- To counteract this, the optimal monetary policy is expansionary, stimulating employment and aggregate income, leading to inflation rising above and beyond the direct effects of tariffs and a positive output gap.
- This result holds across various scenarios, including different tariff types, durations, and terms of trade assumptions.
Identification Strategy
- The paper uses a dynamic, open-economy New Keynesian model with nominal rigidities to characterize Ramsey-optimal monetary policy under government commitment.
- It analytically derives conditions for optimal policy and quantitatively simulates macroeconomic responses to various tariff scenarios (e.g., permanent, temporary, anticipated, with intermediate inputs, and endogenous terms of trade) using a global non-linear algorithm.
Data
The paper uses a calibrated open-economy New Keynesian model, with parameters calibrated to the US economy (e.g., discount factor, elasticities of substitution, Frisch elasticity of labor supply, price-adjustment cost). It does not use empirical data for hypothesis testing or estimation.
Rodrigo Adão, Ana M. Fernandes, Chang-Tai Hsieh, Jose M. Quintero — International Trade and Macroeconomics
This paper analyzes how trade shocks, particularly tariff changes, affect allocative efficiency and welfare in economies with distorted markets, focusing on the role of firm-level importer concentration and its impact on markups.
Finance Application
- This research offers a rich micro-foundation for understanding how trade policy and supply chain dynamics impact firm profitability and market power, which are crucial for asset pricing and corporate finance.
- For asset pricing, importer concentration could be a novel factor explaining cross-sectional stock returns, as firms with higher concentration may exhibit different sensitivities to trade shocks.
- In corporate finance, the findings can inform how firms manage supply chain risk and trade policy uncertainty, influencing their investment decisions, hedging strategies, and access to capital based on their market power in import markets.
Trade ShocksFirm-level DataMarket PowerMarkupsAllocative EfficiencyInternational TradeTariffsSupply ChainsFirm Concentration
Core finding, identification, data
Core Finding
- The paper finds that importer concentration significantly influences the aggregate and distributional effects of tariff changes on allocative efficiency, with magnitudes comparable to neoclassical welfare changes.
- Specifically, trade liberalization episodes often worsen allocative efficiency because tariff reductions are stronger for firms with lower markups (i.e., less concentrated import markets).
- This effect is amplified in poorer and smaller countries due to their higher and more dispersed importer concentration.
Identification Strategy
- The identification strategy leverages firm-level import data across 57 countries to estimate how a firm's import elasticity to tariff changes varies with its import share.
- This elasticity is then mapped to domestic markups using a structural model of oligopolistic importer firms.
- The variation comes from trade liberalization episodes that generate heterogeneous tariff cost changes across origin-good pairs, assuming these changes are exogenous to firm-specific shocks.
Data
The paper uses a unique dataset of firm-level import records for 57 countries spanning 1997-2021, compiled from administrative customs agencies. This data is merged with ad-valorem import tariffs for various origins and 6-digit HS products from Teti (2020).
Christopher Clayton, Antonio Coppola, Matteo Maggiori, Jesse Schreger — International Finance & Macroeconomics
This paper introduces a methodology using large language models (LLMs) to systematically identify the application of and response to geoeconomic pressure from large textual corpora.
Finance Application
- This paper's LLM-based methodology to quantify geoeconomic pressure and firm responses offers rich avenues for finance research.
- In asset pricing, the identified pressure events and firm-level reactions (e.g., R&D, supply chain shifts, profit margin impacts) can be used to construct novel geoeconomic risk factors, analyze their impact on equity and bond returns, and predict firm valuations.
- For the insurance industry, this granular data on specific types of pressure (sanctions, export controls) and their effects on firms and supply chains could significantly improve the modeling and pricing of political risk insurance and supply chain disruption coverage.
GeoeconomicsGeopoliticsLarge Language Models (LLMs)Textual AnalysisFirm ResponsesTariffsSanctionsExport ControlsSupply ChainsR&DInvestmentAsset PricingRisk FactorsFirm ValuationPolitical Risk Insurance
Core finding, identification, data
Core Finding
- The paper reveals that firms respond heterogeneously to geoeconomic pressure, with tariff-affected firms primarily adjusting prices and export control-affected firms increasing R&D and domestic investment.
- This pressure, largely from the US and China against each other, often targets "chokepoint" products, and firm responses vary based on their home country's role.
Identification Strategy
- The paper innovates by using large language models (LLMs) with a two-stage prompt design to systematically extract granular data on geoeconomic pressure and firm responses from earnings calls and analyst reports.
- It further quantifies the measurement uncertainty of this LLM-based classification through prompt and model perturbations.
Data
The paper utilizes large textual corpora, including global earnings call transcripts from Capital IQ (via WRDS) and Orbit's China A-Shares Transcripts, covering 2008-2025. It also incorporates firm and sector-level analyst research reports from J.P. Morgan and Fitch Ratings (BMI) starting from 2011.
Christopher Clayton, Matteo Maggiori, Jesse Schreger — Impulse and Propagation Mechanisms
This paper develops a theory where economic integration, particularly in strategic sectors like financial services, creates dependencies that hegemonic powers exploit for coercion, leading to inefficient global fragmentation as target countries pursue anti-coercion policies.
Finance Application
- This paper offers rich avenues for finance research.
- In asset pricing, the quantified, nonlinear geoeconomic power and 'chokepoint' sectors could be used to model geopolitical risk premia in sovereign bonds, currency markets, and the equities of global financial institutions.
- For instance, how do the valuations of major payment system providers (e.g., Visa, Mastercard, SWIFT-connected banks) reflect their 'chokepoint' status and the risk of fragmentation? In insurance, the framework can inform the pricing of political risk insurance or supply chain disruption insurance, especially for firms operating in sectors or countries identified as vulnerable to coercion.
- The 'fragmentation doom loop' could also be applied to household finance, examining how households in vulnerable countries diversify their savings across different financial systems or asset classes to mitigate geopolitical risk.
GeopoliticsEconomic CoercionFinancial ServicesFragmentationInternational FinanceAsset PricingRisk PremiaSupply ChainsSanctionsDollar DiplomacyIndustrial PolicyInternational Trade
Core finding, identification, data
Core Finding
- The paper finds that gains from integration, driven by economies of scale and specialization, can paradoxically increase a hegemon's power by making alternatives poor substitutes.
- This leads to a 'fragmentation doom loop' where uncoordinated anti-coercion policies by target countries, while individually rational, result in inefficient global fragmentation.
- The paper quantifies this geoeconomic power, showing its nonlinearity and highlighting financial services as a key source of U.S. power, while manufacturing is key for China.
Identification Strategy
- The identification strategy involves a structural model with a Stackelberg game and nested-CES production functions, which yields a 'sufficient statistic' for measuring geoeconomic power.
- This statistic is empirically implemented using input-output tables and bilateral trade data, with elasticities of substitution calibrated from existing literature.
- The model allows for the quantification of power based on expenditure shares and substitution elasticities, and the Russia-Ukraine conflict serves as an illustrative case study for ex-ante anti-coercion policies.
Data
The paper uses goods trade data from BACI, service trade data from the OECD-WTO Balanced Trade in Services (BaTIS), and domestic gross output data from the OECD Inter Country Input Output (ICIO) tables. Elasticities of substitution are drawn from existing literature (e.g., Costinot and Rodríguez-Clare, Rouzet et al.).
Alexander Copestake, Divya Kirti, Maria Soledad Martinez Peria, Yao Zeng — International Finance & Macroeconomics
This paper examines how interoperability in digital payment systems, specifically India's UPI, drives the adoption and usage of digital payments and its downstream effects on lending.
Finance Application
- This research provides a robust framework for valuing fintech firms and payment platforms in asset pricing, by quantifying the 'network premium' derived from interoperability and expanded user bases.
- In household finance, the findings on increased digital payments leading to greater credit access for entrepreneurs and hawkers offer a new channel for studying financial inclusion, credit scoring for underserved populations, and the impact of payment infrastructure on household financial health.
- Insurers could leverage the enhanced transaction data from interoperable systems to develop and price micro-insurance products for small businesses and individuals.
fintechpayment systemsinteroperabilitynetwork effectsfinancial inclusiondigital paymentscredit marketsplatform competitionIndiahousehold financeasset pricing
Core finding, identification, data
Core Finding
- Interoperability significantly increases digital payment usage, particularly in regions with higher initial fragmentation across payment platforms.
- Model-based estimates indicate a more than 50% increase in total digital payment usage in the year after integration, and this also led to increased lending, especially for entrepreneurs and hawkers.
Identification Strategy
- The study uses a natural experiment involving the integration of a major pre-existing digital wallet with India's UPI.
- It employs a heterogeneous adoption design, exploiting regional variation in ex-ante fragmentation across payment platforms.
- Robustness is confirmed by matching districts on observables and instrumenting ex-ante fragmentation using proximity to early-launched 'hub' cities, chosen for marketing reasons unrelated to the integration decision.
Data
The paper utilizes novel data on the universe of UPI payments (value, volume, users, app choices, bank branches), monthly district-level data from a major Indian fintech firm's closed-loop wallet, NPCI data on ATM cash withdrawals, and household borrowing data from the Consumer Pyramids Household Survey (CPHS).
Pablo Cuba Borda, Albert Queralto, Ricardo M. Reyes-Heroles, Mikael Scaramucci — International Trade and Macroeconomics
This paper studies how trade cost shocks, particularly those affecting intermediate goods, drive persistent inflation and impact macroeconomic dynamics, using a combination of empirical analysis and a multi-country general equilibrium model.
Finance Application
- The finding that intermediate input trade costs lead to persistent inflation has significant implications for asset pricing.
- Investors could use this insight to refine inflation risk premia in bond markets, differentiating between transitory and persistent inflation drivers.
- Furthermore, firms with global supply chains heavily reliant on intermediate inputs might face higher and more persistent cost pressures, impacting their long-term earnings and equity valuations, suggesting a 'supply chain risk premium' in equity markets.
- This framework could also inform the design of inflation-hedging strategies for institutional investors and households, particularly in periods of global supply chain disruptions, and help assess the impact of different monetary policy responses to trade shocks on currency and interest rate markets.
MacroeconomicsInternational TradeInflation DynamicsSupply ShocksMonetary PolicyGlobal Value ChainsTariffsInput-Output AnalysisGravity ModelsDSGE ModelsAsset PricingInflation RiskSupply Chain RiskQuantitative Macro
Core finding, identification, data
Core Finding
- Higher trade costs for final goods cause large but short-lived inflation spikes, while increased costs for intermediate inputs trigger more modest but significantly more persistent inflation, lasting up to five years.
- Quantitatively, a 10 percentage point rise in relative import costs of final goods increases inflation by 0.7 percentage points within one year, while the same increase for intermediate inputs leads to a 0.6 percentage point rise in the first year, remaining elevated by 0.2 percentage points for up to five years.
- These shocks also worsen monetary policy trade-offs and contributed to the U.S. inflation surge post-pandemic.
Identification Strategy
- The paper estimates bilateral trade costs using a ratio-type estimator from a static Armington gravity model, exploiting cross-country variation in bilateral trade flows not explained by country-specific characteristics.
- It then identifies the causal effect of these trade cost shocks on inflation using panel local projection methods (Jordà, 2005), controlling for time- and country-fixed effects and other macroeconomic covariates.
- For the post-pandemic analysis, a two-country DSGE model is estimated using Bayesian methods, identifying trade cost shocks through their correlation with novel quarterly U.S. domestic sourcing shares.
Data
The study uses detailed global input-output data from the OECD Inter-Country Input-Output Tables (ICIO) from 1995-2020 (aggregated to 41 countries), supplemented by the World Input-Output Database (WIOD). For the post-pandemic analysis, it constructs novel quarterly data on U.S. domestic sourcing shares for final and intermediate goods, along with standard macroeconomic variables like quarterly real GDP growth, CPI inflation, and nominal interest rates for the U.S. and a Rest-of-World aggregate.
Thomas Bourany — International Trade and Macroeconomics
This paper designs an optimal climate club within an Integrated Assessment Model, considering strategic country participation, heterogeneous climate impacts, carbon taxation costs, and the use of tariffs to incentivize global welfare maximization.
Finance Application
- This research provides a robust framework for quantifying the financial implications of climate policy and geopolitical risk.
- Asset managers could integrate the model's predictions on carbon tax levels and trade tariffs into their valuation models for energy-intensive companies and sovereign bonds, particularly for countries identified as 'outsiders' to climate clubs (e.g., fossil fuel exporters).
- Insurance companies could leverage the welfare decomposition and climate damage functions to refine their pricing of climate-related risks, such as catastrophe bonds, by better understanding the heterogeneous impacts of climate change and the effectiveness of policy instruments like carbon tariffs and transfers in mitigating these risks.
Climate ChangeCarbon TaxTrade PolicyInternational TradeClimate AgreementsIntegrated Assessment ModelsGame TheoryStrategic InteractionHeterogeneityEconomic PolicyEnergy MarketsTariffsCarbon Border Adjustment MechanismWelfare EconomicsAsset PricingRisk ManagementESG
Core finding, identification, data
Core Finding
- The optimal climate club, designed to maximize global welfare while accounting for strategic country participation, would include all countries except major fossil fuel producers (Russia, Saudi Arabia, Nigeria, Iran).
- This club would impose a $110/ton CO2 carbon tax on members and a 50% tariff on goods from non-members, a compromise from the $130/ton globally optimal tax (without free-riding) to incentivize broader participation.
Identification Strategy
- The paper employs a two-step optimization approach within a quantitative Integrated Assessment Model.
- First, for given policy instruments (carbon tax and tariffs), it identifies the optimal coalition of countries that maximizes global welfare, subject to unilateral and sub-coalition Nash participation constraints, using a combinatorial discrete choice problem solved via exhaustive search or a 'squeezing procedure'.
- Second, it uses a grid search over the policy instruments to find the optimal carbon tax and tariff levels for this stable coalition.
Data
The study uses data for 2018-2023, averaged to smooth COVID-19 effects, covering 25 countries and 7 regions. Data sources include GDP per capita from the World Bank (Maddison Project), energy variables (production, consumption, reserves) from the Energy Institute, energy rent from World Development Indicators, country-level temperature data from Burke et al. (2015), and trade flows/gravity variables from CEPII (Conte et al. 2022).
Jorge Miranda-Pinto, Eugenio I. Rojas, Felipe Saffie, Alvaro Silva — International Finance & Macroeconomics
This paper investigates how the structure of production networks influences the severity and propagation of financial crises, particularly Sudden Stops, in emerging and advanced economies.
Finance Application
- This research offers a novel lens for asset pricing by suggesting that a firm's position and its sector's interconnectedness within the production network could be a significant determinant of its risk and expected returns, especially during financial crises.
- For instance, firms in highly interconnected sectors might exhibit lower systemic risk exposure or different sensitivities to aggregate shocks, leading to a 'network risk premium' that could be priced in equity or credit markets.
- In household finance, understanding how network structure amplifies or mitigates crises could inform models of household consumption and savings behavior, particularly in economies vulnerable to Sudden Stops, and help design more resilient financial planning strategies or insurance products against network-induced economic downturns.
Production NetworksFinancial CrisesSudden StopsMacroprudential PolicySystemic RiskInput-Output LinkagesGeneral Equilibrium ModelsEmerging MarketsAdvanced EconomiesAsset PricingRisk PremiaHousehold FinanceInsurance
Core finding, identification, data
Core Finding
- The paper finds that stronger intersectoral linkages in advanced economies act as automatic stabilizers during financial crises by allowing input price adjustments to cushion profit declines.
- Conversely, weaker linkages in emerging markets amplify profit contractions, increasing crisis severity.
- Empirically, countries with more interconnected production structures experience significantly smaller declines in GDP and current account reversals during Sudden Stops.
Identification Strategy
- The paper develops a dynamic, multi-sector small open economy DSGE model with intersectoral linkages and occasionally binding collateral constraints to study this mechanism.
- It conducts a counterfactual experiment where an advanced economy is endowed with an emerging market's network structure.
- Empirically, it constructs a 'distance to the diagonal' metric from industry-level input-output data to quantify network complexity and uses panel regressions with country and year fixed effects to correlate this metric with the severity of Sudden Stops.
Data
The paper uses industry-level input-output data from OECD (2021), macroeconomic outcomes during Sudden Stops from Bianchi and Mendoza (2020), KLEMS sectoral data for volatility and persistence, IMF government gross debt data, Lane and Milesi-Ferretti (2017) for current account to GDP, and World Bank Development Indicators for GDP data.
Tobias Adrian, Domenico Giannone, Matteo Luciani, Mike West — Forecasting & Empirical Methods
This paper introduces a Bayesian framework to systematically integrate judgmental macroeconomic scenarios with statistical model-based risk forecasts, quantifying their concordance and synthesizing them to best match a a statistical reference forecast.
Finance Application
- This methodology could be directly applied in asset pricing to synthesize diverse market expectations (e.g., from surveys, sell-side reports, central bank statements) with quantitative risk models (e.g., VaR, ES models) to forecast asset returns and volatility across different asset classes like equities, fixed income, and commodities.
- For household finance, it could model how individuals integrate their personal economic outlook with official forecasts to make decisions on savings, mortgage choices, or retirement planning.
- In insurance, the framework could be used to combine expert-driven catastrophe scenarios (e.g., climate change impacts, pandemic severity) with actuarial models to better assess tail risks for underwriting, capital allocation, and reinsurance strategies.
macroeconomic forecastingscenario analysisrisk managementBayesian methodspredictive synthesisentropic tiltingasset pricinghousehold financeinsurancefinancial stabilityjudgmental forecasting
Core finding, identification, data
Core Finding
- The paper develops a methodology that reconciles qualitative judgmental scenarios with quantitative statistical forecasts by assigning weights to scenarios based on their concordance with a statistical reference distribution, measured by the expected misclassification rate (EMR).
- This framework provides a systematic way to evaluate and integrate risks from different scenarios, identify scenario set incompleteness, and quantify the relative support for various scenarios.
Identification Strategy
- The methodological innovation lies in combining "entropic tilting" and "Bayesian predictive synthesis" to formally integrate judgmental scenario information with statistical forecasts.
- Entropic tilting is used to transform partial scenario information (e.g., point forecasts or specific percentiles) into full probability distributions, while Bayesian predictive synthesis then combines these scenario-specific distributions into a mixture.
- The optimal weights for this mixture are determined by maximizing the expected misclassification rate (EMR) relative to a statistical reference forecast, effectively identifying the mixture that is "closest" to the reference distribution.
Data
The paper uses published macroeconomic forecasts from the Federal Reserve Board's Tealbook (2007 and 2018) and the NY Fed's Outlook-at-Risk forecasts. These data include point forecasts and various percentiles (e.g., P10, P25, P50, P75, P90) for one-year ahead GDP growth, inflation, and unemployment rates.
Tomás E. Caravello, Alisdair McKay, Christian K. Wolf — Forecasting & Empirical Methods
This paper proposes a hybrid method for evaluating monetary policy counterfactuals, combining empirically estimated short-end policy shock effects with structural models for long-end extrapolation, and finds that behavioral frictions, not market incompleteness, are key for long-horizon effects.
Finance Application
- This hybrid methodology could be applied to forecast the impact of various monetary policy counterfactuals on asset prices.
- For instance, researchers could model how persistent changes in the expected path of short-term interest rates (long-end effects) affect the term structure of government bonds, corporate bond spreads, or equity risk premia.
- The finding that behavioral frictions significantly alter long-horizon extrapolations suggests exploring how investor rationality or limited attention influences the transmission of monetary policy to financial markets, potentially explaining anomalies like the forward guidance puzzle in asset pricing.
Monetary PolicyCounterfactual AnalysisStructural ModelsBehavioral EconomicsYield CurveImpulse Response FunctionsMacro-FinanceAsset PricingTerm StructureForward Guidance
Core finding, identification, data
Core Finding
- The paper finds that evaluating monetary policy counterfactuals for short-term policy changes primarily relies on empirically estimated causal effects of short-end monetary shocks.
- However, for counterfactuals involving persistent policy changes, structural models are necessary to extrapolate long-end effects.
- Crucially, behavioral frictions in these models lead to significantly different long-horizon predictions compared to rational expectations, whereas market incompleteness (HANK vs.
- RANK) has little impact on this extrapolation.
Identification Strategy
- The paper employs a hybrid identification strategy.
- First, it uses empirically estimated causal effects of monetary policy shocks on the short end of the yield curve, derived from VARs and local projections using established monetary shock series (e.g., Romer and Romer, Aruoba and Drechsel).
- Second, for long-end effects, it extrapolates using structural models (RANK, HANK, and their behavioral variants) estimated via impulse response-matching, disciplining these models to match the empirically observed short-end effects.
Data
The paper uses various macroeconomic time series from FRED, including output gap, inflation, federal funds rate, unemployment, investment, consumption, hours, TFP, labor productivity, and labor share. Monetary policy shocks are sourced from Romer and Romer (2004) and Aruoba and Drechsel (2024).
Manish Jha, Jialin Qian, Michael Weber, Baozhong Yang — Forecasting & Empirical Methods
This paper develops a novel "AI Economy Score" by applying Generative AI to corporate conference call transcripts to extract granular managerial expectations about the economy.
Finance Application
- This methodology offers a scalable way to generate novel, forward-looking sentiment indicators that could significantly impact asset pricing.
- For instance, an 'AI Firm Score' could predict individual stock returns, bond yields, or corporate credit spreads by capturing nuanced managerial outlooks on future performance and macroeconomic conditions.
- In household finance, these granular sentiment measures could inform models of consumer confidence or investment behavior, especially if linked to specific sectors.
- For insurance, the AI Economy Score could be integrated into risk models to forecast industry-specific or macroeconomic risks, potentially influencing underwriting decisions or pricing for business interruption insurance.
Generative AILarge Language ModelsEconomic ForecastingManagerial ExpectationsSentiment AnalysisCorporate TranscriptsGDP PredictionAsset PricingCredit RiskBehavioral FinanceUnstructured Data
Core finding, identification, data
Core Finding
- The AI Economy Score, derived from managerial expectations extracted via Generative AI, is a strong and robust predictor of future US economic activities, including real GDP, industrial production, employment, and wages.
- It outperforms traditional survey-based forecasts and maintains predictive power for several quarters, providing additional insights beyond existing benchmark measures.
Identification Strategy
- The methodological innovation involves using Generative AI (e.g., ChatGPT) prompted to act as a finance expert to analyze unstructured text from corporate earnings call transcripts.
- The AI assesses managerial optimism about the US economy based on the text, assigning a score and explanation.
- These firm-level scores are then aggregated to industry and national levels.
- Robustness checks, including masking person, firm, product, and date information, are performed to mitigate potential look-ahead bias.
Data
The paper utilizes corporate conference call transcripts from SeekingAlpha (2006-2020) and Financial Modeling Prep (2021-2023). It also incorporates real economic outcome data from FRED, BEA, Compustat, and CRSP, and survey forecast data from the Survey of Professional Forecasters.
Anton A. Nakov, Mishel Ghassibe — Workshop on Methods and Applications for Dynamic Equilibrium Models
This paper develops a non-linear dynamic general equilibrium framework with disaggregated production networks and state-dependent pricing to analyze how networks dampen or amplify "pricing cascades" depending on the nature of economic shocks.
Finance Application
- The paper's findings on how production networks amplify or dampen pricing cascades under different shock types have significant implications for asset pricing and risk management.
- For asset pricing, the differential sensitivity of sectors to demand versus supply shocks, mediated by their network position (supplier/customer centrality), could explain cross-sectional variations in equity returns and inform supply chain risk premiums.
- For risk management and insurance, understanding these network-driven cascades is crucial for pricing business interruption insurance or supply chain disruption policies, allowing insurers to quantify systemic risk exposure of firms and sectors to various macroeconomic shocks.
production networksstate-dependent pricingmenu costsinflationbusiness cyclessupply shocksdemand shocksmonetary policyEuro Areasectoral heterogeneitypricing cascadesasset pricingrisk managementsupply chain finance
Core finding, identification, data
Core Finding
- The interaction of production networks and state-dependent pricing creates "pricing cascades" where large aggregate movements trigger additional price adjustments.
- Networks dampen these cascades under demand shocks (leading to stronger monetary non-neutrality and muted inflation) but amplify them under aggregate or sector-specific supply shocks (resulting in rapid repricing and large inflationary swings).
- This mechanism quantitatively matches Euro Area inflation and repricing frequency in the post-Covid era.
Identification Strategy
- The authors build a fully non-linear dynamic general equilibrium model featuring a multi-sector structure with empirically realistic production networks and optimal state-dependent pricing decisions.
- They calibrate the model to 39 Euro Area sectors, matching observed sectoral micro-pricing moments and input-output shares.
- The model is then subjected to four structural shock series (aggregate demand, labor wedge, energy, and food prices) to explain post-Covid Euro Area inflation, with counterfactual analyses demonstrating the unique contribution of network-pricing cascade interactions.
Data
The study utilizes Euro Area data, including consumption and input-output shares from the World Input-Output Database (WIOD), sectoral price adjustment frequencies and standard deviations from the PRISMA dataset, labor cost shares from the EU KLEMS database, and macroeconomic series such as Euro Area nominal GDP, nominal hourly earnings, and IMF Energy and Food Price Indices.
Benjamin Moll — Workshop on Methods and Applications for Dynamic Equilibrium Models
This paper argues that rational expectations are unrealistic and computationally intractable in heterogeneous agent macroeconomic models, proposing alternative expectation formation mechanisms like learning and survey expectations.
Finance Application
- This paper offers a rich arbitrage opportunity for finance.
- In asset pricing, replacing RE with learning models (least-squares, reinforcement learning) could explain asset price bubbles, crashes, and anomalies driven by heterogeneous beliefs and learning dynamics, especially in models with aggregate risk and non-linearities.
- For household finance, the framework could be used to model how households form expectations about future interest rates, house prices, or income risk, incorporating empirical evidence from surveys on cognitive biases and heterogeneity in beliefs, thereby improving our understanding of saving, borrowing, and investment decisions (e.g., mortgage choices, stock market participation).
heterogeneous agent modelsrational expectationsbounded rationalitylearningreinforcement learningsurvey expectationsasset pricinghousehold financemacroeconomicsfinancial crisesexpectations formation
Core finding, identification, data
Core Finding
- The central thesis is that rational expectations (RE) in heterogeneous agent (HANK) models are implausible and computationally prohibitive because they require agents to forecast entire cross-sectional distributions to predict equilibrium prices (the "Monster equation").
- The paper advocates replacing RE with simpler, empirically-grounded expectation formation models that are computationally tractable, consistent with empirical evidence, and robust to the Lucas critique.
Identification Strategy
- The paper's methodological innovation is a *critique* of the prevailing rational expectations assumption in heterogeneous agent models and a *proposal for a new research agenda*.
- It outlines three criteria (computational tractability, empirical consistency, endogeneity of beliefs) for evaluating alternative expectation formation models (temporary equilibrium, survey expectations, least-squares learning, reinforcement learning, heuristics).
- The "identification" here is more about a *call for a paradigm shift* in modeling expectations rather than a specific empirical identification strategy.
Data
The paper discusses the use of "survey expectations" data (e.g., New York Fed's Survey of Consumer Expectations) and "hypothetical vignettes" to elicit subjective beliefs, emphasizing the importance of empirical evidence on expectations formation.
Alexander Copestake, Divya Kirti, Maria Soledad Martinez Peria, Yao Zeng — Macro, Money and Financial Frictions
This paper studies how payment interoperability in India increased digital payment adoption and usage by integrating fragmented networks, particularly in regions with higher initial fragmentation.
Finance Application
- This research has direct implications for household finance by showing how payment interoperability can enhance financial inclusion and access to credit, especially for underserved populations like entrepreneurs and hawkers.
- In asset pricing, the increased efficiency and reach of payment networks could affect the valuation of fintech companies and traditional banks, as well as the liquidity and trading activity in local financial markets.
- Insurers could leverage interoperable payment data for more efficient premium collection, faster claims processing, and to develop new micro-insurance products for newly financially included segments.
PaymentsInteroperabilityNetwork EffectsFintechFinancial InclusionDigital PaymentsCredit MarketsIndiaUPIHousehold FinanceAsset Pricing
Core finding, identification, data
Core Finding
- Interoperability significantly boosted digital payment usage, with model-based estimates suggesting an over 50% increase in total digital payment usage nationally within a year.
- This growth was most pronounced in initially fragmented regions and was primarily driven by an increase in the number of users.
- It also led to increased borrowing from non-bank financial companies.
Identification Strategy
- The study leverages a natural experiment: the mandated integration of a major incumbent fintech platform with India's UPI.
- It exploits regional variation in pre-existing payment network fragmentation across districts.
- A heterogeneous adoption design compares more vs. less fragmented districts, and robustness checks include matching and instrumental variables (distance to "hub" cities for the incumbent platform).
Data
The paper uses novel data covering the universe of payments on India's Unified Payments Interface (UPI), all payments on a major pre-existing fintech platform, cross-bank cash withdrawal data from ATMs (as a proxy for cash usage), and household-level borrowing data from the Consumer Pyramids Household Survey (CPHS).
Carolin Pflueger, Pierre Yared — International Economics and Geopolitics
This paper develops a dynamic two-country model where military spending, geopolitical dominance, and government bond prices are jointly determined, revealing feedback loops and potential for self-fulfilling hegemonic transitions.
Finance Application
- This model offers a rich framework for asset pricing research, particularly in sovereign debt and currency markets.
- One could investigate how shifts in perceived geopolitical power, perhaps proxied by defense spending or military alliance changes, impact sovereign bond yields and CDS spreads, especially for countries with varying 'debt capacity.' The concept of 'geopolitical fragility' could be applied to currency markets, exploring how self-fulfilling expectations about a nation's future hegemonic role (e.g., RMB vs.
- USD) could drive exchange rate volatility and capital flows, even without direct conflict.
- Furthermore, the model's insights on 'safe assets' could inform studies on how the safe-haven premium of assets like U.S.
- Treasuries evolves with global power shifts and geopolitical tensions.
Sovereign DebtGeopoliticsInternational FinanceSafe AssetsBond MarketsSelf-Fulfilling PropheciesHegemonyMilitary SpendingPolitical RiskCurrency MarketsAsset PricingFinancial Stability
Core finding, identification, data
Core Finding
- The paper develops a dynamic two-country model where military strength and sovereign bond prices are endogenously linked.
- It demonstrates that hegemons enjoy a funding advantage, which increases with geopolitical tensions, and that war losers face greater debt devaluation.
- Crucially, high debt capacity can lead to 'geopolitical fragility' and multiple steady states, where self-fulfilling bond market expectations can trigger hegemonic transitions even in peacetime.
Identification Strategy
- This paper is primarily theoretical, constructing a dynamic two-country model that endogenizes the interplay between military spending, geopolitical dominance, and government bond prices.
- The model's mechanisms, including feedback loops between military capacity and bond safety, and the emergence of multiplicity and fragility, are designed to be consistent with three stylized empirical facts about historical borrowing costs, geopolitical tensions, and post-war debt devaluations.
Data
The paper uses historical government bond yields for the U.S., U.K., and Netherlands (1700-1970), spreads between developed country yields and U.S. yields alongside a U.S. geopolitical risk index (1980-2023), and bond price/yield data around the 2022 Ukraine invasion. It also examines average 10-year inflation rates post-WWI and post-WWII for the U.S. and defeated countries.
Joel P. Flynn, Antoine B. Levy, Jacob Moscona, Mai Wo — International Economics and Geopolitics
This paper studies how innovation reacts to foreign political risk and shapes its economic consequences, particularly regarding import reliance and trade patterns.
Finance Application
- This research offers a novel mechanism for understanding how geopolitical risk translates into firm-level fundamentals and trade patterns, directly impacting asset pricing.
- Investors could integrate political risk metrics and firm-specific supply chain exposure with innovation data to identify firms better positioned to mitigate supply shocks, informing equity valuation and sector-specific risk premiums.
- The findings on critical minerals could also be used to forecast long-term commodity price trends and volatility, creating opportunities for commodity derivatives and hedging strategies.
- Furthermore, the framework could inform the development and pricing of new political risk insurance products for supply chain resilience, and provide a quantitative basis for ESG investing strategies focused on geopolitical risk management.
Political RiskGeopoliticsInnovationTechnological ChangeSupply ChainsInternational TradeFirm ProductivityAsset PricingRisk ManagementCommodity MarketsESG Investing
Core finding, identification, data
Core Finding
- Sectors and commodities with higher exposure to foreign political risk exhibit significantly greater innovative activity, especially when risk emanates from geopolitical adversaries.
- This directed innovation reduces reliance on risky foreign imports, but can amplify the negative effects of domestic political risk on export performance for countries 'innovated out' of the trade network.
Identification Strategy
- The paper identifies causal effects by exploiting plausibly exogenous foreign fluctuations in import-weighted political risk, interacting these with differential baseline import reliance across sectors.
- Robustness is established through various fixed effects (sector-year, country-year, country-sector), pre-trend analysis showing no anticipation of risk, and placebo tests using export-weighted political risk.
Data
The study uses US patent data (PatentsView), R&D investments (Compustat), country-level political risk (ICRG), geopolitical alliance data (Correlates of War, ATOP, UN voting), trade flows (UN Comtrade, Global Trade Alert), mineral deposit data (USGS), and country characteristics (World Development Indicators).
John S. Becko, Gene M. Grossman, Elhanan Helpman — International Economics and Geopolitics
This paper examines the optimal design of tariffs for large countries that consider both economic welfare and the political allegiance of smaller states in unipolar and bipolar geopolitical scenarios.
Finance Application
- This framework offers a novel way to quantify geopolitical risk and its impact on trade policy, which can be imported into finance.
- In asset pricing, the 'geopolitical alignment' scores (derived from UN voting) could be used to construct a geopolitical risk factor, explaining cross-sectional differences in equity returns or sovereign bond yields.
- For household finance, the model's insights into trade fragmentation could inform how households adjust international portfolio allocations or consumption-saving decisions in response to changing geopolitical blocs.
- In insurance, political risk insurers could leverage these alignment metrics to refine pricing for policies covering trade disruptions, expropriation, or currency risks in countries with varying geopolitical allegiances.
GeopoliticsTrade PolicyTariffsInternational RelationsAsset PricingPolitical RiskGlobalizationSupply ChainsSovereign RiskInternational FinanceRisk Premia
Core finding, identification, data
Core Finding
- The paper finds that when geopolitical concerns are active, optimal tariffs significantly exceed the classic Mill-Bickerdike level.
- In a unipolar world, geopolitical motives can more than double the optimal tariff.
- In a bipolar world, the emergence of a second great power amplifies protectionist pressures and contributes to a retreat from globalization, with optimal tariffs reflecting both economic and political rivalry.
Identification Strategy
- The study develops a theoretical model of optimal tariffs with geopolitical considerations.
- For empirical calibration, it uses UN General Assembly voting patterns as a proxy for geopolitical alignment and estimates alignment costs based on voting similarity, GDP, and vote-buying estimates.
- Great power preferences for alignment are calibrated using military spending data, and the model is applied to both unipolar (1995-1998) and bipolar (2021-2024) geopolitical periods.
Data
The paper utilizes UN General Assembly voting records from 1946-2024, annual GDP data for UN countries from 1970-2023, and military spending data for calibrating geopolitical preference parameters.
Konstantin Egorov, Vasily Korovkin, Alexey Makarin, Dzhamilya Nigmatulina — International Economics and Geopolitics
This paper provides a comprehensive micro-level analysis of the economic impact of trade sanctions imposed on Russia following the 2022 invasion of Ukraine, focusing on their effects on import flows, firm performance, and supply chains.
Finance Application
- The detailed firm-level exposure to sanctions and their impact on revenues, costs, and supply chain linkages could be used to develop novel geopolitical risk factors for asset pricing models, particularly for emerging market equities and corporate bonds.
- Financial institutions could leverage this micro-level data to assess and price trade credit insurance or political risk insurance for companies operating in or trading with sanctioned economies, or to model systemic risk propagation through global supply chains for financial contagion analysis.
- Furthermore, the findings on rerouting and substitution could inform models of capital flight and currency stability in times of geopolitical tension.
sanctionsgeopolitical risksupply chainfirm performanceasset pricingcredit riskinsuranceinternational tradeemerging marketsRussia-Ukraine war
Core finding, identification, data
Core Finding
- Trade sanctions on Russia led to a sharp 60% decline in sanctioned country-product imports and a 31% decline in total sanctioned product imports, even after accounting for substantial rerouting and substitution through friendly countries.
- Firms exposed to these sanctions experienced a 14% drop in revenues, with particularly pronounced effects in high-tech and military-adjacent sectors, and these adverse effects propagated through domestic supply chains.
Identification Strategy
- The study employs two difference-in-differences (DiD) approaches: a pre-post DiD comparing sanctioned versus non-sanctioned country-product varieties before and after the war's onset, and a staggered DiD exploiting variation in the timing of sanctions across country-product pairs.
- Both methods use extensive fixed effects (e.g., product-country, product-quarter, country-quarter, firm, industry-year) to isolate the causal impact of sanctions, assuming parallel trends in the absence of treatment.
Data
The paper uses a manually assembled dataset of sanctions (35 countries, Feb 2022-June 2024, 10-digit product codes), transaction-level Russian customs data, comprehensive Russian firm balance sheets (2017-2023), firm-to-firm domestic railway shipment data, and government procurement contracts (2012-2024) to identify military-related firms.
Jesús Fernández-Villaverde, Yiliang Li, Le Xu, Francesco Zanetti — International Economics and Geopolitics
This paper quantifies the scale and impact of 'dark shipping' (oil tankers disabling AIS transceivers to evade sanctions) on global oil markets and macroeconomic outcomes using a novel machine learning model and satellite data.
Finance Application
- This research offers novel, high-frequency data and a methodology to analyze geopolitical risk in asset pricing and insurance.
- For asset pricing, the dark shipping data could be used to predict oil price volatility, assess the risk exposure of shipping companies (especially those involved in 'shadow fleets'), and evaluate the impact of supply chain disruptions on global equity markets.
- In insurance, the findings on increased risk and non-traditional coverage for dark ships provide a basis for modeling marine insurance premiums, assessing the solvency of alternative P&I clubs, and developing new political risk insurance products for trade routes affected by sanctions.
dark shippingoil sanctionsglobal supply chainscommodity marketsgeopolitical riskmarine insuranceasset pricingmachine learningshipping industryinternational tradesupply chain finance
Core finding, identification, data
Core Finding
- Dark shipping transports an estimated 9.3 million metric tons of crude oil per month, mitigating sanction-induced supply shocks and leading to muted global oil price responses.
- China, as the primary recipient of discounted oil, expands industrial production, transmitting deflationary, supply-driven growth to the U.S. and EU via global supply chains, offsetting some negative impacts of sanctions.
Identification Strategy
- The paper develops a multi-attribute ship clustering model to identify dark ships, combining vessel traits, movement dynamics, and trip-level AIS data.
- It then uses local projections (LPs) to estimate the dynamic causal effects of unexpected oil sanction intensity shocks, identified through a manually curated dataset of new oil-related sanctions, on global oil markets and macroeconomic variables.
Data
The study uses over 330 million crude oil tanker records from satellite-based AIS data (2017-2023), vessel characteristics (age, operator size, flag state ranking), port data, UN Comtrade data for global seaborne oil exports, and macroeconomic series from FRED and Eurostat. It also incorporates a manually curated dataset of oil-related sanctions and validates findings with satellite imagery.
John R. Grigsby, Nathan Zorzi — Economic Growth
This paper develops a life-cycle model with heterogeneous workers, directed search, and retraining to analyze the labor market outcomes for workers in polluting sectors during a climate transition, focusing on how the pace of transition affects lifetime earnings.
Finance Application
- This research offers critical insights for asset pricing and household finance by quantifying the human capital risk associated with climate transition.
- Asset pricing models could incorporate the non-linear and non-monotonic labor cost impacts on firm valuations, especially for companies with significant exposure to 'polluting' labor, leading to new ESG risk factors.
- For household finance, the findings highlight the need for tailored financial planning and insurance products (e.g., 'climate transition unemployment insurance') to mitigate the substantial and uneven human capital losses, particularly for older or less adaptable workers, influencing optimal savings and portfolio allocation decisions.
Climate RiskESGHuman CapitalLabor MarketsHousehold FinanceAsset PricingInsuranceStructural TransformationTransition RiskHeterogeneityNon-linear Effects
Core finding, identification, data
Core Finding
- Workers' lifetime earnings are strongly non-linear and even non-monotonic as a function of the pace of climate transition.
- While a faster transition (e.g., ending in 2050 instead of 2060) can benefit most workers by creating more clean jobs, it also significantly thickens the left tail of earnings losses for the least adaptable workers, posing a trade-off for policymakers.
- Structural unemployment, arising from costly hiring and job search frictions, plays a crucial role in magnifying these earnings losses compared to a frictionless market.
Identification Strategy
- The paper structurally estimates the distribution of worker skill endowments and retraining costs.
- It leverages the block recursivity of competitive search equilibria, allowing for independent simulation of worker 'bins' based on productivity.
- The non-parametric estimation of skill endowments is achieved by matching observed labor reallocation patterns between polluting and clean sectors from 2005-2025, noting that higher-skilled workers reallocate earlier.
- To disentangle skill endowments from retraining costs, the model compares reallocation patterns across different worker cohorts (e.g., 30-year-olds vs. 45-year-olds in 2005), as younger workers have a longer horizon to recoup training investments.
Data
The paper uses US Energy Information Administration (EIA) data on fossil fuel and renewable energy shares (2005-2050) to calibrate the S-curve of the climate transition. Labor market data includes the Quarterly Census of Employment and Wages (QCEW) for employment, US Bureau of Labor Statistics (BLS) Green Goods and Services (GGS) survey for green tasks, and the Current Population Survey (CPS) for job flows, wage changes, and worker demographics (education, sex, race, marital status, geography). US Bureau of Transportation data on car sales (total and alternative fuel) is used to refine manufacturing sector classifications.
Charles I. Jones — Economic Growth
This paper develops a model to quantify the optimal share of GDP that should be spent to mitigate the existential risks associated with advanced Artificial Intelligence, drawing an analogy to the economic response to the Covid-19 pandemic.
Finance Application
- This framework could be directly applied to asset pricing by quantifying an 'AI risk premium' for long-duration assets or technology companies, reflecting the probability and economic cost of AI-induced catastrophic events.
- For insurance and reinsurance, the paper's 'willingness to pay' for risk reduction could inspire novel financial instruments, like AI safety bonds or sovereign risk pools, to fund mitigation efforts against systemic, uninsurable risks.
- In ESG investing, the model provides a quantitative basis for evaluating corporate investments in AI safety, allowing for the development of metrics to assess how firms' risk mitigation strategies impact their valuation and cost of capital, particularly for tech giants.
- Furthermore, it could inform household finance research on how individuals perceive and allocate resources to prepare for low-probability, high-impact systemic risks.
AI riskexistential riskvalue of statistical liferisk mitigationasset pricingESGtail riskcatastrophe risklong-term investinghousehold financeinsurance
Core finding, identification, data
Core Finding
- The model suggests that, for most scenarios and parameter combinations, spending at least 1% of GDP annually to mitigate AI risk is justified, even without considering future generations.
- The average optimal mitigation share across Monte Carlo simulations is 8.1% of GDP, rising to nearly 30% with modest altruism towards future generations.
Identification Strategy
- The paper employs a representative agent model that balances consumption utility against the reduction of existential risk.
- The methodology involves calibrating key parameters such as the value of a statistical life (VOSL), the baseline probability of AI-related extinction, and the effectiveness of mitigation spending.
- Monte Carlo simulations are then used to explore a wide range of parameter values to determine the optimal mitigation spending as a share of GDP, benchmarked against the economic sacrifice made during the Covid-19 pandemic.
Data
The paper uses a calibrated model, drawing on estimates for the Value of a Statistical Life (around $10 million), per capita GDP/consumption (around $56,000), mortality risk from Covid-19 (0.3%), and expert survey estimates for AI existential risk (median 5%, mean >10%). It also considers the economic impact of Covid-19 (4% GDP reduction) and various time horizons for risk realization (5-20 years).
Timo Boppart, Peter J. Klenow, Reiko Laski, Huiyu Li — Economic Growth
This paper develops an endogenous growth model with firm-specific innovation step sizes and R&D efficiency to explain heterogeneous firm growth and P/E ratios, finding that smaller firms contribute disproportionately to aggregate growth.
Finance Application
- This research offers several direct applications to finance.
- In asset pricing, the strong predictive power of P/E ratios for future earnings growth and R&D intensity could be used to construct novel 'innovation' or 'growth potential' factors that explain cross-sectional stock returns, potentially outperforming traditional value factors.
- For corporate finance, the model provides a structural framework to understand how firm-specific innovation opportunities are priced into valuations, informing capital allocation decisions, M&A strategies for acquiring high-growth potential firms, or venture capital investment in early-stage innovative companies.
- Furthermore, the finding that smaller firms drive disproportionately more growth could be explored in active management strategies focusing on small-cap innovation leaders, and tested for market efficiency implications regarding the pricing of these 'idea rents'.
Asset PricingFirm GrowthInnovationR&DP/E RatioStock ReturnsCorporate FinanceMarket EfficiencyFactor InvestingProductivity Growth
Core finding, identification, data
Core Finding
- Empirically, firms with high price-earnings (P/E) ratios exhibit predictably faster subsequent earnings growth and higher R&D intensity.
- The calibrated endogenous growth model, featuring stochastic innovation step sizes and R&D efficiency, implies that the smallest listed firms (accounting for 10% of sales) are responsible for over 60% of aggregate TFP growth, while the largest 10% contribute less than 1%.
Identification Strategy
- The paper constructs an endogenous growth model, building on Klette and Kortum (2004), but crucially introduces stochastic idiosyncratic innovation step sizes and R&D efficiency across firms, and a convex R&D cost function.
- This model is calibrated to match nine empirical moments from Compustat data, including the relationship between P/E ratios and earnings/sales growth, R&D intensity patterns, and sales dispersion.
- The model then infers expected growth contributions of individual firms based on their market observables.
Data
The study uses publicly-listed Compustat firms (all sectors, 1976-2024), stock prices from CRSP, 'street' earnings from I/B/E/S, and sales data from Compustat. R&D data is specifically used for manufacturing firms.
Sheng Cai, Lorenzo Caliendo, Fernando Parro, Wei Xiang — Economic Growth
This paper quantifies how regional growth in China, driven by knowledge diffusion through internal migration and international trade, contributes to aggregate economic growth from 1990 to 2010.
Finance Application
- This paper's insights on spatial knowledge diffusion are highly applicable to finance.
- For asset pricing, the spread of investment strategies or financial innovations (e.g., FinTech) through migration of skilled labor or trade links could explain regional differences in asset returns or market efficiency.
- In household finance, the paper's findings suggest that migration from financially sophisticated regions could drive the adoption of complex financial products (e.g., mutual funds, derivatives) or insurance in less developed areas, impacting household wealth and risk management.
- Researchers could adapt the dynamic spatial growth model to study the diffusion of financial literacy or new insurance products across regions, using patent data on FinTech or insurance innovations as a proxy for knowledge.
spatial economicseconomic growthknowledge diffusionmigrationinternational tradeChinaasset pricinghousehold financeinsuranceFinTechfinancial literacyinstrumental variablesregional development
Core finding, identification, data
Core Finding
- The study finds that internal migration and international trade significantly increase local knowledge accumulation in China.
- Knowledge diffusion, particularly from more developed regions via migration, becomes the dominant driver of both regional and aggregate growth over time, surpassing capital accumulation in later periods.
Identification Strategy
- The paper establishes causality using instrumental variables.
- For migration, it employs Card (2001) and Burchardi et al. (2020) style instruments, leveraging historical migration patterns.
- For trade openness, it uses a five-year lag of import shares as an instrument.
- These instruments address endogeneity concerns in the relationship between migration/trade and knowledge growth.
Data
The paper uses comprehensive province-level data from China (1985-2015), including population census data for inter-province migration flows, macroeconomic indicators (GDP, trade, employment, capital formation), and patent data (applications, granted, stock, flow). It also uses Penn World Tables and WDI for rest of world data.
Wyatt Brooks, Kevin Donovan — Economic Growth
This paper studies the optimal design of fertilizer subsidy policy in developing countries in response to global price shocks, considering financial frictions and heterogeneous farming technologies.
Finance Application
- This paper's insights could be applied to asset pricing by examining how global commodity price shocks, especially for critical inputs, are priced into the equity or debt of firms in emerging markets, considering their financial frictions and heterogeneous production structures.
- In household finance, the framework could inform the design of optimal credit or insurance products for small businesses or farmers facing similar input price volatility and credit constraints, analyzing how such financial instruments affect their resilience and investment decisions.
- For corporate finance, it could shed light on how firms in developing economies manage supply chain risks from imported inputs, and how government interventions (like subsidies or credit guarantees) interact with firm-level financial constraints and hedging strategies.
Optimal PolicyFertilizer SubsidiesGlobal ShocksFinancial FrictionsHeterogeneous AgentsGeneral EquilibriumDeveloping CountriesCommodity PricesSupply Chain RiskHousehold FinanceAsset PricingCorporate FinanceIndustrial Policy
Core finding, identification, data
Core Finding
- The optimal fertilizer subsidy policy in a developing country should "lean into" a global input price shock by decreasing subsidies, reallocating production towards less fertilizer-intensive villages, especially when village production is sufficiently substitutable.
- This contrasts with observed policy responses that typically increase subsidies, leading to significant welfare losses.
Identification Strategy
- The authors use the Russian invasion of Ukraine as an exogenous shock that doubled real fertilizer prices in Rwanda.
- They combine this with household-level panel data from 444 rural Rwandan villages (2020-2024) and a shift-share approach based on baseline village-level fertilizer intensity to identify the causal impact of the shock and estimate key model elasticities.
Data
The paper utilizes primary household-level panel data collected from 444 rural Rwandan villages between 2020 and 2024, as well as vendor-level data from associated markets starting in 2021. This is complemented by comparisons to the nationally representative Rwandan Agricultural Household Survey (AHS).
Tao Wang, Xincheng Qiu, William Du, Adrian Monninger — Behavioral Macro
This paper analyzes how households' sticky and heterogeneous perceptions of unemployment risk influence consumption and saving behavior over business cycles, leading to underinsurance and sluggish economic recoveries.
Finance Application
- This research offers significant insights for household finance, asset pricing, and insurance.
- In household finance, the finding of systematic underinsurance due to sticky beliefs suggests that households may hold suboptimal portfolios or emergency savings, leading to higher financial distress during recessions.
- For asset pricing, this could imply that equity risk premia or credit spreads on consumer debt might not fully reflect the true, objective unemployment risk, but rather the slower-moving perceived risk, potentially explaining anomalies or slow-moving factors.
- In insurance, the underreaction to risk could explain why demand for unemployment or disability insurance is lower than optimal, pointing to market opportunities for products that address these behavioral biases or to the need for policy interventions to improve financial resilience.
Household FinanceBehavioral EconomicsUnemployment RiskExpectationsMachine LearningConsumptionSavingInsuranceBusiness CyclesHeterogeneityAsset PricingCredit Risk
Core finding, identification, data
Core Finding
- The core finding is that perceived job finding and separation expectations are sticky and heterogeneous, underreacting to real-time changes in unemployment risks.
- This belief stickiness attenuates the precautionary saving channel, causing workers to underinsure during recessions and leading to more sluggish economic recoveries.
- The combination of high risk exposure and underinsurance due to belief stickiness operates as a novel amplification mechanism over the business cycle, with low-educated workers exhibiting the stickiest beliefs and being most underinsured.
Identification Strategy
- The paper backcasts subjective expectations on job finding and separation from the Survey of Consumer Expectations (SCE) to 1978 using machine learning (LASSO) trained on the Michigan Survey of Consumers (MSC).
- It then uses real-time machine learning forecasts to proxy objective unemployment risks.
- The identification strategy involves comparing (a) perceived risk, (b) objective risk, and (c) realized transition rates, and then feeding these into a heterogeneous-agent consumption-saving model to simulate aggregate consumption dynamics under different belief assumptions to quantify the impact of belief stickiness.
Data
The paper uses subjective expectations data from the Survey of Consumer Expectations (SCE) and the Michigan Survey of Consumers (MSC), realized job transition rates from the Current Population Survey (CPS), and a pool of 600 real-time macroeconomic indicators for machine learning forecasts.
Jonathan Federle, Cathrin C. Mohr, Moritz Schularick — Behavioral Macro
This paper investigates how unexpected inflation and growth affect voting behavior, particularly increasing support for extremist and populist parties, especially when real wage growth is low.
Finance Application
- This research offers a robust framework for understanding political risk in asset pricing.
- Unexpected inflation leading to increased support for anti-system parties and social unrest could significantly raise political risk premiums in sovereign bond markets, demanding higher yields.
- Furthermore, equity markets might experience heightened volatility and sector-specific revaluations as investors anticipate potential policy shifts (e.g., nationalization, trade protectionism) that could impact corporate earnings and valuations.
- This mechanism could be integrated into models for dynamic asset allocation or stress testing portfolios against political instability shocks.
InflationEconomic SurprisesPolitical RiskPopulismExtremismElectionsReal WagesMacroeconomicsAsset PricingSovereign BondsEquity MarketsPolitical Economy
Core finding, identification, data
Core Finding
- Unexpected inflation (and misery index surprises) significantly increase vote shares for extremist and populist parties, with a 10 percentage point inflation surprise leading to a 15% increase in their vote share.
- This effect is particularly pronounced when real wage growth lags behind the national average, suggesting that real wage erosion is a key mechanism, and it also correlates with increased public protest.
Identification Strategy
- The study uses a panel data regression model with country and time fixed effects, identifying the causal impact through 'inflation surprises' defined as the difference between realized inflation and prior one-year-ahead inflation forecasts.
- This approach isolates the effect of unanticipated economic disappointment, controlling for the direct effects of realized inflation and growth, and treats these surprises as plausibly exogenous shocks.
Data
The paper utilizes a novel long-run cross-country dataset covering 76 years, 18 advanced economies, and 365 elections, combining inflation surprise data from Kim, Ranaldi, and Schularick (2024), macroeconomic data from the Jordà-Schularick-Taylor Macrohistory Database, and party vote shares (extremist/populist) from Funke, Schularick, and Trebesch (2016, 2023), supplemented by OECD forecasts and social unrest data from Banks and Wilson (2014).
Cosmin L. Ilut, Rosen Valchev — Behavioral Macro
This paper develops a bounded rationality framework where agents learn optimal behavior through costly internal reasoning and free accumulated experiences, using Bayesian non-parametric estimation to model subjective uncertainty and its impact on learning and decision-making.
Finance Application
- This framework could be applied to model investor learning about asset return distributions or risk premia, where costly research (reasoning) and observed past returns (experience) interact.
- It could explain persistent behavioral biases or market anomalies if "learning traps" lead to suboptimal portfolio allocations or trading strategies.
- In household finance, it could model how individuals learn about optimal savings or debt management, explaining why some households remain "hand-to-mouth" despite financial education or experience, or why financial literacy interventions might have limited long-term impact if agents fall into stable, suboptimal belief states.
- It could also be used to model how insurance consumers learn about risks and optimal coverage, potentially explaining underinsurance or overinsurance in certain segments.
bounded rationalitycognitive economicsreinforcement learningBayesian learningGaussian Processeslearning trapshousehold financeconsumption-savingsMPCwealth inequalitydecision theorybehavioral finance
Core finding, identification, data
Core Finding
- The model demonstrates that the interaction between costly reasoning and free experience-based learning, coupled with endogenous uncertainty, leads to "learning traps" where agents' beliefs stabilize around suboptimal policies.
- This mechanism can generate empirically relevant phenomena like a large fraction of "Hand-to-Mouth" agents, high marginal propensities to consume (MPCs) for wealthy individuals, and significant wealth inequality, with long-run welfare losses of up to 5% of permanent consumption.
Identification Strategy
- The paper's methodological innovation is a novel framework of bounded rationality that integrates two distinct learning modes: costly deliberative reasoning (internal mental simulation) and free experience-based learning (from observed utility outcomes).
- It uses Gaussian Processes for Bayesian non-parametric estimation of the unknown action value function, allowing subjective uncertainty to dynamically modulate the trade-off between exploitation and experimentation.
- The optimal reasoning strategy is determined by a cost-benefit analysis where the cognitive cost is proportional to the information flow (entropy reduction).
Data
The paper primarily uses a numerical illustration based on an Aiyagari (1994)-style consumption-savings problem for its analysis. It references empirical evidence from Lewis et al. (2019) on MPC heterogeneity and Halvorsen et al. (2024) on wealth accumulation to motivate and validate its findings, but does not use specific external datasets for its own analysis beyond calibration targets.
Marta Leva, Nicola Gennaioli, Raphael Schoenle, Andrei Shleifer — Behavioral Macro
This paper develops and tests a selective recall model to explain how inflation expectations de-anchor, highlighting the role of associative human memory cues (temporal and numerical similarity) in shaping beliefs and creating inconsistencies between different expectation measures.
Finance Application
- This research offers several avenues for finance.
- In asset pricing, the cohort-specific and state-dependent inflation expectations could explain heterogeneous investment strategies across generations, impacting demand for inflation-sensitive assets like TIPS or real estate, and potentially contributing to excess volatility during inflationary regimes.
- For household finance, selective memory could lead to suboptimal saving and consumption decisions, such as younger households under-saving for retirement due to 'forgotten' high inflation, or older households over-reacting to inflation cues.
- The inconsistency between point and density forecasts suggests investors might misprice tail risks (e.g., deflation) in their overall portfolio allocation, only to correct when explicitly prompted, leading to sudden market re-pricing.
behavioral economicsinflation expectationsmemorycognitive psychologyhousehold financeasset pricingsurvey dataintergenerational differencesfinancial decision-makingrisk perception
Core finding, identification, data
Core Finding
- Households' inflation expectations are rigid when inflation is anchored but become highly unstable when surges in inflation trigger the retrieval of forgotten high-inflation episodes, particularly among the elderly.
- This selective recall also explains a state-dependent discrepancy between point and density-based expectations, where deflationary scenarios are neglected in point forecasts but cued in density forecasts.
Identification Strategy
- The paper's identification strategy is to build a model of 'mnemonic beliefs' based on associative human memory principles (primacy, recency, numerical similarity) from cognitive psychology.
- It then tests the specific predictions of this model using micro-level survey data, comparing its explanatory power for observed patterns (e.g., faster de-anchoring by elderly, inconsistency between point and density forecasts) against alternative models like Bayesian learning or simpler experience-based learning.
Data
The study uses micro data from the New York Fed's Survey of Consumer Expectations (SCE) and the University of Michigan's Survey of Consumers (MSC). It also incorporates actual CPI data from Shiller (2005) and compares findings across international surveys from the Bank of England, ECB, and Japan.
Dmitry Taubinsky, Luigi Butera, Matteo Saccarola, Chen Lian — Behavioral Macro
This paper demonstrates that individuals' inflation forecasts are overly sensitive to idiosyncratic household-level events, a bias primarily driven by affect-cued recall.
Finance Application
- This insight is highly relevant for household finance and asset pricing.
- In household finance, it suggests that personal financial decisions (e.g., saving, investment in inflation-hedging assets, or mortgage choices) may be systematically biased by recent life events, leading to suboptimal wealth accumulation.
- For asset pricing, if a significant portion of the population's inflation expectations are driven by idiosyncratic, affect-laden experiences rather than macroeconomic fundamentals, this could introduce non-fundamental noise into aggregate demand and, consequently, into asset prices, creating potential arbitrage opportunities for investors with superior information processing.
Behavioral EconomicsInflation ExpectationsHousehold FinanceAffect-cued RecallIdiosyncratic ShocksConsumer BeliefsRational ExpectationsMemory BiasAdministrative DataSurvey DataAsset PricingMacroeconomics
Core finding, identification, data
Core Finding
- The core finding is that people's inflation forecasts covary much more strongly and negatively with personal income changes (both realized and expected) and adverse health events than with actual inflation.
- These deviations from Bayesian rational expectations are attributed to affect-cued recall, where negative (positive) household-level events trigger negative (positive) recollections, leading to pessimistic (optimistic) forecasts.
Identification Strategy
- The identification strategy leverages a unique linkage between a large-scale Danish consumer expectations survey and comprehensive administrative data.
- This allows for formal tests comparing how actual versus forecasted inflation covaries with household-level variables like income changes and emergency room visits.
- A key natural experiment involves exploiting the random timing of survey participation relative to a family ER visit to isolate the causal impact of such events on beliefs.
Data
The paper utilizes a unique dataset linking the Danish Consumer Expectations Survey (quantitative inflation forecasts and backcasts, qualitative financial situation forecasts) with detailed Danish administrative registry data. This administrative data includes past, present, and future income and assets from the Danish Tax and Customs Authority (SKAT), adverse health events from the Danish National Patient Registry (NPR), and rich demographics.
Menaka Hampole, Dimitris Papanikolaou, Lawrence D.W. Schmidt, Bryan Seegmiller — Economic Fluctuations and Growth Program Meeting
This paper develops novel, granular measures of firm- and task-level AI exposure using natural language processing on resume and job posting data, and analyzes AI's impact on firm productivity, labor demand, and occupational employment shifts.
Finance Application
- The paper's granular, firm-level AI adoption measures, coupled with its robust identification strategy, offer several avenues for finance research.
- In asset pricing, one could investigate whether firm-level AI adoption predicts future stock returns or affects firm valuation multiples, potentially identifying an 'AI factor' or mispricing.
- For corporate finance, the task-level AI exposure and concentration measures could be used to assess firm-specific labor risk, impacting credit risk, bond yields, or the cost of capital.
- In insurance, the paper's insights into AI's role in fraud detection and risk modeling suggest that insurers adopting these technologies could achieve lower loss ratios, which could be linked to their profitability and pricing strategies for various insurance products.
Artificial IntelligenceLabor MarketFirm ProductivityTask ExposureEmploymentInstrumental VariablesNLPAsset PricingCorporate FinanceRisk ManagementInsuranceHuman CapitalTechnological ChangeFirm ValuationCredit Risk
Core finding, identification, data
Core Finding
- AI adoption significantly increases firm productivity (sales, profits, TFP, employment) and leads to task-level labor substitution (mean AI exposure reduces labor demand by 14.5% over 5 years).
- However, this is offset by task reallocation (concentrated AI exposure increases labor demand by 7.5%) and firm-wide productivity gains, resulting in muted net employment effects, especially for high-wage occupations.
Identification Strategy
- The paper uses a Bartik shift-share instrumental variable (IV) strategy to address endogeneity concerns of AI adoption.
- The instrument predicts firm-level AI exposure by interacting university-level AI intensity (share of graduates entering AI-related occupations in 2014-2018) with firm-specific historical hiring shares from universities (2005-2009).
- For occupation-level measures, a shift-share instrument is constructed based on mean/concentration of task exposure to AI across all applications in a period, multiplied by the predicted firm-level AI adoption intensity.
Data
The primary data sources include Revelio Labs (resume and job posting data), O*NET (occupation and task descriptions), Compustat (firm financials), Census Business Trends and Outlook Survey (BTOS) for validation, USPTO AI patent database for validation, LightCast for job posting skills, and BLS-OES for re-weighting.
Josh Feng, Xavier Jaravel — Entrepreneurship
This paper demonstrates significant homophily in hiring by entrepreneurs, where founders are more likely to hire workers from similar social backgrounds, and quantifies the potential wage impacts of increasing entrepreneurial access for underrepresented groups.
Finance Application
- This research has significant implications for venture capital and private equity, as founder-employee homophily could impact firm growth, innovation, and ultimately, exit valuations if it limits access to diverse talent pools.
- For asset pricing, understanding these hiring patterns could inform models of firm valuation and IPO performance, especially for firms founded by underrepresented groups, potentially revealing mispricing or new ESG-related risk factors.
- In household finance, the quantified wage impacts of entrepreneurial access for different demographic groups directly contribute to understanding wealth accumulation, financial inequality, and the effectiveness of policies aimed at fostering inclusive economic growth.
EntrepreneurshipLabor MarketsHomophilyDiversityWage InequalityFirm PerformanceVenture CapitalHousehold WealthESG Investing
Core finding, identification, data
Core Finding
- Entrepreneurs are significantly more likely to hire workers from similar social backgrounds (gender, race, age, education), with these effects being quantitatively large and persistent over the firm's first ten years.
- This homophily is primarily driven by labor demand, as new firms pay higher relative wages to individuals from similar backgrounds to the entrepreneur.
- Counterfactual simulations suggest that reducing access barriers to entrepreneurial careers for women, Blacks, and Hispanics could lead to significant increases in their relative wages (2.77%, 4.36%, and 2.26% respectively).
Identification Strategy
- The paper identifies homophily using detailed fixed effects (industry-by-metro area-by-cohort) to control for confounding factors.
- To disentangle labor supply from labor demand, it employs AKM regressions to extract firm fixed effects for demographic subgroups and then regresses the differences in these firm fixed effects against founder characteristics.
- A CES production function model, calibrated with empirical estimates, is used for counterfactual analysis.
Data
The study primarily uses U.S. Census Longitudinal Employer-Household Dynamics (LEHD) data, complemented by LinkedIn profiles (from Revelio Labs) and Crunchbase data for venture-backed startups. These datasets provide information on firm demographics, worker characteristics, wages, and firm-level attributes.
David Baqaee, Kunal Sangani — Economic Fluctuations and Growth Program Meeting
This paper develops a framework to analyze the effects of quantity-based distortions (quotas) on aggregate output and welfare, characterizing their nonlinear effects and misallocation costs using a small set of sufficient statistics.
Finance Application
- This framework is highly applicable to financial regulation, where many policies are quantity constraints (e.g., bank capital requirements, leverage limits, short-sale bans, foreign exchange controls).
- The concept of 'rents' as a sufficient statistic for the value of relaxing a constraint could be used to quantify the profitability impact of financial regulations on specific institutions or markets.
- The explicit dynamic model and use of asset prices to estimate the effects of announced policy changes (e.g., new capital requirements or trading restrictions) could be directly imported into asset pricing to understand market reactions and the valuation of regulated financial assets.
- Furthermore, the rent-seeking extension could analyze the welfare costs of lobbying and compliance efforts in the financial sector.
Financial RegulationQuantity ConstraintsAsset PricingCapital ControlsMarket MicrostructureEconomic PolicyWelfare AnalysisRent-SeekingDynamic ModelsSufficient Statistics
Core finding, identification, data
Core Finding
- The paper finds that economies with quotas are 'constrained efficient,' meaning allocations maximize output subject to quotas, satisfying macro-envelope conditions and simplifying comparative statics.
- This allows for a nonparametric and nonlinear characterization of how quota and productivity changes affect aggregate output, and how to derive welfare costs of misallocation from an inverse demand system that maps quota prices to quota levels, relying on rents as sufficient statistics.
- For dynamic settings, the effects of announced quota changes can be estimated using asset prices and their changes.
Identification Strategy
- The identification strategy relies on observing quota rents (or prices) and their response to changes in quota levels.
- For first-order effects, only rents are needed.
- For nonlinear effects and misallocation costs, the 'quota demand system' (how rents change with quota levels) is required, which can be estimated directly from data (variation in quotas across markets/time) or built from structural models.
- Natural experiments, such as the H-1B visa lottery, the entry of ride-sharing services, the phase-out of trade quotas, and changes in capital controls, provide exogenous variation.
Data
The paper uses data from various sources for its empirical applications: H-1B visa lottery earnings (Clemens 2013), zoning taxes for 24 U.S. metropolitan areas (Gyourko and Krimmel 2021), New York City taxi medallion prices and vehicle counts (NYC Taxi and Limousine Commission), Chinese textile and apparel export data (OTEXA, Brambilla et al. 2010) and quota license prices (chinaquota.com), and Argentine capital outflow data (Central Bank of Argentina, Refinitiv, Adler et al. 2019).
Robert B. Fluegge, Michael Bailey — Entrepreneurship
This paper uses large-scale social network data and quasi-experimental designs to causally identify how social connections to existing entrepreneurs influence individual firm founding behavior and the aggregate economic spillovers.
Finance Application
- This research offers significant insights for household finance and private markets.
- In household finance, it suggests that social networks could explain heterogeneous participation in private investments (e.g., angel investing, local business funding) or entrepreneurial ventures, with individuals connected to successful founders more likely to invest or start firms.
- For private markets and venture capital, the findings imply that VCs could optimize deal sourcing and portfolio company success by mapping and leveraging entrepreneurial social networks, potentially leading to a 'network premium' in startup valuations or influencing the geographic clustering of investment.
- Furthermore, the industry-specific spillovers could inform strategies for specialized private equity funds.
Social NetworksEntrepreneurshipCausal InferenceQuasi-Experimental DesignHuman CapitalSpilloversPrivate MarketsHousehold FinanceVenture CapitalEconomic GeographyPolicy Targeting
Core finding, identification, data
Core Finding
- The paper finds that social connections to existing firm owners significantly increase an individual's probability of starting a successful new firm, with effects being highly industry-specific and stronger for same-gender connections.
- Modest connection-level effects translate to economically meaningful aggregate social spillovers, implying a substantial 'social multiplier' for entrepreneurship.
- Targeting policy interventions to areas with high spillovers can increase firm entry by 32% compared to targeting low-entrepreneurship places.
Identification Strategy
The study employs two quasi-experimental designs: the 'College design' leverages plausibly exogenous friend-of-friend connections formed during the first weeks of college due to random assignments, and the 'Movers design' exploits the timing of friends' migration across commuting zones using stacked difference-in-differences to isolate causal variation in exposure to entrepreneurial environments.
Data
The research utilizes a novel, privacy-protected dataset linking social network data (128 million U.S. Facebook users and 70 billion friendships) to establishment-level data on new businesses (Customer Graph, 8 third-party sources) and self-reported firm ownership. It also incorporates socioeconomic status estimates, Business Dynamics Statistics, and other public data sources.
Laura Chioda, Paul Gertler, David Contreras-Loya, Dana Carney — Entrepreneurship
This paper uses a randomized controlled trial in Uganda to evaluate the medium and long-term financial returns of entrepreneurial training programs focused on either hard or soft business skills for high school students.
Finance Application
- This paper provides causal evidence on the long-term financial returns to specific human capital investments, which is highly valuable for finance.
- In asset pricing, researchers could use these insights to model how exogenously enhanced entrepreneurial skills (especially soft skills like negotiation) affect the valuation and growth trajectories of private firms, potentially informing venture capital investment strategies or the pricing of private equity.
- For household finance, the findings offer a direct link between skill development and household wealth accumulation through entrepreneurship, enabling studies on how such programs influence financial resilience, savings decisions, and demand for microfinance or business insurance products among entrepreneurial households in emerging markets.
Entrepreneurial FinanceHuman CapitalBehavioral FinanceSmall BusinessEmerging MarketsRandomized Controlled TrialSkill DevelopmentLong-term ReturnsHousehold WealthCredit Markets
Core finding, identification, data
Core Finding
- The 3-week mini-MBA program significantly improved entrepreneurial skills and led to higher quality, larger, and more profitable businesses for participants nine years post-intervention.
- While both hard and soft skills training were effective, the soft skills curriculum was directly linked to improvements in self-efficacy, persuasion, and negotiation.
- Both curricula yielded substantial, cost-effective returns on investment through increased business earnings.
Identification Strategy
The study employs an at-scale randomized controlled trial (RCT) with a 3-arm field experiment design. 4,400 high school students were randomly assigned to one of two treatment groups (hard skills mini-MBA or soft skills mini-MBA) or a control group, allowing for causal inference of the program's impacts over 4 and 9 years.
Data
The paper uses primary data from a field experiment in Uganda, collected through baseline surveys and subsequent medium-term (4-year) and long-term (9-year) follow-up surveys of 4,400 high school students. Data includes demographics, labor market indicators, business performance (revenues, profits, capital), business practices, and measures of hard and soft skills.
Marius Berger, Sara Calligaris, Andrea Greppi, Dmitri Kirpichev Cherezov — Entrepreneurship
This paper investigates whether acquisitions of start-ups by incumbent firms foster or stifle innovation, finding that such deals often lead to a decline in start-up innovation, particularly in cases with high market overlap.
Finance Application
- This research offers a direct application to asset pricing by examining whether public markets efficiently price the long-term innovation consequences of M&A deals.
- An event study could analyze acquirer stock performance around acquisition announcements, differentiating between deals likely to stifle versus scale innovation based on the paper's criteria (e.g., technological proximity, industry overlap).
- Furthermore, venture capital research could investigate how the long-term returns of VC funds are affected by the innovation trajectory of their portfolio companies post-acquisition, potentially revealing an 'acquisition discount' for deals identified as stifling innovation.
M&Aacquisitionsinnovationstart-upspatentsmarket powerantitrustcorporate financeasset pricingventure capitalfirm valuationmarket efficiency
Core finding, identification, data
Core Finding
- The core finding is that acquisitions of innovative start-ups by incumbent firms lead to a substantial decline in the start-ups' patenting activity and patent quality post-acquisition, without a corresponding increase in the acquirer's innovation.
- This negative impact is more pronounced when the acquirer and target operate in the same geographical area or industry, suggesting anti-competitive motives like 'killer acquisitions' rather than innovation scaling.
Identification Strategy
- The study employs a staggered difference-in-differences (DID) framework, using the de Chaisemartin and D'Haultfoeuille (2024) estimator to compare acquired start-ups with not-yet-acquired start-ups.
- To establish causality, it combines this with an exogenous shock identification strategy from Seru (2014), comparing acquired targets to those that received an acquisition bid that was subsequently withdrawn for reasons unrelated to innovation.
Data
The paper uses a comprehensive firm-level cross-country dataset covering 60 countries for the period 2001-2021. This dataset integrates information from the OECD/STI Start-ups Database (Crunchbase and Dealroom), Orbis M&A (for acquisition details), and PATSTAT Global (for patent data), supplemented by Orbis vintage 2022 and OECD/STI IP Database.
Jay Stewart, Cindy Cunningham, Matthew Dey, Lucia S. Foster, Cheryl Grim, John C. Haltiwanger, Rachel L. Nesbit, Sabrina Wulff Pabilonia, Cody Tuttle, Zoltan Wolf — Conference on Research in Income and Wealth
This paper investigates the complex, nonlinear, and industry-specific relationships between establishment-level productivity and the organization of tasks, worker skills, and occupations within manufacturing industries.
Finance Application
- This paper offers significant arbitrage opportunities for finance research by providing a granular, nonlinear understanding of human capital's impact on firm productivity.
- In asset pricing, researchers could develop new human capital-based factors, perhaps distinguishing between 'optimal' and 'suboptimal' task/skill mixes (especially at the extremes of the distribution) to predict cross-sectional stock returns or firm growth, particularly in high-tech sectors.
- For firm valuation, the detailed task/skill measures could be used to better quantify and value human capital as an intangible asset, potentially identifying mispricings by the market that relies on simpler labor input metrics.
- In corporate finance, the findings could inform studies on how a firm's human capital structure influences its capital structure decisions, investment in R&D, or resilience to economic shocks, especially given the observed industry-specific nonlinearities.
human capitalfirm productivitylabor economicsintangible assetsasset pricing factorsfirm valuationnonlinear relationshipsindustry effectsoccupational structuremicrodatacorporate finance
Core finding, identification, data
Core Finding
- The core finding is that explaining productivity dispersion across businesses requires accounting for differences in how tasks, skills, and occupations are organized, revealing strong, highly nonlinear, and industry-specific relationships between establishment-level productivity and these human capital measures.
- These relationships are particularly pronounced at the extremes of the task/skill distribution and for larger establishments, especially in high-tech industries, accounting for a substantial share of within-industry productivity dispersion.
Identification Strategy
- The paper's methodological innovation lies in linking establishment-level data from the Bureau of Labor Statistics Occupational Employment and Wage Statistics (OEWS) survey with productivity data from the Census Bureau's manufacturing surveys (CMP) to create a novel matched dataset.
- While explicitly descriptive rather than causal, the analysis emphasizes uncovering complex, nonlinear relationships through methods like fitting quartic relationships and incorporating industry-specific interactions, moving beyond simple linear adjustments to labor input.
Data
The paper uses establishment-level data from the Bureau of Labor Statistics Occupational Employment and Wage Statistics (OEWS) survey, productivity data from the Census Bureau's manufacturing surveys (Annual Survey of Manufactures and Census of Manufactures) linked through the Longitudinal Business Database (LBD) to form the Collaborative Micro-productivity Project (CMP) data, and task indexes from the Occupational Information Network (O*NET).
Gabriel Chodorow-Reich, Edward L. Glaeser, Xavier Jaravel, Avi Lipton — Conference on Research in Income and Wealth
This paper develops a new, welfare-based price index for housing that incorporates dynamic tenure choice, heterogeneous preferences for owning, and transaction costs, showing it behaves differently from traditional rent-based or user-cost measures, especially in response to mortgage rate changes.
Finance Application
- The model's ability to quantify non-pecuniary benefits (WTP) for homeownership could be used to better understand housing demand, optimal tenure choice under various financial constraints (e.g., liquidity, mortgage rates), and the welfare implications of housing policies.
- This could inform models of household portfolio allocation, particularly the housing share, and how it interacts with other asset classes.
- The "mortgage rate lock" effect and its impact on mobility and housing supply/demand dynamics are directly relevant to mortgage-backed securities (MBS) pricing, as these frictions affect prepayment speeds and housing market liquidity.
- Understanding how different price indices affect perceived inflation could also influence investor expectations and asset pricing.
Household FinanceHousingMortgagesConsumer Price IndexWelfare EconomicsTenure ChoiceBehavioral EconomicsAsset PricingMonetary Policy
Core finding, identification, data
Core Finding
- The paper finds that its new shelter price index, which accounts for non-pecuniary benefits of owning and dynamic tenure choice, responds differently to shocks than traditional measures.
- For instance, a 2 p.p. increase in mortgage rates causes their index to jump 2.6 p.p. above trend, unlike a rent-based index which doesn't change.
- Historically, their index shows less volatility than rent from 2019-2022 and rises less quickly than rent from 2022-2024, reflecting that two-thirds of the population own and experience no direct change when rents fluctuate.
Identification Strategy
- The paper's methodological innovation is the derivation and implementation of a welfare-based price index for housing using a money-metric value function within a dynamic life-cycle model of tenure choice.
- This model is calibrated using a a customized survey that elicits households' willingness-to-pay (WTP) for homeownership, alongside other empirical moments like homeownership rates and transition rates, to capture heterogeneous preferences and dynamic decisions.
- Time-varying "preference shocks" are introduced to match historical homeownership rates while isolating the welfare impact of price changes.
Data
The paper uses a customized survey to elicit households' willingness-to-pay for homeownership. It also incorporates data from the U.S. Consumer Price Index (CPI), Consumer Expenditure Survey (CEX), Survey of Income and Program Participation (SIPP), Survey of Consumer Finances (SCF), American Housing Survey (AHS), Federal Reserve Bank of New York Survey of Consumer Expectations, Freddie Mac House Price Index (FMHPI), Zillow All Homes—Bottom Tier series, and Board of Governors of the Federal Reserve System for mortgage rates and Treasury forward rates.
Ṣebnem Kalemli-Özcan, Muhammed A. Yildirim, Can Soylu — Economic Fluctuations and Growth Program Meeting
This paper develops a New Keynesian Open Economy (NKOE) model incorporating global trade and production networks, input-output linkages, and heterogeneous monetary policy to analyze the macroeconomic impact of trade distortions.
Finance Application
- The finding that 'tariff threats' alone can cause deflation and lower output due to expectation channels could be applied to asset pricing research to study how financial markets (e.g., equity indices, currency markets, commodity prices) react to geopolitical announcements and policy uncertainty, even in the absence of actual policy changes.
- The NKOE Leontief inverse could be adapted to model how supply chain shocks (e.g., from climate events or pandemics) propagate through global production networks to affect firm-specific cash flows and, consequently, credit risk, equity valuations, or the pricing of supply chain interruption insurance.
- This framework also offers a lens to analyze how exchange rate dynamics under different monetary policy regimes impact hedging strategies for multinational corporations.
Global NetworksMonetary PolicyTradeTariffsInput-Output LinkagesInflationExchange RatesMacroeconomicsOpen EconomyNew Keynesian ModelSupply ChainsPolicy UncertaintyExpectationsStagflationInternational Risk SharingAsset PricingCurrency MarketsCredit RiskSupply Chain Finance
Core finding, identification, data
Core Finding
- The macroeconomic impact of tariffs, including stagflationary outcomes, depends critically on global production networks, international risk sharing, and endogenous monetary policy responses.
- Crucially, tariff threats, even without implementation, are self-defeating, leading to deflation and lower output due to the expectations channel and immediate exchange rate adjustments.
Identification Strategy
- The paper develops a novel NKOE Leontief inverse to capture the economy-wide propagation of tariff distortions through global production networks.
- It uses a linearized model for analytical solutions and a non-linear version solved with Dynare using MIT shocks under perfect foresight for quantitative exercises, allowing for analysis of permanent, near-permanent, and transitory tariff shocks, including 'threats' where tariffs are announced but not implemented.
Data
The study uses OECD's Inter-Country Input-Output (ICIO) tables (Yamano et al., 2023) for the global production network (2019 data), and the WTO-IMF Tariff Tracker (WTO and IMF, 2025) for sectoral implemented tariffs (2018 and 2025). Sectoral price rigidities are disciplined by estimates from Nakamura and Steinsson (2008), and monetary policy parameters are based on Carvalho et al. (2021a) and Clarida et al. (2000).
Enghin Atalay, Ali Hortaçsu, Nicole Kimmel, Chad Syverson — Conference on Research in Income and Wealth
This paper argues that conventional measures understate manufacturing productivity growth, especially in durable and computer-related goods, by failing to fully capture quality improvements, leading to a misperception of stagnation.
Finance Application
- If TFP growth in certain manufacturing sectors, particularly tech-related durable goods, is systematically understated by official statistics, then firms in these sectors might be consistently undervalued by investors relying on these data.
- This could create an 'unpriced factor' in asset pricing, where a portfolio of firms in these truly high-growth sectors outperforms.
- For household finance, if the real cost of living is falling faster than official inflation measures suggest due to quality improvements in goods, households' real wealth and purchasing power are higher than perceived, potentially influencing their consumption-savings decisions, retirement planning, and risk tolerance in ways not captured by standard models.
ProductivityTotal Factor Productivity (TFP)Inflation MismeasurementQuality AdjustmentManufacturingDurable GoodsInformation and Communication Technology (ICT)Asset ValuationMacro-FinanceHousehold ConsumptionEconomic Growth
Core finding, identification, data
Core Finding
- We find that nearly all measured TFP growth since 1987 in manufacturing, and its post-2000s decline, comes from a few computer-related industries.
- By contrasting consumer-facing and producer-facing price indices, we estimate that TFP growth is understated by 1.4 percentage points in durable manufacturing and 0.4 percentage points in nondurable manufacturing, and slightly overstated in nonmanufacturing industries, primarily due to uncaptured quality improvements in high-tech products.
Identification Strategy
- The paper introduces a price-based dual approach that contrasts consumer-facing price indices (PCE) with producer-facing (gross output deflators and import price indices) to identify mismeasurement in real output and TFP growth.
- It uses an input-output framework to parse offsetting effects of mismeasured input and output prices, assuming that consumer price indices more comprehensively measure quality improvements.
- The key assumption is that mismeasurement in import price indices is proportionate to mismeasurement in gross output deflators, with a baseline value of ξ=1.5.
Data
The study utilizes Bureau of Labor Statistics (BLS) total factor productivity (TFP) data, Bureau of Economic Analysis (BEA) industry gross output deflators, BLS import price indices, Personal Consumption Expenditures (PCE) price indices from National Income and Product Accounts (NIPA) Table 2.4.4U, and BEA input-output tables. It also references BLS Occupational Employment and Wage Statistics (OEWS) and Economic Censuses for detailed industry and product information.
John Coglianese, Seth Murray, Christopher J. Nekarda — Conference on Research in Income and Wealth
This paper constructs harmonized U.S. population and labor force statistics by reweighting Current Population Survey (CPS) microdata to consistently reflect the latest population estimates, thereby removing discontinuities present in official Bureau of Labor Statistics (BLS) series.
Finance Application
- This harmonized labor market data offers a cleaner signal for financial markets.
- In asset pricing, researchers could re-evaluate the predictive power of labor market indicators (e.g., unemployment, LFPR) for equity risk premia, bond yields, or housing market trends, especially around macroeconomic shocks or policy shifts, free from data discontinuities.
- For household finance, the harmonized microdata weights enable more accurate analysis of how household consumption, savings, and debt decisions respond to true changes in labor force status, improving models of household financial resilience.
- In insurance, more precise demographic and labor force projections, particularly those incorporating immigration, could refine actuarial models for life, health, and unemployment insurance, leading to better risk assessment and premium setting.
Labor MarketsPopulation DemographicsData HarmonizationCurrent Population Survey (CPS)MacroeconomicsEconomic IndicatorsUnemployment RateLabor Force Participation Rate (LFPR)ImmigrationTime Series AnalysisReweightingAsset Pricing ImplicationsHousehold Finance ImplicationsInsurance RiskData Quality
Core finding, identification, data
Core Finding
- The harmonized labor force series, which are free from historical discontinuities, reveal a notably larger labor force shortfall in the post-pandemic period (1.5 million larger than published data indicated).
- The harmonized Labor Force Participation Rate (LFPR) was higher pre-pandemic and recovered more smoothly than official estimates, and the methodology can incorporate alternative population estimates (e.g., immigration-adjusted) before official revisions are published.
Identification Strategy
- The methodological innovation involves constructing 'harmonized' population estimates by combining and smoothing Census Bureau's postcensal and intercensal estimates across five decades, accounting for decennial census breaks and changes in race classification.
- These harmonized estimates then serve as targets for reweighting CPS microdata, replicating the BLS's national coverage and second-stage weighting procedures.
- The approach also incorporates BLS's composite estimation and seasonal adjustment methods, and can integrate alternative population projections like those from the CBO.
Data
The paper primarily uses Current Population Survey (CPS) public-use microdata files (PUMF), various vintages of the Census Bureau's Population Estimates Program (PEP) data, and administrative data on humanitarian migration from the Department of Homeland Security (DHS) and Department of State's Refugee Processing Center. Historical Employment and Earnings reports from the BLS are used for pre-1976 adjustments.
Byron Lutz, David D. Ratner, Louise Sheiner — Conference on Research in Income and Wealth
This paper examines the relative compensation of state and local government workers compared to their private sector counterparts, focusing on the erosion of public sector compensation premiums over the past fifteen years.
Finance Application
- The erosion of competitive compensation for public sector workers, especially skilled ones, could signal declining administrative capacity and fiscal health for state and local governments, impacting the credit quality and pricing of municipal bonds.
- The paper's detailed accrual-based valuation of public pensions and OPEB, particularly its critique of discount rate assumptions, offers a robust framework for municipal bond analysts and asset managers to assess true liabilities and funding risks of public sector entities.
- Furthermore, the findings on changing job stability and compensation trends for public sector employees could inform household finance research on labor income risk, consumption smoothing, and retirement savings decisions for a significant segment of the workforce, potentially influencing demand for various insurance products.
public financemunicipal bondspensionsOPEBcompensationlabor economicsdiscount ratesfiscal stresshousehold financecredit riskasset valuationgovernment employmentactuarial science
Core finding, identification, data
Core Finding
- The paper documents a significant erosion in the public sector compensation premium over the past fifteen years, with total compensation for state and local workers declining from approximately 10 percent higher than comparable private sector workers in 2010 to negative 5 percent by 2023.
- This decline is primarily driven by falling public sector wages relative to private sector wages and is concentrated among college-educated employees, with implications for public sector recruitment and retention.
Identification Strategy
- The paper's methodological innovation lies in augmenting standard compensation data with accrual-based valuations for defined benefit (DB) pensions and retiree health care (OPEB) benefits, and an asset-based valuation for job stability.
- It critiques cash accounting methods and BEA's NIPA methodology for pensions, proposing its own market-based discount rate adjustments for long-term liabilities to better reflect true economic costs.
Data
The paper uses the Employment Cost Index (ECI), Current Population Survey (CPS) microdata, Employer Costs for Employee Compensation (ECEC), National Income and Product Accounts (NIPA), Public Plans Data (PPD), and PEW data on OPEB plans.
W. Erwin Diewert, Chihiro Shimizu, Naohito Abe — Conference on Research in Income and Wealth
This paper uses monthly scanner data on rice purchases in six Japanese Prefectures to construct and compare various multilateral price indexes, addressing chain drift and measuring welfare effects of differing product availability across regions.
Finance Application
- The paper's findings on biased national price indexes due to inter-regional variations and product availability have direct implications for finance.
- In household finance, this suggests that households in less populated areas may face a 'financial choice premium' due to limited access to diverse investment products, insurance, or credit, leading to suboptimal financial outcomes and lower real wealth.
- For asset pricing and real estate, inaccurate national inflation measures could distort real asset returns and valuations, necessitating regionally adjusted price indexes for more precise real estate valuation or location-specific investment strategies.
- Furthermore, the methodology for quantifying welfare effects of product availability could be applied to insurance markets to assess how limited product offerings in certain regions impact risk exposure and welfare.
price indexesscanner datainflation measurementregional economicswelfare effectsproduct availabilityeconometricshousehold financeasset pricinginsurance markets
Core finding, identification, data
Core Finding
- The study finds substantial inter-regional variations in price levels, with lower-population prefectures experiencing significantly higher constant-quality rice prices and welfare costs due to limited product availability compared to larger prefectures.
- These findings suggest that national price indexes, if not adjusted for regional differences and product churn, may be biased, leading to inaccurate measures of real consumption and poverty.
- The paper also demonstrates that the choice of price index methodology significantly impacts results, with KBF preferences providing a superior fit to the data compared to linear or CES models.
Identification Strategy
- The paper's methodological innovation lies in its comprehensive comparison of various multilateral price index construction methods (GEKS, Geary-Khamis, Weighted Time Product Dummy Hedonic, and those derived from econometric estimation of linear, CES, and KBF utility functions) using real-world scanner data.
- A key aspect is the adaptation of Feenstra's (1994) method to quantify the welfare effects of differing product availability across regions by estimating the elasticity of substitution from demand systems.
Data
The paper uses monthly scanner data on purchases of 80 top-selling rice products in six Japanese Prefectures (Hokkaido, Tokyo, Kyoto, Tottori, Kochi, and Kagoshima) over 24 months in 2021 and 2022. This data is derived from the weekly retail sales database, 'SRI+® (Nationwide Retail Store Panel Survey)' by INTAGE Inc., covering 804 supermarkets.
Peter J. Klenow, Ernesto Pastén, Cian Ruane — Conference on Research in Income and Wealth
This paper uses firm-level data from Chile and a general equilibrium model to show that a unilateral carbon tax can increase allocative efficiency, consumption, and welfare by reallocating inputs from low-productivity, fossil-fuel-intensive firms to high-productivity, less-intensive firms.
Finance Application
- This research offers several avenues for finance.
- First, it suggests that firms with higher fossil fuel intensity might be systematically undervalued if markets do not fully price in the risk of future carbon taxes or the allocative inefficiencies they represent.
- This could lead to an "environmental misallocation factor" in asset pricing.
- Second, the paper's finding that carbon taxes can increase overall welfare and consumption could inform sovereign bond pricing for countries implementing such policies, as it implies improved economic fundamentals.
- Third, for household finance, the welfare implications of carbon taxes, particularly their impact on consumption and real wages, could influence household savings behavior and demand for green investment products.
carbon taxmisallocationfirm heterogeneityrevenue productivityESG investingasset pricingclimate risksustainable financecorporate financewelfare economicsChile
Core finding, identification, data
Core Finding
- The core finding is that fossil fuel use is negatively correlated with revenue productivity across Chilean firms, suggesting that higher quality, higher-markup firms are less fossil-fuel intensive.
- Consequently, a unilateral carbon tax can improve allocative efficiency, consumption, and welfare by shifting resources towards these more productive firms, with consumption peaking at a 20% carbon tax.
Identification Strategy
- The paper's identification strategy involves building a static general equilibrium model with heterogeneous firms, where firm-specific production technologies and revenue distortions (representing markups and other misallocations) are backed out to fit observed firm-level data on input choices and revenue.
- Counterfactual carbon tax scenarios are then simulated within this calibrated structural model to assess their impact on output, allocative efficiency, and consumption.
Data
The paper utilizes merged confidential administrative firm-level data from Chile spanning 2015 to 2019, including VAT invoices for firm-to-firm transactions, customs data for firm imports, and annual tax reports for employment, physical capital, sales, and intermediates. Fossil fuel products are identified using CPC v.2 classification codes.
Amanda Michaud, Kathrin D. Ellieroth — Conference on Research in Income and Wealth
This paper develops a framework to estimate labor supply elasticities by disaggregating employment separations into quits and layoffs, and by destination into unemployment and non-participation, highlighting the crucial role of marginal workers.
Finance Application
- This research provides a granular understanding of labor supply decisions, particularly for marginal workers, which is highly relevant for household finance.
- Models of household consumption, savings, and debt accumulation could incorporate the reason for job separation (quit vs. layoff) and destination (unemployment vs. non-participation) to predict financial distress more accurately.
- For instance, understanding how households respond to selective layoffs versus non-selective layoffs could refine predictions of mortgage defaults or credit card delinquencies.
- Insurers could use these insights to better price unemployment insurance or disability products, recognizing that the 'reason' for job loss impacts re-entry into the labor force and thus the duration of claims.
Labor EconomicsHousehold FinanceMacroeconomicsUnemploymentLabor SupplyBusiness CyclesIncome RiskConsumptionSavingsInsurance
Core finding, identification, data
Core Finding
- The paper demonstrates that disaggregating employment separations by reason (quit vs. layoff) and destination (unemployment vs. non-participation) reveals a critical segment of "marginal workers" whose labor supply decisions are highly elastic and significantly amplify business cycle fluctuations in unemployment.
- Specifically, 25-30% of layoffs selectively target these workers, and their procyclical exit from the labor force during recessions exacerbates unemployment volatility.
Identification Strategy
- The study's methodological innovation involves constructing novel labor market flow data from monthly CPS, distinguishing between quits and layoffs into both unemployment and non-participation.
- This granular data, combined with a structural heterogeneous agent model, allows the authors to identify and quantify the behavior of marginal workers and their labor supply elasticities, particularly how they respond to various economic conditions and policy changes.
Data
The paper primarily uses monthly data from the Current Population Survey (CPS) basic monthly files (BMS) from January 1978 to December 2024, and validates findings using the Panel Study of Income Dynamics (PSID) and the Survey of Income Programs and Participation (SIPP).
Bianca He, Lauren I. Mostrom, Amir Sufi — Conference on Research in Income and Wealth
This paper introduces novel measures of customer capital investment using diverse data sources, demonstrates its significant contribution to firm value, and identifies key industry-level determinants.
Finance Application
- This research offers a rich foundation for finance.
- In asset pricing, customer capital could be a new factor explaining cross-sectional stock returns, particularly for service or platform-heavy firms, potentially revealing new risk premia.
- For corporate finance, the detailed measurement and valuation of customer capital can refine M&A due diligence, capital budgeting for intangible investments, and inform optimal capital structure decisions for firms heavily reliant on customer relationships.
- In household finance, understanding how marketing strategies build customer capital could shed light on consumer loyalty and switching costs in financial product markets, impacting pricing strategies for banks and insurers.
intangible capitalcustomer capitalsales and marketingfirm valuationasset pricingcorporate financemachine learningtextual analysisfirm growthM&Acustomer loyaltyplatform business modelsTobin's Q
Core finding, identification, data
Core Finding
- The study finds that investment in customer capital, measured through sales and marketing expenses, sales and marketing salaries, and textual analysis of 10-K filings, is a quantitatively large and growing component of intangible capital.
- Industries with higher customer capital investment exhibit higher Tobin's Q, and sales and marketing expenses strongly predict the value of customer-related intangible assets in acquisitions.
Identification Strategy
- The methodological innovation lies in its comprehensive measurement of customer capital by integrating three distinct data sources: detailed sales and marketing expenses from Capital IQ, sales and marketing employee salary data from Revelio Labs, and quantitative insights extracted from 10-K SEC filings using large language models (Google's Gemini 1.5 Flash) to classify strategies and investment intensity.
- This multi-faceted approach provides a more robust and granular understanding of customer capital.
Data
The paper utilizes financial statement data from Compustat and Capital IQ (1997-2022), employee salary and job category data from Revelio Labs, textual information from annual 10-K SEC filings processed by Google's Gemini 1.5 Flash, and Purchase Price Allocation (PPA) data from the BVR DealStats database for intangible asset valuations in acquisitions.
Laura Alfaro, Paola Conconi, Fariha Kamal, Zachary Kroff — Conference on Research in Income and Wealth
This paper leverages newly linked administrative data to demonstrate that input-output linkages strongly predict intrafirm trade within U.S. multinational enterprises, especially within regional supply chains, correcting for significant measurement error in prior survey-based studies.
Finance Application
- The paper's findings on the prevalence and regional concentration of intrafirm trade, and its sensitivity to IO linkages, offer a novel lens for asset pricing and corporate finance research.
- Researchers could construct firm-level measures of supply chain integration and regional exposure, then examine how these characteristics influence stock returns, firm valuation, and sensitivity to regional trade policy shocks (e.g., tariffs, trade agreements).
- This could explain cross-sectional differences in risk premia.
- For risk management and insurance, the identified vulnerabilities of MNEs to tariffs impacting regional supply chains suggest an avenue to study corporate hedging strategies against trade policy uncertainty, or the demand for specialized supply chain disruption insurance products.
Multinational EnterprisesSupply ChainIntrafirm TradeMeasurement ErrorInput-Output LinkagesTrade PolicyTariffsCorporate StructureFirm BoundariesAsset PricingRisk ManagementInternational Finance
Core finding, identification, data
Core Finding
- Contrary to previous survey-based studies, this paper finds that input-output (IO) linkages are a strong predictor of intrafirm trade.
- A 10 percentage point increase in the IO linkage increases the probability of intrafirm trade by 29% for imports and 21% for exports.
- Furthermore, intrafirm trade is highly prevalent, with over half of foreign affiliates trading with their U.S. parents, and this share rises to three-quarters for affiliates located in North America, underscoring the regional nature of MNE supply chains.
Identification Strategy
- The paper's identification strategy relies on correcting non-classical measurement error in the binary dependent variable (intrafirm trade indicator) by replacing survey data with newly linked administrative customs records from the U.S.
- Census Bureau and MNE production network data from the U.S.
- Bureau of Economic Analysis.
- This allows them to correct for 'false 0s' (unrecorded transactions), 'false 1s' (misallocated trade to primary industries), and missing data due to survey reporting thresholds for smaller affiliates and secondary industries, leading to unbiased and more precise estimates of the IO linkage coefficient.
Data
The study uses the U.S. Census Bureau's Longitudinal Firm Trade Transactions Database (LFTTD) for intrafirm trade, U.S. Bureau of Economic Analysis (BEA) data on MNE production networks (e.g., BE-10 Benchmark Surveys), and BEA's Input-Output accounts for industry linkages, all linked via confidential crosswalks.
Josh Lerner, Namrata Narain, Dimitris Papanikolaou, Amit Seru, Zunda Winston Xu — Conference on Research in Income and Wealth
This paper describes the construction of a new, comprehensive dataset of primary Chinese invention patents, integrating data from multiple sources and addressing unique institutional features of the Chinese intellectual property system.
Finance Application
- This dataset offers a rich opportunity for asset pricing research to study the link between innovation, firm value, and stock returns in China.
- Researchers could examine how patent quality (citations, KPST scores) and quantity predict future stock performance of Chinese listed firms, differentiating between SOEs, private firms, and venture-backed companies.
- The data on PLA-affiliated patents could be used to analyze market reactions to geopolitical events or defense spending announcements for related firms, or to assess the 'national security premium/discount' on their valuations.
- For insurance, the detailed patent data could be used to model innovation-related risks (e.g., IP infringement, technological obsolescence) for Chinese firms, potentially leading to new insurance products or improved underwriting.
Chinese patentsinnovationventure capitalstate-owned enterprisesPLA-affiliated firmspatent qualitytextual analysisfirm-level dataasset pricingcorporate financegeopolitical riskemerging marketsintellectual property
Core finding, identification, data
Core Finding
- The paper's core contribution is the creation and detailed description of a cleaned and disambiguated dataset of Chinese invention patents from 1985 to 2023.
- It reveals the rapid growth of patenting and venture capital in China, the dominance of Chinese assignees, the prevalence of A-type patent publications, and the significant roles of corporate assignees, State-Owned Enterprises (SOEs), and People's Liberation Army (PLA)-affiliated entities in Chinese innovation, alongside measures of patent quality (citations and KPST scores).
Identification Strategy
- The methodological innovation lies in its multi-source data integration and rigorous cleaning process.
- This involves combining patent data from CNIPA, EPO, and Google Patents, de-duplicating records across these sources (handling multiple publications, typos), disambiguating assignee and inventor names using machine learning (XGBoost for sector/country imputation, Levenshtein distance for name matching), and identifying venture-backed, SOE, and PLA-affiliated entities.
- It also constructs patent quality measures like forward citations and Kelly et al. (2021) KPST scores.
Data
The paper primarily uses patent data from the China National Intellectual Property Administration (CNIPA), European Patent Office (EPO) PATSTAT Global dataset, and Google Patents. It also integrates firm-level data from PitchBook (for venture capital firms) and TianYanCha (for Chinese SOEs).
Anmol Bhandari, Paolo Martellini, Ellen McGrattan — Micro Data and Macro Models
This paper develops a theory of firm dynamics and capital reallocation in private firms, using U.S. administrative tax data to analyze the effects of business income, capital, and capital gains taxation on business transfers.
Finance Application
- This paper's insights into tax-induced 'lock-in' effects and capital reallocation frictions in private markets could be directly applied to private equity valuation models, explaining discounts or premiums for illiquid private assets under different tax regimes.
- For household finance, the distinction between transferable and non-transferable entrepreneurial capital, coupled with tax effects on business sales, could inform models of entrepreneurial wealth accumulation, intergenerational wealth transfers, and optimal portfolio choices for business owners.
- The detailed data on private M&A transactions and asset allocation could also be used to build more realistic models of private market liquidity and deal structuring.
Firm DynamicsCapital ReallocationPrivate FirmsBusiness TaxationCapital Gains TaxIncome TaxWealth TaxIntangible AssetsEntrepreneurshipWealth InequalityM&APrivate EquityAsset ValuationLiquidity
Core finding, identification, data
Core Finding
- The study finds that taxing business income is significantly less distortive than taxing capital values or capital gains.
- Capital gains taxes are the most distortive, creating a 'lock-in' effect that discourages trade and leaves capital with less productive owners, leading to a substantial collapse in firm entry and investment.
Identification Strategy
- The model is disciplined by U.S. administrative tax filings of privately-held S corporations, particularly Form 8594 data on business asset acquisitions.
- This unique data provides metrics on the relative size of buyers and sellers and the price-to-size ratio of businesses sold, which are crucial for identifying the output elasticity of intangible capital and the cost of investment in intangibles.
Data
The paper uses U.S. administrative tax filings from the Internal Revenue Service (IRS), including longitudinal panels of privately-held S corporations (business and individual tax forms) and Form 8594 (Asset Acquisition Statement Under Section 1060) for business asset acquisitions. It also references broker data from Pratt's Stats (now DealStats).
Feng Chi — Conference on Research in Income and Wealth
This paper measures the impact of information quality on firms' investment decisions using the U.S. Census as an empirical context, finding that outdated demographic data increases business failure rates.
Finance Application
- This research suggests that the credit risk of small businesses, and thus the pricing of small business loans or regional bank portfolios, could be systematically misestimated if lenders rely on publicly available, but outdated, demographic information.
- For commercial real estate, the valuation of retail properties and the performance of REITs with significant exposure to local retail markets could be affected by these 'information waves,' creating opportunities for investors who can access or infer more timely demographic shifts.
- Furthermore, insurance providers underwriting business interruption or general liability policies for new establishments might misprice risk if their models are based on stale census data, leading to adverse selection or unexpected losses.
Economics of DataFirm InvestmentInformation AsymmetrySmall Business FinanceCredit RiskCommercial Real EstateRetail SectorDemographic DataBusiness FailureLendingUnderwritingAsset Pricing ImplicationsMarket EfficiencyUncertainty
Core finding, identification, data
Core Finding
- Outdated census information significantly increases firm failure rates.
- Specifically, a 10-year gap between decennial censuses raises the failure rate of new establishments by 16 percentage points, representing a 32% increase over the baseline failure rate.
- This effect is stronger in areas with large demographic shifts, for industries relying on precise local information, and for smaller firms.
Identification Strategy
- The study exploits the predetermined, decennial release schedule of the U.S.
- Census data as an exogenous source of variation in information quality.
- It uses an 'excess failure rate' measure, comparing new entrants to existing establishments in the same market and year, and employs a placebo test with randomized census schedules to confirm the effect is not due to chance.
Data
The paper uses establishment-level data from Dun & Bradstreet's National Establishment Time Series (NETS) for Retail Trade and Accommodation/Food Services in New York (1985-2014), combined with demographic data from the U.S. Decennial Census (1980, 1990, 2000, 2010) and the American Community Survey (ACS 2008-2012).
Sung Ah Bahk, John Fitzgerald, Robert A. Moffitt — Conference on Research in Income and Wealth
This paper introduces a new, liquidity-adjusted expenditure-based poverty measure (L-SEPM) that explicitly accounts for the illiquidity of durable goods and in-kind transfers, and re-evaluates the anti-poverty impact of government transfers by considering precautionary saving and insurance effects.
Finance Application
- This paper's emphasis on 'liquidity-adjusted resources' and the illiquidity of assets like housing and vehicles offers a crucial lens for household finance research.
- Researchers could use this framework to model how households manage financial fragility, emergency savings, and debt (e.g., credit card use, payday loans) when a significant portion of their wealth is illiquid.
- The concept of transfers as 'consumption insurance' directly applies to understanding demand for private insurance products (health, unemployment) and how government safety nets influence household risk-taking and portfolio allocation, particularly among low-income and financially vulnerable populations.
Poverty MeasurementHousehold FinanceLiquidity ConstraintsIlliquid AssetsConsumption SmoothingGovernment TransfersPrecautionary SavingDurable GoodsIn-kind TransfersConsumer Expenditure
Core finding, identification, data
Core Finding
- The L-SEPM yields an 8.3% poverty rate in 2022, which is 1.2 percentage points higher than if housing illiquidity were ignored.
- When accounting for precautionary saving and insurance effects, the anti-poverty impact of transfers is estimated to be about 28% smaller than conventional calculations, suggesting that traditional methods may overstate their effectiveness.
Identification Strategy
- The paper's methodological innovation is the L-SEPM, which redefines both resources and poverty thresholds.
- It treats durable goods (like housing and vehicles) and in-kind transfers as illiquid, adjusting the poverty threshold by their service flow (capped at the minimum bundle amount) rather than adding their value to liquid resources.
- It also models the counterfactual impact of removing transfers by considering their role in consumption insurance and effects on precautionary saving, rather than simply subtracting transfer amounts from household resources.
Data
The study primarily uses the 2009-2022 Consumer Expenditure (CE) Survey from the U.S. Bureau of Labor Statistics (BLS). For comparative purposes, it also references data from the Current Population Survey (CPS).
Mark Bils, Bariş Kaymak, Kai-Jie Wu — Micro Data and Macro Models
This paper investigates monopsony power in labor markets through the lens of comparative advantage, finding that employers apply larger wage markdowns to workers with greater comparative advantage, particularly high-wage workers.
Finance Application
- The finding that high-wage workers face larger wage markdowns due to their comparative advantage could be imported into asset pricing by analyzing how this 'labor rent extraction' affects firm valuation and equity risk premiums, especially for firms with a high proportion of skilled labor.
- In household finance, this insight could inform models of human capital valuation and labor income risk for high-income individuals, suggesting that their income is more susceptible to firm-specific monopsony power than previously modeled.
- This could also influence optimal portfolio allocation and demand for income protection insurance for skilled workers.
MonopsonyLabor MarketsComparative AdvantageWage MarkdownsHuman CapitalFirm ValuationLabor RiskHousehold FinanceIncome InequalityBrazil
Core finding, identification, data
Core Finding
- The main empirical finding is that employment growth leads to larger wage changes for an employer's high-wage workers, indicating a less elastic labor supply for them.
- Estimates show markdowns of approximately 25 percent for workers in the top wage quartile, while minimal monopsony power is observed for low-wage workers.
- This suggests that monopsony power is strongest over more productive workers and promotes wage equality within firms.
Identification Strategy
- The paper identifies monopsony power by estimating labor supply elasticities from the relative responses of employment and wages to shifts in labor demand.
- It uses instrumental variables, including employer's life-cycle (age and log employment) and national industry employment growth, to ensure that employment changes are orthogonal to labor supply shifts.
Data
The study uses Brazilian administrative data from the Relação Anual de Informações Sociais (RAIS) for the years 2006-2018. This dataset tracks establishment-worker matches, providing detailed information on employment, earnings, hours, occupations, and worker characteristics.
Jonathan Payne, Adam Rebei, Yucheng Yang — Micro Data and Macro Models
This paper introduces DeepSAM, a novel deep learning method to globally solve and estimate search and matching models with aggregate shocks and heterogeneous agents, characterizing general equilibrium as a high-dimensional partial differential equation.
Finance Application
- The DeepSAM methodology offers a powerful tool for modeling complex financial markets characterized by search frictions, heterogeneous agents, and aggregate shocks.
- Beyond the paper's OTC bond market application, it could be applied to study the pricing and liquidity of illiquid assets like private equity, private credit, or real estate, where matching between buyers and sellers is a key friction.
- Researchers could use it to analyze how different types of institutional investors (e.g., pension funds, hedge funds, insurance companies) with varying mandates and capital constraints interact in these markets, influencing risk premia and market resilience during crises.
- Furthermore, it could model the impact of new financial regulations (e.g., capital requirements, liquidity rules) on market structure, trading activity, and the transmission of shocks across different asset classes or investor segments.
Deep LearningSearch and MatchingHeterogeneous AgentsAggregate ShocksPartial Differential EquationsFinancial MarketsLabor MarketsBusiness CyclesAsset PricingLiquidityYield CurveEstimationSimulated Method of MomentsOTC Markets
Core finding, identification, data
Core Finding
- The core finding is methodological: DeepSAM can globally solve and estimate complex search and matching models with high-dimensional state spaces (including agent distributions and aggregate shocks) that were previously intractable.
- Applied to labor markets, it shows distribution feedback amplifies shocks and weakens positive assortative matching during prolonged expansions, disproportionately benefiting low-wage workers.
- In an OTC bond market, it reveals how financial crises disproportionately impact long-maturity bonds due to investor type switching and liquidity constraints.
Identification Strategy
- The paper's methodological innovation, DeepSAM, approximates the surplus function of search and matching models using a neural network.
- It trains the network parameters by minimizing the average loss in the master partial differential equation (PDE), which explicitly includes the agent distribution as a state variable.
- This approach allows for global solutions across high-dimensional state and parameter spaces, using techniques like pseudo-state variables for parameters and novel sampling strategies for stability.
Data
The paper uses empirical moments from the US labor market (e.g., employment declines during COVID-19 from Cajner et al., 2020; transition rates from Lise and Robin, 2017, and BLS data) for calibration and estimation. For the OTC bond market, it calibrates to average high-grade corporate yield curves (Payne and Szőke, 2024) and crisis haircut rates (Chen, Cui, He, and Milbradt, 2017).
David Baqaee, Ariel Burstein, Yasutaka Koike-Mori — Micro Data and Macro Models
This paper develops a novel method to measure forward-looking welfare using sufficient statistics derived from observed consumption-savings decisions, bypassing the need for explicit knowledge of future prices and outcomes.
Finance Application
- This methodology could be highly valuable in household finance and asset pricing.
- In household finance, it could be used to construct a more accurate, forward-looking measure of household financial well-being, crucial for understanding savings behavior, retirement planning, and responses to policy changes or economic shocks.
- For asset pricing, the inferred 'future prices' or expectations about future economic states (derived from aggregate consumption-savings behavior) could serve as novel factors or sentiment indicators, potentially explaining cross-sectional asset returns or market volatility.
- For insurance, this framework could help quantify the welfare impact of various risks (e.g., health, unemployment) and the value of insurance products in mitigating these forward-looking welfare losses.
welfare measurementsufficient statisticsconsumption-savingshousehold financeasset pricingdynamic modelsexpectationsnon-homotheticityPSIDmoney metricriskincomplete markets
Core finding, identification, data
Core Finding
- The paper demonstrates that dynamic welfare measures, derived from observed consumption-savings behavior, can differ significantly from static measures, especially for low-wealth households and in response to income shocks.
- For instance, a job loss is associated with a substantial reduction in money-metric welfare (e.g., 20%), though this impact is mitigated for older individuals.
Identification Strategy
- The core methodological innovation is to use changes in observed consumption-wealth ratios to infer intertemporal price indices and changes in expectations about the future.
- This approach relies on the assumption of separability of preferences between the present and the future and is extended to handle non-homothetic preferences by comparing consumption-wealth ratios for individuals on the same indifference curve.
- For non-rentiers, the method infers utility from budget shares by matching them to rentiers.
Data
The paper uses bi-annual household data from the Panel Study of Income Dynamics (PSID) between 2005 and 2019, which includes financial net worth, age, and consumption surveys. It also utilizes CPI prices for seven consumption categories.
Sasha Indarte, Raymond Kluender, Ulrike Malmendier, Michael Stepner — Micro Data and Macro Models
This paper develops a novel micro-level approach to measure consumption distortions (wedges) using subjective expectations and transaction data, finding significant heterogeneity and a mix of over- and under-consumption.
Finance Application
- This research offers crucial new empirical moments for household finance models, particularly those incorporating behavioral biases.
- The micro-level consumption wedges and their correlates (MPCs, financial distress, homeownership, consumption commitments) can serve as direct calibration targets for heterogeneous agent (e.g., HANK) models, improving their ability to predict household responses to fiscal and monetary policies.
- For asset pricing, understanding the heterogeneity in consumption distortions could shed light on individual risk-taking and saving behavior, potentially informing models of household portfolio choice and aggregate asset demand.
- In insurance, the findings on over-consumption and financial distress, especially among non-homeowners, could inspire the design of commitment savings products or tailored financial literacy interventions to mitigate behavioral biases.
ConsumptionHousehold FinanceBehavioral EconomicsFinancial ConstraintsSubjective ExpectationsMPCsMicrodataEuler EquationPresent BiasConsumer InertiaWealth Heterogeneity
Core finding, identification, data
Core Finding
- The median absolute consumption wedge is 40% of frictionless consumption, indicating large distortions in consumer behavior.
- Crucially, 51% of consumers over-consume, challenging financial constraints as the sole dominant friction and suggesting present bias, consumer inertia, or consumption commitments play a significant role.
Identification Strategy
- The paper's identification strategy is a new sufficient statistics approach that measures micro-level consumption wedges by comparing actual consumption to a counterfactual 'frictionless' consumption.
- This approach is innovative because it uses subjective expectations data to relax the assumption of Full-Information Rational Expectations (FIRE), thereby isolating the influence of frictions and behavioral preferences from deviations in beliefs.
Data
The paper uses new data linking subjective economic expectations (inflation, income growth, interest rates) from surveys to administrative bank account transactions data (spending, income, balances) for a population of predominantly middle-income US consumers with low liquid wealth, sourced from EarnIn.
Robert Shimer, Liangjie Wu — The Micro and Macro Perspectives of the Aggregate Labor Market
This paper develops a competitive search model with two-sided private information, showing how platforms use fees and matching rates to separate agents based on their willingness-to-pay for partners, under common ranking and supermodularity.
Finance Application
- This framework could be applied to understanding information asymmetry in over-the-counter (OTC) markets for corporate bonds or private equity, where dealers (platforms) match buyers and sellers with private information about asset quality. 'Fees' could be transaction costs or bid-ask spreads, and 'matching rates' could represent market liquidity or trading frequency.
- The model's insight into how WTP dictates screening mechanisms (fees vs. liquidity) could explain why certain asset classes are characterized by high transaction costs and high liquidity (e.g., high-quality, high-demand assets) versus low transaction costs and low liquidity (e.g., niche, illiquid assets where sellers signal quality by accepting fewer matches).
- In household finance, it could model online lending platforms screening borrowers based on credit risk, using interest rates (fees) and loan approval rates (matching rates) to sort different risk types.
information asymmetrymarket designscreeningcompetitive searchassortative matchingtransaction costsliquidityOTC marketsprivate informationcontract theory
Core finding, identification, data
Core Finding
- In markets with two-sided private information, equilibrium must be separating, meaning each platform attracts exactly one type from each side.
- The mechanism for this separation depends on agents' willingness-to-pay (WTP): if more desirable types have higher WTP, they separate through high fees; if they have lower WTP, they separate through low matching rates.
- With positive advertising costs, both mechanisms operate simultaneously, with their direction determined by WTP monotonicity.
Identification Strategy
- The paper's identification strategy is theoretical, building on an extended competitive search model.
- It reformulates the equilibrium as an optimization problem where platforms maximize profits subject to agents' optimal search and incentive compatibility constraints.
- The key methodological innovation is characterizing equilibrium through a system of ordinary differential equations for continuous types, which endogenously determines market utilities, fees, and matching probabilities, ensuring separation and slack upward incentive constraints under common ranking and supermodularity.
Data
This is a purely theoretical paper and does not use any empirical data. It provides parametric examples (marriage, labor, expertise, communicable disease markets) to illustrate its theoretical findings.
Joao Guerreiro, Jonathon Hazell, Chen Lian, Christina Patterson — Micro Data and Macro Models
This paper argues that workers dislike inflation not just due to falling real wages, but also because they incur significant "conflict costs" (e.g., negotiating, job searching) to maintain their real wages, which are typically underestimated in welfare analyses.
Finance Application
- The concept of "conflict costs" introduces a new dimension to labor income risk in household finance, suggesting that households face non-pecuniary costs to maintain their real income during inflation, which could influence saving rates, demand for inflation-hedging assets, or the pricing of income protection insurance.
- In asset pricing, increased labor market friction and conflict during inflationary periods could reduce corporate profitability and cash flows, potentially explaining a higher equity risk premium or lower valuations for labor-intensive firms.
- Furthermore, the survey methodology for eliciting non-pecuniary costs could be adapted to quantify the "hassle factor" or psychological costs associated with various financial decisions, such as switching banks, managing investments, or engaging in complex financial planning.
Labor EconomicsInflationWelfare CostsMenu CostsWage SettingHousehold FinanceAsset PricingLabor Income RiskSurvey DataMacroeconomicsConflict Theory
Core finding, identification, data
Core Finding
- The paper finds that inflation imposes substantial "conflict costs" on workers, who must take costly actions to prevent their nominal wages from eroding.
- A menu-cost style model, calibrated with survey data, demonstrates that these conflict costs more than double the welfare costs of inflation for workers compared to analyses that only consider real wage declines.
- The true welfare cost is determined by "wage erosion" – the real wage decline that would occur if workers did not engage in conflict.
Identification Strategy
- The paper uses a novel survey of 3000 US workers to directly calibrate key model parameters: the median worker's willingness to sacrifice 1.75% of their wages to avoid conflict, and the low degree of default wage indexation to inflation (0.05 percentage points increase in wage offer for every 1% inflation increase).
- This survey-based calibration is complemented by cross-country panel regressions (1964-2022) showing a robust positive correlation between inflation and labor market strikes, validating the model's premise that conflict rises with inflation.
Data
The paper uses a custom survey of 3000 US workers (2024), cross-country panel data on labor strikes (International Labour Organization) and inflation (World Bank, 1964-2022), ASEC-CPS survey data for real wage growth, and inflation expectations from the Survey of Consumer Expectations and Survey of Professional Forecasters.
Benjamin Griffy, Yinghsuan Chao, David G. Wiczer — The Micro and Macro Perspectives of the Aggregate Labor Market
This paper uses a regression discontinuity design and a quantitative equilibrium model to estimate the causal effect of Unemployment Insurance (UI) eligibility on re-employment earnings and decompose its underlying mechanisms.
Finance Application
- This research offers several insights for finance.
- In household finance, the finding that UI eligibility boosts re-employment earnings by enhancing worker bargaining power suggests UI significantly reduces labor income risk and improves consumption smoothing, potentially influencing household savings, debt accumulation, and demand for private insurance products.
- For asset pricing, UI's role in stabilizing labor income could affect the pricing of assets that hedge human capital risk, as changes in UI policy alter the risk profile of a major component of wealth.
- Furthermore, the RDD methodology could be applied to other policy discontinuities affecting household financial decisions or firm behavior in financial markets.
Unemployment InsuranceRegression Discontinuity DesignLabor EconomicsHousehold FinanceHuman CapitalRisk AversionBargaining PowerJob SearchEarningsIncome VolatilitySocial InsuranceAsset Pricing
Core finding, identification, data
Core Finding
- The paper empirically finds that UI eligibility increases re-employment earnings by about 10% per quarter.
- A quantitative model reveals that the "true" causal effect for UI recipients is 15.32%, with nearly all of this effect attributed to workers securing a larger share of the match surplus (higher piece-rate) rather than improved match productivity.
Identification Strategy
- The study employs a regression discontinuity design (RDD) that exploits a near-universal feature of the U.S.
- UI system: a discontinuous monetary eligibility threshold based on workers' past earnings.
- Workers just above this threshold become eligible for UI, creating a quasi-experimental setting to estimate the causal effect of UI eligibility on re-employment outcomes.
Data
The paper uses administrative panel data from the Census Bureau's Longitudinal Employer-Household Dynamics (LEHD) for workers' earnings and employment histories, state-level UI laws, and data from the Survey of Income and Program Participation (SIPP) and Employment and Training Administration (ETA) reports.
Valerie Smeets, Lin Tian, Sharon Traiberman — The Micro and Macro Perspectives of the Aggregate Labor Market
This paper develops and estimates a dynamic framework where college students make forward-looking field-of-study choices, which, along with skill specificity and occupational switching costs, determine how economies respond to labor market disruptions like trade wars and AI adoption.
Finance Application
- The paper's framework could be imported into asset pricing to model human capital as a dynamic asset, where its value and risk premia are influenced by labor market shocks (trade wars, AI) and educational policies.
- This could lead to industry-specific factor models or human capital risk premia.
- In household finance, the findings on how educational choices and flexibility impact lifetime earnings and inequality are crucial for modeling household portfolio allocation, savings, and debt decisions, especially for student loans.
- For insurance, the paper's insights into income risk from skill specificity and labor market disruptions could inform the design and pricing of novel 'human capital insurance' products that protect against technological displacement or trade shocks, with premiums differentiated by field of study and expected labor market flexibility.
Human CapitalLabor Market ShocksSkill SpecificityOccupational MobilityEducation PolicyRational ExpectationsIncome RiskAsset PricingHousehold FinanceInsuranceAITrade WarDynamic Models
Core finding, identification, data
Core Finding
- The paper finds that forward-looking field-of-study choices and educational policy flexibility significantly influence aggregate and distributional outcomes in response to labor market shocks.
- Increased flexibility in field choices boosts aggregate output and reallocates income gains towards lower-income individuals, despite potentially increasing overall income inequality.
- This highlights the critical role of human capital investment and adaptability in mitigating adverse effects of economic disruptions.
Identification Strategy
- The identification strategy leverages the centralized Danish college admissions system, which allows for separating student preferences from selection effects.
- A fuzzy regression discontinuity (RD) design exploits admission cutoffs to identify the causal effect of field choice on occupational sorting.
- The model estimation uses an Expectation-Maximization (EM) algorithm for latent types and a Hotz-Miller inversion for occupational switching costs, utilizing panel data and wage variation.
Data
The paper uses several Danish administrative datasets: the Integrated Database for Labour Market Research (IDA), the Education Register (UDDA), the Coordination Register (KOT) for college applications, the Firm Accounting Statistics Register (FIRE), and the Danish Foreign Trade Statistics Register (UHDI). It also incorporates the ONET Database for occupational tasks and AI technological shock data from Daniel Rock.
Hassan Afrouzi, Andrés Blanco, Andrés Drenik, Erik Hurst — The Micro and Macro Perspectives of the Aggregate Labor Market
This paper develops a macro-labor model with nominal wage rigidities to analyze how unexpected inflation affects labor market dynamics, real wages, and worker welfare, showing that inflation can create the appearance of a tight labor market despite declining real wages.
Finance Application
- The paper's findings offer several avenues for finance research.
- In asset pricing, the mechanism by which sticky wages transfer resources from workers to firms during inflation (increasing corporate profits) could be used to model inflation's impact on equity valuations and corporate bond spreads, particularly for firms in industries with varying wage rigidities.
- For household finance, the quantified welfare losses from real wage declines and costly job search/renegotiation directly inform how inflation affects household consumption, savings, and debt accumulation, potentially explaining differentiated financial distress across income groups.
- This could also motivate research into demand for inflation-indexed financial products or income protection insurance.
InflationLabor MarketsWage RigidityJob SearchVacanciesCorporate ProfitsHousehold WelfareReal WagesBeveridge CurveMacro-Finance
Core finding, identification, data
Core Finding
- Nominal wage stickiness within a match incentivizes workers to engage in costly job-to-job transitions and renegotiations after an unexpected rise in the price level.
- This dynamic leads to a rise in aggregate vacancies relative to unemployment and a decline in average real wages, matching observed patterns during the 2021-2024 inflation period.
- The model implies that recent inflation significantly reduced worker welfare through real wage declines and costly actions, while increasing firm profits due to the transfer of resources from workers.
Identification Strategy
- The paper's primary identification strategy involves a calibrated macro-labor model, using pre-2020 U.S. data, to simulate the effects of unexpected inflation shocks.
- It then tests the model's predictions against observed aggregate and cross-sectional labor market trends during the 2021-2024 inflation period.
- Additionally, it provides historical empirical support by showing systematic increases in vacancies and upward shifts in the Beveridge curve during prior high-inflation periods (1950-2019).
Data
The paper uses data from the Job Openings and Labor Turnover Survey (JOLTS), Current Population Survey (CPS), Atlanta Fed Wage Tracker Index, ADP's Pay Insights, Barnichon's (2010) unified vacancy series, and FRED/BEA for corporate profits.
Joao Guerreiro, Jonathon Hazell, Chen Lian, Christina Patterson — The Micro and Macro Perspectives of the Aggregate Labor Market
This paper argues that workers must take costly actions ('conflict') to have nominal wages catch up with inflation, leading to welfare costs even if real wages do not fall.
Finance Application
- The "conflict costs" faced by workers translate into higher labor costs or reduced productivity for firms, which could impact corporate profitability and, consequently, equity valuations and stock returns, especially in industries with high labor intensity or unionization.
- These welfare costs directly affect household disposable income and financial well-being, influencing savings rates, debt accumulation, and consumption decisions, particularly for households with lower bargaining power.
- The aggregate impact of these conflict costs on labor market dynamics could propagate through the economy, affecting overall economic growth, inflation expectations, and the effectiveness of monetary policy, which are crucial for understanding bond markets and systemic risk.
InflationLabor MarketsWage DynamicsHousehold WelfareSurvey DataMenu CostsConflict CostsMacro-FinanceBehavioral Economics
Core finding, identification, data
Core Finding
- The paper's core finding is that inflation imposes significant welfare costs on workers beyond just real wage declines, due to "conflict costs" workers incur to secure wage increases.
- It quantifies these costs, showing they more than double the welfare costs of inflation compared to considering real wage falls alone, as workers must expend effort to maintain real wages.
Identification Strategy
- The study employs a menu-cost style model of wage setting, calibrated using novel survey data from 3000 US workers.
- This survey directly elicits workers' willingness to sacrifice wages to avoid conflict and their employers' default wage indexation, providing key parameters for the model.
- Observational data on labor strikes and inflation across countries further supports the link between inflation and conflict.
Data
The paper primarily uses novel survey data from 3000 US workers fielded in early 2024. It also incorporates cross-country panel data on labor market strikes from the International Labour Organization and inflation data from the World Bank (1964-2022), and ASEC-CPS data for real wage growth.
Naoki Aizawa, Hanming Fang, Katsuhiro Komatsu — The Micro and Macro Perspectives of the Aggregate Labor Market
This paper develops and estimates an equilibrium labor search model to analyze how labor unions affect employer-provided insurance and job security, and how social insurance policies interact with unionization and labor market outcomes in the U.S.
Finance Application
- This paper offers several insights for finance research.
- First, the findings that unions affect firm-provided benefits, wages, and job security imply that unionization status and the regulatory environment of social insurance are material to firm value and risk.
- Asset pricing models could incorporate union density, social insurance generosity, and their interactions as factors influencing firm-specific or industry-level risk premia, particularly for labor-intensive firms.
- Second, changes in social insurance directly impact households' labor income risk and their demand for private insurance, influencing household savings rates and portfolio allocations.
- Third, the equilibrium labor search model highlights how labor market frictions, influenced by unions and social insurance, affect wages, employment, and firm profitability, which could be linked to aggregate asset pricing phenomena.
labor economicsunionssocial insurancehealth insuranceunemployment insurancefirm-provided benefitswage inequalitylabor market frictionsasset pricinghousehold financefirm valuationrisk premiainsurance demandmacro-finance
Core finding, identification, data
Core Finding
- The paper finds that social insurance expansions (e.g., Medicare, Medicaid, UI generosity) significantly reduce unionization rates by diminishing the perceived value of unions and employer-provided benefits.
- While these expansions can mitigate welfare losses from union decline due to technological changes, they also affect wage inequality, with universal social insurance increasing inequality and targeted social insurance reducing it.
Identification Strategy
- The empirical analysis identifies the causal effects of social insurance expansions on unionization by exploiting plausibly exogenous variations across time and space.
- This includes the staggered introduction of Medicare and Medicaid in the 1960s (using geographic variation in pre-reform coverage and event studies), state-level variations in ACA Medicaid expansion timing (difference-in-differences), and changes in state unemployment insurance generosity.
Data
The paper uses several micro-level datasets: the Current Population Survey (CPS), the Health and Retirement Study (HRS), the Survey of Income and Program Participation (SIPP), and establishment-level data from the Robert Wood Johnson Foundation Employer Health Insurance Survey. It also uses data on state-level union density, NLRB election data, and medical expenditure data (MEPS).
Steffen Altmann, Robert Mahlstedt, Malte J. Rattenborg, Alexander Sebald, Sonja Settele, Johannes Wohlfart — The Micro and Macro Perspectives of the Aggregate Labor Market
This paper conducts a field experiment to study how subjective wage expectations influence unemployed workers' job search behavior and re-employment prospects, using a tailored information intervention in Denmark.
Finance Application
- The findings on miscalibrated wage expectations and spatial search frictions are highly relevant for household finance and local asset pricing.
- Misguided wage expectations could lead to suboptimal household savings, debt accumulation, and investment risk-taking, suggesting that financial literacy interventions incorporating labor market realities could improve household financial health.
- Furthermore, systematic underestimation of local labor market efficiency could contribute to mispricing in local housing markets or regional equity portfolios, as investors might not fully account for the true economic potential or frictions in specific geographic areas.
- This opens avenues for exploring belief-driven factors in asset pricing models.
Household FinanceBehavioral EconomicsExpectationsLabor MarketsJob SearchInformation FrictionsSpatial EconomicsHousing MarketsPrecautionary SavingsUnemployment Risk
Core finding, identification, data
Core Finding
- Unemployed job seekers anchor their wage expectations to pre-unemployment wages, often misperceiving objective re-employment potential.
- A tailored information intervention updates these expectations, leading initially over-optimistic individuals to lower reservation wages and increase search effort (resulting in faster re-employment at lower wages), and initially pessimistic individuals to raise reservation wages and narrow geographic search scope (resulting in faster re-employment at higher wages, by redirecting search to local markets where spatial frictions were underestimated).
Identification Strategy
- The study employs a large-scale field experiment in Denmark where unemployed workers are randomly assigned to a treatment or control group.
- The treatment group receives an exogenous information shock: objective average re-employment wages of comparable workers, derived from administrative data.
- This intervention causally identifies how updated beliefs affect job search behavior and labor market outcomes.
Data
The paper uses a large-scale survey (over 9,000 job seekers) to elicit prior and posterior wage expectations, reservation wages, and planned search effort. This is linked to administrative records from Denmark's central job search platform (jobnet.dk) for click-by-click search behavior and Statistics Denmark registers for employment status, realized wages, and working hours.
Pauline Carry, Benjamin Schoefer — The Micro and Macro Perspectives of the Aggregate Labor Market
This paper investigates whether employers and workers cooperate or engage in conflict during job dismissals in France, focusing on the underutilization of a mutually beneficial separation mode.
Finance Application
- The paper's findings on quantifiable labor market conflict and its impact on separation costs offer rich avenues for finance research.
- In asset pricing, unobservable labor conflict (proxied by industry- or firm-level SMA conversion rates or survey-based conflict indicators) could be a novel factor explaining cross-sectional differences in firm valuations, cost of capital, or stock returns, as higher conflict implies greater operational risk and lower efficiency.
- For household finance, the *nature* of job separation (conflictual dismissal vs. mutual agreement) could differentially predict household financial distress, consumption smoothing, and credit default risk, with conflictual separations leading to worse outcomes due to stigma and longer unemployment spells.
- In insurance, the identified 'deliberate cost-seeking' and 'asymmetric beliefs' could inform the design of new labor dispute insurance products for firms or job loss transition insurance for individuals, where premiums could be tailored based on firm- or industry-specific labor conflict metrics.
labor economicsemployment protectionjob separationsfirm costshuman resourcesconflictbargainingfirm valuehousehold financeunemployment insurancerisk managementcorporate governanceasset pricing
Core finding, identification, data
Core Finding
- Only 12% of potential dismissals in France are resolved through 'separation by mutual agreement' (SMA), a cheaper and more flexible alternative to traditional dismissals.
- A survey of HR directors reveals three main drivers of this low conversion rate, indicating conflict: (i) hostility between employer and employee, (ii) employers using dismissals as a 'discipline device' for other employees, and (iii) asymmetric beliefs about labor court outcomes.
- Removing these factors could increase SMA adoption to 67%.
Identification Strategy
- The paper employs three strategies to estimate SMA conversion rates: time series extrapolation of dismissal declines, a difference-in-differences design across labor market cells with varying SMA take-up, and worker surveys eliciting counterfactual separation outcomes.
- To identify conflict drivers, they use a novel survey of HR directors on specific past dismissals, asking about the role of hostility, discipline, and asymmetric beliefs, and use policy discontinuities for early retirement as a quasi-experimental setting for less conflictual separations.
Data
The paper uses micro data on individual worker flow events (MMO), establishment-level employment records (DADS), Labor Force Survey (LFS) data, administrative SMA records, a worker survey of SMA participants, labor court case records, a worker-manager workplace survey (Reponse), and their own novel survey of HR directors.
Andreas I. Mueller, Damian Osterwalder, Josef Zweimüller — The Micro and Macro Perspectives of the Aggregate Labor Market
This paper uses large-scale high-frequency data on vacancy flows and matched employer-employee data from Austria to document the cyclicality of vacancy flows and their contribution to variation in the vacancy stock, highlighting the crucial role of replacement hires and on-the-job search.
Finance Application
- The micro-level, high-frequency data on vacancy flows and replacement hires could be a powerful leading indicator for firm-specific growth prospects or labor market risk in asset pricing.
- Firms with high replacement hiring rates might signal talent retention issues or wage pressures, impacting future profitability and stock returns.
- For household finance, the cyclicality of on-the-job search and replacement hires directly affects household income uncertainty and job mobility, informing models of consumption, savings, and debt.
- In insurance, understanding the acyclical nature of vacancy lapses and the drivers of replacement hires could refine unemployment insurance models and pricing for employment-related income protection products.
Labor MarketVacanciesBusiness CyclesFirm DynamicsJob SearchReplacement HiresOn-the-Job SearchMacroeconomicsMicrodataAsset PricingHousehold FinanceInsurance
Core finding, identification, data
Core Finding
- The paper documents four key facts: (1) Vacancy inflows explain at least one-third of the cyclical variation in the vacancy stock, with the remainder explained by vacancy fillings. (2) Vacancy lapses are acyclical and do not contribute to stock variation. (3) Replacement vacancies are a key driver of cyclical vacancy inflows. (4) The composition of vacancy inflows varies little over the business cycle and cannot account for cyclical variation in vacancy filling.
- A calibrated search-and-matching model confirms the crucial role of on-the-job search and replacement hires, and the cyclical reposting probability, in explaining vacancy rate fluctuations.
Identification Strategy
- The study employs a decomposition analysis of time series variation in vacancy stocks, similar to Shimer (2012) for unemployment.
- A novel methodology is developed to identify replacement hires using precise daily worker flows from matched employer-employee data, where hires closely preceded by separations of the same worker type in the same firm are classified as replacement hires.
- This allows for a detailed breakdown of vacancy inflows into replacement and new positions.
Data
The paper primarily uses large-scale daily data on vacancy flows from the Austrian 'Arbeitsmarktservice' (AMS) vacancy register database (1987-2019) and matched employer-employee data from the Austrian Social Security Database (ASSD, since 1994). It also compares findings with US JOLTS data and the Austrian 'Offene-Stellen-Erhebung' (OStE) survey.
Benny Kleinman, Ernest Liu, Stephen J. Redding, Xiang Zhang — The Micro and Macro Perspectives of the Aggregate Labor Market
This paper develops a quantitative general equilibrium model with gradual labor mobility and occupation-specific capital accumulation to analyze the dynamics of employment and wages across occupations, particularly in response to technological shocks like automation and AI.
Finance Application
- This framework offers significant potential for finance research.
- In asset pricing, the predicted shifts in labor shares, wage inequality, and capital accumulation due to AI could inform models of future corporate profitability, labor income risk, and thus sector-specific equity returns or risk premiums.
- For household finance, the model's insights into occupation-specific wage dynamics and mobility could be used to analyze household consumption, saving, and debt decisions, particularly for households in occupations highly exposed to AI.
- Insurance researchers could leverage the model's projections of occupation-specific employment and wage dynamics to price unemployment, disability, or human capital-backed insurance products, assessing long-term risk profiles across different occupational groups.
General EquilibriumLabor MarketsOccupational DynamicsTechnological ChangeAutomationArtificial Intelligence (AI)Wage InequalityEmployment PolarizationCapital AccumulationLabor MobilityAsset PricingHousehold FinanceInsuranceLong-term Economic Trends
Core finding, identification, data
Core Finding
- The paper finds that U.S. employment and wage polarization from 1980-2018 was significantly driven by pre-1980 initial conditions, slow adjustment, and productivity, investment, and labor-supply shocks, with automation playing a limited role.
- The model predicts that future AI shocks will initially increase wage inequality and decrease the labor share, but these effects may be transitory, with the economy eventually reaching a new steady state characterized by reallocation to high-skill jobs and no polarization.
Identification Strategy
- The paper's methodological innovation is to invert a dynamic general equilibrium model to infer the underlying paths of fundamental shocks (automation, productivity, investment efficiency, and mobility frictions) that match observed empirical paths of occupation-level employment, wages, capital, and labor shares.
- Key parameters, such as the elasticity of substitution between labor and capital (zeta) and across occupations (sigma), are estimated using model-generated instruments derived from counterfactual scenarios (e.g., without automation or productivity shocks) and moment conditions relating labor shares to wage-rental ratios and procedure shares to prices.
Data
The paper uses U.S. occupation-level data from 1980-2023, drawing from the Decennial Census, American Community Survey (ACS) for employment and wages, the Current Population Survey Annual Social and Economic Supplement (CPS ASEC) for occupational mobility, and BEA industry data for capital and labor shares. It also incorporates external measures of AI exposure from Anthropic (2023).
Edoardo Maria Acabbi, Andrea Alati, Luca Mazzone, Marta Morazzoni — The Micro and Macro Perspectives of the Aggregate Labor Market
This paper develops and empirically tests a model of entrepreneurial team formation, demonstrating that teams with similar overall talent but diverse specialized skills achieve superior firm performance.
Finance Application
- This research offers valuable insights for venture capital and private equity by providing a framework to evaluate startup teams beyond traditional metrics.
- VCs could assess the combination of founders' talent similarity and skill complementarity to predict startup success, inform investment decisions, and potentially improve portfolio returns.
- For public markets, the founding team's human capital composition could be a novel factor in IPO underwriting and valuation, influencing initial pricing and long-term stock performance.
- In household finance, understanding these team dynamics can shed light on entrepreneurial household wealth accumulation, risk-taking, and access to business financing, informing financial planning for founders.
entrepreneurial teamshuman capitalskill complementaritytalentfirm performanceproductivityfirm growthadministrative dataventure capitalprivate equityIPOshousehold financebusiness insurance
Core finding, identification, data
Core Finding
- Entrepreneurial teams exhibit higher sales, productivity, and survival rates compared to single entrepreneurs.
- This superior performance is driven by a dual sorting mechanism: positive sorting on overall talent (team members have similarly high latent talent) and negative sorting on skill specialization (team members possess diverse, complementary skills).
- While matching frictions make it less likely for individuals with highly dissimilar skills to form teams, those teams that do form and exhibit skill heterogeneity achieve better outcomes.
Identification Strategy
- The paper identifies individual talent using AKM fixed effects derived from pre-entrepreneurial wage histories and specialized skills through ESCO occupational classifications.
- It then employs linear probability models and dyadic regressions on rich administrative data to analyze the determinants of team formation and the impact of team composition (average talent, talent dissimilarity, skill dissimilarity) on firm performance, leveraging the longitudinal nature of the data to track careers and firm outcomes.
Data
The study utilizes comprehensive administrative employer-employee records from Portugal (Quadros de Pessoal) spanning 1991-2019, which are linked to private companies' balance sheet data (Sistema de Contas Integradas das Empresas - SCIE) from 2004-2018. Individual skill measures are constructed using the European Skills, Competences, and Occupations (ESCO) database.
Peter J. Klenow, Ernesto Pastén, Cian Ruane — Macroeconomics and Productivity
This paper investigates how unilateral carbon taxes in Chile can improve allocative efficiency, consumption, and welfare by reallocating inputs away from misallocated, fossil-fuel-intensive firms.
Finance Application
- This research has significant implications for asset pricing and corporate finance.
- In asset pricing, the finding that carbon taxes can improve allocative efficiency and firm productivity suggests that equity markets might price in a 'carbon efficiency premium,' where firms with lower fossil fuel intensity and higher revenue productivity (or those that benefit from resource reallocation) could see higher valuations or lower costs of capital.
- For corporate finance, this implies that firms with high fossil fuel intensity and existing misallocation face not only direct carbon costs but also an indirect penalty through reduced allocative efficiency, driving strategic investment decisions towards cleaner production and potentially impacting their access to capital or M&A activity.
- This could also inform ESG investing by highlighting allocative efficiency as a key metric for assessing a company's resilience to climate policy.
carbon taxmisallocationfirm heterogeneityallocative efficiencyChilegeneral equilibrium modelfirm productivityenvironmental policyasset pricingcorporate financeESG investing
Core finding, identification, data
Core Finding
- The core finding is that fossil fuel use is negatively correlated with revenue productivity across Chilean firms, suggesting that higher-quality, higher-markup firms are less fossil-fuel intensive.
- Consequently, a unilateral carbon tax can helpfully reallocate inputs from low-markup, low-quality, fossil-fuel-intensive firms to high-markup, high-quality, less fossil-fuel-intensive firms, leading to an increase in allocative efficiency, consumption, and welfare in Chile, with consumption peaking at around a 20% tax.
Identification Strategy
- The paper uses a static general equilibrium model calibrated to firm-level data from Chile.
- It backs out firm-specific production technologies and revenue distortions to match observed firm data on input choices and revenue.
- Counterfactual carbon taxes are then imposed on this calibrated model to simulate their effects on output, allocative efficiency, and consumption, leveraging empirical correlations between fossil fuel intensity and revenue productivity.
Data
The paper uses merged confidential administrative firm-level data from Chile (2015-2019), including revenue, inputs, prices, quantities, fossil fuel use (from VAT invoices and customs data), employment, physical capital, sales, and intermediates from annual tax reports.
Gert Bijnens, Simon Jäger, Benjamin Schoefer — Macroeconomics and Productivity
This paper investigates the determinants and causal effects of strategy and management consulting on client firms' productivity, labor outcomes, and organizational structure in an advanced economy using comprehensive business-to-business transaction data.
Finance Application
- This paper's insights and data offer significant arbitrage opportunities for finance research.
- In asset pricing, one could examine how public equity markets react to consulting engagements (e.g., through event studies on announcements or a 'consulting intensity' factor) and whether these productivity gains translate into higher stock returns or lower credit risk for publicly traded firms.
- For private equity, the detailed firm-level data on consulting spend and its impact on productivity and organizational changes could be used to quantify the value creation strategies of PE funds, especially for portfolio companies.
- Furthermore, the granular B2B data could inform models of supply chain finance or inter-firm credit risk, where consulting might signal future operational stability or distress.
Corporate FinanceFirm ProductivityManagement ConsultingLabor EconomicsEvent StudiesDifference-in-DifferencesSynthetic ControlB2B TransactionsFirm ValuationCredit RiskPrivate EquityOrganizational Economics
Core finding, identification, data
Core Finding
- Consulting engagements are associated with persistent and positive effects on client firm productivity (e.g., 3.6% increase in labor productivity over five years), alongside a small reduction in employment, stable revenue, increased average wages (2.7%), and a shift towards managerial workers.
- The findings suggest consulting facilitates productivity-enhancing restructuring rather than rent-shifting, with mixed or small effects on profitability.
Identification Strategy
- The study employs a difference-in-differences (DiD) design, identifying 'consulting events' as sharp, significant increases in consulting expenditure (at least 3x prior 3-year average and > EUR 50,000).
- Treated firms are compared to a synthetic control group, constructed using the SDID method (Arkhangelsky et al., 2021), which matches the pre-event paths of treated firms with a weighted average of never-treated firms.
Data
The research leverages unique business-to-business (B2B) transaction data from the universe of Belgian value-added tax (VAT) records (2002-2023), matched with detailed firm-level annual accounts from the Central Balance Sheet Office (including social balance sheets) and administrative firm registries (Crossroads Bank of Enterprises). The analysis focuses on transactions with the 10 largest strategy and management consultancies.
Lorenz KF. Ekerdt, Kai-Jie Wu — Macroeconomics and Productivity
This paper argues that a substantial portion of the observed decline in U.S. research productivity is a transitional composition effect driven by self-selection in researcher ability and the rapid expansion of the research sector.
Finance Application
- The core mechanism of self-selection and diminishing returns to talent in an expanding sector could be directly applied to the asset management industry.
- As the number of hedge funds or quantitative trading firms expands, a Roy-like model could estimate whether the average 'talent' (e.g., skill in generating alpha) of new entrants declines, explaining trends in aggregate alpha generation or the increasing difficulty of finding profitable strategies.
- In household finance, it could analyze self-selection into complex financial products like options or crypto trading, assessing if the average financial literacy or analytical ability of marginal investors declines as these markets expand, leading to systematically worse outcomes for later entrants.
- For insurance, it could model how the expansion of new, complex insurance markets (e.g., cyber risk) affects the average expertise of underwriters, potentially leading to mispricing of risks or increased systemic vulnerabilities for insurers.
self-selectiontalentproductivitydiminishing returnslabor economicsRoy modelmicrodataasset managementhousehold financefinancial literacyskill premiumeconomic growthmarket efficiency
Core finding, identification, data
Core Finding
- The authors estimate that average researcher ability in the U.S. has decreased by 52.2% between 1960 and 2021 due to self-selection.
- This implies that the decline in efficiency-unit adjusted researcher productivity is only half of what is documented in the literature, leading to substantially higher long-run growth rates in semi-endogenous growth models.
Identification Strategy
- The paper estimates a Roy-like model of researcher supply using indirect inference.
- It targets sectoral employment shares and their changes, the sectoral variance of log earnings, and the mean log earnings of workers transiting between research and non-research sectors relative to those who do not.
- Longitudinal moments from the NSCG panel data are crucial for identifying the joint distribution of sector-specific abilities and distinguishing between comparative and absolute advantage.
Data
The paper uses 1960-2021 U.S. microdata, primarily from the U.S. National Survey of College Graduates (NSCG) for detailed work activity and earnings, supplemented by the Decennial Census and the American Community Survey (ACS) for longer aggregate trends. It also uses patent data from the U.S. Patent and Trademark Office's (PTO) PatentsView database and test scores from the National Longitudinal Survey of Youth (NLSY) for motivating evidence.
Jesús Fernández-Villaverde, Yang Yu, Francesco Zanetti — Macroeconomics and Productivity
This paper argues that incumbent firms' defensive hiring of researchers, enabled by an inelastic research labor supply, reduces creative destruction and TFP growth, explaining recent economic trends.
Finance Application
- This paper offers a rich framework for asset pricing and corporate finance.
- The concept of 'defensive R&D' could explain why some firms with high R&D expenditures do not generate commensurate innovation-driven growth, leading to a re-evaluation of R&D's impact on firm valuation and stock returns.
- Investors could potentially identify and price firms engaged in defensive hiring, perhaps leading to a 'defensive R&D discount' or 'market power premium' that is distinct from a pure innovation premium.
- In household finance, the declining elasticity of specialized research labor supply implies increased human capital risk for individuals in these fields, which could influence their optimal savings, portfolio allocation, and demand for career-specific insurance products.
InnovationR&DMarket PowerLabor MarketsCreative DestructionTFP GrowthFirm DynamicsAsset ValuationCorporate StrategyHuman CapitalMonopsony
Core finding, identification, data
Core Finding
- Incumbent firms strategically engage in 'defensive hiring' by over-hiring researchers and offering higher wages to deter potential entrants, especially when the research labor supply is inelastic and ideas are harder to find.
- This mechanism is shown to negatively correlate with creative destruction and sectoral TFP growth, while positively predicting incumbent firm lifespans.
- Empirical evidence confirms that the research labor supply in the US is indeed inelastic and has become more so over time.
Identification Strategy
- The paper identifies the elasticity of research labor supply using a novel dataset of patent applications, inventor data, and stock market returns, instrumenting expected payoffs with unadjusted average patent value and lagged firm size.
- For the impact of R&D on firm entry and TFP, panel regressions are used with industry and year fixed effects, and the causal effect is identified through an interaction term between R&D spending and industry-specific research labor supply elasticity, arguing that effects are stronger where labor supply is less elastic.
- An alternative measure for researcher labor supply uses a Bartik instrument for wage growth.
Data
The paper uses a novel dataset combining US Patent and Trademark Office (USPTO) patent applications (1970-2019) with market values from Kogan et al. (2017). It also uses Compustat Fundamental Annual data (1950-2021) for firm R&D, Bureau of Labor Statistics (BLS) data for sectoral TFP (1987-2019), Business Dynamics Statistics (BDS) from the US Census Bureau for firm entry (1978-2019), and American Community Survey (ACS) via IPUMS and BLS Occupational Employment Survey (OES) for researcher wages and employment.
Hugo Lhuillier — Inequality and Macroeconomics
This paper documents how employer sorting and frictional local labor markets shape spatial wage inequality and job mobility, explaining why larger cities have higher average wages but also greater within-city inequality.
Finance Application
- The spatial heterogeneity in wage inequality and job ladder steepness could inform household finance models by explaining regional differences in savings rates, mortgage debt, and demand for income insurance products.
- Households in cities with steeper job ladders might exhibit different risk-taking behaviors.
- For real estate asset pricing, the explicit modeling of local housing prices driven by employer sorting and labor market tightness offers a micro-foundation for regional property value dynamics and potential mispricing.
- Furthermore, the concentration of productive firms could predict local economic growth, impacting regional equity markets or the credit risk of geographically concentrated corporate portfolios.
spatial economicslabor economicswage inequalityemployer sortingsearch frictionsjob laddershousehold financereal estateincome riskhuman capitalregional economicsasset pricing
Core finding, identification, data
Core Finding
- The paper establishes two novel facts: high-paying jobs are concentrated in large cities while low-paying jobs are dispersed, and wage gains in large cities materialize over time as workers reallocate up job ladders.
- A spatial framework with heterogeneous employers and frictional local labor markets rationalizes these facts, showing that productive employers agglomerate in large cities, intensifying competition and steepening job ladders, leading to higher average wages and greater within-city inequality, even with minimal local TFP gaps.
Identification Strategy
- The identification strategy combines quantitative spatial models and wage-posting settings.
- It uses a mover design à la Abowd et al. (1999) to estimate job fixed effects, abstracting from worker heterogeneity.
- Search frictions are identified from worker flows, and entry cost dispersion is estimated from employer location choices and profit opportunities using 2SLS, instrumenting with wages.
- Local TFPs are recovered as residuals to match average wage premia.
Data
The paper utilizes French matched employer-employee data (DADS) from 2008-2019, including a 4% representative panel and a repeated cross-section of all employed workers. It also incorporates housing price data from the 'Carte des Loyers' (Rental Map) and aggregate employment flows from the 'Enquête emploi en continu' (Labor Force Survey).
Jesus Bueren, Josep Pijoan-Mas, Dante Amengual — Inequality and Macroeconomics
This paper investigates how lifestyles contribute to the education gradient of life expectancy, modeling the joint determination of education and lifestyles early in life, and quantifying the impact of rising labor earnings inequality on health outcomes across cohorts.
Finance Application
- The findings have significant implications for household finance and insurance.
- The education-lifestyle-longevity gradient suggests that optimal retirement savings and decumulation strategies should vary substantially by educational attainment, impacting demand for annuities and long-term care insurance.
- Insurers could leverage the 'latent lifestyle types' to develop more granular risk pricing models for life and health insurance, potentially reducing adverse selection.
- Furthermore, the widening longevity gap across cohorts implies growing disparities in intergenerational wealth transfers and long-term asset demand, which could influence asset prices for duration-sensitive assets.
Health inequalityLife expectancyLifestylesEducationHuman capitalHousehold financeRetirement planningInsuranceSavingsLatent variablesLife-cycle modelEconomic inequalityLongevity risk
Core finding, identification, data
Core Finding
- The study finds that health-protective lifestyles, more prevalent among the educated, explain almost half of the education gradient in life expectancy.
- The widening college wage premium over recent decades has increased this gradient by one year across cohorts born in the 1930s and 1970s, with 60% of this increase attributed to induced changes in the composition of college graduates and high school dropouts due to complementarities between education and lifestyle choices.
Identification Strategy
- The paper employs a novel econometric methodology to estimate unobserved, time-invariant latent lifestyle types (protective vs. detrimental) from panel data on health behaviors and outcomes.
- These types are then integrated into a two-stage life-cycle model where individuals jointly choose education and lifestyle early in life, and subsequently make consumption and saving decisions.
- The model is calibrated to match the joint distribution of education and lifestyles across cohorts, allowing for counterfactual analyses to decompose the effects of income changes and selection.
Data
The study uses panel data from the Health and Retirement Study (HRS) and the Panel Study of Income Dynamics (PSID) for health behaviors (e.g., smoking, drinking, preventive tests) and health outcomes. Additionally, decennial census data from 1940 to 2020 is used to parameterize wage processes.
Jamie Hentall-MacCuish, Eric French, Cormac O'Dea, Uta Bolt — Inequality and Macroeconomics
This paper estimates a dynastic life-cycle model to understand how parental time, education, and money transfers shape intergenerational mobility and assesses the welfare effects of student loan policies on parents and children.
Finance Application
- This model could be directly applied in household finance to analyze how different student loan structures (e.g., income-contingent repayment, parental co-signing) impact household savings for education, children's debt burdens, and subsequent wealth accumulation, informing the design of education-linked financial products.
- The framework for intergenerational wealth transfers could also be extended to study how various forms of transfers (cash, education, housing) influence household portfolio choices, risk-taking, and demand for life insurance, especially under different tax regimes or economic shocks.
- Furthermore, the explicit modeling of human capital risk and returns could inform the pricing of novel insurance products designed to protect against adverse educational or career outcomes.
Intergenerational transfersHuman capitalLife cycle modelStudent loansParental altruismHousehold financeEducation financeWealth accumulationConsumptionSavingsWelfare analysisStructural estimationGMMMethod of Simulated Moments
Core finding, identification, data
Core Finding
- Parental time investments serve a dual purpose: increasing children's human capital and providing direct enjoyment to parents, implying the opportunity cost of time with children is between work and leisure.
- Relaxing intergenerational borrowing constraints through student loans reduces intergenerational persistence and increases parental welfare, but on average negatively impacts children's welfare due to increased debt, with benefits accruing only to marginal students.
Identification Strategy
- The paper employs a two-step estimation strategy.
- First, human capital production functions and wage processes are estimated using GMM (for measurement error in latent skills/investments) and fixed effects (for selection bias in wages) on NCDS data.
- Second, remaining parameters (including altruism and time cost of investment) are estimated via the Method of Simulated Moments, matching simulated life-cycle profiles to observed moments from NCDS, ELSA, and UKTUS data.
Data
The study primarily uses the National Child Development Survey (NCDS) for longitudinal data on individuals born in 1958, complemented by the English Longitudinal Study of Ageing (ELSA) for wealth and transfer data, and the UK Time Use Survey (UKTUS) for detailed time investment measures.
Mary Ann Bronson, Daniel Haanwinckel, Maurizio Mazzocco — Inequality and Macroeconomics
This paper develops and estimates a dynamic lifecycle model of household decisions (labor supply, marriage, divorce, human capital, household production) to evaluate the welfare and distributional effects of different income taxation systems.
Finance Application
- The paper's detailed dynamic model of household decisions, including labor supply, human capital accumulation, marriage, and divorce under various tax regimes, offers a rich framework for household finance.
- It could be used to study how tax policy influences household savings rates, demand for tax-advantaged retirement accounts, and portfolio allocation decisions over the life cycle.
- The impact of tax systems on marriage bonuses/penalties and divorce risk could also inform the design of pre-nuptial agreements, joint vs. separate investment strategies, and the demand for financial planning services around major life events.
taxationhousehold financelabor supplyhuman capitalmarriagedivorcewelfareincome inequalitydynamic modelslimited commitmenthousehold productionpolicy evaluationBush Tax Cutsdifference-in-differences
Core finding, identification, data
Core Finding
- The paper finds that a general joint tax system offers the highest welfare but is complex to implement.
- A simpler income-splitting system augmented with a secondary-earner deduction improves welfare and reduces inequality, primarily by bolstering women's labor force participation and human capital accumulation, without the negative impacts on low-ability women observed under individual taxation.
Identification Strategy
- The paper uses a difference-in-differences analysis, leveraging variations from the 2003 "Bush Tax Cuts," to assess their effects on intra-household specialization and secondary earners' labor force participation.
- They estimate that a 6 percentage point reduction in the average tax rate for secondary earners increased their labor force participation by 4 percentage points, with a stronger response in families with young children.
Data
The study uses data from the Current Population Survey (CPS) (1980-2016), Panel Study of Income Dynamics (PSID) (1980-2010), American Time Use Survey (ATUS) (2000-2011), and Consumer Expenditure Survey (CEX) (1980-2010).
Oliko Vardishvili — Inequality and Macroeconomics
This paper quantifies the macroeconomic costs of college dropouts, identifying unpredictable grant volatility as a key driver, and demonstrates that stabilizing grants can significantly improve social welfare and be fiscally self-financing.
Finance Application
- This research offers several avenues for finance.
- In household finance, the finding that grant volatility acts as an 'uninsurable financial shock' for high-ability, low-wealth students suggests a market for 'education completion insurance' or 'contingent student loans' where repayment terms adjust to unexpected aid reductions.
- This could mitigate human capital misallocation and reduce student loan defaults.
- For the insurance industry, this highlights an untapped market for products that protect against financial aid uncertainty, potentially in partnership with universities or government.
- In asset pricing, the macroeconomic gains from stabilizing grants (increased productivity, reduced skill premium) imply long-term impacts on corporate earnings and economic growth, which could be integrated into models for valuing equity markets or specific sectors reliant on skilled labor, potentially creating a 'human capital risk' factor.
Human CapitalFinancial ConstraintsStudent LoansEducation FinanceInsuranceMacroeconomicsGeneral EquilibriumSkill PremiumLiquidity ShocksDropout RiskSocial Welfare
Core finding, identification, data
Core Finding
- Grant reductions, often unpredictable and unrelated to academic performance, account for 34% of college dropouts, significantly more than academic uncertainty (10%).
- A policy eliminating grant volatility raises social welfare by 2.4% and improves ability sorting, proving fiscally self-financing due to increased aggregate productivity and a broader tax base.
Identification Strategy
- The paper combines empirical evidence from four longitudinal datasets (NLSY79, NLSY97, BPS) with an overlapping generations general equilibrium model.
- Empirically, it identifies grant volatility as an 'uninsurable financial shock' by showing that grant reductions are unpredictable, not offset by other financial sources, and strongly correlate with dropout decisions, especially for high-ability, low-wealth students, even after controlling for eligibility criteria.
- The causal impact is further supported by aligning with regression kink designs around federal EFC thresholds from prior literature.
Data
The paper uses four longitudinal datasets: the National Longitudinal Surveys of Youth (NLSY79 and NLSY97) and the Beginning Postsecondary Students Longitudinal Study (BPS).
Giammario Impullitti, Pontus Rendahl — Inequality and Macroeconomics
This paper develops a framework linking rising markups, slowing productivity growth, and increasing wealth inequality to a common decline in market competition, and analyzes its welfare implications across the wealth distribution.
Finance Application
- This framework offers several applications for finance research.
- In asset pricing, it could inform models where market power (e.g., measured by markups or industry concentration) acts as a fundamental factor explaining the cross-section of stock returns, or how changes in competition affect firms' cost of capital and investment.
- For household finance, the heterogeneous welfare and saving responses to the 'return gap' could be used to analyze how different wealth percentiles adjust their portfolio allocations (e.g., equity vs. bonds) to shifts in market structure, or to evaluate the distributional impact of antitrust policies on household wealth accumulation.
Market PowerWealth InequalityEconomic GrowthAsset ReturnsFirm ValuationHeterogeneous AgentsHousehold FinanceAsset PricingCorporate ProfitsMarkupsCompetitionProductivity Growth
Core finding, identification, data
Core Finding
- Reduced competitive pressure, stemming from increased entry costs, simultaneously raises markups and corporate profits, driving up asset returns and firm valuations, while slowing economic growth by weakening knowledge spillovers.
- This widening 'return gap' (r-g) amplifies wealth inequality, disproportionately benefiting the wealthiest households (top 1%, especially top 0.1%) and leading to welfare losses for most others due to eroded real wages and diminished future wage prospects.
Identification Strategy
- The paper identifies the causal impact of market power by exogenously increasing the 'cost of entry' for firms within an endogenous growth model.
- This shock directly reduces competition and raises markups, allowing the model to trace its propagation through firm dynamics, innovation, aggregate growth, asset returns, and heterogeneous household saving behavior.
Data
The paper motivates its trends using aggregate U.S. data on markups, total factor productivity (TFP), Gini index, and top-10 wealth share from sources like the World Inequality Database and De Loecker et al. (2020). Its incomplete markets model is calibrated using various micro and macro moments from U.S. labor statistics, R&D indicators, and wealth distribution data.
Sarah Moshary, Cailin R. Slattery — Political Economy
This paper provides novel empirical evidence on the causal link between industry consolidation and increased political influence, showing that mergers in the auto retail industry lead to higher lobbying and more favorable legislation.
Finance Application
- This research suggests that industry consolidation can be a leading indicator of future policy changes that benefit the consolidated industry.
- In asset pricing, investors could develop an 'influence factor' by identifying industries undergoing consolidation in politically congruent markets, predicting regulatory tailwinds for firms and impacting their stock prices or bond yields.
- For household finance, policy changes influenced by lobbying (e.g., sales taxes, consumer protection laws) directly affect household budgets and purchasing decisions, offering avenues to study how these shifts impact household consumption or debt.
- In insurance, consolidation in regulated sectors could lead to lobbying for changes in insurance regulations (e.g., liability laws, mandatory coverages), affecting profitability for insurers or demand for specific products.
LobbyingPolitical EconomyMergers and AcquisitionsIndustry ConcentrationRegulatory RiskPolicy InfluenceAsset PricingHousehold FinanceCorporate StrategyEvent Study
Core finding, identification, data
Core Finding
- Mergers in the auto retail industry lead to higher levels of industry lobbying, on the order of +70%, primarily driven by mergers that resolve the collective action problem in politically 'congruent' markets.
- This increased lobbying translates into more favorable legislation, with a 7 percentage point increase in enactment probability for bills car dealers support, implying an annual payoff of $2.80 million for the industry in the year following a merger.
Identification Strategy
- The authors use a difference-in-differences approach, comparing lobbying before and after large auto dealer mergers.
- The key identification strategy leverages variation in the alignment of political and product markets, comparing mergers in 'congruent' states (where product markets are entirely within a single political market) with 'incongruent' states (where product markets cross state borders).
- This allows them to isolate the 'collective action effect' from other factors like market power.
Data
The paper constructs a novel dataset on state-level lobbying clients and compensation from public lobbying records in 27 states. It also uses auto dealership location and ownership data from Infogroup (2021) and bill enactment data from state legislative records.
Jaime Arellano-Bover, Kobi Mizrahi, Shmuel San — Political Economy
This paper studies the long-run cultural assimilation of immigrant children in Israel, focusing on age-at-arrival effects on language acquisition, intermarriage, fertility, and other integration outcomes.
Finance Application
- The findings on cultural assimilation, particularly regarding fertility norms and language acquisition, have direct implications for household finance.
- Immigrant households assimilating to higher fertility norms might exhibit different savings rates, demand for larger housing, and investment in education for more children, impacting long-term wealth accumulation and retirement planning.
- Language proficiency and cultural integration could also influence financial literacy, trust in local financial institutions, and participation in capital markets, leading to differential portfolio choices and risk-taking behavior.
- Furthermore, the out-migration patterns could inform the demand for international financial services or portable insurance products.
Cultural AssimilationImmigrationAge-at-ArrivalLanguage AcquisitionFertilityIntermarriageHousehold FinanceBehavioral FinanceRisk PreferencesSavingsConsumptionHousing MarketsNatural ExperimentFinancial Literacy
Core finding, identification, data
Core Finding
- The study finds that even small differences in age at arrival (between 7 and 17) have significant long-term impacts on cultural assimilation.
- Younger arrivals demonstrate higher Hebrew proficiency, lower out-migration, less residential segregation, increased intermarriage with natives, and assimilation towards higher Israeli fertility norms (more children, later first child).
- Language proficiency is identified as a crucial mediating factor for these assimilation outcomes.
Identification Strategy
- The identification strategy leverages the unexpected lifting of Soviet emigration restrictions in 1989, arguing that the migration of Former Soviet Union (FSU) Jews to Israel in 1990-1991 was driven by 'push' factors (refugee-like conditions) rather than selective 'pull' factors related to children's age, thus providing plausibly exogenous variation in age-at-arrival.
- Robustness checks include intra-family comparisons using sibling fixed effects.
- They also introduce a revealed-preference measure of language acquisition based on the language chosen for a high-stakes university admissions test.
Data
The paper uses population-level Israeli administrative records from 1989-2019, including birth records, migration dates, marriage links, parent-child links, residential location, and Psychometric Entrance Test (PET) test-taking records (including language choice). It also incorporates self-reported education and occupation data upon arrival.
Ying Bai, Ruixue Jia, Jiaojiao Yang — Political Economy
This paper analyzes the long-term effects of state censorship on knowledge production in China, revealing a pattern of initial suppression followed by a significant revival of publications and topic diversity after state control weakened.
Finance Application
- The paper's insights into information control and its dynamic effects could be highly valuable in finance.
- For asset pricing, one could examine how government censorship or control over economic news (e.g., GDP figures, inflation data, corporate scandals) affects market efficiency, price discovery, and the volatility of asset prices in emerging or authoritarian economies.
- In corporate finance, the 'chilling effects' could be analogous to how firms self-censor disclosures or delay innovation in politically sensitive sectors (e.g., AI, defense) due to fear of regulatory backlash or state intervention, impacting their valuation and access to capital.
- For household finance, the study of information intermediaries (publishers) responding to censorship could inform how financial advisors or media outlets alter their advice or reporting under regulatory pressure, influencing household investment decisions and financial literacy.
CensorshipInformation AsymmetryPolitical EconomyKnowledge ProductionChilling EffectsSelf-CensorshipInformation IntermediariesMarket EfficiencyCorporate DisclosureRegulatory RiskChinaHistorical Data
Core finding, identification, data
Core Finding
- Categories subjected to stricter censorship experienced significant declines (20-22%) in publication volume and topic diversity during the seven decades following the bans (1770s-1830s).
- However, political upheavals and the erosion of state control from the 1840s triggered a notable resurgence in these previously restricted categories, with publication trends and keyword stock eventually catching up to pre-censorship levels by the 1910s.
- This dynamic is largely attributed to chilling effects on unbanned topics and the entry/exit behavior of publishers.
Identification Strategy
- The study employs a difference-in-differences approach and an event-study framework.
- The identification relies on the systematic, top-down nature of the Siku Quanshu book-banning campaign (1772-1783) as a sharp exogenous shock to knowledge production.
- This allows for a comparison of publication trends across categories with varying levels of censorship, using the 1760-1772 period as a pre-treatment reference to establish parallel trends.
- A Bartik approach is also used to instrument for the number of publishers.
Data
The paper uses publication data from over 161,000 books spanning 1660s-1940s from the 26-volume General Catalog of Pre-modern Chinese Books, including information on publication category, author, time, reprint status, and publishers. Textual data from book titles is used for keyword analysis and similarity measures. Records of banned books from official and historical sources, and author biographical details from the China Biographical Database Project (CBDB), are also utilized.
Grant Goehring, Walker Hanlon — Political Economy
This paper examines how advocacy groups mobilize constituents to influence political representatives and policy outcomes outside of electoral cycles, using the U.K. women's rights movement and the Contagious Disease Acts as a historical case study.
Finance Application
- This framework could be applied to study the impact of ESG activism or shareholder engagement on corporate policies and asset prices.
- Advocacy groups (e.g., environmental NGOs, consumer protection organizations) could be seen as 'LNA rallies' influencing 'constituent signals' (shareholder proposals, public campaigns, social media sentiment) which, in turn, affect 'MP votes' (corporate board decisions, management strategies).
- The spillover effects could model how activism on one ESG issue (e.g., climate change) influences corporate actions and valuations in related areas (e.g., supply chain ethics, diversity initiatives), even for firms not directly targeted.
- This could also inform household finance research on how consumer advocacy groups influence household consumption, investment in socially responsible funds, or demand for specific insurance products (e.g., climate risk insurance) by solving coordination problems among individual consumers.
political economyadvocacylobbyingsignallingactivismESG investingshareholder activismpolitical riskpolicy uncertaintyhousehold financeconsumer behaviorregulationevent studyinstrumental variables
Core finding, identification, data
Core Finding
- Advocacy efforts by groups like the LNA (Ladies National Association) significantly and persistently increased constituent signalling (petitions) to MPs.
- These signals, in turn, influenced MP votes on the Contagious Disease Acts, particularly in the absence of elections, and had spillover effects on related policy areas like women's suffrage through both coordination and belief-updating channels.
Identification Strategy
- The study employs a staggered-treatment event study analysis to assess the impact of LNA rallies on petitioning activity, observing flat pre-trends to support causality.
- For MP voting behavior, it uses the timing and location of LNA rallies as a quasi-exogenous instrumental variable for constituent petitions, addressing endogeneity concerns that MPs might vote differently for unobserved reasons.
Data
The paper uses a rich dataset of over 300,000 geolocated petitions to Parliament (1864-1883), detailed LNA rally data extracted from digitized historical newspapers (1869-1883), and MP voting records from Eggers and Spirling (2014) for the period 1870-1883. Parliamentary debate texts are also analyzed using ChatGPT.
Youn Baek — Political Economy
This paper demonstrates that American Protestant missionaries played a crucial role in fostering moral universalism and providing expertise, significantly influencing congressional support for U.S. foreign aid and long-term global development efforts.
Finance Application
- The paper's insights into how moral universalism and social influence spread through networks could be applied to understanding the diffusion of ESG (Environmental, Social, Governance) investing preferences or impact investing trends among institutional and retail investors.
- Historically, missionary networks acted as 'information brokers' for foreign countries; this mechanism could be used to study how early cross-border capital flows and foreign direct investment were influenced by non-traditional information channels, potentially predicting the formation of international financial ties or the success of foreign ventures.
- Furthermore, the long-term commitment to global development fostered by missionaries could inform research on the drivers of long-horizon investment strategies by pension funds or endowments towards emerging markets or socially responsible assets, especially in regions with historical ties to these social movements.
International DevelopmentForeign AidSocial InfluenceMoral UniversalismHistorical DataCongressional VotingReligious NetworksInformation DiffusionESG InvestingImpact InvestingCross-Border Investment
Core finding, identification, data
Core Finding
- Exposure to American Protestant missionaries significantly boosted congressional support for major foreign aid bills that initiated the modern era of U.S. development assistance.
- Missionaries influenced public opinion by framing aid in terms of human dignity rather than strategic interests, became key experts on non-Western societies, and their influence contributed to a sustained commitment to global development, reflected in increased Peace Corps participation.
- A 10% decrease in the total number of missionaries reduces the probability of a pro-foreign aid vote by 8 percentage points.
Identification Strategy
- The paper employs a shift-share instrumental variable strategy, where the 'shift' is the share of missionaries per denomination and the 'share' is the proportion of that denomination's members in each congressional district.
- The instrument leverages the plausibly exogenous travel routes of Student Volunteer Movement recruiters (1886-87) to colleges, and further validates exogeneity by exploiting temporal variation in the establishment dates of YMCA chapters, which coordinated with travel secretaries, to isolate the causal effect of missionary exposure.
Data
The study uses a novel dataset compiled from archival records of the Student Volunteer Movement (SVM, 1886-1964), including biographical records of 12,265 missionaries and 24,841 runner-up applicants. It also incorporates U.S. congressional roll-call voting data (VoteView), Census of Religious Bodies (1890), historical YMCA chapter locations, congressional speech data analyzed with natural language processing, bibliometric data (OpenAlex, ProQuest) for publication output, Wikipedia biographies for regional expertise, and Peace Corps volunteer data (1961-2000).
Dan M. Bernhardt, Laurent Bouton, Stefan Krasa, Francesco Squintani — Political Economy
This paper develops a novel spatial voting model where citizens are averse to free-riding, leading to conditional cooperation and complex dynamics of electoral turnout and candidate platform choices.
Finance Application
- The concept of free-riding aversion and conditional cooperation could be directly applied to collective action problems in finance, such as shareholder activism, bondholder committees, or decentralized finance (DeFi) governance.
- For example, it could explain why retail investors participate in proxy votes or DAO governance despite minimal individual impact, driven by a disutility from free-riding on the efforts of large institutional investors or other token holders.
- The model's predictions on how 'voting costs' (e.g., research effort, transaction fees) and 'platform polarization' (e.g., divergent corporate strategies or DeFi proposals) affect participation could inform optimal governance structures and activist strategies in financial markets.
Electoral TurnoutFree-RidingConditional CooperationSpatial VotingCollective ActionBehavioral EconomicsPolitical EconomyShareholder ActivismDeFi GovernancePolarizationVoting Costs
Core finding, identification, data
Core Finding
- The model's central finding is that voter motives (purely selfish vs. purely cooperative) significantly alter the effects of electoral primitives on turnout and candidate platforms.
- For instance, increasing elite polarization boosts turnout for cooperative voters but reduces it for selfish ones, and a 'bandwagon effect' (higher turnout for front-runners) emerges when cooperative motives are strong.
- Furthermore, voting costs can either increase or decrease elite polarization depending on voter motives, challenging conventional wisdom.
Identification Strategy
- The paper is theoretical, introducing a methodological innovation by incorporating 'free-riding aversion' into a canonical spatial voting model.
- This aversion is modeled as a disutility incurred when a citizen abstains but benefits from others' costly voting efforts, thereby endogenously capturing conditional cooperation.
- The model establishes the existence of 'no minimal expansion' equilibria, ensuring robustness to minimal coordination among voters.
Data
This paper is theoretical and does not use primary data. It references existing empirical and experimental literature to ground its assumptions (e.g., conditional cooperation) and to discuss how its comparative static predictions align with or diverge from observed phenomena in political science.
Eva Davoine, Joseph Enguehard, Igor Kolesnikov — Political Economy
This paper quantifies the political costs of tax enforcement in early modern France, focusing on the salt tax and its impact on social conflicts.
Finance Application
- This research offers insights into how political risks stemming from tax policy and enforcement can be quantified and priced.
- In asset pricing, regional tax disparities and enforcement shocks could be factored into municipal bond yields or local equity risk premia, especially for firms sensitive to political stability.
- For household finance, it informs models of tax compliance and household consumption/investment responses to perceived tax unfairness or enforcement intensity.
- Insurers could leverage these findings to better price political risk insurance or property & casualty policies in regions prone to social unrest due to fiscal policies.
TaxationPolitical EconomyHistorical EconomicsState CapacityConflictTax EnforcementRegional DisparitiesFiscal PolicySocial Unrest
Core finding, identification, data
Core Finding
- Increased salt tax enforcement in high-tax regions led to a twenty-fold increase in conflicts between taxpayers and the state, an effect that persisted until the French Revolution.
- The likelihood of conflict was directly correlated with tax differences between neighboring regions, suggesting a quantifiable 'political cost' of taxation, estimated at a 13% revenue loss if conflicts were to be eliminated.
Identification Strategy
- The authors use a spatial difference-in-discontinuities design, comparing municipalities just inside and outside a high-tax region before and after a 1740 reform that created special courts to curb illicit salt smuggling.
- This exploits a discontinuous change in tax enforcement exposure at the salt tax border, combined with time variation from the reform.
Data
The study utilizes a novel digitized parish-level map of 1665 salt tax regions, a comprehensive database of social conflicts (HiSCoD, 1661-1790), and grievance lists submitted to King Louis XVI in 1789, complemented by an alternative dataset on criminal cases against salt smugglers.
Hanming Fang, Ming Li, Guangli Lu — Political Economy
This paper decodes China's industrial policies from 2000 to 2022 by employing large language models (LLMs) to extract and analyze rich information from 3 million government documents across central, provincial, and municipal levels.
Finance Application
- The LLM-based methodology for extracting nuanced policy details and their impact on firm behavior could be applied to analyze the effects of regulatory changes or government interventions on specific financial markets or asset classes.
- For instance, one could use this approach to identify how specific government policies (e.g., subsidies, tax incentives, or environmental regulations) impact the valuation of firms in targeted industries, their access to capital markets, or even the pricing of green bonds.
- The politician data could be used to study how political cycles or leadership changes influence financial market sentiment or policy uncertainty in specific regions, affecting local asset prices or investment flows.
Industrial PolicyLarge Language ModelsChinaFirm BehaviorPolicy DiffusionGovernment InterventionText AnalysisEconomic PolicyPolitical EconomyCorporate FinanceAsset ValuationRegulatory Impact
Core finding, identification, data
Core Finding
- The paper finds that local governments' industrial policy choices are influenced by economic rationale (e.g., revealed comparative advantage), political incentives (e.g., top-down directives, politician career mobility), and administrative capacity.
- While supportive policies boost firm entry and provide monetary benefits, their effect on firm productivity is mixed and short-lived, with policy diffusion leading to inefficiencies like overcapacity and local protectionism.
Identification Strategy
- The core methodological innovation is the use of Large Language Models (LLMs) to systematically extract and classify granular, multidimensional information from a vast, unstructured dataset of 3 million government policy documents.
- This involves careful prompt engineering, multistage extraction and refinement, rigorous verification, and hallucination-robust strategies, enabling the creation of a structured industrial policy dataset previously unavailable.
Data
The paper uses a comprehensive dataset of 3 million government policy documents from 2000-2022. It integrates this with micro-level firm data (firm registration, administrative tax records, Value-Added Tax data) and a manually collected dataset on provincial and city-level politicians.
Lauren Cohen, Umit Gurun, Katie Moon, Paula Suh — Innovation
This paper identifies "patent hunters" who consistently find and develop initially overlooked, "late-blooming" patents, and quantifies the substantial economic benefits these hunters accrue for their firms.
Finance Application
- The market's initial undervaluation of "late-blooming" patents, despite their eventual high impact, presents a potential mispricing opportunity for asset pricing.
- Investors could develop strategies to identify firms with strong "patent hunting" capabilities or those whose patent portfolios exhibit characteristics of "huntable" IP (e.g., technologically peripheral to original writers, in less competitive spaces) to generate alpha.
- For corporate finance, the findings highlight the strategic value of acquiring specific human capital (patent-hunting inventors) and overlooked IP, informing M&A decisions and R&D investment strategies aimed at commercializing neglected innovations.
InnovationPatentsCorporate FinanceAsset PricingFirm ValueR&DHuman CapitalCausal InferenceInstrumental VariablesTechnologyEntrepreneurship
Core finding, identification, data
Core Finding
- The study finds that "patent hunters" generate significant positive value by identifying and commercializing "late-blooming" patents, which are initially overlooked but eventually become highly influential.
- Their adoption of these patents is associated with a 6.7% rise in sales growth, a 1.9% increase in Tobin's Q, and a 4.9% increase in new product offerings, with these hunted patents being technologically distant from the original writers but closer to the hunters' core and in less competitive spaces.
Identification Strategy
- The paper employs an instrumental variable (IV) approach to establish a causal link between patent hunting and firm performance.
- The instrument exploits quasi-random variation in local patent-hunting capacity, leveraging the mobility of patent-hunting inventors from bankrupt neighboring firms to nearby firms within a 100-mile radius.
- Falsification tests confirm that this shock specifically influences late-bloomer patent hunting and not other patenting activities or past firm outcomes.
Data
The study utilizes a comprehensive dataset of over 47 million U.S. patents from Thomson Innovation and PatentsView (1976-2020), merged with S&P Compustat financial data for public firms. It also incorporates new product launch data from Mukherjee et al. (2022) and firm relationship data from FactSet Revere and Compustat.
Yann Algan, Eva Davoine, Thomas Renault, Stefanie Stantcheva — Political Economy
This paper investigates the causal effects of emotions, particularly anger, on citizens' policy views using large-scale social media data and online experiments.
Finance Application
- The causal link between emotions and policy views offers rich avenues for finance research.
- In asset pricing, emotion-driven policy shifts (e.g., protectionism, climate regulations, redistribution) could be incorporated into predictive models for asset returns, volatility, or cross-sectional anomalies.
- For instance, an 'anger index' derived from political discourse could forecast changes in sector-specific equity performance (e.g., energy, materials) or sovereign bond yields.
- In household finance, understanding how induced emotions affect policy preferences could shed light on changes in household savings, investment allocations (e.g., towards ESG funds or local businesses), or demand for certain financial products.
- For insurance, shifts in climate policy driven by public anger could lead to new regulatory risks or opportunities for property and casualty insurers, or influence the demand for climate-related insurance products, which could be modeled using the paper's emotional data.
emotionsangerfearpolicy uncertaintypolitical risksentiment analysissocial mediatextual analysiscausal inferenceexperimentsprotectionismclimate policyredistributionasset pricinghousehold financeinsuranceeconomic policy uncertainty
Core finding, identification, data
Core Finding
- The study documents a significant rise in anger in political discourse since 2016, both from policymakers and citizens, with anger driving higher engagement.
- Experimentally, negative emotions, especially anger, causally increase support for protectionism, restrictive immigration, redistribution, and climate policies, while positive emotions reduce populist inclinations.
- Anger is found to be a much stronger driver of policy views than fear.
Identification Strategy
The paper employs a two-pronged identification strategy: (1) Observational analysis of social media and political speech data (2013-2025) using LLMs for emotion and topic classification, including user fixed effects to control for self-selection. (2) Two nationwide online experiments in the U.S. that randomly assign participants to video treatments designed to induce positive, negative (anger/fear), or neutral emotions, allowing for causal inference on policy views across various domains.
Data
The paper utilizes extensive data including: random samples of X (Twitter) users' tweets (690,000 policy-related, 3 million climate-related tweets), official political party tweets (395,272 tweets), Congressional tweets (1.5 million tweets), Congressional floor speeches (1.8 million turns), and campaign speeches from key political figures (1,992 interventions). Additionally, two large-scale online experiments (Survey A with 3,800 citizens, Survey B with 6,366 citizens) provide experimental data on emotional responses and policy preferences.
Xavier Giroud, Ernest Liu, Holger Mueller — Innovation
This paper demonstrates that innovation spillovers occur not only within local tech clusters but also across geographically distant clusters through firms' internal networks of innovating plants, significantly boosting productivity.
Finance Application
- This research offers several avenues for finance.
- In asset pricing, the 'connectedness' of a firm's innovation network or its location within a high 'social-private innovation wedge' cluster could predict future firm growth, profitability, or risk, informing factor models or investment strategies.
- For corporate finance, the identified under-investment in innovation due to spillovers suggests potential mispricing or opportunities for private equity to acquire and optimize firms by improving internal knowledge transfer.
- In real estate, the productivity gains from cluster interconnectedness could drive local economic growth and asset values, offering insights for regional real estate investment funds.
innovationspilloverstech clustersfirm networksproductivityasset pricingcorporate financereal estategeographic diversificationknowledge diffusionfirm valuationinvestment strategyeconomic geography
Core finding, identification, data
Core Finding
- Larger tech clusters enhance local inventor productivity and also raise the productivity of inventors and plants in other, distant clusters connected via their parent firms' innovation networks.
- Inventors act as 'antennas' for knowledge diffusion within firms, which does not decay with physical distance.
- The social-private innovation wedge, indicating under-investment, is highest in large, well-connected tech clusters.
Identification Strategy
- The study uses merged USPTO-U.S.
- Census Bureau plant-level data, identifying cross-cluster spillovers through within-plant variation in the size of 'connected clusters.' It employs highly granular city × field × year fixed effects and an instrumental variable (IV) design where changes in the number of inventors in 'third' cities (where the parent firm has no presence) predict the size of connected clusters, isolating exogenous variation.
Data
The paper utilizes the United States Patent and Trademark Office (USPTO) patent database, merged with confidential establishment-level data from the U.S. Census Bureau's Longitudinal Business Database (LBD), Census of Manufactures (CMF), and Annual Survey of Manufactures (ASM). BEA economic areas define 'cities' for cluster analysis.
Daniel P. Gross, Bhaven N. Sampat — Innovation
This paper demonstrates how the U.S. Committee on Medical Research during World War II acted as a "big push" to transform the nascent biomedical sector into a robust, integrated innovation system with long-lasting effects on science, drug development, and medical practice.
Finance Application
- The "big push" theory and the concept of a coordinated, integrative R&D policy transforming a nascent sector into a high-growth innovation system could be applied to understanding the development of new financial markets or technologies.
- For instance, one could study how early government or central bank interventions (e.g., in the early days of derivatives markets, securitization, or FinTech like blockchain) acted as a "big push" to establish key infrastructure, foster interdependencies between different financial actors (banks, tech firms, regulators), and catalyze long-term innovation and growth in specific financial products or services.
- This could involve analyzing the impact of early regulatory sandboxes, government-backed consortia, or specific policy mandates on the subsequent growth and systemic integration of FinTech, potentially using patent data or new product introductions as outcomes.
innovation systemsfinancial innovationFinTechpolicy impactnatural experimentmarket developmentR&Deconomic historybiomedicinecausal inference
Core finding, identification, data
Core Finding
- The CMR's coordinated, use-oriented R&D policy during WWII, despite its modest budget, triggered a systemic transformation of U.S. biomedicine.
- It led to rapid, sustained growth in postwar scientific publications, fueled new drug development (especially in previously underexplored areas), and integrated new knowledge into medical practice, establishing the foundations for the modern U.S. biomedical innovation ecosystem.
Identification Strategy
- The paper leverages the exogenous shock of World War II military medical needs, which were distinct from pre-war civilian research priorities.
- It compares the long-run growth trajectories of biomedical research subjects that received CMR funding against those that did not, particularly distinguishing between "established" and "embryonic" (pre-1940 underexplored) fields.
- A placebo test using World War I research trajectories further strengthens the causal inference by showing no similar "big push" effect from that earlier conflict.
Data
The paper uses newly-collected archival data on 590 CMR research contracts (1941-1945), mapped to Medical Subject Headings (MeSH). It also utilizes biomedical research publications (1930-1970) from Microsoft Academic Graph, Web of Science, and PubMed, a list of new drugs introduced (1940-1975) from de Haen, digitized medical textbooks, and NIH extramural research grants (1948-1970).
Ronan C. Lyons, Allison Shertzer, Rowena Gray — CRIW Pre-Conference, Summer 2025
This paper constructs a new, methodologically consistent shelter price series and an alternative Consumer Price Index (CPI) for the U.S. from 1914-2006, revealing significantly higher historical inflation than official estimates.
Finance Application
- A revised, higher historical inflation series would necessitate a re-evaluation of real returns on various asset classes (stocks, bonds, real estate) over the long run, potentially explaining historical asset pricing puzzles like the equity premium puzzle or re-calibrating long-term asset allocation models.
- In household finance, the implied lower real income and wealth accumulation could inform research on household savings behavior, retirement planning adequacy, and the real burden of mortgage debt.
- For insurance, an underestimated historical inflation rate means that real liabilities for long-term products like annuities were higher than assumed, prompting better modeling of future inflation risk for product design and pricing.
InflationCPIHousing PricesReal ReturnsAsset PricingHousehold FinanceEconomic HistoryHedonic ModelsCost of LivingReal EstateMeasurement Error
Core finding, identification, data
Core Finding
- The official BLS Rent of Primary Residence (RoPR) series understated rental inflation, growing by 2.6% per year compared to the new Historical Housing Prices (HHP) rental index's 3.7%.
- This leads to an alternative CPI that grew by 3.6% per year instead of the official 3.3%, implying a significant understatement of living costs and an overstatement of real income and living standards over the 20th century, particularly between 1914 and 1987.
Identification Strategy
- The paper's methodological innovation involves constructing a new HHP rental index from over one million historical newspaper rental listings across 30 U.S. cities (1914-2006).
- This index uses a flexible hedonic approach with rolling two-year windows to adjust for observable housing characteristics and allow implicit prices to evolve, thereby mitigating bias from unobserved quality drift and vacancy nonresponse.
- Additionally, a revised set of shelter weights is constructed, consistently applying the Owners' Equivalent Rent (OER) concept across time by anchoring to 1987 CPI OER weights and using BEA national accounts and BLS expenditure surveys for earlier periods.
Data
The paper primarily uses historical newspaper rental listings from 30 U.S. cities (1914-2006) to construct its HHP rental index. It also utilizes official Bureau of Labor Statistics (BLS) Consumer Price Index (CPI) data, Rent of Primary Residence (RoPR) series, Consumer Expenditure Survey (CE) weights, and Bureau of Economic Analysis (BEA) national accounts data for housing components.
Ina Ganguli, Jeffrey Lin, Vitaly Meursault, Nicholas F. Reynolds — Innovation
This paper develops and validates a pipeline for selecting robust Natural Language Processing (NLP) models to measure invention similarity from patent text, demonstrating how model choice critically impacts economic measurements.
Finance Application
- The rigorous NLP model selection and validation pipeline could be directly applied to financial texts, such as earnings call transcripts, 10-K/Q filings, or analyst reports, to derive more robust measures of firm-specific risk, sentiment, or innovation signals for asset pricing.
- For instance, one could validate NLP models against market reactions to earnings announcements or credit rating changes.
- The finding of declining innovation similarity could inform corporate finance by linking firms' 'spreading out' in idea space to their cost of capital, growth options, or M&A strategies, potentially revealing how diversification or novelty in innovation affects firm valuation and investment decisions.
NLPText AnalysisPatentsInnovationModel SelectionMachine LearningEconomic HistorySimilarityAsset PricingCorporate Finance
Core finding, identification, data
Core Finding
- The paper's validated NLP representations (GTE and PaECTER) reveal a secular decline in contemporaneous patent similarity over the past century and a half, suggesting inventors are 'spreading out' over an expanding knowledge frontier.
- This finding is independently corroborated by declining rates of historical patent interferences, while alternative, unvalidated NLP models yield ambiguous or opposing trends.
Identification Strategy
- The methodological innovation is a comprehensive pipeline for NLP model selection, emphasizing domain-specific validation tasks.
- These tasks include expert legal judgments from historical patent interference cases (1998-2014), non-expert human annotations of patent similarity (1880-1920), and patent office technology classifications (1850-2023).
- The long-run decline in patent interference rates (1864-2014) serves as an independent, out-of-sample confirmation of the GTE-based similarity trends.
Data
The study uses the full text of US utility patent claims from 1836–2023 (from ProQuest and PatentsView), patent metadata, modern patent applications, historical and modern patent interference data, non-expert human annotations, and Google Books Ngrams data.
Gabriele Cristelli — Innovation
This paper examines how institutional grants from the NSF Science Development Program impacted university research capacity and local innovation in the U.S. from 1960 to 1990.
Finance Application
- This paper's findings suggest that localized innovation spillovers from university institutional funding could be a novel factor in asset pricing.
- Firms located in commuting zones with high university R&D intensity and strong absorptive capacity might exhibit higher growth opportunities, leading to different stock valuations or returns.
- In household finance, the increased local innovation and human capital could drive local economic growth, impacting household wealth accumulation, consumption patterns, and real estate values in those regions, which could be relevant for mortgage and insurance markets.
University researchInnovationPatentsLocal economic developmentDifference-in-differencesHuman capitalR&D spilloversScience policyFirm growthGeographic factors
Core finding, identification, data
Core Finding
- The NSF Science Development Program's institutional grants to universities significantly boosted university research capacity (faculty size, PhDs awarded, publications) and local innovation, leading to an 18-32% increase in patenting by incumbent firms in host commuting zones.
- This effect was particularly strong in technology fields with higher exposure to local university research, driven by increased reliance on scientific knowledge and local PhD graduate employment in industrial R&D.
Identification Strategy
- The paper employs a difference-in-differences research design, exploiting the National Science Foundation's policy of excluding top-ranked universities from the Science Development Program as a comparison group.
- It compares outcomes for universities receiving grants and their host commuting zones against those of the excluded top-ranked institutions and their host regions, with evidence of parallel pre-treatment trends.
Data
The study uses a large-scale dataset combining historical scientific publications (OpenAlex), PhD dissertations (ProQuest), and patent records (USPTO, Patstat, PatentCity, Google Patents) from 1960-1990. It also incorporates patent-science citation links, inventor disambiguation, and County Business Patterns data for local economic indicators.
Samuel Goldberg, Tai Lam — Innovation
This paper examines the causal impact of generative AI (GenAI) on product production, firm entry, sales, quality, and variety within a creative goods marketplace using a difference-in-differences design.
Finance Application
- The findings on GenAI's impact on market structure, competition, and product attributes (quality, variety, substitution) are highly relevant to finance.
- This framework could be applied to study how AI-driven fintechs disrupt traditional financial services, affecting the valuation of incumbent firms (especially those with proprietary data/IP), the competitive landscape in areas like wealth management or lending, and the labor market for financial professionals.
- The crowd-out effect on traditional artists could mirror job displacement for financial analysts or advisors, while increased variety and quality could translate to more personalized and efficient financial products for households.
Generative AIMarket StructureCompetitionIntellectual PropertyLabor MarketsFintechAsset ValuationHousehold FinanceInsurance
Core finding, identification, data
Core Finding
- GenAI acts as a substitute for non-GenAI products, leading to crowd-out of traditional content and artists, particularly in niche markets.
- However, substantial GenAI firm entry increases overall product quality and variety, expands sales, and benefits some consumers, implying a net market expansion despite adverse effects on traditional producers.
Identification Strategy
- The study employs a difference-in-differences design, leveraging a policy change by a stock image platform in December 2022 that allowed GenAI content into 'non-branded' markets (treated group) but prohibited it in 'branded' markets (control group) due to legal concerns.
- This quasi-experimental setup allows for causal inference of GenAI's market impact.
Data
The paper utilizes image-level data from a large stock image platform, including information on authors, titles, publication dates, sales status, tags, and image embeddings (1046-dimension from Google's Vertex AI). It explicitly identifies whether an image was produced using GenAI and uses a random forest model to proxy for image quality based on sales probability.
Edward W. Chen, Reagan Lengefeld, Omar Asensio — CRIW Pre-Conference, Summer 2025
This paper evaluates the long-run impact of infill development on surrounding property values in Savannah, Georgia, finding significant positive spillovers that are geographically concentrated and cost-effective, particularly in climate-disadvantaged communities.
Finance Application
- The findings on localized, density-dependent property value spillovers could inform real estate investment strategies, particularly for REITs specializing in urban redevelopment or distressed assets, by identifying 'tipping point' neighborhoods for higher returns.
- For household finance, mortgage lenders could refine risk models in climate-vulnerable areas, potentially lowering borrowing costs or increasing access to capital for homeowners in revitalizing zones, as the paper demonstrates financial viability despite climate risks.
- Property insurers could also leverage this granular spatial data to better price policies by incorporating the positive externalities of infill development on neighborhood quality and property values, especially in areas previously deemed high-risk.
Urban EconomicsReal EstateProperty ValuesSpilloversInfill DevelopmentClimate RiskCausal InferenceDifference-in-DifferencesSynthetic ControlNeighborhood RevitalizationPublic PolicyAgglomeration Economies
Core finding, identification, data
Core Finding
- Infill development generates substantial market appreciation, with properties within 25 meters experiencing a long-term 11% increase in appraised values, corresponding to a 3.61 $/sqft premium.
- These positive spillovers decay with distance and persist for at least six years, peaking in years two-three with a 4.81-17.17% premium.
- The effects are most pronounced in city blocks with higher infill density (16.7-76.9% infill properties), indicating threshold effects for agglomeration economies.
- The program is highly cost-effective, yielding a tax revenue-to-cost ratio of 2.42 and a payback period of 5.76 years when considering averted public costs.
Identification Strategy
- The study employs a propensity score weighted difference-in-differences (DiD) design, specifically the Sun and Abraham staggered DiD estimator, to identify spillover effects by comparing properties within a specified distance from infill sites to those beyond.
- Genetic matching is used to ensure covariate balance between treated and control groups, and partially pooled synthetic control methods (SCM) are utilized for validation, especially for block-level analysis.
Data
The paper uses historical data for 396 publicly-funded infill projects (2004-2019), Chatham County Tax Assessor Database (2001-2022) for housing characteristics and appraised values, Savannah Area Geographic Information System (SAGIS) for parcel locations, 2019-2023 5-year American Community Survey for socioeconomic data, and the Climate and Economic Justice Screening Tool (CEJST) for identifying climate-disadvantaged areas.
Friedrich Geiecke, Xavier Jaravel — Innovation
This paper introduces and evaluates a simple, open-source platform for conducting large-scale qualitative interviews using AI-led agents, demonstrating its robustness and versatility across various research topics.
Finance Application
- This platform and methodology could be transformative for household finance by eliciting nuanced mental models of financial decision-making, such as savings, investment, and debt, going beyond standard surveys.
- For instance, AI-led interviews could uncover how subjective states like 'meaning in life' or 'trust in institutions' influence retirement planning or insurance purchases.
- In asset pricing, it could provide rich qualitative data on retail investor sentiment, behavioral biases, and reactions to policy changes or market events, offering a deeper understanding of non-pecuniary factors driving investment choices in specific asset classes like ESG or crypto.
AI-led interviewsqualitative researchhousehold financebehavioral economicsdecision makingsubjective statesmental modelssurvey methodologyLLMssentiment analysisfinancial literacyrisk perceptionpolitical economyconsumer behavior
Core finding, identification, data
Core Finding
- AI-led qualitative interviews are found to be robust, comparable to human experts (especially with voice interaction), and yield richer, more detailed data than traditional open-text survey fields.
- The platform effectively elicits deeply personal subjective states, political views, decision-making factors, and mental models of public policies, revealing significant heterogeneity across socio-demographic and political groups.
Identification Strategy
- The methodological innovation lies in developing a simple, versatile, open-source AI-led interview platform using a single LLM agent with an adjustable system prompt that incorporates established sociological best practices.
- Robustness is assessed through comparisons to human expert ratings of transcripts, respondent-based quality metrics (e.g., preference for AI, confidence in responses, word count), and expert evaluation of content depth between AI-led interviews and open-text fields.
Data
The paper utilizes transcripts from AI-led qualitative interviews conducted with thousands of respondents across various topics, including 466 US respondents for 'meaning in life', 422 French voters for political views, 107 US respondents for educational/occupational choices, and 390 US respondents for mental models of inflation. It also includes survey responses on interview quality and human expert ratings of transcripts.
Daniel Björkegren, Junho Choi, Divya Budihal, Dominic Sobhani, Oliver Garrod, Paul Atherton — Digital Economics and Artificial Intelligence
This paper investigates whether an AI-powered WhatsApp chatbot can provide cost-effective and superior access to information for teachers in low-connectivity environments like Sierra Leone, compared to traditional web search.
Finance Application
- The paper's insights into cost-effective information delivery in low-connectivity, low-income environments could be applied to household finance.
- An AI chatbot could deliver tailored financial literacy education or micro-insurance product information to underserved populations via low-bandwidth platforms like WhatsApp, reducing data costs for users and increasing access to vital financial knowledge.
- This could improve financial decision-making and uptake of financial services in remote areas, potentially impacting savings rates, debt management, and insurance penetration.
AIChatbotInformation AccessLow-ConnectivityEmerging MarketsCost-EffectivenessFinancial LiteracyHousehold FinanceDigital DivideWhatsApp
Core finding, identification, data
Core Finding
- The study finds that an AI-powered WhatsApp chatbot is significantly more cost-effective and provides higher quality information than traditional web search for teachers in Sierra Leone.
- AI responses consume 3,107 times less bandwidth and are 98% less expensive than loading a web page, even with AI compute costs.
- Independent evaluations show AI responses are rated as more relevant, helpful, and correct, with fewer inaccuracies, compared to web search results.
Identification Strategy
- The study uses a quasi-experimental approach by providing access to an AI chatbot to a sample of teachers and comparing their usage and the quality/cost of AI responses against traditional web search.
- The methodological innovation involves using language models to categorize queries, compare AI responses to web content, and conducting blinded evaluations by independent teachers to rate the quality of responses from both sources, while meticulously measuring bandwidth usage and cost.
Data
The paper uses 40,350 queries submitted by 529 Sierra Leonean teachers to an AI-powered WhatsApp chatbot over 17 months. It also uses data from a random subsample of these queries submitted to google.com.sl, with the top 5 search results scraped and analyzed, and survey data from teachers on internet usage and AI awareness.
Gorkem Turgut Ozer, Brad N. Greenwood, Anand Gopal — Digital Economics and Artificial Intelligence
This paper empirically investigates the social and human costs of legalized online sports betting, focusing on its impact on problem gambling hotline calls and suicide rates in US states.
Finance Application
- This research offers significant arbitrage opportunities for household finance and insurance.
- In household finance, researchers could investigate the impact of online sports betting legalization on household debt levels (e.g., credit card debt, personal loans), bankruptcy filings, and credit scores at the state or zip code level, using a similar DID approach.
- For insurance, the observed increase in suicide rates directly impacts life insurance claims, prompting studies on how online betting legalization affects claim frequency and severity, and how insurers adjust risk models and premiums.
- The broader financial distress could also be linked to local economic indicators and asset performance.
problem gamblingonline sports bettingsuicidedifference-in-differencesnatural experimenthousehold financeinsurancemental healthconsumer behaviordigitizationgamificationpolicy
Core finding, identification, data
Core Finding
- The core finding is that the legalization of online sports betting is strongly correlated with a significant increase in calls to the National Problem Gambling Hotline (29% increase) and a rise in suicide rates (2.75% increase), whereas the legalization of physical sportsbooks shows no such significant correlation.
- This suggests that the digital nature and gamification elements of online betting exacerbate negative social outcomes.
Identification Strategy
- The paper uses a difference-in-differences (DID) approach to exploit the phased state-wide legalization of sports betting across the US following the 2018 Supreme Court decision (Murphy v.
- National Collegiate Athletic).
- States that legalized sports betting (physical or online) are treated, while those that had not yet legalized are controls.
- The DID model includes state and year-month fixed effects, and various robustness checks are employed.
Data
The paper uses data on suicide from the Centers for Disease Control and Prevention (CDC), legislative changes from the American Gaming Association (AGA), and calls to the National Problem Gambling Hotline from the National Council on Problem Gambling (NCPG). Control variables are sourced from the American Community Survey of the U.S. Census, with the panel spanning January 2017 to December 2022 at the state-month level.
Nikhil Agarwal, Alex Moehring, Alexander Wolitzky — Digital Economics and Artificial Intelligence
This paper develops and validates a sufficient-statistic approach for designing human-AI collaboration in classification tasks, accounting for human biases and effort responses, finding that humans under-respond to confident AI predictions due to overconfidence.
Finance Application
- This framework can be directly applied to financial decision-making where humans interact with AI.
- In asset pricing, it could optimize how AI-generated market forecasts or trading signals are integrated by human portfolio managers, identifying when to automate trades (high AI confidence) versus when human oversight is critical, especially if managers are overconfident in their own models.
- For household finance, the approach can design optimal interfaces for robo-advisors, determining how AI advice on savings, investments, or debt management should be presented to retail investors to mitigate under-response or effort crowd-out.
- In insurance, it could refine human-AI collaboration in underwriting or claims processing, automating clear-cut risk assessments or fraud detections while providing human adjusters with optimal AI information for complex, uncertain cases, accounting for human biases in risk perception.
Human-AI CollaborationDecision MakingBehavioral EconomicsInformation DesignExperimental EconomicsOverconfidenceEffort Crowd-OutAI NeglectClassificationFinancial TechnologyRobo-AdvisorsAlgorithmic Trading
Core finding, identification, data
Core Finding
- Humans under-respond to AI predictions and reduce effort when presented with confident AI assessments.
- This under-response primarily stems from human overconfidence in their own signal precision, rather than under-confidence in the AI.
- The optimal human-AI collaboration policy automates cases where the AI is confident and delegates uncertain cases to humans with full disclosure of AI predictions, with the additional benefit of assisting humans being negligible beyond selective automation.
Identification Strategy
- The paper's identification strategy centers on estimating a 'sufficient statistic' function, V(x), which represents the probability of correct human classification given a calibrated AI assessment x.
- This function is estimated from data generated in Stage 1 of an incentivized online fact-checking experiment, where AI assessments are fully disclosed.
- The approach is then validated in Stage 2 by testing optimal and benchmark policies derived from the estimated V(x) in a within-participant experiment.
- A structural model of belief updating is used to decompose observed biases into overconfidence and AI neglect.
Data
The study uses data from an incentivized online fact-checking experiment conducted on the Prolific platform. Participants classify statements from the FEVEROUS database, which provides curated ground-truth labels. AI assessments are generated using OpenAI's GPT-40.
Kristina McElheran, Mu-Jeung Yang, Zachary Kroff, Erik Brynjolfsson — Digital Economics and Artificial Intelligence
This study examines the productivity dynamics of artificial intelligence (AI) in American manufacturing, finding J-curve-shaped effects with initial productivity losses followed by later gains, driven by costly adjustment and varying across firm characteristics.
Finance Application
- The J-curve effect of AI adoption suggests a potential mispricing in asset markets: firms undergoing initial productivity dips due to AI investment might be undervalued by short-term focused investors, creating an arbitrage opportunity for long-horizon investors.
- This could be explored by examining stock returns of AI-adopting firms, especially those with characteristics (young, growth-oriented) predicted to weather the dip better.
- For corporate finance, the costly adjustment (capex, labor shedding) implies significant financing needs and operational risk, impacting capital structure decisions and the pricing of debt and equity, or even the demand for specialized insurance products covering business interruption during technological transitions.
AIProductivityJ-curveTechnology AdoptionManufacturingFirm PerformanceOrganizational ChangeLabor MarketsCapital InvestmentAsset PricingFirm ValuationCorporate FinanceOperational RiskMispricing
Core finding, identification, data
Core Finding
- The paper finds significant initial productivity losses for industrial AI adopters, which are attributed to costly adjustment processes like increased work-in-progress inventory, investment in industrial robots, and labor shedding.
- However, early AI adopters exhibit stronger growth over time, with younger, growth-oriented firms faring better than older incumbents, confirming a J-curve pattern at the micro-level.
Identification Strategy
The study employs three identification strategies: selection-on-observables using rich controls for organizational factors, an Instrumental Variable (IV) approach using the 'no lack of AI expertise' barrier as an instrument, and within-firm fixed effects using panel data to control for time-invariant confounders.
Data
The paper uses two main datasets: the 2021 Management and Organizational Practices Survey (MOPS) linked to the Annual Survey of Manufactures (ASM) for detailed establishment-level data, and a panel of 55,000 manufacturing firms from the 2018 Annual Business Survey (ABS) combined with the Economic Census of Manufacturing (CMF) from 2012 to 2017.
Friedrich Geiecke, Xavier Jaravel — Digital Economics and Artificial Intelligence
This paper introduces and validates an open-source platform for conducting robust, AI-led qualitative interviews at scale, demonstrating its versatility across various social science applications and providing a pipeline for textual analysis.
Finance Application
- This methodology could revolutionize behavioral finance and household finance research by enabling large-scale qualitative data collection.
- For instance, AI-led interviews could probe individual investors' mental models of market volatility, inflation expectations, or specific asset class performance (e.g., real estate, cryptocurrencies), informing theories of asset pricing and investor behavior.
- In household finance, the platform could uncover the subjective factors influencing savings rates, debt decisions, or retirement planning across diverse demographics, providing granular insights into financial well-being and policy effectiveness.
- The textual analysis pipeline could also be adapted to process earnings call transcripts or analyst reports to identify emerging themes and sentiment shifts impacting corporate valuations and market movements.
Qualitative ResearchLarge Language ModelsAI InterviewsBehavioral FinanceHousehold FinanceMental ModelsDecision MakingSurvey MethodologyTextual AnalysisInvestor SentimentRisk PerceptionAsset Allocation
Core finding, identification, data
Core Finding
- The AI-led qualitative interviews, conducted using a simple, adaptable open-source platform and a single LLM agent, reliably elicit people's views at scale.
- They achieve performance ratings comparable to human experts across diverse topics and yield richer, more confident data than open-text fields, enabling the identification of detailed mental models and heterogeneity patterns.
Identification Strategy
- The robustness of the AI-led interviews is assessed through multiple methods: comparisons by trained sociologists to hypothetical human experts (both text-based and face-to-face), and five respondent-based quality metrics (preference for AI vs. human, preference for AI-led vs. open text fields, perceived accuracy of content, confidence in responses, and word count).
- The study also employs an LLM-based pipeline for automated textual analysis and concept coding, with its accuracy validated against human labelers.
Data
The paper uses interview transcripts and open-text field responses from respondents recruited via the Prolific platform. Samples include a representative U.S. population for 'meaning in life' (466 respondents), French voters for 'political views' (422 respondents), U.S. respondents for 'educational/occupational choices' (107 respondents), and U.S. respondents for 'mental models of policies' (800 respondents for Trump administration policies, 390 for inflation causes).
Kunal Handa, Alex Tamkin, Miles R. McCain, Saffron Huang, Esin Durmus, Sarah Heck, Jared Mueller, Jerry Hong, Stuart Ritchie, Tim Belonax, Kevin K. Troy, Dario Amodei, Jared Kaplan, Jack Clark, Deep Ganguli — Digital Economics and Artificial Intelligence
This paper introduces a novel empirical framework using LLM-based classification of millions of AI conversations to measure real-world AI usage patterns across economic tasks and occupations, distinguishing between automation and augmentation.
Finance Application
- This methodology offers a powerful tool for analyzing unstructured data in finance.
- For asset pricing, one could apply a similar LLM-based classification to earnings call transcripts or corporate reports to identify firms' AI adoption strategies (e.g., focus on augmentation vs. automation, specific AI-impacted tasks).
- This could then be linked to firm-level stock returns, volatility, or future growth prospects, potentially creating new factors.
- In household finance, the task-level analysis of AI impact on wages and job zones could inform models of labor income risk, affecting household savings, consumption, and portfolio allocation decisions, especially for different skill groups.
- For insurance, the framework could be adapted to analyze claims data or policyholder communications to detect emerging risks or opportunities related to AI adoption, such as changes in business interruption risk for firms heavily investing in AI automation, or new types of professional liability risks.
Artificial IntelligenceLabor MarketsTask-based ApproachLLM ClassificationEconomic ImpactAutomationAugmentationO*NETUnstructured DataFirm StrategyLabor Income RiskAsset PricingHousehold FinanceInsurance
Core finding, identification, data
Core Finding
- AI usage primarily concentrates in software development and writing tasks, with approximately 36% of U.S. occupations using AI for at least a quarter of their associated tasks, but deep integration (>=75% tasks) is rare (~4% of occupations).
- The usage patterns show a mix of augmentation (57%) and automation (43%), and peak in occupations in the upper quartile of wages (e.g., software developers) but drop off at both very high and very low wage extremes.
Identification Strategy
- The paper's methodological innovation is using a privacy-preserving LLM-based system (Clio, powered by Claude.ai) to classify millions of real-world human-AI conversations.
- This allows for granular mapping of these conversations to specific tasks, occupations, and skills from the U.S.
- Department of Labor's O*NET Database, and further categorizing the nature of AI interaction (automation vs. augmentation).
Data
The primary data source is over four million Claude.ai Free and Pro conversations from December 2024 to January 2025. This conversational data is mapped to the U.S. Department of Labor's O*NET Database, which provides information on tasks, occupations, skills, median wages, and job zones.
Susan Athey, Dean Karlan, Emil Palikot, Yuan Yuan — Digital Economics and Artificial Intelligence
This paper investigates how fixed "type" features (e.g., gender, age) and mutable "style" features (e.g., smiling in a photo) in online profiles influence funding outcomes and disparities on a micro-lending platform, using a combination of observational data, computer vision, and generative AI experiments.
Finance Application
- This research offers a powerful framework for analyzing subjective decision-making and disparities in various financial contexts.
- In credit markets, it could be applied to study how "style" features in loan applications (e.g., perceived professionalism in a photo, language in a personal statement) influence credit decisions and interest rates, even if not predictive of default, potentially creating biases against certain demographic "types." For venture capital or equity crowdfunding, the methodology could investigate how "style" features of founders (e.g., perceived charisma, visual presentation of a pitch deck) affect funding outcomes, and whether generative AI could be used to debias investor perceptions or create more equitable pitching environments.
FintechMicrofinanceCredit MarketsBehavioral FinanceMachine LearningComputer VisionGenerative AIDiscriminationFairnessOnline PlatformsCrowdfundingHousehold FinanceESG
Core finding, identification, data
Core Finding
- The study finds that "style" features, such as smiling or avoiding sunglasses/body-shots, significantly impact funding success on the Kiva micro-lending platform.
- These style choices are often correlated with "type" features (e.g., men smile less than women), exacerbating existing type-based disparities in funding.
- Crucially, style features do not predict loan repayment probability, suggesting lenders are influenced by psychological perceptions (e.g., trustworthiness) rather than financial considerations.
Identification Strategy
- The paper employs a two-step identification strategy.
- First, it uses Augmented Inverse Propensity Weighting (AIPW) on observational data, combined with computer vision, to estimate average treatment effects of style features.
- Second, it conducts randomized survey experiments where generative AI (GANs) creates synthetic profile images that exogenously vary one feature at a time (e.g., smile, gender) to obtain internally valid causal estimates.
- A natural experiment from a Kiva website redesign is also used to check the stability of coefficients.
Data
The research uses publicly available observational data from the Kiva micro-lending platform (500,000+ borrowing campaigns, 2006-2020), augmented with computer vision algorithms (CNNs) to extract over 100 "type" and "style" features from profile images. It also utilizes generative AI (StyleGAN) to create synthetic images for randomized experiments and a deep learning model (Peterson et al., 2022) to estimate psychological traits from images.
Jack W. Fisher — Digital Economics and Artificial Intelligence
This paper quantifies worker welfare in the gig economy by modeling gig work as a usage good, accounting for misperceptions and learning in labor supply and insurance choices.
Finance Application
- The framework for modeling misperceptions and learning about job characteristics (pay, costs, flexibility) can be applied in household finance to understand how individuals make labor supply decisions, affecting income volatility, savings, and debt.
- For insurance, the use of fixed vs. variable premiums to identify misperceptions is directly transferable to usage-based insurance (UBI) or other markets, informing pricing strategies that account for behavioral biases and reduce churn.
- In asset pricing, aggregate misperceptions and churn in the gig economy could influence labor income risk and consumer spending, impacting equity risk premiums for consumer-facing industries or the valuation of gig economy platforms.
Gig EconomyLabor EconomicsBehavioral EconomicsMisperceptionsLearningWelfareInsuranceHousehold FinanceLabor SupplyPolicy Evaluation
Core finding, identification, data
Core Finding
- The paper finds a substantial average monthly surplus for gig workers (£1,066), but misperceptions lead one in six participants to be worse off, explaining high churn.
- Policies like California's Proposition 22, aimed at full-time workers, are shown to likely harm overall worker welfare due to costs passed on as lower wages, especially impacting part-time workers who generate most of the total surplus.
Identification Strategy
- The identification strategy models gig work as a usage good, leveraging workers' endogenous exit decisions and misperceptions about the value of gig work.
- Crucially, it uses the choice between fixed and variable insurance premiums, combined with observed hours worked and dynamic behavior (exit rates, changes in hours), to infer heterogeneous outside options and the speed of learning about true valuations.
Data
The paper uses administrative data from a UK vehicle insurer covering food delivery drivers, including policy choices (hourly vs. monthly premiums), hours worked across platforms, and worker demographics. This is complemented by a survey on worker experiences and expectations.
Guy Aridor, Tevel Dekel, Rafael Jiménez-Durán, Ro’ee Levy, Lena Song — Digital Economics and Artificial Intelligence
This paper provides the first systematic analysis of the magnitude and drivers of election-related content consumption on smartphones during the 2024 U.S. election campaign, using novel, granular data on observed screen content.
Finance Application
- The granular, objective data on cross-app political content consumption could be invaluable for finance.
- Researchers could link this data to household financial records (e.g., brokerage accounts, credit card data) to study how differential exposure to political news (e.g., partisan, low-quality, or from specific platforms) influences investment decisions, risk-taking, or consumption patterns.
- In asset pricing, this data could shed light on how fragmented information environments affect market efficiency, asset prices, and trading volumes for politically sensitive sectors (e.g., defense, energy, tech), especially around political events or celebrity endorsements.
- The variance decomposition method could be adapted to analyze the drivers of information acquisition for financial news, separating individual preferences from platform algorithmic biases.
smartphone datanews consumptionpolitical economysocial mediainformation diffusionheterogeneitynatural experimentvariance decompositionhousehold financeasset pricinginvestor behaviormarket efficiencypolitical riskinformation asymmetrysentiment analysis
Core finding, identification, data
Core Finding
- The median American consumed limited and stable election content, primarily from personalized content apps like social media (less than 10% from dedicated news apps).
- Heterogeneity in consumption is mostly driven by individual differences (e.g., swing-state residents, news app users, celebrity followers) rather than the specific applications used, though app algorithms (e.g., X vs.
- Facebook) also play a role.
Identification Strategy
- The paper uses novel data collected via an SDK embedded in a smartphone app, which passively checks for pre-specified keywords on-screen every three seconds across all applications.
- This allows for objective, cross-app, fine-grained measurement of content consumption.
- A variance decomposition (AKM-style) is employed to disentangle the relative importance of individual-specific effects versus application-specific effects in driving content exposure.
- The Taylor Swift endorsement of Kamala Harris serves as a quasi-natural experiment to study the impact of non-political figures on political news consumption.
Data
The primary data comes from Screenlake, a proprietary dataset of observed keyword occurrences across all apps on individuals' smartphones for several thousand Americans during the 2024 U.S. election campaign. This includes 532 distinct election-related keywords and thousands of control keywords. Ground-truth labels for keyword classification are derived from Fox News and New York Times articles, and Reddit posts. The study also uses demographic data and app usage benchmarks from Google Play Store, industry reports, and a Prolific sample.
Malika Korganbekova, Aliya Korganbekova, Alibek Korganbekov, Vinny DeGenova, Yasaman Khazaeni, Cole Zuber — Digital Economics and Artificial Intelligence
This paper uses two large-scale field experiments on a major e-commerce platform to evaluate how platform-side interventions, specifically algorithmic re-ranking, can reduce shipping costs and environmental impact without negatively affecting consumer outcomes.
Finance Application
- This research offers significant insights for asset pricing by identifying a novel, algorithm-driven source of operational efficiency and ESG performance for e-commerce firms.
- Researchers could investigate whether firms that successfully implement such demand-side algorithmic nudges exhibit higher stock valuations, lower cost of capital, or superior ESG ratings, potentially leading to a 'logistics efficiency factor' in asset pricing models.
- For insurance, the quantified impact of shipping strategies on incident rates and returns provides a basis for designing dynamic supply chain insurance products, offering lower premiums to retailers adopting efficient re-ranking algorithms that reduce damage and delays.
- In household finance, the findings could inform how algorithmic nudges influence household budgeting and consumption decisions, particularly for large purchases, by altering perceived costs and environmental trade-offs.
e-commercelogisticssupply chainshipping costsconsumer behaviorplatform algorithmsfield experimentsESGfirm valuationhousehold consumptioninsurancerisk managementasset pricing
Core finding, identification, data
Core Finding
- The core finding is that simply removing fast-shipping promises backfires, leading to increased shipping distances, costs, and product returns.
- However, an alternative approach of algorithmic re-ranking, which prioritizes well-liked and cheaper-to-ship products, significantly reduces shipping distances, increases centralized fulfillment, lowers shipping costs by up to 7.3%, and reduces incident rates, all without degrading conversion rates or increasing product returns.
Identification Strategy
- The identification strategy relies on two large-scale randomized controlled field experiments conducted on a major e-commerce platform.
- The first experiment randomly removed 25-75% of fast-shipping badges and associated ranking boosts for treated consumers.
- The second experiment randomized consumers into control, 'Substitute' (expensive items replaced with cheaper-to-ship substitutes), and 'Substitute+Re-Rank' (same substitution, but re-ranked to prioritize lower shipping costs) groups.
Data
The paper uses rich, proprietary data from Wayfair, including consumer clickstream data, transaction and shipping records (with FedEx/UPS tracking), detailed product characteristics, seller characteristics, and granular data on prices, wholesale costs, shipping costs, product returns, and incident rates.
Raveesh Mayya, Rohit Aggarwal, Harpreet Singh — Digital Economics and Artificial Intelligence
This paper demonstrates that a structured 'Metadata Prompting' approach significantly mitigates language barriers for non-native English-speaking programmers using Generative AI tools, improving productivity and user preference.
Finance Application
- The finding that structured 'metadata prompting' significantly reduces cognitive load and performance disparities for non-native English speakers interacting with AI has direct applications in household finance and insurance.
- AI-powered financial advisors or fintech platforms could implement structured input interfaces (e.g., dropdowns, specific fields for 'InvestmentGoal', 'RiskTolerance', 'IncomeSource') instead of free-form natural language.
- This would enable non-native English-speaking households to more effectively articulate their financial needs and understand complex products, reducing financial exclusion and improving financial literacy.
- Similarly, insurance companies could use structured prompting for AI-assisted claims processing or policy selection, ensuring clarity and accuracy for diverse customer bases and potentially reducing information asymmetry in complex financial contracts.
Generative AILarge Language ModelsProductivityLanguage BarriersDigital DivideField ExperimentHuman-AI InteractionPromptingCognitive LoadFintechHousehold FinanceInsuranceGlobal MarketsInformation Asymmetry
Core finding, identification, data
Core Finding
- Programmers with low English proficiency experience significantly lower performance (higher task abandonment, longer completion time, lower code accuracy) when using traditional natural language prompting with GenAI.
- However, implementing 'Metadata Prompting' (a JSON-like structured input) effectively neutralizes these language-based disparities, making performance metrics statistically insignificant across proficiency levels.
- All developers, particularly those with lower English proficiency, strongly prefer the structured prompting method.
Identification Strategy
- The study employs a 2x2 crossover field experiment embedded in a recruitment process at a technology firm in India. 965 candidates were randomized to complete two coding tasks using either traditional or metadata prompting, with the order of tasks and prompting techniques also randomized.
- Baseline English proficiency (GRE-style test) and coding ability were assessed, and task requirements were communicated in Hindi to control for understanding, isolating the effect of prompting technique on AI interaction.
Data
The paper uses primary data from a field experiment involving 965 full-stack engineer candidates in India. Data includes individual-level characteristics (English proficiency, coding ability, academic performance, programming experience, AI tool usage, gender, education) and task-level outcomes (completion time, code accuracy measured by unit tests, task abandonment rates).
Yuval Lidany — Digital Economics and Artificial Intelligence
This paper quantifies the causal effect of cross-product compatibility on consumer willingness to pay for smartphones and analyzes its implications for market power, welfare, and cross-market mergers in the smartphone and laptop industries.
Finance Application
- The concept of 'ecosystems' and 'lock-in' is highly relevant to financial services, where institutions offer bundled products (e.g., banking, investment, insurance).
- This paper's methodology could be applied in household finance to measure consumers' willingness to pay for integrated financial platforms or seamless data transfer between different financial products from the same provider.
- This could inform how financial institutions leverage cross-product compatibility to create market power and how regulators should evaluate mergers between complementary financial service providers (e.g., a bank acquiring a wealth management firm) in terms of consumer welfare and market concentration.
Industrial OrganizationConsumer BehaviorEcosystemsCompatibilityLock-inMarket PowerMergers and AcquisitionsWillingness to PayExperimental EconomicsStructural ModelDemand Estimation
Core finding, identification, data
Core Finding
- The study finds that compatibility significantly increases consumers' willingness to pay for smartphones by 9% of the retail price.
- Open ecosystems increase consumer surplus but have varying effects on firm profits and market concentration depending on the hardware quality gap between competitors.
- A cross-market merger between Samsung and HP reduces smartphone market concentration but raises Samsung prices, disadvantaging consumers who value compatibility less.
Identification Strategy
- The causal effect of compatibility is identified using a novel experimental design where participants' willingness to pay (WTP) for a smartphone is elicited conditional on being awarded a laptop of varying brands, thus exogenously varying compatibility levels.
- This experimental WTP difference, net of a brand-matching fixed effect, is then used as a micro-moment in a structural demand model to discipline consumer heterogeneity.
Data
The paper uses three data sources: a novel experiment to elicit WTP for compatibility, repeated cross-sectional market data from IDC's Tracker Database (2018-2023) on smartphone and laptop sales and characteristics, and a proprietary survey on product ownership and demographics.
Namrata Kala, Elizabeth Lyons — Digital Economics and Artificial Intelligence
This paper uses a randomized control trial to investigate how digital worker surveillance and the managerial justification for its use causally impact worker performance and job satisfaction in an online labor market.
Finance Application
- The core insight—that the *justification* of monitoring, not just its presence, critically impacts performance and satisfaction—is highly applicable to finance.
- In asset management, new monitoring tools for portfolio managers, if implemented without clear performance-based justification, could reduce alpha generation or increase turnover among top talent.
- For household finance, gig economy platforms could improve worker retention and effort by transparently justifying their extensive monitoring (e.g., for optimizing routes or ensuring fair pay).
- In insurance, usage-based insurance (UBI) or wellness programs could see higher adoption and more positive behavioral changes if policyholders understand *why* their data is being monitored (e.g., to reduce premiums for safe driving or healthy lifestyles), rather than perceiving it as arbitrary oversight.
Digital surveillanceManagerial clarityPerformanceJob satisfactionRemote workRandomized control trialGig economyBehavioral financeCorporate governanceInformation asymmetryInsuranceAsset management
Core finding, identification, data
Core Finding
- Digital surveillance itself does not significantly affect worker performance on average.
- However, when managers fail to explain the presence or elimination of digital surveillance based on worker performance, it significantly reduces worker output (by approximately 17%).
- The study suggests that worker perceptions of managerial capabilities, rather than monitoring alone, drive performance, and that removing surveillance as a reward for good performance is not valued relative to pay increases.
Identification Strategy
- The study employs a randomized control trial (RCT) on UpWork, an online labor market.
- Workers are initially monitored for two hours to establish baseline productivity (high or low).
- Subsequently, they are randomly assigned to treatment groups that vary both the presence of digital surveillance and whether its use or removal is explicitly justified based on their baseline productivity, allowing for causal inference.
Data
The paper uses data from 434 remote workers hired on UpWork to perform an 8-hour data entry task. Data includes worker characteristics from UpWork profiles (e.g., geographic location, gender, advertised wage), worker productivity (number of correct cell entries), compliance with surveillance, and post-task survey responses measuring job satisfaction and willingness to accept future jobs.
Yikun Jiang — Digital Economics and Artificial Intelligence
This paper uses a large-scale field experiment on Stack Overflow to causally identify how non-monetary incentives, specifically social recognition and instrumental privileges, drive user contributions to online knowledge platforms.
Finance Application
- The findings on social and instrumental motivation for content contribution are highly applicable to financial markets.
- In asset pricing, this framework could explain incentives for retail investors or financial analysts to share investment insights on platforms like StockTwits or Seeking Alpha; social validation (likes, followers) or 'expert' badges could drive information production, impacting price discovery and market efficiency.
- In household finance, gamified apps could use non-monetary rewards (e.g., 'saving streaks,' 'investment milestones') or social leaderboards to motivate better financial behaviors, especially for less experienced users.
- For insurance, peer-to-peer models could incentivize safer behavior through social recognition for low claims or instrumental privileges like reduced deductibles, fostering trust and participation.
Field ExperimentSocial MotivationInstrumental MotivationReputationUser-Generated ContentNon-monetary IncentivesOnline PlatformsInformation ProductionBehavioral Economics
Core finding, identification, data
Core Finding
- A single anonymous upvote, increasing reputation by 10 points, significantly boosts an individual's probability of contributing additional answers by approximately 15% and the number of answers by about 6%, with effects lasting over four months.
- The impact is most pronounced for users with low-to-moderate experience or reputation, and those nearing privilege thresholds.
- Structural estimates reveal that social motivation plays a more central role than instrumental motivation in driving contributions, and platform strategies enhancing social recognition can increase contributions by 15-25%.
Identification Strategy
- The study employs a large-scale field experiment on Stack Overflow, involving 12,182 individuals.
- The treatment consists of exogenously assigning a single anonymous upvote to a recent eligible answer, thereby increasing the recipient's reputation by 10 points.
- This randomized intervention allows for causal inference by shifting both social and instrumental motivations, and subsequent behavior is tracked daily for several months.
Data
The paper uses comprehensive daily behavioral data from 12,182 individuals on Stack Overflow, a leading online question-and-answer platform for programmers. Data was collected over four and a half months (August 2023 - January 2024) and tracked until July 2024, including information on answers, questions, comments, profile details, reputation points, and voting behavior.
Leon Musolff — Digital Economics and Artificial Intelligence
This paper examines the impact of algorithmic pricing tools on price dynamics, competition, and tacit collusion in e-commerce using a novel dataset from Amazon Marketplace.
Finance Application
- The findings on algorithmic price wars and tacit collusion could be directly applied to high-frequency trading (HFT) and market making in financial markets.
- Algorithmic 'resetting' strategies could explain patterns in bid-ask spreads or flash crashes, where algorithms might coordinate to temporarily widen spreads or reset prices.
- This could also inform research on the impact of algorithmic trading on market efficiency, liquidity, and the potential for implicit collusion in asset pricing, or even the pricing of complex insurance products where algorithms optimize premiums.
Algorithmic PricingE-commercePrice WarsTacit CollusionEvent StudyMarket MicrostructureHigh-Frequency TradingAlgorithmic TradingMarket EfficiencyGame Theory
Core finding, identification, data
Core Finding
- Firms adopting repricing tools initially drop prices by 16.93%, leading to a 9.67% market price decrease.
- However, 'resetting' strategies, which involve periodically raising prices, effectively coax competitors to increase prices by 11.4% in less competitive markets.
- The paper finds that an equilibrium in delegated strategies, where the average price approaches the monopoly price, better explains observed price cycling than traditional Edgeworth cycles.
Identification Strategy
- The study employs an event study design around the activation of repricing tools by third-party sellers on Amazon.
- It exploits the fact that signing up for a repricer and activating it are separate actions, allowing for the estimation of dynamic treatment effects using a specification with offer and day fixed-effects, and robustness checks with the Sun and Abraham (2021) approach.
Data
The paper uses a novel, proprietary e-commerce dataset from a repricing company, covering price change notifications for Amazon third-party sellers from 2018-2020. It also uses historical Keepa data for the 10,000 best-selling products to analyze long-term trends in price wars and cycling.
Saharsh Agarwal, Uttara M. Ananthakrishnan — Digital Economics and Artificial Intelligence
This paper investigates the causal impact of TikTok adoption on users' screen time, sleep patterns, and app substitution using a high-frequency, individual-level app usage dataset.
Finance Application
- The findings on increased screen time and sleep deprivation, particularly among heavy users, could be applied to household finance to study the impact on financial decision-making, risk tolerance, and susceptibility to financial scams or misinformation.
- For insurance, granular phone usage and inferred sleep data could inform novel underwriting models for health, life, or auto insurance, as sleep patterns are linked to health and cognitive function.
- In asset pricing, the 'attention economy' and substitution effects could be used to model retail investor attention shifts between traditional financial news and 'finfluencer' content on platforms like TikTok, potentially predicting trading volumes or meme stock phenomena.
Social MediaBehavioral FinanceHousehold FinanceSleepScreen TimeDigital Well-beingConsumer BehaviorAttention EconomyRisk-takingDecision-makingInsuranceFintech
Core finding, identification, data
Core Finding
- TikTok adoption significantly increases overall screen time for heavy users (221 minutes/week), primarily due to TikTok itself, while simultaneously causing a strong substitution effect away from other social media and gaming apps.
- This increased screen time is most pronounced during late-night hours, leading to delayed bedtimes (85 minutes/week for heavy users) and a moderate reduction in sleep duration (36 minutes/week).
Identification Strategy
- The study employs a Difference-in-Differences (DiD) framework, leveraging the staggered adoption of TikTok at the individual level.
- It addresses endogeneity concerns by comparing early adopters to matched late adopters (with similar usage intensity), excluding the COVID-19 period, and using dynamic lead and lag models to rule out pre-existing trends.
Data
The paper uses a comprehensive mobile usage dataset from approximately 250,000 opt-in Android users in the US, tracking real-time app usage activity (opening/closing times to the millisecond) for all apps. Sleep patterns (duration, bedtime, wake-up time) are inferred from the longest uninterrupted screen-off sequences.
Lea Bignon — Digital Economics and Artificial Intelligence
This paper develops and estimates a structural model of demand and supply for insulin to analyze how Continuous Glucose Monitors (CGMs) impact pharmaceutical demand, pricing, and innovation incentives by providing patient-specific information and accelerating physician learning.
Finance Application
- The paper's insights into how high-frequency, individual-specific data from digital devices (CGMs) reshapes demand, pricing, and innovation incentives in healthcare has direct parallels to the insurance industry.
- Wearable health technologies (smartwatches, fitness trackers) provide insurers with granular, real-time data on policyholders' health and behavior, reducing information asymmetry.
- This could lead to more dynamic, risk-based pricing of health or life insurance premiums, incentivize product innovation (e.g., wellness programs, personalized policies), and alter competition among insurers based on their ability to leverage such data.
- The structural model could be adapted to analyze how the adoption of such wearables affects insurer learning, premium setting, and the profitability of different insurance product designs.
HealthcareDigital HealthInformation AsymmetryLearningTechnology AdoptionPharmaceutical MarketsPricingInnovationStructural ModelInsuranceWearable TechnologyRisk AssessmentDynamic Pricing
Core finding, identification, data
Core Finding
- CGMs' patient-specific information steers insulin demand towards newer products, fostering physician learning, with limited information spillover to nonusers.
- Manufacturers of drugs benefiting from higher perceived quality due to newly observable attributes could negotiate higher prices, and these attributes shift the relative profitability of drug innovation strategies, thereby shaping future pharmaceutical innovation.
Identification Strategy
- The identification strategy leverages a policy change in France that expanded CGM coverage, boosting adoption among insulin-dependent diabetic patients.
- A structural model of demand and supply is estimated, incorporating patient-specific learning through digital devices, physician-level learning about new products, and price bargaining between manufacturers and a regulator.
- CGM adoption is assumed exogenous to insulin choice, and the model uses a flexible specification for patient-product match values and GMM for supply estimation.
Data
The paper uses comprehensive medical claims data from the French health insurance system from 2015-2021. This includes records of prescription reimbursements, patient and prescriber IDs, prescription dates, drug characteristics, patient demographics, and medical conditions. CGM adoption and attrition are inferred directly from these claims data.
Lea Bignon — Industrial Organization
This paper develops and estimates a structural model of demand and supply for insulin to analyze how Continuous Glucose Monitors (CGMs) influence patient choices, physician learning, and pharmaceutical pricing, ultimately shaping innovation incentives in the insulin market.
Finance Application
- The insights on how new information technologies reduce information friction and accelerate learning about product performance are highly relevant for insurance and fintech.
- Insurers could use data from new digital health devices (e.g., smartwatches, genetic tests) to offer personalized health or life insurance premiums, similar to how CGMs influence insulin pricing.
- In fintech, the framework could model how AI-driven investment platforms or personalized financial planning apps accelerate household learning about investment products, impacting demand and the profitability of financial innovation.
- The regulatory bargaining framework could also be applied to understand how financial regulators negotiate fees or disclosure requirements with financial institutions in the presence of new information technologies.
Digital HealthInformation AsymmetryLearningPharmaceuticalsInsulin MarketCGMPricingInnovation IncentivesConsumer WelfareStructural ModelNatural ExperimentInsuranceFintechHousehold FinanceAsset PricingRegulatory Economics
Core finding, identification, data
Core Finding
- CGMs' patient-specific information steers insulin demand towards newer products, fostering physician learning, especially for drugs that perform well on newly observable attributes (e.g., low overnight glucose levels).
- This allows manufacturers of such drugs to negotiate higher prices (up to +4.4%), shifting the relative profitability of innovation strategies and increasing consumer welfare for users.
Identification Strategy
- The paper leverages a natural experiment: a policy change in France that expanded CGM coverage, boosting adoption.
- It identifies the impact by comparing insulin choices of similar patients with and without CGMs.
- To address endogenous CGM adoption, it uses a flexible specification for patient-product match values, controlling for product-specific unobserved heterogeneity and assuming remaining within-group heterogeneity is independent of CGM drivers.
- Physician learning is modeled dynamically based on experience signals from patients, and the pricing model uses a Nash bargaining framework between manufacturers and the regulator, internalizing demand-side learning.
Data
The paper uses comprehensive medical claims data from the French health insurance system (2015-2021), which includes patient and prescriber IDs, prescription dates, drug characteristics, and inferred CGM adoption and attrition for the entire population.
Leon Musolff — Industrial Organization
This paper empirically investigates how algorithmic pricing tools affect price dynamics, competition, and tacit collusion among third-party sellers on Amazon Marketplace.
Finance Application
- The findings on algorithmic price wars, 'resetting' strategies, and tacit collusion have direct implications for financial markets.
- Researchers could investigate whether similar algorithmic behaviors lead to supra-competitive spreads or price cycling in high-frequency trading (HFT) or algorithmic trading in less liquid asset classes.
- In household finance, the paper's insights could predict how dynamic pricing of retail financial products (e.g., credit card rates, insurance premiums) might evolve, influencing consumer behavior and welfare.
- Regulators could use this evidence to design market mechanisms that prevent algorithmic collusion in financial markets, especially in less transparent OTC or bond markets.
Algorithmic PricingPrice WarsTacit CollusionE-commerceDynamic PricingMarket DesignEvent StudyMarket MicrostructureAlgorithmic TradingHousehold FinanceRegulationIndustrial Organization
Core finding, identification, data
Core Finding
- Firms adopting algorithmic repricing tools initially drop their prices by 16.93%, causing market prices to fall by 9.67%.
- However, 'resetting' strategies employed by these algorithms effectively induce competitors to raise prices, leading to an 11.4% increase in both competitor and market prices in less competitive markets.
- The observed price cycling, while resembling Edgeworth cycles, is better explained by a model of equilibrium in delegated strategies, which suggests average prices can reach near monopoly levels.
Identification Strategy
- The study employs two event studies.
- First, a 'repricer activation' event study exploits the delay between a merchant signing up for a repricer and actually activating it for an offer, using this activation as a treatment event.
- Second, a 'resetting strategy activation' event study examines the impact of merchants turning on these specific price-raising strategies.
- Both analyses use two-way fixed effects and are robust to alternative estimators like Sun and Abraham (2021).
Data
The paper utilizes a proprietary, high-frequency e-commerce dataset from a repricing company, covering price change notifications for Amazon third-party sellers from August 2018 to March 2020. This dataset includes detailed information on listing prices, shipping costs, Buybox ownership, shipping details, and seller characteristics for over 100,000 products and 55,000 merchants.
Yasmine van der Straten — Household Finance
This paper develops a macro-financial model to analyze how climate change affects housing markets, mortgage credit, and private adaptation, highlighting the role of financial constraints and the potential for rental markets to improve resilience.
Finance Application
- This paper's findings are directly applicable to asset pricing by informing how climate risk should be capitalized into real estate values, potentially leading to higher prices due to scarcity.
- For household finance, it highlights how credit constraints create an 'adaptation gap' that worsens wealth inequality, suggesting research into targeted financial products or policies.
- In insurance, the explicit modeling of moral hazard from climate risk insurance provides a framework for designing optimal insurance contracts that balance risk transfer with adaptation incentives, and could be used to price flood insurance more accurately.
Climate RiskHousing MarketsMortgage CreditAdaptationFinancial ConstraintsWealth InequalityInsuranceReal EstateMacrofinance
Core finding, identification, data
Core Finding
- Climate change, by degrading land and increasing scarcity, paradoxically raises house prices over time, despite weakening housing demand.
- While price signals can incentivize efficient adaptation in frictionless markets, credit-constrained households underinvest in adaptation, exacerbating wealth inequality and accelerating habitat loss.
- A shift to a rental model with unconstrained landlords, or the moral hazard inherent in insurance, can lead to more efficient adaptation.
Identification Strategy
- The paper is a theoretical model with overlapping generations, simulating the long-run effects of climate change on housing and financial markets.
- It uses counterfactual analysis based on IPCC climate scenarios and Florida coastal data to quantify welfare effects under different adaptation and credit constraint scenarios.
Data
The paper uses IPCC (2023) climate change scenarios (SSP1-1.9 to SSP3-7.0), data on Florida coastal housing at risk of flooding (Dahl et al., 2018; Bernstein et al., 2019), NASA Sea Level Projection Tool (Garner et al., 2021), and U.S. economy data from the Federal Reserve Board, BEA, and U.S. Census Bureau for parameterization. It also uses the Zillow Home Value Index (ZHVI) for Miami-Dade County.
Trevor Bakker, Stefanie DeLuca, Eric English, James S. Fogel, Nathaniel Hendren, Daniel Herbst — Household Finance
This paper constructs new population-level linked administrative data to analyze how credit access and repayment vary by race, class, and hometown in the U.S., revealing significant disparities rooted in childhood environments and algorithmic biases.
Finance Application
- The findings on algorithmic bias in credit scoring could be directly applied to insurance underwriting, investigating whether similar biases lead to discriminatory pricing or access for auto, home, or life insurance products.
- The causal link between childhood environments and repayment behavior suggests new avenues for household finance research on early financial education interventions or social capital-building programs to improve long-term financial stability and reduce default risk across various financial products.
- Furthermore, persistent credit constraints and differential access to credit by demographic groups could influence household portfolio allocation and risk-taking, offering insights into asset demand and potential mispricing in asset markets where participation is stratified by credit access.
credit accesscredit scoresalgorithmic biasracial disparitiesclass disparitieschildhood environmentsocial capitalfinancial literacyhousehold financedelinquencyrepaymentadministrative datacausal inferencemortgagestudent loansauto loansinsurance underwritingasset pricing
Core finding, identification, data
Core Finding
- Significant and persistent disparities in credit scores and credit access exist by race, class, and hometown, emerging early in life.
- Credit scores exhibit both calibration bias (understating delinquency for disadvantaged groups) and balance bias (assigning lower scores to disadvantaged groups even with perfect repayment).
- These gaps are not fully explained by income or wealth but are strongly predicted by parental credit scores and childhood exposure to local environments, suggesting a causal role for social capital and learned financial behaviors.
Identification Strategy
- The paper employs a 'childhood movers design' to identify the causal effects of childhood exposure to places on later-life repayment outcomes, exploiting variation in the timing of parents' cross-county moves.
- It also rigorously defines and measures 'calibration' and 'balance' biases in credit scoring algorithms using observed delinquency outcomes and credit scores, and uses OLS regressions to control for various factors to isolate the impact of race, class, and hometown.
Data
The study utilizes new population-level linked administrative data combining credit bureau records with Decennial Censuses, Census surveys, and federal income tax returns for over 25 million individuals. It also incorporates data from the Survey of Income and Program Participation (SIPP), the Survey of Consumer Finances (SCF), a bespoke Prolific Academic survey, and the National Financial Capability Study (NFCS).
Jess Cornaggia, Kimberly J. Cornaggia, Han Xia — Household Finance
This paper investigates how financial shocks, specifically natural disasters, differentially impact the academic performance and human capital investment of male and female college students.
Finance Application
- This research offers significant insights for household finance and insurance.
- In household finance, the finding that women exhibit greater resilience in human capital investment during financial shocks, especially for lower-income families, suggests that financial literacy and emergency savings programs could be tailored to leverage or enhance these gender-specific strengths.
- For insurance, the differential impact on human capital (e.g., males discontinuing education) highlights a potential market for education-related insurance products (e.g., tuition protection, student loan payment protection) that could be designed with gender-specific risk profiles in mind.
- The role of self-efficacy as a non-cognitive trait influencing financial resilience also opens avenues for behavioral finance research on how such traits affect investment decisions and risk-taking during adverse economic events.
Gender GapFinancial ShocksNatural DisastersHuman CapitalHousehold FinanceResilienceSelf-efficacyEducationProductivityBehavioral FinanceInsurance
Core finding, identification, data
Core Finding
- The study finds a 'reverse gender gap' where female students are more resilient to financial shocks than males in terms of academic performance and college enrollment, particularly in non-STEM fields and for lower-income families.
- Males experience greater productivity losses, lower grades, and are more likely to discontinue college enrollment following such shocks, a difference partly explained by self-efficacy.
Identification Strategy
- The paper employs a triple-differences (DDD) empirical design.
- It compares treated students (from counties affected by financially disruptive natural disasters) to control students (from the same university, same courses, same originating state, but unaffected by disasters), separately for males and females, and then contrasts these gender-specific differences.
- This approach controls for various confounding factors related to student, course, and institutional characteristics.
Data
The primary data sources are the NCES Beginning Postsecondary Students Longitudinal Study (BPS12) for detailed student demographics, academic transcripts, family financial information (FAFSA), and enrollment status. This is supplemented by Federal Emergency Management Agency (FEMA) datasets for natural disaster declarations and associated damage costs, and the National Student Loan Data System (NSLDS) for federal loan histories.
Christopher Eaglin, Apoorv Gupta, Filippo Mezzanotti, Jonathan Zinman — Household Finance
This paper uses a randomized controlled trial (RCT) on vehicle-collateralized small business loans in South Africa to study how liquidity constraints and moral hazard affect borrower responses to loan modifications.
Finance Application
- This research offers direct insights for household finance and fintech lending.
- Lenders and policymakers can design more effective loan modification programs for distressed consumers (e.g., auto loans, mortgages) by targeting payment reductions to high-equity, liquidity-constrained borrowers, and debt write-downs to less liquidity-constrained ones.
- The novel use of GPS data to proxy entrepreneurial effort could be adapted by fintech lenders to dynamically assess and price credit risk, or to offer personalized loan terms and modifications in asset-backed lending markets where asset utilization reflects borrower commitment.
loan modificationdebt restructuringentrepreneurial effortliquidity constraintsmoral hazarddebt overhangcollateralized debtrandomized controlled trialsubprime lendingGPS datahousehold financesmall business lendingfintech
Core finding, identification, data
Core Finding
- The study finds that for distressed small business loans collateralized by vehicles, liquidity constraints are the primary driver of default.
- Payment reduction (liquidity relief) significantly improves repayment performance and increases entrepreneurial effort, especially for borrowers with higher baseline equity in their collateral.
- Debt reduction (debt overhang relief) is generally ineffective for most borrowers but can improve outcomes for those who are less liquidity-constrained.
Identification Strategy
- The paper employs an RCT that randomizes nearly all of a publicly-traded lender's poorly performing loans into three arms: a control group receiving standard modification (capitalizing arrears, extending maturity), a payment reduction treatment (further maturity extension to lower monthly payments), or a debt reduction treatment (interest write-down to reduce total debt).
- It also uses a machine learning approach to identify less liquidity-constrained borrowers for heterogeneous treatment effect analysis.
Data
The study uses administrative data on vehicle-collateralized debt from a publicly-traded lender, unusually rich administrative data on entrepreneurial effort derived from GPS devices installed in the financed vehicles (measuring driving activity), and credit bureau data (Experian South Africa) to track outside borrowing and credit access.
Jacqueline Lane, Simon Friis, Tianxi Cai, Michael E. Menietti, Griffin Weber, Eva C. Guinan — Science of Science Funding
This paper uses a field experiment to show that focusing evaluator attention solely on project feasibility leads to lower scores, different rankings, and better prediction of future productivity compared to multi-criteria evaluations.
Finance Application
- The core finding that evaluation format significantly alters perceived feasibility and predictive power has direct implications for venture capital, private equity, and corporate innovation.
- VC firms could implement a 'feasibility-only' stage for early-stage startup evaluations to rigorously identify critical flaws and inconsistencies before committing capital, potentially reducing failure rates.
- Similarly, corporate R&D departments could use this approach for internal project greenlighting, especially for high-risk, high-reward initiatives, to ensure robust implementation plans.
- In insurance, this could inform underwriting processes for complex, novel risks, where a focused feasibility assessment might better predict project completion and risk exposure than a broader evaluation.
innovationevaluationdecision-makingfield experimentfeasibilityattention allocationcognitive biasventure capitalprivate equitycorporate innovationgrant fundingpeer reviewproject managementresource allocation
Core finding, identification, data
Core Finding
- A field experiment demonstrates that when experts evaluate early-stage projects, a single-criterion (feasibility-only) review format leads to significantly lower feasibility scores and a substantial reordering of project rankings compared to multi-criteria (impact, novelty, feasibility) reviews.
- This focused evaluation also proves more predictive of subsequent research productivity (publications).
- The mechanisms driving these differences are critical flaw discovery (broader identification of sub-criteria issues) and coherence scrutiny (deeper integration of component compatibility).
Identification Strategy
- The paper employs a randomized controlled field experiment where expert evaluators were randomly assigned to either a multi-criteria review condition (assessing impact, novelty, and feasibility) or a single-criterion, feasibility-only review condition for grant proposals.
- This exogenous variation in evaluation format allows for causal inference on how attention allocation affects assessment outcomes, with a triple-blind setup to minimize bias.
Data
The study uses grant proposals submitted to a U.S. medical school's translational science program, evaluator scores (on a 4-point scale for impact, novelty, and feasibility), qualitative comments from evaluators (coded using a human-LLM hybrid approach), and subsequent publication data (2023-2024 publications by principal investigators, matched via keywords).
Christopher Esposito — Science of Science Funding
This paper assesses the resilience of the U.S. knowledge supply chain by tracing multi-generational citation paths from NSF-funded research to U.S. patents and simulating the impact of cross-border knowledge flow restrictions.
Finance Application
- The findings suggest that firms heavily reliant on global scientific knowledge (identifiable through patent citations to foreign papers or authors) may face increased geopolitical risk, which could be incorporated into asset pricing models as a "knowledge supply chain risk factor." Companies with a higher proportion of their innovation paths passing through international intermediaries might experience lower innovation productivity and longer R&D cycles under protectionist policies, impacting their future earnings and investment decisions.
- Insurers could develop new products to hedge against "knowledge flow disruption risk" for R&D-intensive companies, covering losses from delayed patenting or increased R&D costs due to international scientific collaboration barriers.
innovationgeopolitical risksupply chainpatentsscientific collaborationR&Dknowledge flowasset pricingcorporate financeinsurance
Core finding, identification, data
Core Finding
- The paths linking NSF-funded research to U.S. patents rely heavily on foreign contributions, with 52% of intermediary papers produced outside the U.S.
- Restricting knowledge flows reduces connectivity, extends path length, lowers innovation productivity, and traps promising knowledge trajectories outside the U.S., leading to significant economic loss in terms of NSF investments.
Identification Strategy
- The paper constructs shortest-path networks connecting NSF-funded papers to U.S. patents via citations.
- It then simulates disruptions by probabilistically removing citation edges that cross the U.S. border (from 0% to 100% exclusion rates) and allows the network to reconfigure, comparing outcomes (path length, innovation productivity, captured paths) under these various restriction scenarios to the status quo.
Data
The paper uses patent records from USPTO (PatentsView) and PatStat, scientific publication data from OpenAlex, author affiliation information (including location) from SciSciNet, and NSF funding acknowledgments from SciSciNet. It focuses on "triadic" patents (USPTO, JPO, EPO) that cite 5+ scientific articles.
David Popp, Daniel Acuna, Myriam Gregoire-Zawilski, Lizhen Liang — Science of Science Funding
This paper uses topic modeling and matched controls to causally evaluate how government funding influences the research topics pursued by academic scientists in clean energy.
Finance Application
- This methodology offers a powerful framework for finance research.
- In asset pricing, topic modeling could analyze academic finance papers, financial news, or corporate filings to identify emerging themes (e.g., climate risk, AI in finance).
- A matched control event study could then assess how specific funding initiatives (e.g., grants from large asset managers, government agencies, or foundations) causally influence the direction of academic finance research, potentially revealing how funding shapes financial innovation or market efficiency.
- For household finance, it could study how funding from consumer protection agencies or financial literacy programs impacts research on household financial decisions, identifying potential biases or gaps in knowledge generation.
- In insurance, the method could track how funding from industry or regulators influences research in actuarial science or risk modeling, particularly for emerging risks like cyber or climate change, and its impact on new product development.
topic modelingcausal inferenceevent studymatched controlsresearch fundingresearch directioninnovationtext analysisacademic researchscience policyasset pricinghousehold financeinsurance
Core finding, identification, data
Core Finding
- Government funding, particularly from targeted and open energy-specific programs, significantly shifts the research focus of funded scientists towards the grant topic.
- This change is persistent over time and is not observed in matched control groups, with ARPA-E open calls showing the largest shifts by attracting scientists whose prior research was dissimilar to the grant topic.
Identification Strategy
- The study employs a quasi-experimental design using matched controls and an event study regression.
- It leverages topic modeling (Latent Dirichlet Allocation, TF-IDF, NMF) to quantify the similarity of researchers' publication portfolios to grant abstracts and Funding Opportunity Announcements (FOAs) over time.
- Control scientists are meticulously matched based on their prior research trajectory, career stage, publication productivity, and institutional quality, ensuring they were equally plausible candidates for funding.
- The event study regression includes researcher-by-grant fixed effects and cohort-specific year fixed effects to control for pre-treatment differences and time-varying trends.
Data
The paper utilizes the Dimensions database for 146 million scientific publications, data on 699 awards from select Department of Energy (DOE) and National Science Foundation (NSF) programs between 2009-2016, and abstracts of funded projects and FOAs. Grant information is sourced from USAspending.gov, OSTI.gov, the DoE website, and the Federal Reporter database. Institutional quality is assessed using Carnegie classification and Times Higher Education rankings.
Enrico Berkes, Aiday Sikhova, Bruce A. Weinberg — Science of Science Funding
This paper examines how the gender composition of principal investigators (PIs) on biomedical research teams influences credit attribution and productivity for female faculty.
Finance Application
- This research offers a compelling framework for analyzing gender dynamics in finance, a heavily male-dominated field.
- One could investigate how the gender of senior partners or fund managers affects the career progression, credit allocation (e.g., deal lead, client ownership), and productivity of junior female professionals in investment banking, private equity, or asset management.
- The finding that female leaders are particularly impactful in 'heavily-male fields' could be directly tested by examining female leadership in trading desks or quantitative finance, and its effect on female talent retention and advancement.
- Furthermore, the observation of increased within-group inequality among women could shed light on 'queen bee' phenomena or other complex dynamics in finance careers.
GenderLeadershipCredit AttributionProductivityTeamsAcademiaSTEMFixed EffectsCareer ProgressionInequality
Core finding, identification, data
Core Finding
- Research teams with more women PIs significantly increase female faculty's authorship, particularly on high-impact papers and in heavily-male fields, and foster continued collaborations.
- While this narrows the overall gender gap, female leaders tend to support highly productive women more, potentially increasing within-group inequality among female faculty.
Identification Strategy
- The study employs an enhanced Abowd et al. (1999)-style design, leveraging faculty working on multiple projects.
- It uses researcher-year fixed effects to control for time-varying unobserved characteristics (motivation, ability) and paper fixed effects to account for unobserved field/paper characteristics (quality, norms), allowing for causal interpretation.
Data
The paper uses unique, linked data from UMETRICS (payroll transactions, job titles, effort allocation, NIH grants from 31 universities) and Author-ity (disambiguated MEDLINE publication data on authors and grants).
Giorgio Tripodi, Xiang Zheng, Yifan Qian, Dakota Murray, Benjamin Jones, Chaoqun Ni, Dashun Wang — Science of Science Funding
This paper examines how academic tenure, a significant career milestone, influences faculty research productivity, impact, and the pursuit of novel, high-risk ideas across various disciplines.
Finance Application
- This research provides a powerful framework to analyze career incentives and performance in finance.
- One could study how 'up-or-out' promotion systems in investment banking or private equity influence junior professionals' risk-taking, short-term performance, and innovation versus conformity.
- Similarly, examining the career paths of hedge fund or mutual fund managers after achieving 'star' status or a certain AUM (analogous to tenure) could reveal whether job security leads to a shift towards more novel, unconventional investment strategies that might initially yield lower alpha but potentially higher long-term, breakthrough returns, or simply a decline in effort.
- The observed disciplinary heterogeneity could be mapped to different investment styles (e.g., quantitative vs. fundamental) or firm types.
TenureCareer TrajectoriesIncentivesProductivityResearch OutputImpactNoveltyRisk-TakingMoral HazardScreeningAcademic Labor MarketsOrganizational BehaviorScience of SciencePersonnel EconomicsExecutive CompensationFund Manager PerformanceInvestment BankingPrivate EquityHuman CapitalRisk ManagementInnovation
Core finding, identification, data
Core Finding
- Faculty publication rates peak just before tenure, with post-tenure trends varying by discipline: lab-based fields maintain high output, while non-lab fields decline.
- After tenure, scholars increasingly pursue novel, high-risk research, but this shift is associated with a decline in research impact, resulting in fewer highly cited papers.
- These changes are sharply tied to the tenure year, not career age, and are distinct from control groups without tenure.
Identification Strategy
- The study uses a 'sharp break' identification strategy, comparing research trajectories immediately before and after the tenure year.
- It employs several control groups, including non-tenured researchers, academics in different institutional contexts (Europe, government labs), and non-tenure-eligible faculty, to isolate the effect of tenure.
- Fixed-effects regressions and various robustness checks are used to control for individual heterogeneity and other confounding factors.
Data
The paper integrates seven large-scale datasets, including faculty rosters from Academic Analytics Research Center (AARC), internal HR records from two R1 universities, SciSciNet (Microsoft Academic Graph), Scopus, Dimensions, and ProQuest dissertation data, covering over 12,000 researchers across 15 disciplines.
Justine Boudou, John McKeon — Science of Science Funding
This paper examines how relaxing resource constraints in supercomputing allocations affects the quantity and direction of scientific research, finding a trade-off between exploratory innovation and immediate impact.
Finance Application
- The core mechanism of resource allocation influencing innovation direction and impact is highly relevant to finance.
- For asset pricing, one could study how capital allocation within quantitative hedge funds or investment banks (e.g., to different research teams or trading strategies) affects the novelty and profitability of their models, and how this is reflected in fund performance or firm valuation.
- In household finance, the allocation of financial advisory resources (e.g., time, analytical tools) could influence whether advisors offer conventional or more exploratory investment advice, and the subsequent impact on client outcomes and retention.
- For insurance, the study's findings could inform how R&D budgets for actuarial science or risk modeling are allocated; concentrating resources might lead to novel risk assessment methods for emerging risks (e.g., climate change, cyber threats) with initially lower adoption but higher long-term resilience, versus incremental improvements to existing models with immediate, measurable returns.
Resource AllocationInnovationR&DScientific ProductivityExploration vs. ExploitationTrade-offsQuasi-ExperimentFintechAsset ManagementInsurance Risk Modeling
Core finding, identification, data
Core Finding
- Relaxing resource constraints significantly increases the number of scientific publications and shifts research towards less popular, newer, broader, and more exploratory topics, moving beyond researchers' prior expertise.
- However, these directional shifts, particularly for the most exploratory work, are associated with fewer citations, suggesting a trade-off between frontier-expanding innovation and immediate scientific impact, despite no change in journal quality.
Identification Strategy
- The study leverages the quasi-experimental design of the XSEDE program, where researchers' requests for supercomputing resources are evaluated by experts (XRAC) for objective need, but then systematically reduced by an algebraic formula due to system-wide capacity constraints.
- This creates plausibly exogenous variation in the share of resources allocated relative to the recommended amount, allowing for causal inference on the impact of resource constraints on scientific output and direction.
Data
The paper uses a novel dataset combining internal XSEDE data (2015-2022) on 710 scientific grants, including detailed records of requested, recommended, and allocated computing resources (Service Units), with publication data from Dimensions AI and XSEDE's internal publication database. Dimensions AI concepts are used to characterize the topics, novelty, and breadth of research.
Charles Ayoubi, James M. Zumel Dumlao, Misha Teplitskiy — Science of Science Funding
This paper demonstrates that academic peer review acts as a significant knowledge acquisition channel for reviewers, increasing their likelihood of subsequently citing reviewed manuscripts, with this learning effect moderated by geographic and intellectual distance.
Finance Application
- This paper's insights on learning from structured evaluation could be highly relevant to venture capital or private equity.
- For instance, do VC partners who conduct deep due diligence on a startup (analogous to peer review) make more informed subsequent investment decisions or identify better follow-on opportunities in related sectors? The 'geographic distance' could be the physical distance between the VC firm and the startup, while 'intellectual distance' could be the thematic difference between the VC's prior investment focus and the startup's technology.
- A similar study could compare VCs who actively led due diligence on a deal versus those who were invited but declined due to 'unavailability,' with 'citation' proxied by subsequent investments in similar companies or improved portfolio performance.
knowledge transferinformation asymmetrydue diligenceventure capitalprivate equityfinancial analystslearninggeographic proximityintellectual distanceevaluationabsorptive capacity
Core finding, identification, data
Core Finding
- The study finds that actively evaluating a manuscript more than doubles a reviewer's likelihood of citing that knowledge within three years compared to qualified evaluators who were invited but unavailable.
- This learning effect is moderated by both geographic and intellectual proximity; reviewing partially compensates for geographic barriers to knowledge transfer, and greater intellectual similarity between the reviewer and manuscript content leads to stronger knowledge gains.
Identification Strategy
- The identification strategy employs a quasi-experimental design by comparing two groups of invited reviewers: those who agreed to and completed a review (treatment group) versus those who declined the invitation explicitly stating unavailability (control group).
- This approach aims to establish causality by treating unavailability as an exogenous reason for non-participation, thus creating a plausible counterfactual for knowledge acquisition.
Data
The study utilizes administrative data from the Institute of Physics Publishing, comprising 104,306 reviewer-manuscript pairs across 55 physical sciences journals for submissions between 2018-2019. This data is matched with OpenAlex records to track subsequent citations and extract author information and abstracts for intellectual distance measures.
Todd D. Gerarden, Bryan K. Bollinger, Kenneth Gillingham, Drew S. Vollmer, Daniel Xu — Environmental & Energy Economics
This paper examines the effects of U.S. solar panel tariffs on production relocation, prices, welfare, and employment, comparing these impacts to manufacturing subsidies.
Finance Application
- This research offers critical insights for asset pricing by demonstrating how trade policies induce supply chain reconfigurations and affect firm profitability, which can be used to model policy risk in renewable energy sector valuations.
- For household finance, the findings on consumer surplus and price impacts can inform models of household investment in solar energy and their sensitivity to government subsidies (like the ITC).
- Insurers could use the detailed analysis of production relocation and cost structures to better assess supply chain resilience and operational risks for solar project financing and insurance policies, particularly for green bonds or infrastructure funds exposed to renewable energy assets.
Trade PolicyTariffsSubsidiesSolar EnergySupply ChainProduction RelocationWelfare EconomicsEmployment ImpactsEvent StudyStructural ModelRenewable Energy FinancePolicy Risk
Core finding, identification, data
Core Finding
- U.S. solar panel tariffs led to firms relocating production to tariff-free countries, increased domestic prices relative to other markets, resulted in modest gains for domestic manufacturers but larger losses in consumer surplus and environmental benefits, and ultimately reduced domestic solar industry employment and wages on net.
- Conversely, subsidizing solar panel manufacturing could increase domestic production, employment, and welfare.
Identification Strategy
- The paper uses event studies to causally identify production offshoring by comparing Chinese manufacturers exposed to tariffs to those not exposed.
- It then develops a structural model of oligopolistic competition, estimating demand using 2SLS with global input prices (silver, aluminum) and foreign market solar panel prices as instruments, and supply using within-firm variation in production across locations to identify cost efficiencies.
Data
The study utilizes IHS Markit data for global solar supply chain, production by country, and U.S. shipments. It also incorporates Federal Register data for tariff details, U.S. Customs and Border Protection data for collected duties, UN Comtrade and USITC DataWeb for trade flows, EIA, SEIA, and LBNL for solar adoption data, and market data for silver and aluminum prices.
Kimberly A. Clausing, Jonathan M. Colmer, Allan Hsiao, Catherine Wolfram — Environmental & Energy Economics
This paper evaluates the global impacts of Carbon Border Adjustment Mechanisms (CBAMs) on climate action, competitiveness, and distributional effects using a quantitative trade model and detailed plant-level data for steel and aluminum industries.
Finance Application
- This research offers rich insights for asset pricing by informing how CBAMs and carbon pricing impact firm valuations, particularly for carbon-intensive industries like steel and aluminum.
- Finance researchers could analyze how CBAM announcements or changes in carbon prices affect the stock returns, credit spreads, and ESG ratings of exposed firms, potentially leading to new 'carbon risk' factors.
- For household finance, the quantified consumer surplus changes due to CBAMs could be linked to household consumption patterns, savings, and wealth inequality across different income groups and regions.
- The regulatory risk introduced by CBAMs also presents an opportunity for insurance research to develop new products hedging against carbon price volatility or trade policy changes.
climate financeESGcarbon pricingtrade policyfirm valuationregulatory risksupply chaininternational economicshousehold consumption
Core finding, identification, data
Core Finding
- CBAMs can significantly facilitate collective climate action by improving domestic competitiveness, reducing emissions leakage, and encouraging other countries to adopt carbon taxes, while largely avoiding disproportionate burdens on lower-income countries.
- They also foster green investment and can be effective even with country-level average emissions data for implementation.
Identification Strategy
- The paper employs a quantitative trade model calibrated with detailed plant-level data.
- For demand elasticity, it instruments for world aluminum and steel prices using Australia's share of global bauxite and iron ore extraction as supply shifters.
- For supply elasticity, it uses group fixed effects, comparing plants within narrowly defined groups (same country, production mode, ownership, and plant age quartile) to mitigate endogeneity concerns from prices and unobserved technology.
Data
The study utilizes detailed plant-level data for primary aluminum (Wood Mackenzie) and steel (Climate TRACE, Global Steel Plant Tracker) covering production, capacity, costs, and Scope 1 & 2 emissions. It also uses global consumption data (World Bureau of Metal Statistics), historical production data (USGS), carbon pricing data (World Bank's Carbon Pricing Dashboard), plant ownership data, and shipping cost estimates (UNCTAD).
Stefano Baruffaldi, Pietro Santoleri, Eugenia Shevtsova — Science of Science Funding
This paper causally estimates the effects of competitive European mobility grants on academic researchers' mobility, scientific productivity, and collaboration networks using a Regression Discontinuity Design.
Finance Application
- The RDD methodology and findings on human capital mobility could be directly applied to finance.
- For instance, one could study the causal impact of competitive scholarships or early-career grants on finance PhDs' career trajectories, research output in top finance journals, or subsequent employment in prestigious financial institutions.
- Alternatively, the framework could assess how competitive seed funding for fintech startups (analogous to grants) impacts their innovation (e.g., patenting new financial products), growth, and long-term success, especially distinguishing between funding from top-tier venture capitalists versus smaller funds.
researcher mobilitygrantscausal inferenceregression discontinuity designscientific productivitycollaboration networkshuman capitalinnovationpolicy evaluation
Core finding, identification, data
Core Finding
- Marie Skłodowska-Curie Individual Fellowships significantly increase the likelihood of researchers moving to their intended destination country by 31-34 percentage points.
- While there is no average positive effect on publication quantity or quality, grants supporting extra-European mobility, longer research stays, or moves to higher-quality institutions lead to increased publication output, quality, and expanded collaboration networks.
Identification Strategy
- The study employs a Fuzzy Regression Discontinuity Design (RDD) by exploiting the discontinuity in grant assignment.
- Grant allocation is based on a competition-specific evaluation score threshold, creating a quasi-random assignment for applicants just above and below this cutoff, allowing for causal inference on individual researchers.
Data
The paper uses data on all applicants to the Marie Skłodowska-Curie Individual Fellowships from the Seventh Framework Programme (2007-2013), sourced from the COmmon Research DAta Warehouse (CORDA). This administrative data is linked with researchers' affiliations and publication outcomes obtained from Elsevier's Scopus database.
Yuqi Song, Joseph E. Aldy — Environmental & Energy Economics
This paper analyzes selection biases and over-crediting in US forest-based carbon offset projects by comparing regulated and voluntary carbon markets using geospatial and forest inventory data.
Finance Application
- This research highlights critical integrity issues in carbon offset markets, directly impacting ESG investing and corporate climate commitments.
- Finance researchers could investigate how over-crediting affects the pricing of carbon credits, the financial performance of firms relying on these offsets for ESG targets, and the credibility of green bonds or sustainability-linked loans.
- The methodology for identifying 'greenwashing' could also be adapted to evaluate other ESG claims, informing asset managers about true environmental impact versus reported metrics.
Carbon MarketsESGGreenwashingClimate FinanceOffset CreditsMarket EfficiencyCorporate IncentivesAsset PricingRisk ManagementEnvironmental Economics
Core finding, identification, data
Core Finding
- The study finds significant selection biases: regulated markets attract projects with historically increasing carbon stocks, while voluntary markets attract projects with historically decreasing carbon stocks.
- Both markets suffer from substantial over-crediting due to generous baseline settings, with voluntary market projects issuing about three times more offset credits than justified by business-as-usual baselines.
Identification Strategy
- The identification strategy compares IFM projects' pre-market biomass trends with matched control plots using Nearest Neighbor Matching on geophysical and land-use characteristics.
- A multinomial logistic regression links these pre-market trends to market entry choices, and simulations verify strategic market selection and quantify over-crediting under different baseline scenarios.
Data
The paper links a novel geospatial dataset of 62 US Improved Forest Management (IFM) offset projects (34 regulated, 28 voluntary) with ground-truth above-ground carbon storage measurements from the US Forest Service Forest and Inventory Analysis (FIA) program (1999-2021).
David A. Keiser, Bhashkar Mazumder, David Molitor, Joseph S. Shapiro — Environmental & Energy Economics
This paper analyzes national trends in US drinking water pollution, quantifies the impact of Safe Drinking Water Act loans on pollution reduction and mortality, and estimates the significant net benefits of these public health investments.
Finance Application
- The findings have direct implications for municipal bond markets, as improved water quality and health outcomes could reduce credit risk for water utilities and municipalities, affecting bond yields and ratings.
- In real estate, cleaner water could increase property values in affected areas, influencing real estate investment decisions and mortgage lending risks.
- For the insurance sector, reduced mortality and morbidity rates due to improved water quality could lower payouts for life and health insurers, potentially leading to adjustments in premium pricing and the development of new environmental risk insurance products.
Environmental EconomicsPublic HealthInfrastructureGovernment SpendingCausal InferenceDifference-in-DifferencesEvent StudyInstrumental VariablesWater QualityPollutionMortalityHealth OutcomesMunicipal FinanceReal EstateInsuranceHousehold Finance
Core finding, identification, data
Core Finding
- US drinking water pollution, particularly violations of health standards, declined by half from 2003-2019.
- Safe Drinking Water Act loans causally reduce pollution and significantly decrease mortality rates among older Americans, with an estimated cost of $124,000 per premature death avoided and a benefit/cost ratio of 19.6.
- Pollution levels are higher in low-income areas, with complex patterns for Black and Hispanic communities.
Identification Strategy
- The study primarily uses difference-in-differences and event study designs to estimate the causal impact of Safe Drinking Water Act loans.
- It compares pollution and mortality outcomes in areas receiving loans versus those that do not, accounting for heterogeneous treatment timing and using system-by-pollutant, state-by-year, and zip code fixed effects.
- Instrumental variables (loans as an instrument for pollution) are also employed to estimate concentration-response functions.
Data
The paper uses 230 million drinking water pollution readings on 1,800 pollutants from 48 states (obtained via FOIA), linked to new service territory maps. It also utilizes confidential administrative Medicare data (2009-2019) on health outcomes for US adults aged 65 and older, linked by zip code, and detailed information on 9,200 subsidized Safe Drinking Water Act loans.
Michael Kremer, Mauricio Romero, Santiago Saavedra — Environmental & Energy Economics
This paper uses a randomized controlled trial in Bogotá schools to evaluate the causal impact of portable HEPA air filters on student learning and indoor air quality.
Finance Application
- This research offers insights for household finance, insurance, and real estate asset pricing.
- In household finance, it suggests that investments in home or school air purification could be viewed as human capital investments, potentially impacting long-term income, savings, and educational spending.
- For insurance, improved indoor air quality could lead to lower health-related claims, influencing health and life insurance premiums.
- In real estate, properties (residential, commercial, or school facilities) with superior indoor air quality or filtration systems might command a valuation premium, affecting real estate investment strategies and REIT performance.
Air QualityEducationHuman CapitalHealthRandomized Control TrialHousehold FinanceInsuranceReal EstateAsset PricingEnvironmental Economics
Core finding, identification, data
Core Finding
- Air filters significantly improved student test scores by 0.03 standard deviations in 2023, while reducing indoor PM2.5 pollution by 0.47 µg/m³.
- Although no effect was found in 2024, the expected benefits, including learning gains and averted pandemic school closures, are estimated to exceed costs, particularly if the test score-wage correlation reflects human capital accumulation.
Identification Strategy
- The study employs a randomized controlled trial (RCT) design, assigning schools to receive filters, monitors, both, or neither.
- It uses an ANCOVA-style specification for test scores and instrumental variables (filter assignment) to estimate the causal effect of PM2.5 levels on learning, leveraging the random allocation of filters.
Data
The paper uses student-level data on standardized test scores (Saber 11) from over 42,000 students annually, school and student socio-demographic information, and real-time indoor and outdoor air quality measurements (PM2.5, PM10) from Plantower PMS7003 monitors and the city's SISAIRE network for 2023 and 2024.
Marie Briere, James M. Poterba, Ariane Szafarz — Workshop on Aging
This paper analyzes the impact of a 2019 French reform (Loi Pacte) that introduced tax-deductible voluntary contributions to employer-sponsored retirement plans on employees' saving behavior.
Finance Application
- This research directly informs household finance by providing empirical evidence on how tax incentives shape retirement saving decisions and wealth accumulation, especially highlighting heterogeneous responses across income, age, and prior saving experience.
- For asset pricing, understanding these shifts in voluntary contributions and withdrawals can shed light on the demand for long-term versus short-term assets and the flow of capital between different financial products (e.g., employer-sponsored plans vs. individual accounts).
- Insurers can leverage these insights to design and market retirement income products, such as annuities, by understanding which demographic segments are most responsive to tax advantages and how new policy changes might influence product demand and competitive landscapes.
Tax incentivesRetirement savingHousehold financeNatural experimentPolicy impactVoluntary contributionsDefined contribution plansSaving behaviorHeterogeneityFrance
Core finding, identification, data
Core Finding
- The introduction of pre-tax deductibility for voluntary long-term (LT) retirement contributions significantly increased overall contributions, particularly among higher-income, older workers, and those who were already making voluntary contributions.
- While there was little substitution between LT and medium-term (MT) saving plans, the reform also led to increased withdrawals from existing accounts, likely due to transfers to other financial institutions offering new retirement products.
Identification Strategy
- The identification strategy employs a difference-in-differences (diff-in-diff) model, exploiting the staggered adoption of the new pre-tax voluntary saving option by firms.
- The study compares the saving behavior of workers at firms that adopted the new option before and after its introduction, using workers at firms that would eventually adopt but had not yet done so as a control group.
Data
The paper uses administrative panel data from Amundi, a large retirement plan administrator in France, covering voluntary saving choices of approximately 1.4 million workers across 2,679 French firms between 2017 and 2022.
Oleksandra Cheipesh, Yarine Fawaz, Pedro Mira — Workshop on Aging
This paper investigates whether response times in cognitive tests can predict future cognitive decline and financial decision-making in older adults, leveraging automatically recorded data from large-scale surveys.
Finance Application
- This research provides a novel, low-cost behavioral signal (response times) that could be incorporated into models predicting financial decision-making capacity and wealth trajectories.
- For household finance, it could identify individuals at high risk of poor financial decisions (e.g., suboptimal investment choices, susceptibility to fraud) due to undiagnosed cognitive impairment, informing targeted financial literacy interventions or protective measures.
- In insurance, it could help assess longevity risk or the need for long-term care insurance by providing an early, objective indicator of cognitive decline and associated financial vulnerability, potentially leading to more personalized product offerings or risk assessments.
response timescognitive declinefinancial decision-makinghousehold financeagingwealth accumulationbehavioral economicssurvey datalongitudinal studiesrisk assessment
Core finding, identification, data
Core Finding
- Slower response times (RTs) are strong, independent predictors of future cognitive decline, health deterioration (frailty, mortality), and wealth loss, even after controlling for baseline cognition.
- Crucially, RTs also predict unawareness of cognitive decline, a state linked to significant financial vulnerability, suggesting they capture a dimension of cognitive vulnerability not fully reflected in standard cognitive measures or self-reported awareness.
Identification Strategy
- The study leverages automatically recorded response times (RTs) from Computer-Assisted Personal Interviews (CAPI) in the Survey of Health, Ageing, and Retirement in Europe (SHARE) waves 8 and 9.
- This provides a low-cost, non-intrusive measure of cognitive processing speed, which is then used in regression models to predict future cognitive, health, and financial outcomes, controlling for baseline states, demographics, and fixed effects.
Data
The paper uses longitudinal data from the eighth and ninth waves (2019-2022) of the Survey of Health, Ageing, and Retirement in Europe (SHARE), which includes automatically recorded time stamps from Computer-Assisted Personal Interviews (CAPI) and an expanded cognitive module.
Matilde Bombardini, Frederico Finan, Nicolas Longuet-Marx, Suresh Naidu, Francesco Trebbi — Environmental & Energy Economics
This paper studies how climate change and mitigation-related employment changes influence U.S. partisan voting and political competition over environmental policy, projecting future political equilibria and legislative outcomes.
Finance Application
- The paper's findings offer a novel framework for assessing climate transition risk in asset pricing and insurance.
- The demonstrated link between climate shocks, local employment shifts, and political outcomes (e.g., increased probability of carbon pricing) can be used to model how policy uncertainty and regulatory changes affect the valuation of 'brown' assets (fossil fuels, carbon-intensive industries) versus 'green' assets (renewables, clean tech).
- Insurers could leverage these insights to better price long-term climate risks in property and casualty insurance, incorporating political economy factors into their catastrophe models and underwriting decisions.
- For household finance, the localized effects of climate and job transitions on voting behavior could inform models of regional housing market dynamics and household wealth accumulation, particularly in areas undergoing green transitions or facing severe climate impacts.
Climate ChangePolitical EconomyVoting BehaviorEnvironmental PolicyGreen JobsBrown JobsAsset PricingTransition RiskHousehold FinanceInsuranceClimate RiskPolicy Uncertainty
Core finding, identification, data
Core Finding
- The paper finds that Democratic vote share increases with exogenous changes in local climate and green transition employment.
- Specifically, a one standard deviation increase in extreme heat days (8.13 days) increases the Democratic margin by 0.90 percentage points, while a similar increase in green jobs boosts it by 0.34 percentage points and brown jobs decrease it by 0.62 percentage points.
- Under a worst-case climate scenario (SSP5-8.5), the probability of the House passing a carbon-pricing bill is projected to be 9 percentage points higher in 2050 than in 2020, even as both parties shift slightly rightward on climate policy.
Identification Strategy
- The identification strategy combines precinct-level voting data with high-resolution climate and employment data, using block-group and congressional district-by-year fixed effects to isolate localized effects of climate and employment shocks on voter demand.
- To address endogeneity in local job composition, a shift-share instrumental variable approach is employed.
- A structural model of political competition is then estimated using GMM, incorporating candidate policy positions and their responsiveness to expected vote shares and local party trends.
Data
The paper uses 2000-2020 precinct-level voting data, high-resolution temperature and precipitation data (updated Schlenker and Roberts, 2009), census block-group level measures of 'green' and 'brown' employment shares (constructed from BLS GGS, LODES, CBP, QCEW), and congressional candidate positions on environmental policy (Longuet-Marx, 2024, using campaign websites and Project Vote Smart surveys). Future projections utilize NASA Earth Exchange (NEX) Global Daily Downscaled Projections (GDDP) for climate and BLS Occupational Outlook Handbook (OOH) for jobs.
Tatyana Deryugina, Julian Reif — Environmental & Energy Economics
This paper develops a new framework to quantify the long-run effects of air pollution exposure on life expectancy by combining short-run quasi-experimental estimates with a demographic model.
Finance Application
- This research provides robust causal estimates of the long-term health costs of air pollution, which has direct implications for several finance areas.
- In **asset pricing**, these estimates can inform the valuation of firms in polluting industries by quantifying regulatory and litigation risks, and the impact on labor productivity.
- For **real estate**, the long-run health effects of regional pollution could be capitalized into property values, allowing for studies on how climate-related health risks are priced in housing markets. **Insurance companies** can use these long-run survival estimates to refine actuarial models for life and health insurance products, better pricing policies and managing long-term liabilities related to environmental health risks.
environmental riskhealth economicsmortalitylife expectancyair pollutioninstrumental variablesasset pricingreal estateinsurancehousehold financeclimate riskexternality
Core finding, identification, data
Core Finding
- The study finds that acute (1-day) SO2 exposure leads to both short-term mortality displacement among frail individuals and longer-term accelerated aging among healthier individuals, with the latter dominating.
- A permanent 10% reduction in SO2 is estimated to increase life expectancy at birth by 1.1 years, with 90% of these survival benefits accruing after age 50, significantly exceeding naive extrapolations of short-run effects.
Identification Strategy
- The paper uses daily changes in wind direction as an instrumental variable for observed changes in SO2 concentrations.
- This quasi-experimental approach allows for causal inference by controlling for contemporaneous weather conditions and a rich set of fixed effects, assuming wind direction affects mortality only through its impact on air pollution.
Data
The study combines daily death records (1972-1988) from the National Vital Statistics, daily air pollution data (SO2, NO2, TSP, O3, CO) from the EPA's Air Quality System, and daily weather data (wind speed/direction, temperature, precipitation, humidity) from ERA5 reanalysis and PRISM Climate Group. County-level population estimates are from the SEER Program.
Jingyuan Wang, Jianwei Xing — Environmental & Energy Economics
This paper examines how consumer subsidies affect the growth of the nascent Chinese electric vehicle industry, considering the role of low-quality 'lemon' entrants and their impact on industry reputation and welfare.
Finance Application
- The paper's insights into 'lemon' firms and collective reputation externalities are highly relevant for understanding ESG investing and green finance.
- If a 'green' sector attracts firms engaging in 'greenwashing' (lemons), this could depress the valuation of genuine green firms, affecting green bond pricing or ESG fund performance.
- This framework could also be applied to household finance, studying how consumer perceptions of quality for new financial products (e.g., FinTech, crypto) or 'green' investment vehicles influence adoption rates and investor trust, especially if early entrants are 'lemons.' For insurance, the 'lemon' problem and reputation for product failures (like EV fires) directly impact risk assessment and pricing for new technology insurance, such as auto insurance for EVs or product liability for emerging tech companies.
Industrial PolicySubsidiesElectric VehiclesReputationLemonsFirm EntryDynamic ModelsConsumer BehaviorGreen FinanceESGFinTechInsurance RiskInformation Asymmetry
Core finding, identification, data
Core Finding
- The net welfare impact of the Chinese EV subsidy from 2012-2018 is nearly zero, with the reputation impact reducing subsidy benefits by 10.8%.
- Poorly designed subsidies attract 'lemon' entrants with low, imperfectly observed quality, undermining collective reputation and dampening industry growth.
- Optimal subsidy design involves a conservative level and attribute-based stringency to effectively filter out lemons and mitigate reputation losses.
Identification Strategy
- The paper develops a structural model of vehicle demand, firm entry, and EV reputation dynamics.
- It identifies reputation externalities using a difference-in-differences (DID) design for EV fires and instrumental variables (IVs) for lemon share (distance from lemon firms interacted with central subsidy) and prices (central/local subsidies, battery supplier/capacity).
- Firm entry decisions are modeled using a finite-period dynamic discrete choice model with partially oblivious equilibrium.
Data
The study uses Chinese EV market data from 2012 to 2018, including vehicle registration data, model-level attributes, government policies, and charging station data. It also leverages consumer reviews from the Autohome platform and complaints/repair data from the Car Quality Network to identify 'lemon' firms and measure reputation.
Karl M. Aspelund — Environmental & Energy Economics
This paper evaluates the efficiency and distributional impacts of trade restrictions, such as segmented trading and individual production requirements, in Iceland's fisheries permit market.
Finance Application
- The paper's quantification of efficiency costs versus redistributive benefits of regulations in a real asset market (fishing quotas) offers a rich avenue for asset pricing research.
- Investors pricing fishing quotas or the equity of fishing firms would need to account for these regulatory-induced 'social costs' and their impact on firm cash flows and risk profiles.
- For household finance, the detailed analysis of how these regulations affect worker earnings and labor demand directly informs studies on labor income risk, consumption smoothing, and wealth inequality among households in regulated industries.
- This could lead to research on how households adjust savings or portfolio choices in response to such policy-induced income shifts.
environmental economicsregulationpermit marketsfisherieslabor economicsincome redistributionfirm productivityasset valuationhousehold financepolitical riskregulatory risk
Core finding, identification, data
Core Finding
- The study finds that while permit trading increases the harvest share of productive boats and shifts income to higher-income workers, trade restrictions are binding and lower productivity.
- Segmentation is more efficient at increasing labor demand, whereas production requirements are more effective at redistributing income to low-income workers, with a combined approach balancing both objectives.
Identification Strategy
- The paper uses a difference-in-differences analysis comparing small boats entering trading in 2001 to large boats already trading to identify the causal impacts of permit trading.
- It also leverages 'bunching' of firms around regulatory thresholds (e.g., 50% production requirement) to demonstrate that these restrictions are binding.
- A structural model of fishery production and permit trading is developed to simulate counterfactual market equilibria and quantify trade-offs.
Data
The paper utilizes detailed daily harvest and permit trading data, linked to administrative records on worker employment and earnings (pay-slips, tax returns, census records), as well as boat characteristics and crew registry data from Iceland's fisheries.
Lucia E. Torres Frasele — Workshop on Aging
This paper causally estimates the impact of natural menopause on women's health and employment outcomes using genetic predisposition for menopause timing as an instrumental variable.
Finance Application
- This research offers crucial insights for household finance and insurance.
- The causal link between menopause, health deterioration, and reduced labor supply directly impacts women's lifetime earnings, retirement savings adequacy, and the timing of Social Security benefit claims (e.g., SSDI).
- Insurers can leverage this understanding to refine risk models and product design for health, disability, and long-term care insurance, particularly for middle-aged women.
- Furthermore, the predictable health shock could influence household consumption smoothing and asset allocation decisions, potentially leading to a shift towards less risky assets or increased precautionary savings as women approach menopause.
Household FinanceLabor SupplyHealth ShocksRetirement PlanningInsuranceGenetic DataInstrumental VariablesWomen's HealthAgingHuman Capital
Core finding, identification, data
Core Finding
- Menopause significantly accelerates health conditions (more than tripling pre-menopause trends) and substantially reduces the likelihood of working for pay by nearly 2 percentage points annually, leading to an 18 percentage point reduction over ten years.
- The diagnosis of an additional medical condition post-menopause reduces the likelihood of working for pay by 49-77 percentage points, primarily affecting full-time work and conditions like arthritis, cancer, and lung disease.
Identification Strategy
- The study employs an instrumental variable (IV) approach, using the polygenic risk score (PGS) for the timing of menopause as an instrument for reported age at natural menopause.
- This method isolates the genetically determined component of menopause timing, which is plausibly independent of confounders and measurement error, and uses the menarche PGS as an exclusion restriction to address selection bias from hysterectomies.
Data
The paper utilizes detailed longitudinal data from the Health and Retirement Study (HRS), including self-reported age of natural menopause, health conditions, labor market activity, and genetic data (polygenic risk scores derived from DNA saliva samples) for women of European ancestry.
Paul Brimble, Rob Garlick, Kate Orkin — Development Economics
This paper uses experimental and panel analysis to show how unconditional cash transfers alleviate market frictions, improve production decisions, and enhance economic welfare for poorer households in rural Kenya.
Finance Application
- This research offers several insights for finance.
- In household finance, it suggests that liquidity constraints and market frictions significantly distort household investment and labor supply decisions; cash transfers can mitigate these, potentially increasing demand for productive assets and formal financial services (e.g., micro-loans, business insurance).
- For insurance research, the findings highlight how alleviating frictions can reduce reliance on informal risk-sharing or increase the capacity to pay for formal insurance, allowing households to undertake riskier, higher-return investments.
- This could inform the design of financial inclusion programs that combine transfers with access to credit or insurance to maximize welfare gains.
cash transfershousehold financemarket frictionsexperimental economicseconomic developmentcapital accumulationlabor supplyinsuranceliquidity constraintsKenya
Core finding, identification, data
Core Finding
- Unconditional cash transfers to poorer households in rural Kenya alleviate market frictions, enabling them to increase capital stock, diversify production, supply more labor, and make more efficient production decisions.
- This leads to improved economic welfare through both direct wealth effects and indirect efficiency gains, contradicting the idea that such transfers lead to reduced labor supply.
Identification Strategy
- The paper employs experimental analysis, specifically using large, unconditional cash transfers as an intervention, combined with panel data analysis.
- This design allows for causal inference on how these transfers impact household production decisions, capital accumulation, and labor supply in the presence of market frictions.
Data
The paper uses primary household-level panel data collected in rural Kenya, likely from an experimental setting where some households received unconditional cash transfers.
Samuel W. Arenberg, Nicholas F. Reynolds, Sam Stripling — Workshop on Aging
This paper argues that the recent stagnation in midlife mortality in the US is a delayed consequence of a similar stagnation in early-life mortality for the same birth cohorts, a phenomenon termed "scarring."
Finance Application
- This research has direct implications for insurance and household finance.
- For **insurance**, actuaries could integrate cohort-specific "scarring" effects into life insurance and annuity pricing models, adjusting for the persistent health disadvantages of certain birth cohorts to avoid mispricing long-term liabilities.
- In **household finance**, the findings suggest that individuals from "scarred" cohorts may face higher healthcare costs and shorter working lives, influencing their optimal savings rates, retirement planning, and demand for health and long-term care insurance products.
- This could lead to cohort-specific financial advice and product offerings.
MortalityCohort EffectsLife ExpectancyHealth EconomicsScarring HypothesisLongevity RiskInsurance PricingHousehold SavingsRetirement PlanningPublic Health
Core finding, identification, data
Core Finding
- The paper empirically demonstrates a striking alignment: cohorts experiencing a stagnation in midlife mortality (post-2000) are precisely the same cohorts that experienced a stagnation in early-life mortality decades earlier (around 1947).
- This "scarring" effect, combined with the opioid crisis, can explain almost all of the recent stagnation in midlife mortality in the US, a finding supported by cross-country comparisons where countries without midlife mortality stagnation also lacked earlier child mortality stagnation.
Identification Strategy
- The paper's identification strategy relies on a "cohort alignment" approach, observing the temporal concordance of mortality trends across different life stages (early-life vs. midlife) for the same birth cohorts.
- It compares these patterns across the US, a pool of European countries, Switzerland, and other Anglo countries to support the scarring hypothesis, and quantifies the contribution of early-life stagnation to midlife stagnation through an "accounting exercise."
Data
The study primarily uses data from the Human Mortality Database (HMD), supplemented by Social Security Administration cohort life tables for earlier cohorts, post-neonatal mortality rates from the National Center of Health Statistics (Lindner & Grove, 1943), and overdose death data from the National Vital Statistics System.
Anne Katrine Borgbjerg, Esben Agerbo, Nabanita Datta Gupta, Timothy Halliday — Workshop on Aging
This paper analyzes how genetic risk for Alzheimer's Disease (AD) affects labor market outcomes among Danes, revealing significant gender differences.
Finance Application
- This research offers crucial insights for household finance and insurance.
- In household finance, it suggests that individuals, particularly women, with higher genetic predispositions for AD may need to adjust retirement savings and investment strategies to account for earlier labor market exit and potential long-term care needs.
- Financial advisors could develop tailored advice incorporating genetic risk factors (ethically and legally).
- For the insurance industry, these findings could inform the pricing of long-term care and life insurance policies, potentially leading to gender-differentiated premiums or product designs, while navigating genetic discrimination regulations.
Alzheimer's Diseasegenetic risklabor supplyretirementdisability pensionhousehold financeinsurancecognitive declinegender differencespolygenic scoreslong-term carehuman capital
Core finding, identification, data
Core Finding
- Higher genetic risk for AD increases dementia diagnoses and GP visits for both genders.
- For women aged 45-65, it reduces labor participation and raises disability pension uptake, especially near retirement, with these effects weakening for those with high polygenic scores for education.
- For men, AD genetic risk shows no employment impact, highlighting that AD's economic costs manifest well before advanced stages and disproportionately affect women's labor supply.
Identification Strategy
- The study employs a proxy-phenotype design, using children's genetic variants (APOE-e4 carrier status and AD polygenic score) to predict parents' labor market outcomes.
- To address measurement error bias inherent in this design, they use instrumental variable (IV) regressions, instrumenting the primary AD polygenic score with another AD polygenic score (ga3646 for women, ga3635 for men) under the assumption that the instrument only affects outcomes via the endogenous variable.
Data
The paper utilizes administrative and genetic data from over 200,000 Danes, combining genetic information from the iPSYCH study with full population Danish registers. These registers provide detailed data on labor market outcomes, education, and demographic variables for parents aged 45-65.
Marcella Alsan, Yousra Neberai — Workshop on Aging
This paper quantitatively examines how the American Medical Association's (AMA) public relations campaign in the post-WWII era successfully defeated national health insurance and entrenched private health insurance in the U.S.
Finance Application
- This research offers direct insights for insurance studies, particularly on how lobbying and public relations campaigns can shape market structure and demand for private insurance products.
- Similar methodologies could be applied to investigate how financial institutions (e.g., life insurers, asset managers, fintech firms) use PR to influence regulatory outcomes, public perception of financial risks, or the adoption of specific investment/savings products.
- For household finance, the framework for quantifying campaign exposure and its impact on enrollment decisions could be adapted to analyze how marketing or advocacy efforts influence household financial choices, such as retirement plan participation, investment in ESG funds, or responses to debt relief programs, especially when public and private options compete.
- The paper's use of text analysis on legislative records could also inform asset pricing research by analyzing corporate lobbying efforts or regulatory communications for signals predicting policy shifts or market movements.
insurance market structurelobbyingpublic relationspublic opinionhealth insurancehousehold financepolitical economyevent studytext analysishistorical dataregulatory captureprivate insuranceasset pricing implications
Core finding, identification, data
Core Finding
- A one standard deviation increase in the AMA's campaign exposure explains approximately 20% of the increase in private medical insurance enrollment and led to a 6 percentage point decline in public opinion support for national health insurance.
- This indirect lobbying effort also influenced legislative rhetoric and voting behavior, demonstrating its significant and persistent impact on market structure and policy outcomes.
Identification Strategy
- The study employs a difference-in-differences design, comparing changes in private health insurance enrollment and public opinion before and after the AMA's campaign, across areas with varying campaign intensity.
- Campaign exposure is constructed from archival data on mass advertising circulation (scaled by local newspaper readership) and physician outreach (pamphlets distributed, scaled by AMA membership).
- The authors control for historical confounders like unionization and income, and utilize spatial and temporal fixed effects, arguing that campaign exposure was uncorrelated with pre-existing trends.
Data
The paper uses novel archival data from Campaigns, Inc. (the PR firm) and National Archives, historical directories (American Medical Directory, American Hospital Directory, N.W. Ayer & Son's Directory of Newspapers and Periodicals), insurance enrollment reports, Gallup public opinion data, and digitized Congressional Record text and voting records.
Dora Costa, Lars Olov Bygren, Benedikt Graf, Martin Karlsson, Joseph Price — Workshop on Aging
This paper investigates the multigenerational impact of historical harvest variability on longevity, particularly for male-line grandsons, suggesting an epigenetic mechanism.
Finance Application
- This research offers a novel perspective on long-term, intergenerational risk in household finance and insurance.
- Insurers could develop new actuarial models that incorporate ancestral environmental shocks, such as harvest variability, to price life, health, and long-term care insurance more accurately, especially for populations with deep historical records.
- In household finance, the findings suggest that inherited health vulnerabilities, potentially linked to historical climate events, could influence intergenerational wealth transfers, precautionary savings, and retirement planning decisions, prompting a need for financial products that address these long-horizon, epigenetically transmitted risks.
- For asset pricing, understanding these long-term human capital impacts from climate shocks could inform regional economic growth forecasts and the valuation of local businesses or real estate.
EpigeneticsIntergenerational transmissionLongevity riskClimate riskHuman capitalHousehold financeInsuranceHealth economicsEconomic history
Core finding, identification, data
Core Finding
- Using two multigenerational datasets for Sweden, we show that grandsons' longevity was strongly linked to spatial shocks in paternal grandfathers' yearly harvest variability when agricultural productivity was low and market integration was limited.
- The elimination of harvest variability was as important for life expectancy as decreasing contemporaneous infectious disease rates for Swedish men reaching age 40 between 1870 and 1919.
Identification Strategy
- The study uses two multigenerational datasets for Sweden and establishes causality by controlling for grandparents' birth year and county fixed effects, and by examining second cousin samples where grandfathers who were brothers experienced differential harvest swing exposure during epigenetically sensitive years (ages 9-12).
- This within-family comparison helps isolate the effect of harvest swings from unobserved family characteristics.
Data
The paper uses crowd-sourced genealogies from Family Search linked to Swedish censuses, and multigenerational data from the SwedPop database (from parish records linked to national death records). It also constructs annual county-level harvest indices and county-level infant mortality rates for Sweden.
Erica M. Field, Seema Jayachandran, Munir Squires — Development Economics
This paper studies how out-marriage in Oman, driven by the availability of marriageable male cousins, reduces women's social insularity and leads to more gender-egalitarian attitudes, despite being associated with seemingly worse socioeconomic outcomes.
Finance Application
- The paper's findings on how social insularity and kin networks facilitate trust, informal credit, and wealth preservation have direct implications for household finance and insurance.
- Researchers could investigate how out-marriage, by weakening these kin networks, affects household savings behavior, demand for formal insurance products, and intergenerational wealth transfers.
- For instance, out-married women might exhibit higher participation in formal financial markets or different asset allocation strategies due to reduced informal risk-sharing and increased exposure to diverse financial norms.
Social NetworksCultural ValuesHousehold FinanceInformal CreditRisk SharingIntergenerational WealthGender EconomicsEmerging Markets
Core finding, identification, data
Core Finding
- Out-marriage, influenced by the availability of marriageable male cousins, significantly reduces women's social insularity and leads to more gender-egalitarian attitudes in Oman (0.6 SD increase).
- This shift in attitudes occurs despite out-marriage being associated with outcomes typically linked to reduced female empowerment, such as earlier marriage and lower educational attainment.
- The study highlights social exposure as a distinct driver of cultural change, independent of conventional development markers.
Identification Strategy
- The paper uses an instrumental variable (IV) approach, leveraging the quasi-random variation in the 'age-weighted number of marriageable male cousins' available to a woman.
- This instrument serves as a supply-side determinant of whether a woman marries a cousin, exogenously influencing the likelihood of out-marriage and subsequent changes in social insularity and attitudes.
Data
The study uses primary survey data collected from 1,538 women aged 25-50 in the Dakhliya region of Oman (Bid Bid and Nizwa districts) in 2007 and 2009. The survey gathered detailed information on extended families, cousin availability, marriage patterns, social interactions, socioeconomic outcomes, and gender attitudes.
Deivy Houeix — Development Economics
This paper demonstrates that digital technologies, while reducing information frictions and improving efficiency through embedded observability, can deter adoption among lower-ability workers who fear losing informational rents, highlighting a trade-off between efficiency and diffusion.
Finance Application
- This research offers profound insights for FinTech, microfinance, and insurance.
- In lending, digital payment data could enable more precise credit scoring and dynamic loan pricing for small businesses, but lenders must address the 'reluctant adopter' problem by designing privacy-preserving options to attract informal firms fearing loss of informational rents.
- For insurance, real-time observability of behavior (e.g., driving patterns for car insurance, health data for life insurance) could lead to more efficient, personalized products, but similar privacy concerns could hinder adoption among high-risk or low-income individuals.
- The 'privacy-by-default' approach could be critical for promoting financial inclusion and welfare gains in these markets.
Digital paymentsInformation asymmetryMoral hazardContract theoryTechnology adoptionInformal marketsFinancial inclusionRelational contractsObservabilityPrivacySME lendingMicroinsurance
Core finding, identification, data
Core Finding
- Digital payments significantly reduce cash-related costs and reshape relational contracts by mitigating moral hazard, increasing driver effort, and improving firm performance.
- However, this embedded observability deters adoption, especially among poorer, low-performing drivers who anticipate losing informational rents.
- Removing observability nearly doubles adoption and, within a relational contract framework, increases total welfare.
Identification Strategy
- The study employs two field experiments in Senegal's taxi industry.
- The 'impact experiment' randomized access to digital payments and transaction observability for taxi owners among drivers willing to adopt.
- The 'adoption experiment' re-offered the technology to reluctant drivers, randomly varying owner observability *before* their adoption decision, to causally identify the effect of anticipated observability on adoption.
Data
The paper uses five rounds of surveys with taxi owners and drivers over nearly two years to track informal contract data. It also includes mystery passenger audits to measure driver effort and pricing behavior, and administrative mobile money data covering daily payment and transfer transactions at the driver level.
Niclas Moneke, Torsten Figueiredo Walter — Development Economics
This paper provides causal evidence that data collectors manipulate survey samples by excluding high-cost subjects, leading to systematic bias in aggregate statistics and erroneous inference across various human-collected data types.
Finance Application
- The insights into data collector incentives and endogenous sample selection have direct applications in finance.
- In household finance, interviewers for surveys like the SCF or PSID might systematically under-report complex financial portfolios or less financially literate households to reduce effort, biasing estimates of wealth inequality or financial fragility.
- In corporate finance and asset pricing, if data providers (e.g., company personnel, auditors, or third-party aggregators) manipulate firm-level data (e.g., revenue, employees, R&D) to avoid scrutiny or reduce reporting burden, it could distort observed firm size distributions and other characteristics, impacting asset pricing factors and corporate investment studies.
- Similarly, credit rating analysts or loan officers might selectively report information, biasing credit risk assessments and capital allocation.
Data qualityMeasurement errorSurvey biasIncentivesMoral hazardHousehold financeCorporate financeAsset pricing anomaliesCredit riskData collectionSelection biasFirm size
Core finding, identification, data
Core Finding
- Data collectors systematically exclude high-cost subjects, leading to non-random sample selection where marginalized populations are under-represented.
- This manipulation introduces systematic bias in aggregate statistics, such as fertility measures, and can distort distributions, like the 'missing middle' in firm size, undermining microeconomic and macroeconomic analysis.
- The bias is exacerbated by higher effort costs (e.g., extreme temperatures) and reduced by increased detection probability (e.g., mandatory audits).
Identification Strategy
- The paper primarily uses two strategies: first, exploiting the random assignment of individual questionnaires (man's questionnaire) across households in 181 surveys to create exogenous variation in data collector effort.
- Second, for women, comparing survey data to contemporaneous population censuses, assuming censuses lack the same data collector incentives.
- Additionally, it leverages exogenous variation in day-to-day temperature as a proxy for effort cost and policy changes in the Indian Economic Census's eligibility thresholds to identify firm size manipulation.
Data
The paper uses microdata from the Demographic and Health Survey (DHS) and Multiple Indicator Cluster Survey (MICS) across 181 surveys in 73 countries, matched with contemporaneous population censuses. It also incorporates weather data from the Princeton Meteorological Forcing Dataset, data from the Indian Economic Census (1998, 2005, 2013), and the US National Longitudinal Survey of Youths 1997 (NLSY97) alongside the Current Population Survey's Annual Social and Economic Supplement (CPS ASEC).
Nadia Ali, Giacomo De Giorgi, Aminur Rahman, Eric Verhoogen — Development Economics
This paper presents the first randomized evaluation of a government financial-support program (Tasdir+) in Tunisia, which offered matching grants for fixed market-access costs to promote exports.
Finance Application
- The finding that fixed-cost export subsidies significantly boost exports on the intensive margin for existing relationships provides a direct input for asset pricing and corporate finance.
- Investors could develop strategies to identify and invest in firms (especially non-totally exporting and quality-certified ones) in countries implementing similar RCT-backed export promotion programs, anticipating positive impacts on revenue and firm value.
- For trade finance and export credit insurance, the increased volume and stability of existing export flows due to these subsidies could reduce perceived credit risk for beneficiary firms, potentially leading to lower financing costs or insurance premiums, and informing product design for these financial services.
Export PromotionGovernment SubsidiesRandomized Control TrialFirm PerformanceInternational TradeIntensive MarginExtensive MarginCorporate FinanceAsset PricingCredit RiskTrade FinanceEmerging Markets
Core finding, identification, data
Core Finding
- The program led to a positive and statistically significant increase in exports, primarily on the intensive margin (increased sales within existing destinations and products), but had limited impact on the extensive margins (number of destinations or exported products).
- This contrasts with workhorse trade models that predict fixed costs primarily affect extensive margins.
- The effects were driven by non-totally exporting firms and those with quality certifications.
Identification Strategy
- The study employs a randomized evaluation (RCT) design.
- Eligibility for matching grants was randomized across five application rounds (2018-2019) among 487 firms.
- The treatment group received matching grants, while the control group did not.
- The analysis uses ANCOVA and Poisson Pseudo Maximum Likelihood (PPML) specifications, controlling for baseline values and fixed effects, to estimate causal effects.
Data
The paper utilizes several administrative data sources: the Repertoire National des Entreprises (RNE) from the Tunisian national statistical agency (INS) for annual domestic sales, exports, employment, and wages (2015-2021); transaction-level customs records from the Tunisian customs authority for detailed export transactions (2017-2022); and administrative data from the Tasdir+ program itself, including firm applications and reimbursement records. Researchers also conducted baseline and follow-up surveys.
Eoin F. McGuirk, Nathan Nunn — Development Economics
This paper investigates how agricultural development projects in traditionally pastoral areas of Africa lead to increased civil conflict and uneven economic outcomes, particularly for pastoral communities.
Finance Application
- This research offers a robust framework for assessing political risk and local economic shocks in emerging markets. **Asset pricing** models could incorporate geocoded project and ethnographic data to predict conflict-induced disruptions, affecting the valuation of firms operating in these regions or sovereign bond spreads. **Insurance** companies could leverage these insights to develop more granular political risk or business interruption insurance products, pricing premiums based on the specific type of development project and the local socio-economic context (e.g., pastoral vs. agricultural areas).
- For **household finance**, the evidence of diverging wealth outcomes due to mismatched projects highlights a source of idiosyncratic risk, informing studies on household risk management, asset accumulation, and the effectiveness of microfinance in conflict-prone areas.
Political RiskEmerging MarketsGeocoded DataConflictDevelopment EconomicsAsset ValuationInsuranceHousehold FinanceCausal InferenceAfrica
Core finding, identification, data
Core Finding
- Implementing agricultural projects in traditionally pastoral areas leads to an almost two-fold increase in the risk of conflict (9.7 p.p., p<0.01), while in non-pastoral areas, it reduces conflict (-1.9 p.p., p<0.01).
- Despite increasing cropland coverage and nighttime luminosity in pastoral areas, the economic gains are concentrated in non-pastoral households, leading to significantly lower wealth for pastoral respondents.
Identification Strategy
- The paper employs a difference-in-differences (DD) strategy at the 0.5-degree cell-year level, controlling for cell fixed effects and country-by-year fixed effects, and a triple difference approach by interacting agricultural projects with a transhumant pastoralism index.
- Robustness is confirmed through event study plots using the Callaway and Sant'Anna estimator and a natural experiment exploiting a "supply shock" of World Bank agricultural projects (2006-2008) due to the World Development Report on agriculture.
Data
The study uses geocoded conflict data (UCDP, ACLED), World Bank development project data (AidData), ethnographic information on transhumant pastoralism (Ethnographic Atlas), land cover classification (Copernicus Climate Change Service), nighttime luminosity (DMSP-OLS Nighttime Lights), household survey data (DHS), and political power data (Ethnic Power Relations).
Oriana Bandiera, Ananya Kotia, Ilse Lindenlaub, Christian Moser, Andrea Prat — Development Economics
This paper quantifies the role of worker-job matching in economic development across 28 countries, attributing differences to endowments, technology, and idiosyncratic matching frictions.
Finance Application
- The paper's quantification of 'idiosyncratic matching frictions' (σ/A) directly relates to uninsurable labor income risk for households; higher frictions imply greater wage dispersion for similar skills, influencing household savings, portfolio choices, and demand for precautionary assets.
- In asset pricing, firms operating in countries with lower matching frictions (higher meritocracy) might exhibit higher productivity and lower labor costs, potentially leading to higher valuations or lower cost of capital, which could be explored as a factor in cross-sectional stock returns.
- The model's framework could also inform human capital valuation, where the 'technology' parameters (skill complementarities) affect the returns to education and optimal human capital investment decisions.
labor economicsmatching modelshuman capitaleconomic developmentwage inequalityproductivitycross-country analysismicrodatasortinglabor market frictionshousehold financeasset pricinglabor income riskhuman capital investment
Core finding, identification, data
Core Finding
- The paper finds that improvements in worker-job matching due to reduced idiosyncratic matching frictions account for only a small share of cross-country income differences.
- However, improved worker-job matching is crucial for unlocking the gains from economic development generated by adopting frontier endowments and technology, suggesting that meritocracy is more a consequence than a source of economic development.
Identification Strategy
- The paper develops an equilibrium matching model with multidimensional skill heterogeneity.
- It identifies the extent of idiosyncratic matching frictions in each country based on residual wage dispersion conditional on worker skills and job skill requirements (following Salanié, 2015).
- Once frictions are pinned down, technology parameters are identified based on observed sorting patterns between workers and jobs.
Data
The primary data source is the Survey of Adult Skills (PIAAC) from the OECD, covering over 120,000 individuals across 28 countries, which provides direct measures of worker skills (numeracy, literacy) and job skill requirements. Additional data include GDP per capita (World Bank) and labor shares (OECD, FRED).
Varun Kapoor — Development Economics
This paper examines how frictions arising from wage theft, worker reneging, and liquidity constraints affect labor supply and demand in informal labor markets in India using three field experiments.
Finance Application
- The findings have direct implications for household finance and small business lending.
- Workers' strong preference for daily pay due to liquidity constraints highlights a demand for highly liquid savings products and short-term income insurance.
- For firms, their preference for back-loaded contracts and sensitivity to worker reneging suggests a market for flexible working capital loans and operational risk insurance products.
- Fintech solutions leveraging digital platforms for reputation building could also address the trust and information asymmetry issues, potentially reducing credit risk for lenders and improving access to finance for both workers and informal businesses.
informal labor marketswage theftliquidity constraintscontract theoryfield experimentinsurancecreditcommitmentworker renegingfirm behaviorhousehold financesmall business financefintech
Core finding, identification, data
Core Finding
- Workers strongly prefer daily payment contracts over back-loaded ones, driven by fears of wage theft (28%), liquidity constraints (22%), and a demand for flexibility to break contracts (50%).
- Conversely, firms prefer back-loaded contracts due to their own liquidity constraints (37%) and costs associated with worker reneging (63%).
- These misaligned preferences lead to significant welfare losses, with 72% of workers rejecting job offers ultimately earning less income than the offers would have provided.
Identification Strategy
- The study employs three field experiments.
- A worker-side experiment cross-randomizes job offers with different payment structures (daily, smooth back-loaded, steep back-loaded) and an insurance contract against wage theft to isolate the effects of liquidity, wage theft concerns, and flexibility.
- A firm-side experiment offers firms different contract types, credit contracts (to alleviate firm liquidity constraints), and guarantor contracts (to compensate for worker reneging).
- Finally, a matching experiment observes the behavior of matched worker-firm pairs under these contracts.
Data
The paper uses data from field experiments conducted in Patna, India, involving 1,360 workers and 349 informal construction firms. Data sources include surveys with workers and firms (pre- and post-experiment), panel data on job outcomes at labor stands, and detailed observations on contract completion, firm reneging, and hours worked from matched worker-firm pairs.
Peter Deffebach, David Lagakos, Yuhei Miyauchi — Development Economics
This paper investigates how the spatial distribution of income within cities varies with a country's development level, using a novel granular dataset and a quantitative spatial model.
Finance Application
- The findings have direct implications for real estate asset pricing, particularly in cross-country or emerging market contexts, by highlighting how property values are driven by different factors (job access vs. amenities) depending on a city's development level.
- For household finance, the non-homothetic preferences and varying commuting costs can explain heterogeneous household savings, debt, and mortgage decisions across urban areas and income groups, especially in developing economies where job access is paramount.
- This could inform models of financial inclusion or default risk.
- Furthermore, understanding spatial income distribution can aid in valuing local businesses or regional equity markets, as consumer demand and labor supply are spatially concentrated.
Urban economicsSpatial economicsIncome distributionCity developmentCommuting costsAmenitiesReal estateHousehold financeAsset pricingEmerging marketsDeveloped marketsQuantitative spatial model
Core finding, identification, data
Core Finding
- The paper documents that spatial income gradients and amenity valuations differ significantly between developed and less-developed cities.
- Specifically, less-developed cities show steeper income declines with distance from city centers and associate natural amenities with poorer neighborhoods, while developed cities exhibit flatter gradients and associate amenities with richer areas.
- These patterns are primarily driven by nonhomothetic preferences for amenities, higher commuting costs, and more centralized jobs in less-developed contexts.
Identification Strategy
- The authors employ a quantitative spatial model with non-homothetic preferences, calibrated to U.S. cities, and then conduct counterfactual simulations.
- They estimate city-level commuting costs using a Pseudo-Poisson Maximum Likelihood estimator on model-implied commuting gravity equations, leveraging granular data on residential and workplace locations to identify the impact of distance and urban features on income distribution.
Data
The paper uses a new dataset covering 145,000 neighborhoods across 127 cities in 26 countries, combining household-level travel surveys (JICA) for developing countries, census/tax data for developed countries, LODES data for US commuting flows, and geographic data (World Settlement Footprint, OpenStreetMap, Amazon Web Services Terrain Tiles, HydroSHEDS) for amenities and city characteristics.
Jeanne Sorin — Development Economics
This paper estimates the net returns of road improvements in Kampala, Uganda, accounting for both benefits and heterogeneous land acquisition costs influenced by property rights, and then models optimal infrastructure investments under various institutional settings.
Finance Application
- The paper's insights are highly relevant for real estate asset pricing and infrastructure project finance in emerging markets.
- Investors in real estate or infrastructure funds in these regions must explicitly price in 'property rights risk' and the potential for non-market-value land acquisition, which can significantly alter project costs and returns.
- For sovereign and municipal finance, the 'tax revenue wedge' and high cost of domestic funds directly impact the creditworthiness of entities undertaking infrastructure projects.
- Financial institutions lending to or investing in bonds issued by such entities should incorporate these fiscal constraints and associated land acquisition risks (e.g., project delays, social unrest) into their credit risk models.
- Furthermore, the impact of property rights on household wealth and collateral could inform models for microfinance and development finance, while the quantification of land acquisition uncertainty could lead to demand for specialized political risk or project delay insurance products.
Infrastructure InvestmentProperty RightsLand Acquisition CostsReal Estate ValuationEmerging MarketsUrban EconomicsFiscal PolicyProject FinanceHousehold WealthCredit RiskSovereign DebtDevelopment EconomicsAsset Pricing
Core finding, identification, data
Core Finding
- The paper finds that road improvements in Kampala yielded significant net welfare gains ($119 per resident) primarily because weak property rights allowed for land acquisition below market value, circumventing high fiscal costs associated with market-value compensation.
- If land had been acquired at market value, net welfare gains would have been negative.
- The optimal allocation of road improvements is significantly influenced by the spatial heterogeneity of property rights and the tax revenue wedge, leading to higher welfare gains under the current 'voluntary land take' approach compared to full market compensation.
Identification Strategy
- The paper estimates local benefits by exploiting variation in the timing of road improvements, comparing traffic speeds on roads upgraded early versus late using Google Maps data, and analyzing property value changes in parishes receiving improvements using a retrospective panel of broker transactions.
- For land acquisition, it models the probability of landowners negotiating for compensation based on the market value of affected land and the local property rights regime, assuming the timing of upgrades was not based on predicted local speed impacts.
Data
The paper uses Google Maps API data (traffic speed, trip times), ride-hailing company data (motorcycle trips, commuting patterns), and two novel surveys: one with 377 real estate brokers (land market values, transaction data, appraisal exercise) and another with 548 landowners (actual land acquisition costs, negotiation behavior, property rights). It also incorporates public data from Open Street Map, KCCA, and Ugandan national surveys.
Dev Patel — Development Economics
This paper examines how Bangladeshi rice farmers form beliefs about soil salinity from ambiguous environmental signals, and how these beliefs drive adaptation decisions and economic outcomes, using natural and field experiments.
Finance Application
- The paper's insights into how ambiguous signals are interpreted based on prior beliefs could be applied to how investors and households perceive climate risk in financial markets.
- For instance, ambiguous news about climate events (e.g., 'extreme weather' without specifics) might be interpreted differently by investors with prior exposure to specific risks (e.g., coastal real estate vs. drought-prone agriculture), leading to heterogeneous asset pricing or insurance demand.
- Insurers could leverage the 'default hypothesis' concept to understand why policyholders might under- or over-estimate specific climate risks, informing product design and risk communication to encourage optimal adaptation investments in resilient homes or businesses.
Environmental BeliefsClimate ChangeAdaptationAmbiguous SignalsBayesian LearningNatural ExperimentsField ExperimentsSoil SalinityTechnology AdoptionReal EstateInsuranceHousehold FinanceClimate Risk
Core finding, identification, data
Core Finding
- Farmers endogenously interpret ambiguous environmental signals (like low yield) based on their prior beliefs about specific threats (e.g., salinity vs. fungus), leading to persistent belief errors.
- Salient shocks (saltwater floods) significantly increase perceived salinity and drive adaptation, while subtle shifts (underground salinity intrusion) have no detectable impact on beliefs despite similar true salinity changes.
- Misaligned beliefs lead to suboptimal technology adoption and reduced agricultural profits, with correct beliefs potentially increasing profits by 16%.
Identification Strategy
- The paper uses two natural experiments: saltwater floods, identified via a difference-in-differences design comparing floods with more saline surface water; and underground salinity intrusion, identified via a triple difference-in-differences design using sea level rise, ocean salinity, and distance to the coast.
- It also employs two field experiments: an information experiment randomly providing farmers with true soil salinity data to shift beliefs, and a seed experiment randomly offering salinity-tolerant seeds for free to study adoption and profit impacts.
Data
The study uses primary survey data from 2,279 rice farmers in Bangladesh, including their beliefs and planting behavior. It collects agronomic measurements of true soil salinity on farmers' plots, and integrates satellite data for ocean salinity, sea level elevation, and flood detection, along with river station measurements and a custom-built database of rice seed characteristics.
Sigurd Galaasen, Andreas R. Kostøl, Joan Monras, Jonathan Vogel — Labor Studies
This paper investigates the effects of immigrant-induced demand shocks on native workers' labor-market outcomes, distinguishing them from traditional supply-side effects.
Finance Application
- This research offers significant insights for asset pricing and household finance.
- In asset pricing, the identified immigrant-induced demand shocks could predict differential stock returns for firms in demand-exposed sectors and regions, particularly for locally-focused companies.
- For household finance, the documented income gains for native workers could be linked to changes in their savings behavior, local real estate investments, or demand for specific financial products like mortgages, providing a micro-foundation for aggregate consumption and investment patterns.
- The granular electronic payments data could also serve as a leading indicator for sector-specific consumption, informing investment strategies.
ImmigrationLabor MarketsDemand ShocksConsumptionWage IncomeSectoral AnalysisRegional EconomicsInstrumental VariablesElectronic PaymentsHousehold FinanceAsset PricingCredit RiskSectoral Returns
Core finding, identification, data
Core Finding
- The study finds large, positive, and persistent effects of demand exposure to EU expansion on native worker income.
- Specifically, natives in market-sectors at the 75th percentile of demand exposure experienced an annual real earnings increase of over 6,000 krone (1.3% of average earnings) compared to those at the 25th percentile between 2003 and 2015.
Identification Strategy
- The paper employs an instrumental variable (IV) approach, using a variant of the Card instrument.
- It predicts local labor market immigrant-induced labor supply shocks based on the distribution of immigrants from EU accession countries across local labor markets prior to the shock (2003) and aggregate immigrant inflows (excluding local inflows).
- These predicted shocks are then interacted with initial local immigrant intensities of production (for supply exposure) and consumption (for demand exposure) across sectors, leveraging the natural experiment of EU expansions in 2004 and 2007.
Data
The study combines employer-employee data (administrative records of employment histories and tax income from 2000-2015) with a novel dataset of electronic payments for all Norwegian residents (debit card and online bank wire payments from 2006 onwards), which captures sector and location of expenditure, and consumer nationality.
Sydnee Caldwell, Ingrid Haegele, Jörg Heining — Labor Studies
This paper uses discrete choice experiments to estimate the money-metric value of non-wage amenities for German workers, analyzing their impact on inequality and the gender pay gap.
Finance Application
- This research could be applied in household finance to understand how non-pecuniary benefits influence investment decisions, such as the willingness to accept lower financial returns for ESG-compliant investments or for platforms with superior user experience.
- In asset pricing, firms offering highly valued non-wage amenities might experience lower employee turnover and higher productivity, leading to more stable cash flows and potentially commanding a premium in equity valuations or a lower cost of capital, thus explaining part of the 'ESG premium.' For insurance, understanding the money-metric valuation of amenities like health benefits or flexible work could inform the design and pricing of employer-sponsored insurance plans, as firms might pay more for benefits highly valued by employees, influencing demand for private insurance.
Labor EconomicsBehavioral EconomicsHousehold FinanceAsset PricingESGInequalityGender Pay GapValuationDiscrete Choice ExperimentsNon-pecuniary Benefits
Core finding, identification, data
Core Finding
- The study finds that non-wage amenities have significant money-metric valuations for marginal workers, which vary widely across firms and even among workers within the same firm.
- Importantly, these valuations are non-negatively correlated with wages, implying that non-wage amenities do not mitigate inequality.
- A substantial portion of the gender pay gap is explained by gender differences in the valuation of these non-wage amenities.
Identification Strategy
- The identification strategy relies on discrete choice experiments embedded in a large-scale survey of German workers.
- By randomizing wages across hypothetical job offers from firms that respondents identify as likely application targets, the researchers are able to recover money-metric valuations for the set of marginal workers.
Data
The paper utilizes a large-scale survey of German workers, with responses linked to administrative data by the IAB. Financial support for the research was provided by the National Science Foundation (SES 2242439, "Firm Wages and Amenities").
Andrew C. Barr, Jonathan Eggleston, Steven Mello, Alexander A. Smith — Labor Studies
This paper investigates the long-term economic consequences of modest financial shocks, specifically those stemming from minor car crashes, on individuals' employment and earnings.
Finance Application
- This research offers significant insights for household finance and insurance.
- It highlights the fragility of household balance sheets to seemingly small, unexpected expenses, suggesting a need for better emergency savings products or micro-insurance tailored to cover such events.
- For insurance, it underscores the value of coverage beyond direct repair costs, informing optimal deductible design and the pricing of policies that account for the downstream economic consequences of insured events.
- In asset pricing, the aggregate impact of these widespread 'small' shocks on consumer spending and labor supply could influence macroeconomic factors relevant for equity risk premiums or credit risk for lenders exposed to vulnerable populations.
Household FinanceFinancial ShocksEconomic ResilienceLabor SupplyEarningsInsuranceAdministrative DataCausal InferenceVulnerable Populations
Core finding, identification, data
Core Finding
- Minor car crashes, acting as modest financial shocks, lead to persistent declines in employment and earnings, particularly among financially vulnerable populations.
- These negative effects are substantially mitigated during periods of higher liquidity and among individuals with access to greater social supports.
Identification Strategy
- The paper leverages minor car crashes as an exogenous, modest financial shock to identify causal impacts.
- By combining car crash reports with Census and tax records, the authors estimate the effects on employment, earnings, and social well-being, focusing on the financial rather than physical consequences of these events.
Data
The study uses a unique dataset combining car crash reports with administrative Census and tax records.
Zihan Hu, Hyejin Ku, Xuan Wang, Xinjue Yao — Labor Studies
This paper investigates how transitioning from temporary to permanent contracts affects worker productivity, focusing on the incentive effect of tenure uncertainty.
Finance Application
- This insight into how job security affects effort and productivity has significant implications for household finance and corporate governance.
- In household finance, it suggests that individuals facing employment uncertainty (e.g., gig workers or those on short-term contracts) might exhibit different savings rates, debt accumulation, or investment behaviors (e.g., more conservative) compared to those with permanent employment, potentially relaxing their financial discipline once job security is achieved.
- In corporate governance, this mechanism could explain variations in executive or fund manager performance over their tenure; managers might exert greater effort during initial, less secure periods to 'earn' their permanence, influencing firm valuation or fund returns.
- This could inform optimal executive compensation design, linking incentives to tenure uncertainty.
labor economicsincentivesproductivitycontractstenureemploymenthousehold financecorporate governanceexecutive compensationrisk-takingsavings
Core finding, identification, data
Core Finding
- The study reveals a significant 5% drop in worker productivity immediately after transitioning from fixed-term to permanent contracts, with the decline intensifying over time.
- This suggests workers exert additional effort under fixed-term contracts to secure permanent employment, particularly those with weaker outside options.
- The absence of this decline in a comparable plant with guaranteed permanent contracts reinforces that tenure uncertainty drives this effort.
Identification Strategy
- The authors leverage high-frequency production data from a Chinese garment factory where piece-rate workers experience no changes in tasks or wages upon attaining tenure, allowing them to isolate the pure incentive effect of tenure.
- A comparison with a Vietnamese plant where workers are guaranteed permanent contracts after two fixed-term contracts serves as a crucial counterfactual.
Data
The paper uses high-frequency production data from a Chinese garment factory focusing on piece-rate workers, and data from a comparable plant in Vietnam.
Itzik Fadlon, Briana Sullivan, Vedant Vohra — Labor Studies
This paper uses administrative data to analyze the Black-White earnings gap, focusing on the asymmetric impact of job ladder transitions and firm pay policies, revealing a significant racial penalty in career setbacks driven by differential sorting and liquidity constraints.
Finance Application
- The finding that Black workers face a higher racial penalty in career setbacks, partly due to greater reliance on premature retirement account withdrawals driven by liquidity constraints, has direct implications for household finance research on racial wealth gaps, emergency savings, and retirement planning.
- This insight could inform the design of targeted financial products or policy interventions aimed at improving financial resilience for vulnerable households.
- In asset pricing, the differential sorting of Black workers into firms with varying pay penalties could be integrated into ESG (Environmental, Social, and Governance) models, suggesting that firms with better diversity and inclusion practices might have more stable human capital and lower labor-related risks, impacting their valuation.
- For insurance, the higher propensity for early withdrawals among Black workers highlights a greater demand for or value of unemployment insurance or other liquidity-providing insurance products, which could influence pricing and product development in the personal lines insurance market.
Labor EconomicsRacial InequalityEarnings GapsJob MobilityFirm EffectsLiquidity ConstraintsHousehold FinanceWealth InequalityESGUnemployment InsuranceRetirement SavingsDiscrimination
Core finding, identification, data
Core Finding
- The paper finds a systematic race-specific asymmetry in earnings determination, where losses from downward job transitions are meaningfully larger than gains from upward transitions.
- Specifically, for a $1 earnings increase in upward transitions, downward transitions impose a $1.25 loss for White workers and a $1.50 loss for Black workers.
- A key finding is the 'racial penalty' where Black workers lose an additional $0.24 for every $1 decrease in White workers' pay during downward transitions, which is attributed to differential sorting into firms that penalize pay at a higher rate, rather than unequal pay within the same firms.
Identification Strategy
- The identification strategy leverages worker mobility, comparing movers across cohorts with differential 'treatment intensity' (the difference in average pay between origin and destination firms).
- It employs a flexible event-study framework with individual fixed effects and job-history-dependent firm effects, relaxing traditional exogenous mobility assumptions.
- The model's validity is assessed using 'Parallel Pre-Trends' and 'Parallel Post-Trends' tests to ensure comparability across mover cohorts and account for residual sorting.
Data
The paper uses population-wide IRS tax records linked with the Census Numident and the Longitudinal Business Database (LBD) for employer-employee matches from 2005-2019. It also incorporates Form 1099-R data to identify retirement account withdrawals and IRS 1040 filings for household income and spousal information.
Pauline Carry, Benny Kleinman, Elio Nimier-David — Labor Studies
This paper uses establishment mobility to disentangle the contributions of inherent "location effects" from the spatial sorting of workers and firms in explaining spatial wage disparities.
Finance Application
- This research offers a powerful framework for understanding regional differences in financial outcomes.
- In asset pricing, one could decompose local stock market returns or firm valuations into 'location effects' (e.g., inherent regional advantages like infrastructure) versus 'sorting effects' (e.g., the concentration of high-quality firms and skilled labor).
- This could help identify whether regional 'hot spots' are fundamentally more productive or simply attract better firms and talent, impacting investment strategies.
- For household finance, understanding if local wealth disparities are due to inherent location advantages or the sorting of high-earning individuals and firms could inform models of local housing bubbles, mortgage default risk, or regional savings rates.
- Insurers could use this to better price regional risks, distinguishing between exogenous local hazards and the endogenous concentration of high-risk/low-risk economic agents.
Spatial EconomicsLabor EconomicsFirm DynamicsWage InequalityIdentification StrategyFixed EffectsRegional EconomicsAsset PricingHousehold FinanceInsuranceSorting Models
Core finding, identification, data
Core Finding
- The study finds that location effects (e.g., local geography, infrastructure, agglomeration) account for only 2-4% of spatial wage disparities across French commuting zones.
- The majority of wage differences are instead driven by the spatial sorting and co-location of workers and firms, with worker composition explaining 30% and establishment composition 17%.
- The elasticity of local wages to population density, often cited as evidence of agglomeration, drops significantly from 0.06-0.08 to 0.007 when controlling for worker and firm composition.
Identification Strategy
- The paper leverages a 'double-mover' design, combining establishment mobility across locations with worker mobility across establishments and locations, to estimate separate worker, establishment, and location fixed effects.
- This allows for a decomposition of wage variance.
- The identification is supported by event studies showing no pre-trends, placebo tests for within-commuting zone relocations, and robustness checks using plausibly exogenous relocations driven by personal preferences.
Data
The study primarily uses French administrative records, including the Register of Establishment Relocation (1993–2021) for firm moves, linked employer-employee data (DADS Postes, 2002-2016) for worker wages and mobility, and the SINE survey of entrepreneurs. It also uses US commercial data from Dun & Bradstreet and SEC 10-K filings for publicly listed firms to validate relocation patterns.
Zoe B. Cullen, Bobak Pakzad-Hurson, Ricardo Perez-Truglia — Labor Studies
This paper investigates the causes and consequences of salary negotiations in the U.S. tech sector through field experiments and a theoretical model, highlighting the role of information frictions.
Finance Application
- This research offers significant insights for household finance and corporate finance.
- In household finance, it suggests that many consumers may not negotiate financial product terms (e.g., mortgage rates, credit card APRs, insurance premiums) due to similar information frictions or fear of 'backlash' from financial institutions.
- Simple informational nudges could be designed to encourage households to negotiate, potentially leading to substantial savings or better terms.
- In corporate finance, the findings imply that negotiation dynamics in private markets (e.g., M&A, private equity deals, debt restructuring) could be influenced by perceived 'negotiability' and behavioral biases of participants, leading to suboptimal deal terms if information frictions are not addressed.
negotiationinformation frictionsbehavioral economicsfield experimentlabor marketscompensationhousehold financecorporate financeinsurancenudgesdecision makingwage gap
Core finding, identification, data
Core Finding
- A light-touch informational intervention (encouragement treatment) significantly increased both negotiation attempts and compensation gains, particularly for individuals who underestimated negotiation success.
- In contrast, a heavily discounted negotiation coaching service did not meaningfully affect negotiation attempts, suggesting that extensive-margin uncertainty (e.g., fear of backlash or belief that offers are non-negotiable) is a more significant barrier than intensive-margin uncertainty (lack of skills or information on how to negotiate).
Identification Strategy
- The study employs two randomized controlled field experiments.
- The 'encouragement treatment' randomly assigned participants to receive a message encouraging negotiation with factual data on success rates.
- The 'coaching treatment' randomly offered a deep discount on a negotiation coaching service.
- These random assignments allow for causal identification of the effects of addressing extensive-margin and intensive-margin information frictions, respectively.
Data
The paper utilizes survey and experimental data collected from over 3,100 active job seekers in the U.S. tech sector, obtained in partnership with levels.fyi. This includes detailed information on participants' backgrounds, prior beliefs about negotiation, negotiation attempts, and changes in compensation terms.
Anders Humlum, Mette Rasmussen, Evan K. Rose — Labor Studies
This paper develops a new survey-based approach to measure non-wage amenities and compensating differentials in the labor market, finding that higher-paying firms generally offer slightly worse non-pay amenities.
Finance Application
- This research offers significant arbitrage opportunities for finance.
- In household finance, the quantified trade-offs between wages and amenities (e.g., layoff risk, stress, flexibility) can refine models of household labor supply, savings, and investment decisions, as non-pecuniary benefits directly impact effective income and financial planning.
- For asset pricing and ESG investing, firms offering superior amenities (e.g., lower layoff risk, better work-life balance) could exhibit lower labor turnover, higher employee productivity, and reduced litigation risk, translating into higher firm valuations or lower cost of capital, providing a micro-foundation for 'S' (social) factors in ESG models.
- The explicit measurement of 'layoff risk' could also inform pricing of unemployment insurance products.
Labor EconomicsCompensating DifferentialsNon-wage AmenitiesFirm EffectsHousehold FinanceAsset PricingESG InvestingLabor SupplyJob SearchSurvey DataAdministrative DataHuman Capital
Core finding, identification, data
Core Finding
- The study finds a significant negative correlation between firm wage effects and amenity values.
- Specifically, a 10% higher pay in a job is associated with a 5% reduction in the value of amenities for job movers.
- This implies that worse amenities at high-paying firms offset more than half of their wage advantage, and the variance in pay across firms overstates the variance in utility by 50%.
Identification Strategy
- The paper elicits workers' reservation wages to return to their previous jobs using a survey of 20,000 Danish job movers.
- It uses a two-way fixed effects model (AKM-like) to estimate firm-wide premia and match effects in amenity values.
- To address measurement error and potential correlation, firm wage effects are instrumented with split-sample estimates from an AKM model on the full Danish population excluding the survey sample.
Data
The paper uses a survey of 20,000 job movers in Denmark, eliciting information on specific amenities (flexibility, job security, perks, stress, respect) and reservation wages. This survey data is linked to administrative matched employer-employee data from Denmark's E-Income Register and demographic data from the Population Register (BEF).
Mary Kate Batistich, Timothy N. Bond, Sebastian Linde, Kevin J. Mumford — Labor Studies
This paper models college major selection under incomplete information and statistical discrimination, finding that black students choose more difficult majors due to employer biases, leading to lower observed returns to their human capital investments.
Finance Application
- The paper's insights into statistical discrimination and human capital investment have strong implications for household finance and credit markets.
- Financial institutions (lenders, insurers) may statistically discriminate against certain demographic groups, relying more on observable proxies (e.g., education, job type) if they perceive less precise information about underlying creditworthiness or risk for these groups, leading to disparities in loan approvals, interest rates, or insurance premiums.
- This framework could also be used to model how individuals from discriminated groups make human capital investment decisions (e.g., choosing riskier education paths) and how these choices impact their lifetime wealth accumulation, savings, and financial risk-taking behavior, contributing to understanding racial wealth gaps.
statistical discriminationhuman capitalsignalinginformation asymmetryracial inequalitymajor choicelabor economicshousehold financecredit marketsinsurancewealth inequality
Core finding, identification, data
Core Finding
- The paper theoretically shows that statistical discrimination (employers having noisier signals for black workers) pushes black students to choose more difficult majors, while student information frictions (noisiness of beliefs about aptitude for black students) push them towards less difficult majors.
- Empirically, using administrative and survey data, they find that statistical discrimination dominates: black students with similar academic preparation select and graduate in more difficult majors but face lower observed returns to major difficulty, and the racial wage gap increases with major difficulty.
Identification Strategy
- The identification relies on a theoretical model that generates testable predictions about major choice and labor market outcomes under varying degrees of statistical discrimination and student information frictions.
- These predictions are tested by comparing major selection and wage returns between black and white students, controlling for academic preparation (SAT scores) and other characteristics, and observing how these racial differences vary with major difficulty and academic preparation.
Data
The study uses administrative student transcript records from 12 large public universities (state schools sample), the American Community Survey (ACS) from 2011-2021 for labor market outcomes, and the Baccalaureate and Beyond (B&B) 2008/18 longitudinal study of college graduates for major choice, wages, and GPA.
Iacopo Morchio, Christian Moser — Labor Studies
This paper develops an equilibrium search model with endogenous firm pay, amenities, and hiring, using Brazilian linked employer-employee data to identify microeconomic sources of the gender pay gap and assess policy consequences.
Finance Application
- This paper's structural quantification of firm-level gender pay gaps, amenity provision, and 'gender wedges' (a proxy for taste-based discrimination or comparative advantage) offers novel insights for ESG investing.
- These metrics could serve as new 'S' (Social) factors, potentially influencing firm valuation, cost of capital, or stock returns.
- For household finance, understanding how gender-specific amenity valuations and labor market frictions shape lifetime earnings and job mobility can inform models of household saving, investment, and insurance demand, especially for products related to work-life balance or career flexibility.
Gender pay gapLabor economicsEquilibrium search modelCompensating differentialsFirm heterogeneityWorkplace amenitiesESG investingAsset pricingHousehold financeLabor market frictions
Core finding, identification, data
Core Finding
- The study reveals a significant gender pay gap (13.3 log points) in Brazil, largely driven by women sorting into lower-paying employers.
- Compensating differentials related to workplace amenities explain half of this gap, as women value amenities more.
- Counterfactual equal-treatment policies, while reducing pay gaps, lead to lower worker welfare and output due to adverse firm incentives.
Identification Strategy
- The model parameters are identified using a constructive proof based on linked employer-employee data.
- This involves: (1) using a log-additive wage equation (akin to AKM) to separate worker and gender-specific firm pay components; (2) identifying gender-specific employer ranks through revealed preferences based on firm size and worker flows; and (3) pinning down economy-wide elasticities of vacancy and amenity costs by matching aggregate labor share, the firm pay-profit gradient, and the aggregate amenity cost share.
- This allows for the recovery of firm-level utility offers, amenity values, and gender wedges.
Data
The primary data source is the Brazilian linked employer-employee register Relação Anual de Informações Sociais (RAIS) from 2007 to 2014, covering all tax-registered workers and firms. It includes detailed worker characteristics, job spells, and earnings. Firm productivity measures are augmented with Bureau van Dijk's Orbis Historical data.
Jing Cai, Sai Luo, Shing-Yi Wang — Labor Studies
This paper uses a field experiment to causally examine how financial incentives (signing bonuses) and workplace monitoring affect worker effort, performance, and retention in a Chinese automobile manufacturing firm.
Finance Application
- The nuanced trade-offs between compensation and monitoring for worker retention and performance offer rich avenues for corporate finance and ESG research.
- In corporate finance, these findings could inform models of optimal human capital investment and its impact on firm valuation, particularly for labor-intensive firms, by analyzing how different incentive structures affect labor costs, productivity, and ultimately, firm cash flows and stock returns.
- For asset pricing and ESG, the causal links between compensation/monitoring and employee well-being/turnover could be used to develop more robust 'social' metrics, investigating whether firms with retention-focused compensation strategies exhibit lower idiosyncratic risk or higher valuation multiples.
- In insurance, the finding that monitoring increases quit rates suggests a potential 'monitoring premium' or adverse selection problem, where intrusive monitoring might drive away lower-risk policyholders, thus informing optimal contract design to balance moral hazard reduction with customer retention.
field experimentlabor economicshuman capitalcompensationmonitoringworker effortemployee retentionmoral hazardadverse selectioncorporate financeESGhousehold financeincentive designChina
Core finding, identification, data
Core Finding
- Both financial incentives and monitoring increase worker effort, but through distinct channels: signing bonuses lead to longer working hours and reduced quit rates without improving performance evaluations, while enhanced monitoring improves performance as evaluated by managers but increases quit rates and does not affect hours worked.
- A cost-benefit analysis reveals that bonuses are cost-effective due to improved retention, whereas monitoring is not due to the high costs associated with increased turnover.
Identification Strategy
- The study employs a randomized controlled trial with new hires.
- For financial incentives, job applicants were randomly assigned to receive either a standard compensation package or one with a signing bonus (capturing selection and moral hazard).
- A subset of the original control group later received a surprise bonus (isolating moral hazard).
- For monitoring, workers in randomly selected production-line stations were assigned to increased oversight via additional visits from an independent monitoring team.
Data
The paper utilizes administrative data from the firm (August 2023-January 2024) including monthly worker performance metrics (evaluation scores, performance bonuses), earnings, hours worked, and employment dates. It also uses weekly station-level monitoring records and baseline/endline survey data on worker satisfaction, well-being, and social networks.
Lars Johannessen Kirkebøen, Edwin Leuven, Magne Mogstad, Jack Mountjoy — Labor Studies
This paper investigates how college enrollment, particularly in elite programs, causally influences assortative mating patterns, household formation, and economic outcomes in Norway.
Finance Application
- The causal link between elite education and the formation of high-earning, homogamous households has significant implications for household finance.
- These 'elite households' may exhibit distinct patterns in wealth accumulation, savings rates, and investment choices (e.g., higher demand for wealth management services, different risk tolerance for long-term assets) compared to other households.
- For insurance, the findings on delayed fertility and potentially altered risk profiles (due to higher human capital and stable incomes) could inform underwriting models for life, disability, and long-term care insurance, allowing for more precise risk assessment and product design for different educational attainment groups.
Assortative MatingHuman CapitalHousehold FormationIncome InequalityEducationCausal InferenceRegression DiscontinuityFertilityWealth AccumulationFinancial PlanningInsuranceLabor Markets
Core finding, identification, data
Core Finding
- The study finds that college enrollment, rather than pre-selection, primarily drives assortative mating by institution and field of study, especially among high earners from elite programs.
- Elite education significantly boosts female applicants' own earnings and their likelihood of matching with higher-earning elite partners, leading to elite household formation, higher household earnings, but delayed fertility.
- For men, the effects on partner earnings and fertility are negligible, and these match-making effects are concentrated among students attending the same institution at the same time, and are larger where opposite-sex peers are more abundant, suggesting significant search costs in the marriage market.
Identification Strategy
- The study exploits admission discontinuities in Norway's centralized college application system, which effectively randomizes applicants near unpredictable admission cutoffs into different programs.
- A fuzzy regression discontinuity design is employed, where crossing the admission threshold into a preferred program serves as an instrument for actual enrollment, allowing for the identification of causal effects of enrollment versus selection.
Data
The paper uses detailed Norwegian population register data, including specific education types (institution and field), labor market outcomes, marriage/cohabitation records, and application data from the Norwegian Universities and Colleges Admission Service for the years 1998-2004.
Andrew C. Barr, Jonathan Eggleston, Steven Mello, Alexander A. Smith — Public Economics
This paper investigates the long-term economic consequences of modest financial shocks, specifically those stemming from minor car crashes, on individuals' employment and earnings.
Finance Application
- This research has direct implications for household finance, highlighting the pervasive financial fragility where small, unexpected shocks can have long-lasting negative effects on household balance sheets and human capital.
- It could inform models of emergency savings, optimal insurance design (e.g., income replacement policies beyond vehicle repair), and the demand for short-term credit or liquidity.
- For asset pricing, understanding how these micro-shocks aggregate could shed light on labor income risk, consumption volatility, and the pricing of assets sensitive to household financial health or liquidity preferences.
Household FinanceFinancial FragilityUnexpected ShocksLiquidityInsuranceLabor IncomeConsumptionAdministrative DataCar Accidents
Core finding, identification, data
Core Finding
- The study finds that even minor car crashes, acting as modest financial shocks, lead to persistent declines in employment and earnings, particularly for financially vulnerable populations.
- These negative effects are significantly mitigated when individuals have higher liquidity or access to greater social supports.
Identification Strategy
- The paper leverages minor car crashes as an exogenous, quasi-random shock to identify causal impacts.
- By combining car crash reports with comprehensive Census and tax records, the authors are able to isolate the effects of these shocks on economic outcomes, controlling for other factors.
Data
The research utilizes a unique dataset that merges car crash reports with U.S. Census and tax records, providing detailed information on individuals' employment, earnings, and other socio-economic indicators.
Itzik Fadlon, Briana Sullivan, Vedant Vohra — Public Economics
This paper uses administrative data to analyze the racial penalty in earnings during job ladder transitions, finding that Black workers face larger losses in downward moves due to differential sorting driven by liquidity constraints and signaling.
Finance Application
- The findings have direct implications for household finance, particularly regarding wealth accumulation and financial resilience.
- The racial penalty in earnings during career setbacks, coupled with higher reliance on retirement withdrawals for liquidity among Black households, suggests that financial products and advice should be tailored to address these race-specific vulnerabilities.
- For credit risk modeling, lenders could incorporate these differential earnings dynamics to better assess default probabilities and recovery rates for different racial groups, potentially leading to more equitable lending practices or targeted support.
- Insurance providers (e.g., for income protection) could refine their actuarial models by accounting for the higher earnings volatility and liquidity stress experienced by Black workers during job transitions.
Racial inequalityLabor economicsJob mobilityEarnings dynamicsLiquidity constraintsHousehold financeCredit riskRetirement savingsSocial safety nets
Core finding, identification, data
Core Finding
- First, the paper finds a race-specific asymmetry in earnings determination, where losses from downward job transitions are significantly larger than gains from upward transitions; for a $1 earnings increase in upward moves, downward moves impose a $1.25 loss for White workers and a $1.50 loss for Black workers.
- Second, Black workers experience an additional "racial penalty" of $0.24 for every $1 decrease in White workers' pay during downward transitions, which is not due to differential pay within the same firms but rather to differential sorting into 'worse' jobs, driven by greater liquidity constraints and adverse signaling effects.
Identification Strategy
- The identification strategy leverages worker mobility and an event-study approach, using differences in average firm pay between origin and destination firms as the "treatment intensity" of a job move.
- It analyzes transitions up and down the job ladder, estimating race-specific passthrough rates.
- The model allows for individual fixed effects, job-history-dependent firm effects, and differential selection into upward/downward transitions by race, validated by "parallel pre-trends" and "parallel post-trends" tests.
Data
The paper uses population-wide IRS tax records linked with the Census Numident and the Longitudinal Business Database (LBD) for employer-employee matches from 2005-2019. It also incorporates Form 1099-R extracts to study retirement account withdrawals.
Anna Chorniy, Amy Finkelstein, Matthew J. Notowidigdo — Public Economics
This paper empirically examines how cash versus in-kind government transfers affect the consumption of temptation goods and explores the normative implications for optimal transfer design under self-control problems and mental accounting.
Finance Application
- This research offers insights into how liquidity shocks and mental accounting affect consumption patterns, particularly for 'temptation goods.' In household finance, these findings could inform the design of 'labeled' savings accounts or restricted spending mechanisms (e.g., for emergency funds or retirement contributions) to help individuals with self-control problems manage their finances more effectively, mimicking the non-fungibility of in-kind transfers.
- For asset pricing, understanding the cyclical consumption of temptation goods following cash payouts could reveal seasonal demand patterns for specific industries (e.g., alcohol, entertainment, or even certain healthcare services) that could be incorporated into factor models or trading strategies.
- Insurers could also leverage these insights to better price health and life insurance products by accounting for the behavioral impact of income timing and form on health-related behaviors and associated claims.
Household FinanceBehavioral EconomicsLiquidity ConstraintsConsumption SmoothingMental AccountingSelf-Control ProblemsIncome ShocksSocial WelfareHealth EconomicsPublic Finance
Core finding, identification, data
Core Finding
- The study finds that cash transfers (SSI) significantly increase emergency department visits for drug and alcohol use (20-30%) and new prescription drug fills (40-100%), while in-kind food transfers (SNAP) do not.
- A theoretical model attributes this non-fungibility to mental accounting and self-control problems, suggesting that a paternalistic social planner would optimally include a positive share of SNAP, which increases as self-control issues worsen.
Identification Strategy
The identification strategy exploits within-month variation in benefit payout timing: for SNAP, it uses the staggered monthly schedule based on the last digit of recipients' case numbers, and for SSI, it analyzes outcome changes around the first-of-the-month payout, employing a difference-in-differences approach comparing SSI recipients to other low-income adults.
Data
The paper utilizes two decades (1998-2019) of linked administrative data from South Carolina, including records of cash (Supplemental Security Income) and in-kind (Supplemental Nutrition Assistance Program) benefit receipt, detailed health care utilization (emergency department and hospital records), and Medicaid prescription drug fills.
Marco Palladino, Antoine Bertheau, Alexander Hijzen, Astrid Kunze, Cesar Barreto, Doğan Gülümser, Marta Lachowska, Anne Sophie Lassen, Salvatore Lattanzio, Benjamin Lochner, Stefano Lombardi, Jordy Meekes, Balazs Murakozy, Oskar Skans — Gender in the Economy
This paper quantifies the role of gender-specific firm wage premiums in explaining the private-sector gender wage gap across 11 countries, decomposing it into sorting and pay-setting channels.
Finance Application
- This research offers significant insights for asset pricing and corporate finance by highlighting firm-specific labor market dynamics.
- Firms with larger gender wage gaps or less equitable pay-setting mechanisms (e.g., lower rent-sharing for women) could be perceived as having higher ESG risks, potentially leading to lower valuations, higher cost of capital, or underperformance.
- In household finance, these firm-level disparities directly impact household income and wealth accumulation, particularly for female-headed households, influencing their savings, investment decisions, and demand for financial products like retirement plans or mortgages.
- This could also inform insurers about income stability risks for different demographic groups.
Gender wage gapFirm wage premiumsLabor economicsSortingPay-settingRent-sharingMatched employer-employee dataESGHousehold financeCorporate governanceHuman capitalWage inequality
Core finding, identification, data
Core Finding
- Firm wage premiums account for a substantial share (15-32%) of the gender wage gap, predominantly driven by women sorting into lower-paying firms, especially those with high part-time work incidence.
- Additionally, women receive only 89% of the rent-sharing benefits men receive within the same firms, with this pay-setting disparity being largest in high-wage firms.
Identification Strategy
- The study employs a harmonized research design across countries, applying the Kitagawa-Oaxaca-Blinder decomposition to gender-specific employer wage premiums.
- These premiums are estimated using the Abowd, Kramarz, and Margolis (AKM) two-way fixed effects model, which separates wage differences into portable worker fixed effects and firm fixed effects, allowing for the decomposition into 'sorting' (between-firm) and 'pay-setting' (within-firm) components.
Data
The paper uses harmonized matched employer-employee administrative data from 11 advanced economies: 10 European countries (Denmark, Finland, France, Germany, Hungary, Italy, the Netherlands, Norway, Portugal, and Sweden) and Washington State, USA, primarily covering the period 2010-2019, including high-quality information on work hours.
Jeffrey Clemens, Anwita Mahajan — Public Economics
This paper causally estimates the impact of federal pandemic aid to U.S. state and local governments on population health outcomes, including mortality, hospitalizations, testing, and vaccination rates.
Finance Application
- The paper's identification strategy, using political representation as an instrument for federal aid, could be adapted to study the political economy of financial interventions.
- For example, researchers could investigate how states or industries with greater Congressional representation receive disproportionate shares of federal financial bailout funds or regulatory relief during financial crises, impacting asset prices, firm valuations, or credit availability.
- Furthermore, the findings on health disparities and the cost-benefit analysis of life years saved could inform ESG investing by providing a framework to quantify the financial value of social impact from public health interventions, potentially influencing investment decisions in healthcare or social infrastructure bonds.
Public HealthPandemicCOVID-19Fiscal PolicyGovernment AidInstrumental VariablesPolitical EconomyHealth DisparitiesMortalityVaccinationTestingState and Local Government
Core finding, identification, data
Core Finding
- For every $1,000 increase in federal fiscal aid per state resident, states experienced 38 fewer deaths from all causes per 100,000 residents, with two-thirds of this reduction attributable to lower COVID-19 mortality.
- This aid also significantly reduced COVID-19 related hospitalizations and emergency room visits, primarily by increasing testing and vaccination rates, and disproportionately benefited non-Hispanic Black Americans, thereby reducing health disparities.
Identification Strategy
- The study employs an instrumental variables (IV) strategy, leveraging the fact that federal aid distributions were more generous to states with higher 'excess representation' in the U.S.
- Congress (measured as senators/representatives per million residents).
- This political representation serves as an exogenous instrument for federal aid, allowing the authors to isolate its causal effect on health outcomes.
Data
The paper utilizes state-level data from the CDC Wonder database for mortality, Kaiser Family Foundation (KFF) for all-cause hospitalizations, HHS COVID-19 Reported Patient Impact and Hospital Capacity for COVID-19 specific hospitalizations, Hopkins CSSE for COVID-19 cases and tests, and CDC for vaccination rates, supplemented by political and demographic covariates.
Nadja Dwenger, Anna Gumpert — Public Economics
This paper causally estimates the short-term and long-term effects of secondments from high-capacity to low-capacity public administrations on output quantity and quality, using German reunification as a natural experiment.
Finance Application
- This research offers valuable insights for understanding knowledge transfer and capacity building within large, complex financial organizations or regulatory bodies.
- For asset managers, secondments could be analyzed for their impact on deal flow (quantity) and investment decision quality (e.g., lower error rates, higher risk-adjusted returns).
- In household finance, one could study how secondments of experienced staff to new branches or acquired entities affect loan origination volume (quantity) and default rates or customer satisfaction (quality).
- For insurance, secondments could be used to transfer expertise in complex claims processing or underwriting to new markets or product lines, impacting claims efficiency and accuracy, ultimately affecting profitability and customer trust.
knowledge transferorganizational economicshuman capitalpublic administrationstate capacitysecondmentsproductivityqualityGerman reunificationquasi-experimentinstrumental variablesfinancial institutionsregulatory economicsmergers and acquisitionsinsurance operations
Core finding, identification, data
Core Finding
- Secondments significantly increased short-term output quantity (10.6% increase in declarations assessed per employee for a one-standard deviation increase in secondments) and had lasting positive effects on long-term output quality (a 0.28 standard deviation reduction in objections).
- The effects are more pronounced for complex tasks, suggesting successful knowledge transfer, and also indicate increased taxpayer trust.
Identification Strategy
- The study exploits a quasi-experimental setting from German reunification, where East German tax offices were partnered with West German tax offices for capacity building.
- The variation in secondments is exogenous to the East German tax offices' needs, driven by West German partner characteristics: the local real value of financial incentives for seconded officials, distance between partner offices (proxy for commuting costs), and relative size of the West German partner office (proxy for supply of volunteers).
- These instruments are uncorrelated with pre-existing differences in the former GDR.
Data
The paper uses a novel, major data collection effort, assembling a unique dataset with detailed information on secondments (workdays, individual-level data for a subset) at the East German tax office level for 1990-1996 (up to 2000 for some). It also uses official statistics on tax office performance (number of cases assessed for corporate, personal income, and wage tax; number of objections raised, granted, and rejected), local prices for building land, and tax office addresses.
Lucy Msall, Ole-Andreas Naess — Public Economics
This paper studies how the removal of "stepped-up basis" for inherited assets in Norway affected capital gains realizations and intergenerational wealth transfers, using novel individual-level data on unrealized capital gains.
Finance Application
- The findings offer rich insights for household finance and behavioral finance, particularly in modeling intergenerational wealth transfers and bequest motives under varying tax regimes.
- Wealth managers and financial planners could leverage this to design optimal tax-aware strategies for high-net-worth clients, guiding decisions on asset realization timing and inter-vivos gifts versus bequests in anticipation of policy changes.
- Furthermore, the observed surge in realizations could inform asset pricing and market microstructure research on how large-scale policy-induced selling pressure impacts liquidity, price discovery, and market depth for specific asset classes, especially illiquid ones like private businesses.
Capital Gains TaxInheritance TaxStepped-Up BasisCarryover BasisHousehold WealthIntergenerational TransfersRealization BehaviorTax PolicyQuasi-ExperimentDifference-in-DifferenceMicrodataNorwayBehavioral EconomicsWealth Management
Core finding, identification, data
Core Finding
- The removal of stepped-up basis for inherited stock in Norway led to a 31% increase in capital gains tax collections due to higher taxable realizations and a reduction in intergenerational transfers of highly-appreciated stock.
- This behavioral response was concentrated among wealthy households with highly appreciated stock portfolios and older investors with children, indicating strong altruistic bequest motives and sensitivity to tax incentives.
Identification Strategy
- The paper employs two difference-in-difference (DiD) strategies.
- The first exploits cross-sectional variation in portfolio composition, comparing individuals with positive unrealized capital gains in stock (treatment) to those with zero/negative gains (control).
- The second uses investor age variation, comparing older age groups (treated) to younger ones (control, 30-49), assuming stronger bequest motives in older groups.
- Both rely on parallel trends assumptions and use Poisson Pseudo Maximum Likelihood regression.
Data
The study uses novel individual-level data on unrealized capital gains in Norway, constructed from comprehensive transaction and wealth data from the Norwegian Tax Authority (1992-2018). This includes detailed income, wealth, real estate transactions, and listed/unlisted stock transactions, supplemented by external market data for valuation and validation.
Sakina Shibuya, Felipe Parra — Gender in the Economy
This paper investigates the causal impact of military base presence on sexual violence, fertility, and child support disputes in Colombian municipalities using an event-study approach.
Finance Application
- The causal link between military base presence (specifically drafted soldiers) and increased sexual violence, independent of general security or demographic shifts, presents several finance research opportunities.
- In asset pricing, this localized social risk could be a factor in municipal bond yields, reflecting higher social costs or potential legal liabilities for local governments, or in local real estate valuations, as households might discount properties in affected areas.
- For household finance, it could drive migration decisions, increase demand for personal security expenditures, or influence household savings behavior in response to heightened local social risk.
- Insurers could explore pricing this specific risk into personal injury, health, or even property insurance policies for businesses and residents in municipalities hosting such military bases, potentially leading to mispricing if this specific driver of risk is not adequately captured by existing models.
Military presenceSexual violenceColombiaEvent-studyCausal inferenceLocal economiesSocial riskHousehold financeReal estateInsuranceConflict studies
Core finding, identification, data
Core Finding
- The presence of military bases significantly increases registered sexual violence by 72% over 16 years, primarily driven by municipalities hosting standing units with drafted soldiers.
- This increase is not attributed to changes in population, security conditions, or reporting behavior, and it does not significantly affect fertility or child support disputes.
- Some evidence suggests spillover effects of increased sexual violence and decreased child support disputes in neighboring municipalities.
Identification Strategy
- The study employs an event-study approach, utilizing the staggered introduction of military bases across Colombian municipalities from 1998 to 2016.
- It uses a classical two-way fixed effects model, a modified version with division jurisdiction-year fixed effects, and primarily the de Chaisemartin and D'Haultfœuille (dCdH) estimation to address heterogeneous treatment effects and ensure robust causal inference.
Data
The paper constructs a novel municipality-year dataset on military base presence from national and local newspaper articles, army organizational charts, historical records, congressional reports, and legislative documents. It also uses sexual violence and child support lawsuit data from the Office of the Attorney General, fertility data from DANE birth certificates, and demographic and violence/security data from DANE and the Conflict and Violence module.
Erica M. Field, Rob Garlick, Kate H. Vyborny — Gender in the Economy
This paper experimentally investigates how commuting barriers, particularly safety and propriety concerns, constrain women's labor supply in urban Pakistan by randomizing offers of gender-segregated or mixed-gender transport services at varying prices.
Finance Application
- The findings on how non-pecuniary factors (safety, propriety) significantly influence labor supply decisions, even more than price, could be applied to understanding household portfolio choices, savings behavior, and human capital investment.
- For instance, perceived safety in financial markets or digital platforms might be a non-pecuniary barrier to participation for certain demographics, similar to mobility constraints for labor.
- This could explain differential adoption rates of fintech or investment products.
- The value placed on "safe" commuting options could translate into a premium for real estate in areas with better (or perceived safer) transport infrastructure or proximity to employment centers, especially for female-headed households or households with female workers, informing models of housing prices and urban development.
- This study also highlights safety concerns as a major barrier, which could inspire research into insurance products that mitigate perceived safety risks (e.g., ride-sharing safety insurance, personal safety devices linked to insurance) and how such products might impact labor force participation or consumption patterns.
labor supplygender economicsmobilitytransportationrandomized controlled trialnon-pecuniary factorshousehold financehuman capitalemerging marketssafetyproprietary concernsjob searcheconomic development
Core finding, identification, data
Core Finding
- Offering transport increases women's job application rates by 70%.
- This effect is almost entirely driven by women-only transport (150% increase), while mixed-gender transport has minimal impact (40% increase).
- Women value the women-only service more than large price discounts for mixed-gender services, highlighting the importance of safety and propriety concerns over pecuniary costs in labor decisions, even for "latent jobseekers."
Identification Strategy
- The study uses a randomized controlled trial (RCT) design.
- It randomizes offers of commuting services (control, mixed-gender, women-only) and price discounts (20%, 60%, 80%) at the enumeration block and jobseeker levels.
- The causal effect on job applications is identified by comparing application rates across these randomized treatment arms, instrumenting offered transport with assigned transport.
Data
The paper uses administrative data from a large job search and matching platform ("Job Talash") in Lahore, Pakistan. This includes data on 2,653 female jobseekers (recruited door-to-door, including non-participants), 172 job adverts from 120 firms, jobseeker demographics, education, work experience, occupational preferences, self-reported safety concerns, commuting distances, and posted salaries.
Ludovica Ciasullo, Martina Uccioli — Gender in the Economy
This paper finds that policy-induced changes in work arrangements, specifically increased schedule regularity, significantly reduce the child penalty on women's labor supply and earnings, but do not alter the gendered division of home production.
Finance Application
- This research offers a clean shock to household income stability and composition, which could be used to study household portfolio choice, savings behavior, and debt decisions.
- For instance, do households with higher and more stable female income shares exhibit different risk-taking in equity markets or demand for insurance products? The null result on home production division could also inform models of intra-household financial bargaining and consumption smoothing.
Child PenaltyLabor SupplyWork ArrangementsGender EconomicsHousehold FinanceIncome StabilityPolicy ImpactQuasi-ExperimentAustraliaTime UseHousehold IncomeLabor Market
Core finding, identification, data
Core Finding
- The Australian 2009 Fair Work Act, by allowing new mothers to maintain regular schedules while reducing hours post-childbirth, reduced the child penalty on their hours worked from a 47% to a 38% drop.
- This led to a significant increase in the female share of household income, yet surprisingly, the female share of home production remained unchanged.
Identification Strategy
- The study employs a quasi-experimental difference-in-differences approach, leveraging variation in the timing of the Australian 2009 Fair Work Act, the timing of childbirth, and differential exposure to the law across occupations and industries.
- A triple-difference strategy is used, comparing mothers before and after the act, across terciles of 'casual prevalence' in their pre-birth jobs.
Data
The paper primarily uses the Household, Income, and Labour Dynamics in Australia (HILDA) dataset from 2001-2019, which provides rich longitudinal data on work arrangements, time use, and family linkages for Australian households.
Ketki Sheth, Shanthi Manian, Shibiru Ayalew — Gender in the Economy
This paper uses a large-scale field experiment in Ethiopia to investigate whether loan officers discriminate against female business owners in capital allocation decisions and finds no evidence of such discrimination.
Finance Application
- This paper's findings challenge the common assumption of gender discrimination in credit markets, particularly for established small businesses, which is highly relevant for household finance.
- It suggests that observed gender gaps in credit access might stem from factors other than lender bias, redirecting policy focus towards demand-side constraints or business support.
- For entrepreneurial finance and venture capital, the machine learning analysis provides a robust benchmark, suggesting that gender is not a valuable signal for predicting business success.
- This could prompt research into whether VC or angel investors, who often rely on heuristics in early-stage funding, exhibit biases that lead to mispricing or missed opportunities for female-led startups.
Household FinanceCredit MarketsDiscriminationFinancial InclusionEntrepreneurial FinanceMachine LearningBehavioral EconomicsLending
Core finding, identification, data
Core Finding
- The study finds no evidence that loan officers discriminate against female-owned businesses in high-stakes capital allocation decisions, neither for competition prizes nor for loan consideration.
- Loan officers' beliefs about future business performance were accurate, and gender was not a significant predictor of business success, implying no trade-off between gender equity and effective capital allocation.
Identification Strategy
- The identification strategy relies on a field experiment embedded in a national business plan competition.
- Business-owner gender was randomly assigned to loan officers evaluating real business applications.
- This causal identification is complemented by incentivized belief elicitation to distinguish between taste-based and statistical discrimination, and machine learning algorithms to benchmark the predictive power of gender for business outcomes.
Data
The paper uses data from 3,696 evaluations of 916 real business applications by 84 loan officers from 13 financial institutions in Ethiopia. A follow-up survey 18 months after the competition provides actual business performance outcomes.
Namrata Kala, Madeline McKelway — Gender in the Economy
This paper uses a field experiment in rural India to demonstrate that communication training for married women causally increases their labor supply and earnings, particularly when they initially have stronger preferences for working than their husbands.
Finance Application
- This research suggests that intra-household communication frictions can significantly impede optimal household economic decisions.
- In household finance, this implies that financial literacy programs might be more effective if coupled with communication training, enabling spouses to better convey financial information and preferences.
- It also suggests that gender-specific financial products or advice could be tailored to account for communication dynamics, potentially improving household savings, investment risk-taking, or insurance uptake by empowering women to articulate their financial needs and goals more effectively within the household.
Household FinanceBehavioral EconomicsGenderCommunicationField ExperimentLabor SupplyFinancial Decision-MakingIndia
Core finding, identification, data
Core Finding
- The communication training led to a 53% increase in total earnings over 10 months (and 124% by month 10) for women who were more interested in working than their husbands at baseline.
- These effects were persistent for at least one year and were driven by women persuading their husbands to change their preferences regarding female employment, rather than shifting bargaining power or women's own interest in working.
Identification Strategy
- The study employs a randomized controlled trial (field experiment) with 1,540 married women in rural India.
- Women were randomly assigned to either a communication training group or an active control group.
- Causal effects are identified by comparing outcomes between these groups, with heterogeneity analyzed based on pre-specified spousal disagreement on female employment at baseline.
Data
The paper utilizes administrative data from a carpet manufacturer on program applications, daily attendance, productivity, and monthly earnings for 10 months. It also uses extensive survey data from women (at baseline, five weeks, and six months post-treatment) and husbands (six months post-treatment), including vignettes to assess communication styles and lab-in-the-field games with couples.
Todd R. Jones, Ezra Karger, Valerie Michelman — Gender in the Economy
This paper uses Australian longitudinal data and quasi-experimental methods to show that the 'child penalty' and associated time/income costs are highly nonlinear, with the first child incurring significantly larger marginal costs than subsequent children.
Finance Application
- This research offers crucial insights for household finance and insurance.
- The finding that the first child represents a large 'fixed cost' in terms of forgone income and time, while subsequent children have much lower marginal costs, implies that financial planning, savings rates, and debt accumulation strategies should differ significantly before and after the first birth.
- It suggests a higher demand for life and disability insurance around the first birth to hedge against this initial human capital shock.
- Moreover, understanding these nonlinear costs can inform the design of child-related financial products, parental leave policies, and housing market dynamics, as families' housing needs and mortgage decisions are influenced by these scaling effects.
Household FinanceChild PenaltyFertilityLabor SupplyTime UseConsumptionInsuranceLife CycleSavingsDebtHuman CapitalQuasi-Experiment
Core finding, identification, data
Core Finding
- The marginal cost of children is not constant; the costs of a first child along key dimensions of time and income are much larger than those of subsequent children.
- This pattern is robustly identified using event studies, pregnancy loss, and twin births, highlighting significant returns to scale in child-rearing, particularly in home production of childcare.
Identification Strategy
- The study employs several identification strategies: event studies around the first birth, pregnancy loss as an instrument for a first or second child, and twin births as an instrument for a second child.
- It also compares short-run effects among individuals with the same completed fertility to address selection concerns.
Data
The research primarily uses the Household, Income and Labour Dynamics in Australia (HILDA) Survey, a long-running individual-level longitudinal dataset (2001-present) that includes detailed information on time use, income, expenditures, family structure, and fertility.
Virginia Minni, Kieu-Trang Nguyen, Heather Sarsons, Carla Srebot Roeder — Gender in the Economy
This paper demonstrates how managers' gender attitudes causally shape workplace culture, reduce gender pay gaps, and increase promotion rates for women within multinational firms, with lasting and spillover effects.
Finance Application
- This research offers a novel, quantifiable metric for corporate culture and its impact on human capital, highly relevant for ESG investing.
- Asset managers could develop an 'Expat Manager Gender Norms Index' for multinational firms, testing if those with more progressive scores exhibit superior long-term stock returns, lower cost of capital, or reduced tail risk, thereby informing new ESG factors.
- In household finance, the improved career prospects and pay equity for women under progressive managers could be linked to different household savings, investment, and retirement planning behaviors.
- Insurers could also incorporate a firm's 'gender culture score' into group benefits pricing, as better retention and employee well-being might correlate with lower claims for group life, health, or disability policies.
Corporate CultureGender Pay GapDiversity and InclusionHuman CapitalMultinational FirmsManagerial EconomicsESGLabor EconomicsCultural TransmissionAsset PricingHousehold FinanceInsurance
Core finding, identification, data
Core Finding
- Managers from countries with one standard deviation more progressive gender attitudes reduce the gender pay gap by 5 percentage points (18%), primarily by increasing women's promotion rates.
- These effects persist after managers rotate out, spill over to other local managers (both peers and subordinates), and are strongest in more conservative destination countries, indicating a significant cultural transmission mechanism.
Identification Strategy
- The study leverages quasi-exogenous cross-country manager rotations within a multinational firm, using a triple differences design.
- It compares male and female workers before and after exposure to expat managers whose gender attitudes are proxied by World Values Survey data for their home country and birth cohort.
- This allows for controlling worker and manager fixed effects, as well as pre-trends, to establish causality.
Data
The paper uses detailed personnel records (2011-2021) from a multinational firm operating in over 100 countries, including data on performance, pay, job rotations, promotions, and supervisory relationships. It also incorporates employee survey responses (2017-2021) and World Values Survey data for gender attitudes, and for external validity, Brazil's Relação Anual de Informações Sociais (RAIS) employer-employee data (2009-2021).
Chiara Campana, Pierre Koning, Maarten Lindeboom — Economics of Social Security
This paper examines the long-term health and employment outcomes of Disability Insurance (DI) applicants in the Netherlands following a 2003 reform that increased screening intensity.
Finance Application
- The Chronic Disease Index (CDI) and Work Ability Index (WAI) developed in this paper, derived from medical prescription data, offer a novel approach for life and disability insurers to assess long-term mortality and morbidity risk.
- These indices could be integrated into underwriting models for more precise risk classification and pricing, potentially reducing adverse selection.
- Furthermore, the finding that stricter screening disproportionately impacts applicants with mental health conditions provides critical insights for designing mental health coverage and risk management strategies in both private and social insurance programs.
- For household finance, the study highlights how changes in social insurance eligibility can alter individual financial planning, influencing demand for private insurance products or retirement savings behavior.
Disability InsuranceHealth OutcomesEmployment OutcomesPolicy ReformRegression DiscontinuityScreeningTargetingChronic Disease IndexWork Ability IndexMedical PrescriptionsMortalityInsurance UnderwritingHousehold FinanceSocial SecurityRisk Management
Core finding, identification, data
Core Finding
- The 2003 'Gatekeeper Protocol' reform, which tightened DI screening, led to a 33% reduction in DI applications, primarily from individuals with hard-to-verify conditions (e.g., mental impairments).
- Compliers (those screened out) exhibited higher survival rates, lower medical expenditures, and higher employment rates up to 18 years post-application, indicating improved targeting.
- The study constructs a Chronic Disease Index (CDI) and Work Ability Index (WAI) from medical data, finding that screening was more effective in targeting 'health disability' than 'work disability,' largely driven by changes in the demographic composition of applicants (fewer women, more older workers).
Identification Strategy
- The identification strategy employs a Regression Discontinuity-in-Time (RDiT) design, leveraging the sharp discontinuity created by the January 2003 implementation of the 'Gatekeeper Protocol' reform.
- This reform significantly increased the costs and intensity of screening for DI applications, allowing for a comparison of applicant cohorts immediately before and after the policy change.
Data
The paper uses detailed administrative data from the National Social Insurance Institute (NSII) on DI applications and awards, linked with Statistics Netherlands (CBS) databases. This includes individual demographics, annual earnings (1999-2022), annual medical spending (2009-2021), and yearly medical prescription data (2006-2022) which are mapped to 22 chronic disease conditions.
Kristy Kim, Caleb K. Wroblewski — Economics of Social Security
This paper analyzes how a shift in federal government retirement benefits from defined benefit (DB) to a mixed DB-DC plan impacted employee retention and workforce composition.
Finance Application
- This research offers crucial insights for household finance by detailing how retirement plan design influences individual labor supply, savings rates, and optimal retirement timing, which can be integrated into life-cycle consumption and investment models.
- For corporate finance, it highlights the impact of non-wage compensation on human capital retention and firm productivity, informing executive compensation strategies and M&A due diligence regarding employee benefits.
- In asset pricing, these shifts could influence the demand for different asset classes (e.g., fixed income for DB liabilities vs. equities for DC accounts) and the valuation of companies based on their pension liabilities and human capital management.
retirement benefitsdefined benefitdefined contributionlabor supplyemployee retentionhuman capitalregression discontinuityhousehold financecorporate financepension plansnon-wage compensationadverse selectionimplicit taxes
Core finding, identification, data
Core Finding
- The transition from backloaded DB pensions to more portable mixed DB-DC plans significantly reduced federal career length by about nine months and decreased the likelihood of staying until retirement by 3 percentage points.
- This effect was concentrated among highly productive, educated, and higher-paid workers with better outside options, leading to adverse selection and a decline in average worker value.
- Conversely, the reduction in implicit taxes on continued work under the new system increased labor supply for older workers after retirement eligibility.
Identification Strategy
- The study employs a cohort-based regression discontinuity (RD) design with local randomization, leveraging a 1984 policy change that retroactively altered retirement coverage for newly hired federal workers.
- Workers hired just before and after the policy cutoff (between 1984 and 1986) were unaware of the new system's exact terms, ensuring comparability of cohorts and isolating the causal effect of the benefit change, as confirmed by no significant changes in worker characteristics or wages around the cutoff.
Data
The paper utilizes comprehensive payroll data from the Office of Personnel Management (OPM), obtained through FOIA requests, covering quarterly records of federal government workers from 1973 to 2022. The main sample focuses on full-time, non-seasonal workers hired within a narrow window around the 1984 policy change.
Anders Humlum, Pernille Plato — Economics of Social Security
This paper demonstrates that effective reskilling programs for workers injured in accidents lead to significant mental health benefits for both the workers and their partners, alongside improved labor market outcomes.
Finance Application
- This research offers direct insights for the insurance industry, particularly disability and workers' compensation.
- Insurers could integrate the quantified mental health and partner benefits of reskilling into their actuarial models, potentially leading to more accurate long-term claims forecasting, reduced mental health-related payouts, and improved return-to-work rates.
- In household finance, the findings on partner spillover effects (employment, relationship stability) highlight how health shocks and interventions impact household financial resilience, suggesting avenues to study how access to such programs influences household savings, debt, and consumption decisions post-shock.
InsuranceHousehold FinanceHuman CapitalHealth ShocksLabor Market OutcomesMental HealthSpillover EffectsDisability
Core finding, identification, data
Core Finding
- The core finding is that reskilling injured workers prevents one case of depression for every three participants, with equally large spillover effects on partners.
- These mental health benefits are greatest during schooling.
- Additionally, reskilling leads to higher partner employment and increased relationship separation, suggesting it frees partners from costly commitments.
- Overall, these mental health and partner benefits add 83% to the direct labor earnings gains from reskilling.
Identification Strategy
- The identification strategy exploits institutional variation in Denmark where similar vocational degrees grant different access to higher education programs, creating a quasi-random assignment to reskilling opportunities for injured workers.
- The authors use a triple-difference approach, combining a difference-in-differences design for work accidents with an inverse probability weighting (IPW) strategy to compare workers with and without access to reskilling.
Data
The paper uses comprehensive Danish register data from 1995 to 2017, linking individual-level records on work accidents, reskilling activities, labor market outcomes, health care utilization (including prescription drugs), and partner relationships.
Lindsay Jacobs — Economics of Social Security
This paper models the heterogeneous impacts of increasing Social Security retirement ages (Early Eligibility Age and Full Retirement Age) on work, claiming, disability, savings, and welfare across blue-collar and white-collar occupations in the U.S.
Finance Application
- The paper's findings on occupational heterogeneity in response to retirement policy and health shocks offer rich avenues for household finance and insurance research.
- In household finance, this framework could model how different occupational groups (e.g., physically demanding vs. sedentary jobs) optimally adjust private savings, investment portfolios, and annuity purchases in anticipation of, or reaction to, Social Security reforms, considering their differential health trajectories and productivity declines.
- For insurance, the observed higher SSDI application rates and welfare losses for blue-collar workers suggest a greater underlying demand for private disability or long-term care insurance among these groups, allowing insurers to develop more targeted products and refined pricing strategies based on occupational risk profiles and policy sensitivity.
Social SecurityRetirement AgeOccupational HeterogeneityLabor SupplySavingsDisability InsuranceHousehold FinancePublic PolicyDynamic ModelsHealth ShocksWelfare Analysis
Core finding, identification, data
Core Finding
- Increasing the Early Eligibility Age (EEA) has larger labor supply and disutility effects for blue-collar workers, leading to greater SSDI applications and welfare loss for this group.
- Conversely, increasing the Full Retirement Age (FRA) primarily affects white-collar workers' labor supply and increases savings for all, with blue-collar workers experiencing greater welfare loss due to steeper productivity decline with age and less margin to respond to policy changes.
Identification Strategy
- The paper employs a dynamic structural model of work, savings, and Social Security decisions, estimated using Health and Retirement Study (HRS) data linked with O*NET occupational task data.
- The model's parameters are identified by matching simulated behavior to observed responses to past FRA increases in the HRS data, and then used to predict responses to hypothetical future EEA/FRA changes, explicitly accounting for preference and state heterogeneity across occupations.
Data
The study uses data from the Health and Retirement Study (HRS) for U.S. males born between 1931-1950, covering biennial interviews from 1992 to 2018. It also integrates O*NET data on occupational tasks to classify jobs by physical intensity.
Sydney Gordon — Economics of Social Security
This paper examines how a decrease in Social Security Administration (SSA) field office staffing during the Reagan era affected enrollment in Old-Age, Survivors, and Disability Insurance (OASDI) and Supplemental Security Income (SSI) programs.
Finance Application
- This research offers a valuable framework for household finance and insurance.
- Reduced access to public safety nets due to administrative barriers could compel households to increase private savings or demand for private disability and life insurance, influencing optimal portfolio allocation and retirement planning.
- Insurers could incorporate government administrative efficiency as a factor in pricing private policies, as it affects the overall risk exposure of individuals and the potential for 'crowd-in' effects in private markets.
Government policySocial SecurityDisability InsuranceSSIAdministrative burdenHousehold financeRetirement planningPrivate insuranceLabor supplyPublic economics
Core finding, identification, data
Core Finding
- A 10% decrease in SSA field office employees in a county led to a 0.06% decrease in OASDI enrollment (primarily driven by disability insurance) and a 0.32% decrease in SSI enrollment.
- These reductions imply a national shortfall of 79,027 enrollments, disproportionately affecting lower-income individuals and SSI recipients, suggesting administrative capacity significantly impacts benefit access.
Identification Strategy
- The study employs a long-differences approach, comparing changes in benefit enrollment between 1984 and 1990 to changes in field office staffing over the same period.
- The key identifying assumption is that the downsizing, achieved through employee attrition rather than layoffs, provides exogenous variation in staffing levels, avoiding endogeneity issues related to local program demand.
- Controls for local economic conditions, population changes, and state fixed effects are included.
Data
The paper uses federal government employee microdata from the Office of Personnel Management (OPM) for SSA employees (1974-2014), Social Security beneficiary data at the county-year level for OASDI and SSI programs, and 1980/1990 decennial census data for demographic controls. Placebo tests utilize program expenditure data from the US Bureau of Economic Analysis (BEA).
Eric Zitzewitz — Economics of Education
This paper examines how a college (UCLA) navigated the conflicting pressures of an affirmative action ban and diversity targets by adopting holistic admissions, and quantifies the 'cost' of this approach.
Finance Application
- This paper's insights into the 'cost' of achieving diversity targets under legal constraints could be highly relevant for ESG investing and corporate finance.
- Firms increasingly face pressure to enhance diversity (e.g., on boards, in leadership) while adhering to 'race-neutral' hiring laws.
- The methodology of quantifying the trade-offs between diversity goals and traditional performance metrics (e.g., academic qualifications in this paper, or profitability/efficiency in a corporate context) could be adapted to evaluate the financial implications of corporate diversity initiatives or ESG mandates.
- It also has implications for household finance, as changes in college admissions policies directly impact human capital formation, career trajectories, and intergenerational wealth accumulation for different demographic groups.
affirmative actiondiversitycollege admissionshuman capitalESGcorporate social responsibilityhousehold financepolicy evaluationquasi-experimentlabor markets
Core finding, identification, data
Core Finding
- UCLA's shift to holistic admissions in 2006, in response to pressure for increased Underrepresented Minority (URM) enrollment under an affirmative action ban, raised the URM share of admitted students by 3 percentage points.
- This increase was achieved primarily by altering admissions decisions *within* racial/ethnic groups, favoring applicants with URM-like characteristics (e.g., grades over SAT scores, attendance at low-test-score high schools), and was roughly 4-5 times more 'costly' in terms of UCLA's pre-2006 preferences than a simple reallocation of slots.
Identification Strategy
- The paper employs a quasi-experimental design, analyzing UCLA's admissions data from 2004-2009, which spans a significant policy change in 2006.
- This change involved UCLA's adoption of a holistic admissions process in direct response to protests over low URM enrollment, while operating under an existing state-level affirmative action ban (Proposition 209).
- The identification relies on comparing admissions criteria and their 'costs' before and after this exogenous policy shift.
Data
The paper uses UCLA undergraduate admissions data for the College of Letters and Science from 2004-2009, obtained via a Public Records Act request. This dataset includes applicant characteristics such as test scores (SAT Math, Reading, Writing, Subject Tests), GPAs (unweighted and weighted), socioeconomic indicators (family income, parental education, high school API decile), and racial/ethnic background. It also references IPEDS data for overall enrollment trends.
C. Kirabo Jackson — Economics of Education
This paper investigates a Chicago policy that granted increased autonomy to school principals, finding positive average effects on student outcomes, significant heterogeneity driven by principal quality and school-specific needs, and high cost-effectiveness.
Finance Application
- The findings on optimal decentralization and the importance of agent quality and local context could be directly applied to corporate governance and asset management.
- For instance, how does delegating more autonomy to branch managers in retail banking or portfolio managers in investment firms affect their performance, risk-taking, and client outcomes? The observed heterogeneity suggests that such delegation should be selective, favoring high-quality managers aligned with firm objectives and those operating in diverse markets with unique needs.
- This framework could also inform the design of incentive structures in insurance, where agent autonomy in tailoring policies to heterogeneous client risk profiles might lead to more efficient outcomes and higher retention, similar to the paper's 'stability channel' for principals.
organizational economicscorporate governanceagency theorydecentralizationleadershipheterogeneitycausal inferenceeducation economicshuman capitalasset managementhousehold financecost-effectiveness
Core finding, identification, data
Core Finding
- The Chicago Independent School Program (ISP), which granted principals greater autonomy, improved math and English passing rates by 0.16 standard deviations, comparable to more resource-intensive interventions but at a minimal cost of under $50 per pupil.
- The effects are highly heterogeneous, with high-performing principals and schools with atypical student populations benefiting most, suggesting that local capacity, aligned incentives, and heterogeneity are key to successful decentralization reforms.
Identification Strategy
- The study employs a differences-in-differences (DiD) approach with matched comparison schools.
- Each treated ISP school is matched to a set of non-ISP schools based on pre-ISP characteristics using Mahalanobis distance, allowing for each treated school to have its own counterfactual time path.
- Event-study plots confirm parallel trends before treatment, and permutation tests rule out selection on anticipated future gains.
- Deconvolution kernel density estimation is used to recover the distribution of true effects, accounting for sampling variability.
Data
The paper uses publicly available school-level data from Chicago Public Schools (CPS) websites, linked to the Common Core of Data (CCD), Illinois Standards Achievement Test (ISAT), Partnership for Assessment of Readiness for College and Careers (PARCC), and NWEA district tests. It also incorporates 5Essentials survey data for school climate and leadership scores, and personnel files for principal turnover and spending categories.
Barbara Biasi, Minseon Park, John D. Singleton, Seth D. Zimmerman — Economics of Education
This paper uses a regression discontinuity design to causally identify how school board members' ideologies, rather than just their demographic identities, affect organizational outcomes and policy changes in California school districts.
Finance Application
- This research offers a powerful framework for corporate governance studies.
- LLMs could extract 'ideologies' (e.g., growth vs. value, risk-aversion, ESG commitment) from corporate directors' public statements, proxy statements, or past professional histories.
- Applying a similar RD design to close elections for corporate board seats or other quasi-experimental settings could causally identify how specific board ideologies, beyond simple demographic diversity, impact firm investment, capital structure, executive compensation, M&A activity, or even stock market performance and risk-taking.
- The distinction between identity and ideology is critical for understanding the true drivers of corporate governance effectiveness.
Board GovernanceCorporate GovernanceIdeologyIdentityRegression DiscontinuityCausal InferenceText AnalysisLarge Language ModelsLLMsPolicy OutcomesDecision MakingESGExecutive Compensation
Core finding, identification, data
Core Finding
- Board members' stated policy agendas (ideologies) significantly impact organizational outcomes, with a mean effect of 0.16σ across tested policy domains, through pivotal votes and agenda-setting.
- In contrast, demographic identities are poor proxies for ideology, and their limited governance effects are fully explained by differences in policy priorities.
Identification Strategy
- The study employs a regression discontinuity (RD) design, leveraging exogenous variation from narrowly-decided elections.
- It compares outcomes in years following elections where a candidate with a specific attribute (identity or ideology) just wins or loses against a candidate without that attribute, ensuring causal inference.
Data
The paper uses a novel dataset of 22,426 school board candidates in California (1998-2022), linking them to policy priorities extracted from candidate platforms using Large Language Models (LLMs). It also uses LLMs to analyze board meeting minutes (2016-2023) and combines these with public datasets on school district financial, workforce, and student outcomes.
Mary Kate Batistich, Timothy N. Bond, Sebastian Linde, Kevin J. Mumford — Economics of Education
This paper models college major selection under incomplete information and statistical discrimination, finding that Black students choose more difficult majors but face lower returns due to employer discrimination, while also having less precise beliefs about their aptitude.
Finance Application
- This research offers significant insights for household finance and credit/insurance markets.
- The finding that statistical discrimination leads to 'optimal mismatch' where certain groups undertake riskier human capital investments but face lower returns could be directly applied to financial investment decisions.
- For instance, do minority investors, facing perceived or actual discrimination in wealth management or entrepreneurial funding, choose riskier or less liquid assets to over-signal their financial acumen, only to achieve lower risk-adjusted returns due to market biases? In credit markets, statistical discrimination could explain why certain demographic groups receive less favorable loan terms or are pushed towards subprime products, even with similar objective creditworthiness, impacting their wealth accumulation and financial stability.
- Insurers could also be statistically discriminating in pricing, leading to suboptimal insurance choices or higher costs for certain groups, impacting their financial resilience.
Statistical DiscriminationHuman CapitalSignaling TheoryInformation AsymmetryRacial InequalityHousehold FinanceCredit MarketsInsurance PricingWealth AccumulationInvestment Decisions
Core finding, identification, data
Core Finding
- The paper finds strong evidence that statistical discrimination dominates student information frictions in major choice.
- Black students, with similar academic preparation to White students, select more difficult majors but experience lower observed labor market returns to major difficulty.
- This outcome is consistent with employers statistically discriminating, leading Black students to 'mismatch' by over-signaling their ability, and also confirms that Black students have less precise beliefs about their aptitude.
Identification Strategy
- The identification strategy relies on a theoretical model of major choice under incomplete information and statistical discrimination, which generates testable predictions about racial differences in major selection and labor market outcomes.
- Empirically, these predictions are tested by examining how major difficulty (measured by wage return, percentile return, and STEM course credits) and its interaction with race affect major choice and log earnings, controlling for academic preparation (SAT scores) and institutional quality.
Data
The paper uses administrative student transcript records from 12 large public universities (state schools sample), the American Community Survey (ACS) from 2011-2021 for labor market outcomes, and the Baccalaureate and Beyond 2008/18 (B&B) longitudinal study for college graduates. It also incorporates home zip code socioeconomic status characteristics (median income, education, and income mobility).
Oliver Schlenker — Economics of Health
This paper examines how nursing shortages, induced by cross-border wage differentials after the 2011 Swiss franc stabilization, affect healthcare provision and patient outcomes in German hospitals.
Finance Application
- This paper's findings have direct implications for insurance and asset pricing.
- Increased mortality and reduced healthcare quality due to labor scarcity could significantly impact health and life insurance claims, necessitating adjustments in premium pricing and risk models for insurers operating in affected regions.
- For asset pricing, the declining health outcomes and potential long-term economic stagnation in regions facing healthcare labor scarcity could affect the valuation of local businesses, real estate, and municipal bonds, introducing a novel regional risk factor for investors.
- Furthermore, the fixed-price and wage-rigid environment of German hospitals provides a unique setting to study how labor supply shocks propagate through firms with limited pricing power, offering insights for valuing firms in other regulated or human-capital-intensive sectors.
Labor scarcityHealthcare economicsNatural experimentDifference-in-differencesMortality riskLife expectancyHealth insuranceLife insuranceHuman capitalRegional economicsFixed pricesWage rigidityHospital financeHousehold riskAsset pricing
Core finding, identification, data
Core Finding
- German border hospitals experienced a 12.5% reduction in nursing staff, leading to decreased care intensity (e.g., a 12% drop in surgery likelihood) and a significant increase in mortality rates, particularly for older and urgent cases (e.g., 0.4%-points for 75+ urgent cases, 2.35% for sepsis, 1.56% for heart attacks).
- These effects translated into a stagnation in regional life expectancy, highlighting the fragility of healthcare systems to labor market regulations and scarcity.
Identification Strategy
- The paper employs a multi-period difference-in-differences ("event study") design, leveraging the exogenous shock of the 2011 Swiss franc stabilization.
- This event created a significant and stable cross-country wage differential, making cross-border commuting to Switzerland attractive for German registered nurses.
- Treatment is assigned to German counties near the Swiss border, and control counties are selected using propensity score matching based on pre-event demographic and economic characteristics.
Data
The study uses rich administrative data from Germany and Switzerland, including the Swiss Cross-Border Commuter Statistic, German Hospital Statistic, Integrated Employment Biographies, German Patient Statistic (18-19 million observations annually), and regional demographic and economic data from INKAR and the German Federal Statistical Office.
Marcella Alsan, Crystal Yang — Economics of Health
This paper uses a randomized controlled trial to evaluate the causal impact of health care accreditation on quality standards and inmate mortality in US jails.
Finance Application
- The demonstrated causal link between accreditation and reduced mortality has direct implications for **insurance** markets, particularly for health and life insurance provided to correctional facilities; insurers could use accreditation status to refine risk assessment and pricing, potentially offering lower premiums to accredited jails.
- For **ESG investing** and **social impact bonds**, accreditation provides a robust, verifiable metric for social impact, enabling the design of financial products that reward improvements in correctional health outcomes.
- Furthermore, the paper's model of information frictions and quality assurance could be adapted to study the impact of external audits or certifications on **operational risk** and compliance within financial institutions, examining how such mechanisms affect internal coordination and performance.
Health EconomicsPublic EconomicsRandomized Control TrialCausal InferenceInsuranceSocial ImpactESGOperational RiskPublic HealthCorrectional FacilitiesMortalityQuality Standards
Core finding, identification, data
Core Finding
- The study finds that health care accreditation significantly improves health care quality standards, particularly in personnel training, patient care, and prevention and safety protocols.
- Crucially, this accreditation causally reduces inmate mortality by approximately 70% (ITT), with these reductions concentrated among preventable and acute deaths.
Identification Strategy
- The paper employs a Randomized Controlled Trial (RCT) design, randomizing 44 US jails into a treatment group (22 jails receiving immediate, subsidized health care accreditation) and a control group (22 jails receiving accreditation after the study period).
- This rigorous experimental setup allows for a clean causal identification of accreditation's impact on health outcomes and quality standards.
Data
The study collects primary data through comprehensive facility surveys (including protocols and procedures), staff surveys, detailed medical audits (with baseline look-back and cause of death analysis), and qualitative interviews with jail leadership and incarcerated individuals. Independent media reports are also used to validate death logs.
Jonathan Arnold, Atul Gupta, Alexander L. Olssen, Tong Liu — Economics of Health
This paper investigates the effects of hospital consolidation on prices and costs, and the underlying mechanisms, focusing on acquisitions of independent versus system-owned hospitals in California.
Finance Application
- The finding that hospital M&A is largely driven by 'pricing arbitrage' (leveraging superior bargaining power) rather than operational efficiencies could inform M&A strategies and valuation in other regulated industries, such as regional banking, utilities, or other healthcare sub-sectors (e.g., physician groups, nursing homes).
- This could lead to new factors in M&A event studies, predicting deal success or returns based on the bargaining power differential between acquirer and target.
- For private equity, this suggests a key value creation lever in healthcare investments, prompting analysis of how PE-backed consolidations exploit such arbitrage opportunities and their implications for long-term returns and regulatory risk.
- Furthermore, the quantified pass-through of cost savings to insurers (around 50%) could be a valuable input for insurance pricing models and for assessing the impact of consolidation on insurance premiums in various markets.
M&AHealthcareHospital IndustryBargaining PowerCost SynergiesPrice EffectsIndustrial OrganizationPrivate EquityInsuranceMarket Power
Core finding, identification, data
Core Finding
- Hospital systems acquiring independent hospitals lead to significant net price increases and cost savings, but the price increase is an order of magnitude higher than cost savings.
- Over 90% of this price increase is due to the target hospital benefiting from the acquirer's superior bargaining ability (pricing arbitrage), rather than increased market power or passed-through cost synergies.
- Acquisitions of system-owned targets show no detectable effects on prices or costs.
Identification Strategy
- The study employs a staggered difference-in-differences (DiD) research design to estimate causal effects, comparing trends for acquired hospitals against non-acquired controls, using an estimator robust to staggered treatment and heterogeneity.
- This is complemented by a structural model of patient demand and hospital-insurer price negotiations to disentangle the contributions of market power, bargaining ability, and cost synergies.
Data
The paper utilizes novel proprietary hospital outpatient billing claims data (2013-2020) for California, providing detailed patient-level claims and negotiated prices for privately insured patients. It also uses detailed administrative data on hospital operating costs from annual reports submitted to California's Department of Health Care Access and Information (HCAI) for the same period.
Rebekah Dix, Kelsey Moran, Thi Mai Anh Nguyen — Economics of Health
This paper quantifies the direct and allocative effects of electronic health record (EHR) system interoperability on patient outcomes and hospital patient flows in the US healthcare market.
Finance Application
- This research offers several avenues for finance applications.
- First, the impact of interoperability on patient flows and costs could inform M&A strategies in fintech or insurtech, where mergers that enhance data compatibility might yield significant value.
- Second, the observed network effects and market dominance of Epic due to superior interoperability could be studied in asset pricing to understand how investors value firms with strong network advantages or proprietary data ecosystems (e.g., payment networks, alternative data providers).
- Third, the quantification of 'technological frictions' and their welfare costs could inspire research on how information asymmetry or incompatible data standards affect market efficiency, liquidity, and asset pricing in specific financial markets, such as cross-border investments or ESG data integration.
- Finally, for insurance, the direct link between EHR interoperability and reduced healthcare costs could lead to studies on how health insurers price policies based on provider network interoperability, or how investments in health tech by insurers impact their profitability and risk management.
healthcare economicsEHRinteroperabilitytechnological frictionsnetwork effectsdifference-in-differencesinstrumental variablespatient outcomeshospital competitionwelfare analysisinformation economicsmarket powerfintechinsurtechM&Aasset pricinginformation frictions
Core finding, identification, data
Core Finding
- The study finds that when hospitals align on the same EHR vendor, patient charges and readmissions decrease (e.g., 4% drop in charges for transfers, 11% decline in 60-day readmissions for referrals), and patient transfers/referrals between them increase by 8-10%.
- Eliminating these interoperability frictions could reallocate 7.5% of transfer patients and increase welfare by 21%, equivalent to a 57-kilometer reduction in travel distance, with most gains coming from direct utility effects rather than allocative efficiency.
Identification Strategy
- The primary identification strategy uses a difference-in-differences (DiD) design, comparing patient outcomes and flows for hospital pairs before and after they switch to the same EHR vendor, using non-switching pairs as controls.
- To address endogeneity of EHR choices and measurement error, an instrumental variables (IV) approach is employed, leveraging 'system pressure'—the share of beds among other hospitals in the same system (but outside the focal hospital's local market) that use a particular EHR vendor—as an exogenous shifter for vendor adoption.
Data
The paper utilizes hospital-level data on EHR vendor choices and interoperability from the American Hospital Association (AHA) Annual Survey and IT Survey (2005-2019), supplemented by the Health Information and Management Systems Survey (HIMSS) (2005-2017). Patient-level outcomes and flows are derived from Centers for Medicare and Medicaid Services (CMS) Medicare Claims data (MedPAR, Outpatient, Carrier files), and patient sharing relationships are captured by DocGraph Hop Teaming data (2013-2019).
Silvia H. Barcellos, Leandro Carvalho, Kenneth Langa, Sneha Nimmagadda, Patrick Turley — Economics of Health
This paper causally links education to reduced incidence and delayed onset of Alzheimer's Disease and Related Dementias (ADRD), showing that education can mitigate genetic risk for ADRD.
Finance Application
- This research offers significant insights for household finance and insurance.
- In household finance, the causal link between education and reduced dementia risk implies that human capital investment (education) can be a powerful tool for mitigating future healthcare costs and preserving cognitive function for financial decision-making in old age.
- This could influence optimal lifecycle savings, retirement planning, and the demand for long-term care insurance, especially for individuals aware of their genetic predispositions.
- For the insurance industry, these findings suggest that educational attainment, particularly when combined with genetic risk information (if ethically and legally permissible), could be incorporated into more refined underwriting and pricing models for long-term care and disability insurance products.
- Insurers might offer lower premiums to highly educated individuals, reflecting their reduced ADRD risk, thereby creating new market segments or pricing strategies.
EducationDementiaAlzheimer's DiseaseGenetic RiskRegression DiscontinuityHuman CapitalHousehold FinanceInsuranceLong-Term CareHealth EconomicsLifecycle Planning
Core finding, identification, data
Core Finding
- The study finds that one additional year of schooling causally reduces ADRD cumulative incidence by approximately 1 percentage point.
- Furthermore, the compulsory schooling reform weakened the relationship between genetic predisposition and ADRD incidence, particularly for individuals with higher genetic risk, suggesting that genetic risk is not immutable and can be modified by social policy like education.
Identification Strategy
- The identification strategy employs a regression discontinuity design (RDD) by exploiting a natural experiment in the UK.
- In 1972, a compulsory schooling reform (ROSLA) increased the minimum school-leaving age from 15 to 16, affecting only students born on or after September 1, 1957.
- This created a sharp discontinuity in educational attainment based on birthdate, allowing for causal inference on the effect of education on dementia incidence.
Data
The study uses data from the UK Biobank (UKB), a large population-based study. Dementia outcomes are identified using hospital records, mortality data, and self-reported medical history. The UKB also provides molecular genetic data, including polygenic indices (PGI) and APOE variants, for participants.
Matthew V. Zahn — Economics of Health
This paper develops and estimates a structural model of health insurer entry and product competition in Medicare Advantage to analyze how firms strategically respond to policy changes and consumer behavior, and to evaluate the welfare implications of different subsidy designs.
Finance Application
- This structural model of firm entry, product design, and consumer sorting in a subsidized, regulated market is highly transferable to other financial sectors.
- For instance, it could analyze how financial advisors or mutual funds strategically enter markets and design products (e.g., fee structures, risk profiles) in response to regulatory incentives or client demographics, impacting market competition and investor welfare.
- The methodology for estimating fixed costs using moment inequalities could also be applied to study entry/exit decisions of financial institutions or investment funds, especially in markets with high regulatory hurdles or information asymmetries that lead to risk selection.
insurance marketsMedicare Advantagecompetitionfirm entryproduct differentiationrisk selectionmoral hazardsubsidiesstructural estimationmoment inequalitieshealth economicsindustrial organizationpolicy evaluation
Core finding, identification, data
Core Finding
- Firms strategically use market entry and product repositioning to engage in risk selection, avoiding markets with higher cost patients and more competitors.
- Accounting for these strategic responses is crucial, as models that abstract from them mispredict the direction and magnitude of welfare outcomes.
- A targeted subsidy policy that incentivizes high-risk seniors into Medicare Advantage can reduce government expenditures by 1% ($10 billion nationally) and reverse historical positive selection, while maintaining similar market entry and enrollment.
Identification Strategy
- The identification strategy for demand and healthcare utilization relies on plausibly exogenous variation in county-level CMS benchmarks, which influence the number of plans and their financial generosity.
- Endogeneity of premiums and supplemental benefits is addressed using instrumental variables, including the plan's marginal revenue around benchmarks and demographics of non-overlapping rival counties.
- Fixed costs of entry are identified using moment inequalities derived from revealed preference assumptions, combined with an exclusion restriction that assumes unobserved fixed costs are constant across adjacent markets within a service area.
Data
The paper uses novel administrative data from the Medicare program (2016-2018), including Medicare Beneficiary Summary Files, Traditional Medicare claims, and Medicare Advantage encounter data. These are supplemented with MA plan characteristics, CMS bid templates, DRG InterStudy data on other insurance products, and various public sources for provider supply and market demographics.
Catherine E. Ishitani — Economics of Health
This paper investigates how intermediary buyers in the generic pharmaceutical market use quality disclosures to discipline manufacturers and improve drug quality, finding that intermediaries significantly increase equilibrium quality.
Finance Application
- The paper's insights into how informed intermediaries discipline quality in markets with consumer information asymmetry could be highly relevant for finance.
- In asset pricing, this framework could analyze how institutional investors use disclosures (e.g., ESG ratings, regulatory fines, past performance issues) to select asset managers or specific securities, weighing fees against 'quality' factors.
- For household finance, it could shed light on whether financial advisors effectively steer clients towards higher-quality financial products or away from predatory ones based on disclosures, and quantify the premium they might pay for such quality.
- In insurance, the scoring auction model could be adapted to study how brokers choose insurance policies, balancing premium costs against insurer quality metrics like solvency, claims efficiency, or customer satisfaction scores, especially in response to regulatory disclosures or past penalties.
Information asymmetryIntermediary behaviorQuality signalingAuction theoryInstitutional investorsFinancial advisorsInsurance marketsRegulatory impactConsumer welfareEmpirical industrial organization
Core finding, identification, data
Core Finding
- Generic drugs are not perfect substitutes, with pervasive quality failures documented by FDA disclosures.
- Intermediary buyers, such as wholesalers and pharmacies, significantly respond to recall disclosures by reducing purchases of low-quality drugs by 60% for a decade.
- These intermediaries are willing to pay a 2% premium for every 10% reduction in recall risk, driving a 34% increase in equilibrium quality and improving overall welfare by forcing manufacturers to compete on quality.
Identification Strategy
- The paper employs a generalized difference-in-differences design, exploiting variation in FDA disclosure timing to estimate the causal effects of quality disclosures on drug purchases.
- Additionally, it develops a structural model of scoring auctions for generic drug procurement, which non-parametrically identifies buyer preferences and manufacturer costs using only winning price data, extending existing methods to handle incomplete price information and multiple sources of asymmetry.
Data
The paper constructs a comprehensive dataset by combining FDA manufacturing quality disclosures (2000-2022) obtained via FOIA requests, drug labels, import manifests, and administrative records to identify manufacturing locations. It also uses drug price data (average acquisition costs for retail pharmacies, 2010-2022) and administrative claims data from UnitedHealth (2000-2022) to track drug purchases and patient adherence.
Andrew Goodman-Bacon, Adriana Lleras-Muney, Joseph Price, Dahai Yue — Economics of Health
This paper estimates the causal effects of Medicare's introduction on mortality rates and life expectancy for its early recipients using linked historical census and death records.
Finance Application
- The causal estimates of Medicare's impact on life expectancy have direct and significant implications for the insurance and pension industries.
- Life insurers and annuity providers could use these findings to refine mortality tables, adjust pricing for long-term products, and better model the impact of future social policies on longevity risk.
- Pension funds, facing longer payout periods due to increased life expectancy, could incorporate such causal policy effects into their asset-liability management and long-term solvency projections.
- Furthermore, the methodology of linking historical census data to death records to identify causal policy impacts on longevity could be adapted to study other long-term financial risks or opportunities related to demographic shifts.
Medicarelife expectancymortality ratescausal inferencedifference-in-differencesinterrupted time seriesstructural modelhistorical datasocial policyhealth economicsinsurancepension fundshousehold financelongevity risk
Core finding, identification, data
Core Finding
- Medicare causally increased life expectancy at age 65 for men born between 1885 and 1915 by approximately one year on average.
- This effect is robust across three distinct identification strategies and is larger for cohorts with more exposure, with similar gains across socio-economic status groups for men.
- The effects for women, however, are inconclusive.
Identification Strategy
- The paper employs three complementary causal inference strategies: a structural model of cohort mortality, an interrupted time-series (ITS) design assuming log-linear mortality rates, and a staggered difference-in-differences (DiD) design relying on parallel trends across cohorts.
- All methods assume no anticipation of Medicare's effects before eligibility, and the DiD design carefully selects comparison cohorts to ensure valid parallel trends.
Data
The study constructs a novel dataset of over 18 million white individuals by linking the 1940 full-count US Census records to death records from the FamilyTree database at FamilySearch. It also uses Social Security Administration (SSA) cohort life tables for comparison and structural model estimation.
Zack Cooper, Stuart V. Craig, Aristotle Epanomeritakis, Matthew Grennan, Joseph R. Martinez, Fiona Scott Morton, Ashley T. Swanson — Economics of Health
This paper empirically analyzes the effects of non-horizontal mergers between hospitals and physician practices on competition and pricing, finding significant price increases without quality improvements.
Finance Application
- This research offers direct insights for insurance companies, as it quantifies how provider mergers impact the prices they pay, which can inform premium setting, risk modeling, and negotiation strategies.
- For asset pricing, the findings could be used to analyze how these anticompetitive effects and rising healthcare costs affect the valuations and stock returns of publicly traded hospital systems, health insurers, and physician practice management companies.
- In household finance, the documented price increases directly impact household budgets, providing a basis to study the effects on medical debt, savings behavior, and demand for different health insurance products.
Healthcare EconomicsMergers and AcquisitionsVertical IntegrationMarket PowerPricingHealth InsuranceAntitrustMachine LearningDifference-in-DifferencesHospital IndustryPhysician PracticesHousehold FinanceAsset Pricing
Core finding, identification, data
Core Finding
- Hospital acquisitions of physician practices lead to average price increases of 3.3% for hospitals and 15.1% for physicians two years post-integration, with no discernible effects on quality.
- These price increases are larger in transactions with greater scope for foreclosure, recapture, and increased horizontal concentration in physician markets.
Identification Strategy
- The authors employ a difference-in-differences event study strategy, comparing treated providers (those involved in mergers) to carefully constructed control groups.
- Control providers are identified using a discrete choice model of patient demand to ensure they are not close substitutes or exposed to spillover effects, and propensity score matching is used for observable characteristics.
Data
The study combines novel integration event data (2008-2016) identified using machine learning algorithms applied to Medicare Data on Provider Practice and Specialty (MD-PPAS), AHA surveys, and SEC filings, with claims data from a large national insurer (UnitedHealthcare, 2011-2016) focusing on childbirth admissions.
Renuka M. Diwan, Paul J. Eliason, Riley League, Jetson Leder-Luis, Ryan C. McDevitt, James W. Roberts — Economics of Health
This paper investigates the ambiguous effect of competition on fraud in Medicare's durable medical equipment (DME) market, finding that competitive bidding disproportionately benefits fraudulent firms by driving out legitimate ones.
Finance Application
- This paper offers rich insights for asset pricing and insurance research.
- In asset pricing, the findings suggest that regulatory shocks like competitive bidding can significantly alter firm-specific risk and market structure, potentially creating a 'fraud risk premium' for firms in affected sectors.
- Researchers could examine how stock prices or bond yields of publicly traded healthcare suppliers react to such regulatory changes, especially considering the scale advantage of larger firms.
- For insurance, the paper directly addresses the trade-off between cost-cutting (via competitive bidding) and fraud in a large public insurance program (Medicare).
- This framework could be applied to private health insurers or other P&C lines (e.g., auto repair networks, home services) to analyze how different competitive bidding designs impact fraud rates, claims costs, and ultimately, insurance premiums and profitability.
healthcare economicsfraudcompetitive biddingregulationmarket structurefirm behaviornatural experimentMedicareDMEinsurance markets
Core finding, identification, data
Core Finding
- The study finds that after the staggered rollout of competitive bidding in Medicare's DME procurement, fraudulent firms increased their market share by 10 percentage points, primarily because legitimate firms exited the market due to reduced margins.
- Fraudulent firms, often larger, were better positioned to absorb administrative costs and compete on price, without significantly altering their bidding behavior or product quality.
Identification Strategy
- The paper exploits the staggered rollout of competitive bidding across geographic regions and DME product categories in Medicare.
- This natural experiment allows for a difference-in-differences (DiD) approach to identify the causal effects of increased competition on fraud, market structure, and firm behavior.
Data
The research uses Medicare claims data for durable medical equipment (DME) from 2008 to 2019, covering the universe of patients. It also hand-collects data on firms subject to anti-fraud enforcement (civil litigation, criminal lawsuits, administrative exclusion) and identifies 'suspicious' firms through connections to sanctioned entities (common ownership, shared address, referrer links). Competitive bidding data from FOIA requests are also used.
Zihan Hu, Hyejin Ku, Xuan Wang, Xinjue Yao — Personnel Economics
This paper investigates how transitioning from temporary to permanent employment contracts affects worker productivity due to changes in incentives.
Finance Application
- This insight could inform asset pricing models by suggesting that firms with a higher proportion of temporary workers might exhibit higher short-term productivity but potentially face long-term human capital risks or moral hazard issues once employees gain permanence, impacting firm valuation or risk premia.
- In household finance, the uncertainty associated with temporary contracts and the subsequent change in effort upon securing permanence could influence household savings, debt accumulation, and investment decisions, as job security affects perceived future income and financial planning.
- For insurance, the findings could be relevant for designing unemployment or disability insurance, considering that workers under temporary contracts might face different risk profiles due to higher effort levels, and the moral hazard of reduced effort post-permanence could be incorporated into employee benefit insurance contracts.
labor economicsincentivesmoral hazardcontract theoryproductivityemployment contractsfirm valuationhousehold financehuman capitallabor risk
Core finding, identification, data
Core Finding
- The study finds a significant 5% drop in worker productivity immediately after transitioning from temporary to permanent status, with the decline intensifying over time.
- This suggests that workers exert additional effort under temporary contracts to secure permanent employment, and the decline is more pronounced for workers with weaker outside options, highlighting the role of tenure uncertainty.
Identification Strategy
- The study leverages high-frequency production data from a Chinese garment factory where piece-rate workers experience no changes in tasks or wages upon attaining tenure, allowing them to isolate the pure incentive effect of tenure.
- A comparable plant in Vietnam, where permanent status is guaranteed after two temporary contracts, serves as a control for the effect of tenure uncertainty.
Data
The paper uses high-frequency production data from a Chinese garment factory and a comparable plant in Vietnam.
Ingrid Ellen, Daniel Hartley, Jeffrey Lin, Wei You — Real Estate
This paper examines the unintended consequences of 1960s Fair Access to Insurance Requirements (FAIR) plans, finding they inadvertently fueled urban disinvestment and neighborhood decline.
Finance Application
- This research offers crucial insights for modern insurance and real estate finance, particularly concerning climate risk.
- The findings on moral hazard from over-insurance and lack of risk-based pricing in FAIR plans can be directly applied to evaluate current government-backed insurance programs (e.g., NFIP, state-backed wind pools) in climate-vulnerable areas, predicting their impact on property values, maintenance, and potential for disinvestment.
- For household finance, the study highlights how insurance policy can profoundly affect household wealth (home equity) and credit access in urban areas, suggesting avenues to research mortgage defaults or property tax revenues in regions facing changing insurance landscapes due to climate change.
- The clever use of digitized city directories to measure financial service withdrawal could also be adapted to study the impact of bank branch closures or fintech adoption on local economic outcomes and asset prices.
InsuranceMoral HazardUrban EconomicsReal EstateDisinvestmentRedliningPolicy EvaluationClimate RiskProperty InsuranceHistorical DataHousehold Finance
Core finding, identification, data
Core Finding
- FAIR plans, designed to address insurance redlining, led to significant housing disinvestment, with affected neighborhoods losing an average of 313 pre-war housing units per census tract (29.8% of the 1950 stock).
- This disinvestment was concentrated in multi-family and rental buildings, contributing to declines in population and income, and increases in the Black population share, primarily due to moral hazard stemming from over-insurance and diluted underwriting incentives.
Identification Strategy
- The study employs a triple-difference design, comparing outcomes before and after FAIR plan authorization (1968), between neighborhoods with and without likely FAIR access (measured by private insurer withdrawal from 1940-1967 using digitized city directories), and across states that did versus did not adopt FAIR plans early.
- This approach estimates intent-to-treat effects, controlling for unobserved confounding differences and ensuring parallel pre-trends.
Data
The paper uses a balanced panel of consistent-boundary census tracts from 1950-1990 across 26 major US cities. It digitizes city directories from 1940 and 1967 to measure private property and casualty insurer market access. Additionally, it uses fire data from National Fire Protection Association reports (1938-1969), a 1978 US Department of Justice survey, and the National Fire Incident Reporting System (1980-1988).
Lei Ma — Real Estate
This paper develops and estimates an equilibrium model of segmented housing markets to quantify the causes and distributional consequences of the shift towards larger homes and evaluates housing policies aimed at improving affordability.
Finance Application
- The paper's structural model and quantification of zoning costs and demand heterogeneity could be applied to asset pricing by developing more granular real estate investment trust (REIT) valuation models that account for local regulatory risks and demographic shifts.
- In household finance, the model could inform studies on how zoning-induced housing supply constraints and affordability impact household savings, mortgage debt decisions, and wealth accumulation across different income and demographic groups.
- For insurance, understanding how local regulations influence property values and construction types could refine property insurance pricing models and risk assessments in specific geographic segments.
housing marketshousing supplyhousing demandzoning regulationsaffordabilityhousing subsidiesequilibrium modelheterogeneous preferencesdevelopment costsasset pricinghousehold financereal estatepolicy evaluationurban economics
Core finding, identification, data
Core Finding
- The paper finds that the trend towards larger, more expensive homes is driven by both high-income demand and, more significantly, by zoning density restrictions that limit smaller home construction.
- While demand-side subsidies benefit recipients, they can raise prices for non-recipients, whereas supply-side subsidies for small homes increase their construction but crowd out larger homes, leading to modest, untargeted welfare gains.
Identification Strategy
- The paper identifies household preference parameters using a mixed logit demand model with an instrument based on the availability of similar housing nearby to address price endogeneity.
- Supply-side regulatory costs are identified using a residual approach, inferring them from the gap between housing prices and observed production costs, exploiting variation in development choices across parcels and house types.
Data
The paper uses microdata from CoreLogic for property characteristics and values, L2 for individual-level demographics and addresses, and various sources for neighborhood characteristics. Housing development costs are derived from web-scraped building permits, land values from Davis, Larson, Oliner, and Shui (2021), and zoning codes are manually collected from municipal documentation.
James D. Macek — Real Estate
This paper quantifies the welfare and urban structural impacts of minimum lot size regulation across the US, focusing on housing affordability, income sorting, and land values.
Finance Application
- The paper's quantification of land value changes (e.g., 17% loss from deregulation) offers direct insights for real estate asset pricing, particularly for REITs and private equity funds with geographically diversified portfolios, allowing for better pricing of regulatory risk.
- For household finance, the differential welfare impacts on renters versus homeowners, and across income groups, can inform models of mortgage demand, default risk, and household wealth accumulation under various regulatory scenarios.
- Furthermore, the findings on cities' disincentives to deregulate unilaterally could explain persistent local market inefficiencies and their implications for municipal bond yields and local government fiscal health.
Housing economicsLand use regulationUrban economicsIncome sortingGentrificationWelfare analysisReal estate valuationHousehold wealthMortgage marketsMunicipal financeRegulatory risk
Core finding, identification, data
Core Finding
- Complete deregulation of minimum lot sizes would lead to significant progressive welfare gains for renting households (9% of income) but substantial losses for landowners (17% of land values).
- This deregulation would also cause gentrification in high-density neighborhoods of productive cities, with minimal aggregate productivity gains (0.25%) and exacerbation of neighborhood choice externalities.
Identification Strategy
- The paper estimates the elasticity of neighborhood amenity values to income using a 'donut strategy' instrumental variable approach.
- Terrain slopes in a surrounding buffer zone (0.75-1.25km) are used as instruments for local average income, correcting for endogeneity from unobserved amenities and reverse causality.
Data
The study utilizes CoreLogic property assessment and transaction data, 2016-2020 American Community Survey (ACS) household income distributions and demographics, 2020 Census housing counts, National Neighborhood Data Archive (NaNDA) for amenities, and USGS EDNA for terrain slopes.
Joseph Gyourko, Sean E. McCulloch — Real Estate
This paper quantifies suburban homeowners' preferences for housing unit density and the associated welfare losses from increased density, using a boundary discontinuity design and a structural hedonic model.
Finance Application
- This research offers critical insights for household finance and real estate asset pricing.
- Lenders could integrate 'densification risk' into mortgage underwriting, as significant welfare losses from increased density can depress property values and elevate default risk, particularly in affluent, low-density areas.
- For real estate investors and REITs, understanding these localized preferences is vital for valuing development projects and existing portfolios, as strong NIMBYism can impose substantial political costs and limit supply growth, impacting returns.
- Furthermore, the quantified welfare losses provide a basis for assessing how regulatory changes or unexpected densification events could affect household housing wealth and broader balance sheet stability.
Housing EconomicsUrban EconomicsHedonic PricingBoundary Discontinuity DesignZoning RegulationsProperty ValuesWelfare AnalysisHousehold PreferencesReal EstateMortgage RiskHousehold FinanceReal Estate Asset PricingNIMBYismLocal Government
Core finding, identification, data
Core Finding
- Suburban homeowners generally exhibit a distaste for increased housing density, experiencing an average welfare loss of about $9,500 for a 1/2 unit per acre increase.
- This aversion is significantly higher for increases in rental unit density (5-6 times larger) and is particularly pronounced among affluent households in low-density neighborhoods, suggesting a long, left tail of those with extreme aversions.
Identification Strategy
- The study employs a boundary discontinuity design, exploiting sharp changes in minimum lot size restrictions at municipal borders to identify causal effects on housing density and prices.
- This systematic variation in density exposure, induced by regulatory discontinuities, is then integrated into a structural hedonic model to estimate individual homeowners' preferences for density.
Data
The research utilizes parcel-level CoreLogic tax assessment files for single-family homes (including sales prices, lot size, structure size, age) and 2010 U.S. Census block data for housing unit counts (total, owner-occupied, renter-occupied) to construct density exposure measures. It also incorporates land use regulation data from WRLURI surveys and various geographic/amenity data.
Anders Frederiksen, Louis Junker-Jensen — Personnel Economics
This paper quantifies the 'part-time penalty' in Denmark, showing that part-time work leads to lower earnings and slower human capital accumulation, with the penalty widening over a career.
Finance Application
- This paper's findings have significant implications for household finance, particularly regarding retirement planning and investment decisions.
- The persistent and widening part-time penalty, predominantly affecting women, implies lower lifetime earnings and thus potentially lower retirement savings accumulation.
- Future research could investigate how households with part-time earners adjust their savings rates, asset allocation (e.g., equity exposure), and demand for insurance products (life, disability) to mitigate this human capital deficit and associated income risk.
- It could also explore whether financial literacy interventions or policy changes (e.g., subsidized skill development for part-time workers) could alleviate the long-term financial consequences.
Labor EconomicsHuman CapitalGender Wage GapPart-time WorkLifetime EarningsHousehold FinanceRetirement SavingsInvestment BehaviorInsurance
Core finding, identification, data
Core Finding
- The part-time penalty for nurses in Denmark is 9-14% at the beginning of their career, increasing by 0.5% annually.
- This widening gap is driven by differential human capital accumulation, as full-time workers gain competencies faster, leading to higher productivity and wages.
- Part-time workers cannot fully overcome this penalty by switching to full-time work later, as their accumulated human capital remains lower.
Identification Strategy
- The study leverages highly homogeneous Danish nurses, a regulated profession with known graduation dates, to establish a clean career trajectory and control for worker characteristics.
- It uses the high persistence of work hours to treat part-time and full-time groups as distinct and employs an instrumental variable approach (using children and partner status) to address measurement error in hours.
- Direct evidence on human capital accumulation is provided through hospital personnel records detailing 'competency supplements' linked to hours worked and seniority.
Data
The paper uses comprehensive Danish register data (1994-2022) for earnings and demographics, the Danish Labor Force Survey for work hours (actual and usual), and detailed hospital personnel records (2017-2021) for nurse-specific wage components and competency supplements.
Alessandra Fenizia, Tom Kirchmaier — Personnel Economics
This paper examines the individual-level productivity effects of working from home (WFH) in public sector jobs, finding a significant increase in productivity primarily due to reduced distractions.
Finance Application
- This research offers valuable insights for financial services, where many roles involve knowledge work and require high concentration.
- The finding that reduced distractions significantly boost productivity could inform WFH policies for equity analysts, portfolio managers, or quantitative researchers.
- Furthermore, the paper's emphasis on managerial task allocation could guide financial firms in optimizing hybrid work models by strategically assigning tasks (e.g., deep research vs. collaborative meetings) to maximize team performance.
- Reduced absenteeism from WFH could also translate to more consistent staffing in critical financial operations like trading or customer support.
working from homeremote workproductivitypublic sectorincentivesmanagerial economicslabor economicspersonnel economicsdistractionsabsenteeismtask allocationfinancial serviceshuman capital
Core finding, identification, data
Core Finding
- Working from home increases worker productivity by 12% on average in public sector jobs, despite weak performance incentives.
- These gains are mainly driven by reduced distractions and are not attributable to changes in work quality, hours, absenteeism, or task characteristics.
- When tasks are assigned by a supervisor who can match them to work location, productivity gains nearly double.
Identification Strategy
- The study exploits quasi-exogenous variation in work location from a deterministic rotation schedule, instrumenting actual WFH with assigned WFH.
- Task allocation is also plausibly random after a specific date, allowing for a clean comparison of productivity between WFH and in-office work for the same individuals.
Data
The paper uses novel administrative data from the Crime Recording and Resolution Unit (CRRU) of the Greater Manchester Police, including daily records of cases processed, time spent, and work quality. It also incorporates personnel files with demographic and work schedule information, and a brief survey on workers' perceived benefits and drawbacks of WFH.
Siddharth E. George, Martin Mattsson — Personnel Economics
This paper examines the impact of a large salary increase for frontline bureaucrats in India on their performance, honesty, corruption, and retention.
Finance Application
- This research offers several insights for finance.
- In household finance, the findings challenge the assumption that higher pay automatically leads to better service or ethical conduct from financial advisors or insurance agents, suggesting that compensation design needs to consider factors beyond just absolute pay to prevent mis-selling or improve client outcomes.
- For corporate finance and ESG, the paper's revelation that expert predictions on pay-performance links were inaccurate could inspire studies on the efficiency of 'governance prediction markets' and how investor expectations about executive compensation (e.g., linking pay to ESG metrics) align with actual firm performance and ethical behavior.
- The retention effect without performance improvement is also critical for human capital management in financial institutions, questioning whether high salaries merely retain average talent or truly incentivize top performers.
Public Sector PayBureaucracyCompensationIncentivesRetentionPerformanceCorruptionBehavioral EconomicsPersonnel EconomicsExpert PredictionsHousehold FinanceCorporate GovernanceBehavioral FinanceHuman Capital
Core finding, identification, data
Core Finding
- A 91% salary increase for frontline bureaucrats in Telangana, India, had no average impact on their performance, honesty, or corruption.
- However, it significantly reduced quit rates by 2.4 percentage points (16%) after two years, without improving the average quality of bureaucrats.
- Expert forecasters incorrectly predicted that higher salaries would improve both performance and selection.
Identification Strategy
- The study uses a difference-in-differences (DiD) design, exploiting a sudden and retroactive policy change in Telangana, India.
- One group of frontline bureaucrats (Junior PSes) received a 91% pay increase, while a comparable group (Regular PSes) performing the same job did not, allowing for a causal comparison of outcomes.
Data
The paper uses administrative data from Telangana's Department of Panchayat Raj and Rural Development (PRRD), including supervisor evaluations, third-party auditor assessments, citizen surveys, daily activity reports (for effort), personnel rosters (for retention), and financial expenditure data (for corruption proxies). It also incorporates expert forecasts from the Social Science Prediction Platform and baseline covariates from SHRUG data.
Colleen Chien, Jillian Grennan, Jason Sandvik — Personnel Economics
This paper examines the impact of a targeted mentorship program within a large technology firm on innovation productivity, employee retention, and corporate culture.
Finance Application
- This research offers significant insights for asset pricing and firm valuation.
- Firms that strategically invest in human capital through mentorship programs, leading to sustained innovation, higher employee retention, and improved corporate culture, might be undervalued by the market.
- This could form the basis of an alpha factor: going long firms with strong mentorship indicators (e.g., high Glassdoor scores for 'community' and 'collaboration', low Revelio-based turnover) and short firms lacking these.
- Furthermore, mentorship programs represent valuable intangible assets, aligning with ESG investing, where investors could use these metrics to identify firms with superior long-term value creation potential.
Human CapitalInnovationCorporate CultureEmployee RetentionESGAsset PricingFirm ValuationIntangiblesLabor EconomicsDifference-in-Differences
Core finding, identification, data
Core Finding
- The mentorship program significantly increases mentees' innovative output in both the short and long term, with larger gains for underrepresented groups.
- It also generates positive spillovers to non-mentored collaborators, improves corporate culture perceptions (community, collaboration, integrity), and reduces engineer turnover, leading to a substantial return on investment for the firm.
Identification Strategy
- The study uses a staggered rollout of a mentorship program within a superstar tech firm.
- It employs two-way fixed effects (TWFE) and stacked difference-in-differences (DID) estimators, exploiting variation in treatment timing and using never-mentored and not-yet-mentored innovators as control groups.
- Parallel trends tests support the empirical strategy.
Data
The paper uses proprietary firm data including the complete innovation pipeline (idea disclosures to patent applications) and detailed demographic characteristics for 29,211 innovators and 66,397 ideas. It also incorporates Glassdoor employee reviews (over 13 million) and Revelio labor market intelligence platform data on employee movements across firms.
Achyuta Adhvaryu, Parker Howell, Andrea Neyra Nazarrett, Anant Nyshadham, Jorge A. Tamayo — Personnel Economics
This paper examines how the allocation of managerial attention, both personal and through subordinates, impacts productivity in a large retail firm.
Finance Application
- This research offers a novel lens for asset pricing by suggesting that 'managerial attention quality' could be a factor explaining cross-sectional stock returns or firm valuation, as it directly impacts operational efficiency and sales stability.
- In household finance, understanding how managerial attention influences product availability and pricing could inform models of consumer spending behavior and household budgeting under varying market conditions.
- For insurance, the findings on reduced stockouts and leaner inventory could be used to refine risk models for business interruption insurance or supply chain insurance, potentially leading to more accurate premium pricing for retail businesses.
Managerial EconomicsFirm ProductivityRetail IndustryInventory ManagementPricing StrategyOrganizational BehaviorQuasi-ExperimentFixed EffectsSupply Chain ManagementOperational Risk
Core finding, identification, data
Core Finding
- High-performing managers significantly boost sales and productivity by strategically focusing their attention on reducing stockouts, optimizing inventory levels (both value and inventory-to-sales ratio), and making more effective pricing decisions.
- This operational efficiency is supported by an organizational restructuring that reallocates personnel towards back-of-store functions like pricing and inventory management.
Identification Strategy
- The study leverages a quasi-random policy of rotating middle managers across stores within a large Colombian retail firm.
- It uses an Abowd-Kramarz-Margolis (AKM) model to estimate manager and store fixed effects on log productivity, classifying managers as 'high type' if their fixed effects are above the median.
- Robustness checks for endogenous mobility and pre-trends, including event-study designs, support the causal interpretation of manager effects.
Data
The paper utilizes extensive administrative data from a large Colombian retailer, covering over 200 stores and 20,000 workers from 2017-2020. This includes scanner-level sales, detailed inventory records, supplier characteristics, transaction data, and employee-level information (wages, promotions, attendance). A complementary managerial survey provides insights into management styles and traits.
Yuen Ho, Yihong Huang — Personnel Economics
This paper uses field experiments in a Tanzanian garment factory to study the benefits and costs of supervisor discretion in worker promotion decisions, finding that discretion crowds in private information but also introduces biases and reduces application rates.
Finance Application
- The findings on the trade-offs of discretion in managerial selection have direct implications for understanding firm value and risk in asset pricing.
- Firms that effectively balance the informational benefits of discretion with its costs (bias, reduced applicant pool) in human capital management could exhibit superior productivity and lower turnover, leading to better stock performance or lower cost of capital.
- In household finance, the worker's aversion to discretion could inform the design of financial products or advisor compensation structures; for instance, perceived subjectivity in advisor recommendations might reduce client trust and engagement, even if it leverages private information.
- In insurance, the framework could be applied to claims adjustment or underwriting, where discretion might improve accuracy but also introduce biases or reduce perceived fairness, impacting customer retention or regulatory scrutiny.
human capitalmanagementlabor economicsfield experimentdiscretionincentivesselectionsortingbiasfavoritismfirm performanceorganizational economicspersonnel economics
Core finding, identification, data
Core Finding
- Supervisors possess valuable private information about workers' managerial quality, which they leverage when incentivized, leading to the selection of workers with significantly higher measured managerial ability compared to objective methods.
- However, discretion also introduces biases (gender, favoritism) and is disliked by workers, reducing the pool of high-quality applicants.
Identification Strategy
- The paper employs two main field experiments: a 'Supervisor Referral Experiment' that randomizes financial incentives for supervisors based on referral quality, and a 'Worker Application Experiment' that randomly varies whether supervisor referrals are emphasized in the promotion selection process.
- These designs allow for causal inference on the effects of incentives on referral quality and perceived discretion on worker application behavior.
Data
The study uses primary data from field experiments in a large Tanzanian garment factory, combined with extensive administrative data on worker demographics, performance (output, attendance, KPIs, bonuses), and a specially designed leadership test measuring managerial potential (including firm-specific knowledge and soft skills).
Anat Admati, Nathan Atkinson, Paul Pfleiderer — Law and Economics
This paper theoretically analyzes how corporate governance and managerial compensation interact with weak law enforcement to enable profitable corporate misconduct, showing that common enforcement policies can perversely exacerbate social harm.
Finance Application
- This paper's findings have significant implications for ESG investing, suggesting that current ESG metrics and compliance programs might be ineffective or even counterproductive if firms can profit from misconduct and receive fine discounts for 'cooperation.' For insurance, the explicit discussion of D&O insurance and indemnification highlights how these mechanisms can reduce managerial liability, potentially increasing moral hazard and the likelihood of corporate misconduct, impacting D&O policy pricing and risk assessment.
- In corporate valuation, the paper implies that firms engaged in profitable misconduct might be mispriced if markets underestimate the true social costs or the perverse incentives created by weak enforcement, offering potential arbitrage opportunities for sophisticated investors.
Corporate GovernanceLaw EnforcementMisconductManagerial CompensationESGD&O InsuranceRegulatory ArbitrageShareholder ValueOptimal DeterrenceCorporate Crime
Core finding, identification, data
Core Finding
The paper demonstrates that when law enforcement is weak, corporations can adjust managerial compensation (e.g., stock-based pay, indemnification) to incentivize managers to engage in profitable but harmful misconduct, effectively treating fines as a 'cost of doing business.' It further shows that policies like voluntary compliance programs or self-reporting, which offer discounted fines, can perversely increase social harm by making misconduct more profitable and thus more prevalent, even if detected earlier.
Identification Strategy
- The paper develops a principal-agent theoretical model where a corporation sets managerial compensation to maximize profits, and the manager chooses activity levels (beneficial vs. harmful).
- The model incorporates various law enforcement policies (fixed fines, variable fines, managerial fines, compliance programs, self-reporting) and analyzes how the corporation's optimal compensation structure and activity choices respond to these policies, especially when enforcement is weak and misconduct is profitable.
- Numerical examples illustrate the theoretical results.
Data
The paper is theoretical and uses numerical examples for illustration, but does not use empirical data.
Bocar A. Ba, Patton Chen, Tony Cheng, Martha C. Eies, Justin E. Holz — Law and Economics
This paper evaluates Durham, North Carolina's HEART program, a civilian crisis response initiative, to assess its impact on crime, public trust, and cost-effectiveness compared to traditional policing.
Finance Application
- This research offers insights for insurance pricing, particularly for property and casualty insurers, by demonstrating how effective public safety programs can reduce risk exposure in specific geographic areas, potentially leading to refined premium calculations.
- In household finance, the findings could inform studies on how perceived public safety and trust in local services influence housing demand, property values, and household investment in local businesses.
- For municipal bond markets, the fiscal sustainability and positive social impact of such programs could be a factor in credit ratings and bond yields, acting as an ESG-like metric for local government efficiency and stability.
public safetycrime reductionsocial programsmunicipal financeinsurance riskproperty valuespublic trustESGlocal economicscost-effectiveness
Core finding, identification, data
Core Finding
- The HEART program significantly reduces crime reports, arrests, and emergency response times, primarily through civilian-led interventions.
- It also fosters public trust, evidenced by an increase in future 911 calls, and is found to be a fiscally self-sustainable intervention based on a contingent valuation survey and marginal value of public funds framework.
Identification Strategy
- The study employs a difference-in-differences design to evaluate the causal impact of the HEART program.
- It compares outcomes in areas where the program was implemented against control areas, isolating the program's effect on crime, arrests, response times, and public trust.
Data
The paper utilizes data from Durham, North Carolina, including 911 call records, crime reports, arrest data, and emergency response times. It also incorporates an original contingent valuation survey to assess the public's willingness to pay for the program's benefits.
Eric Budish, Maya Durvasula, Benjamin N. Roin, Heidi L. Williams — Law and Economics
This paper demonstrates that intellectual property rights that are unenforceable in practice lead to significant underinvestment in socially valuable research and development, specifically in finding new uses for existing drugs.
Finance Application
- This paper's insights could be applied to asset pricing by examining how the enforceability and duration of intellectual property rights are priced into the valuation of pharmaceutical and biotech firms.
- Investors might systematically undervalue firms with a strong pipeline of 'new uses' if the market anticipates enforceability issues, creating an 'innovation premium' or discount.
- Furthermore, the concept of 'missing markets' due to IP enforceability could be generalized to other innovation-intensive sectors (e.g., software, green tech) to identify underinvested R&D areas and their impact on long-term firm growth and risk premia.
- For insurance, the social cost of missing health innovations could inform models of healthcare cost inflation and the pricing of health and longevity insurance products.
Intellectual PropertyInnovationR&DPharmaceuticalsMarket ExclusivityEnforceabilityFirm ValuationAsset PricingPolicyHealthcare Costs
Core finding, identification, data
Core Finding
- The authors find that when intellectual property rights become unenforceable (e.g., after generic drug entry), research investment and commercialization for new drug uses nearly cease.
- Their estimates suggest that 200-800 new uses for existing drugs have been missed due to these incentives, with a social cost on the order of several trillion dollars.
Identification Strategy
- The identification strategy leverages a quasi-random variation in the enforceability of intellectual property rights over time for new uses of existing drugs.
- This variation arises because patents on new uses are perfectly enforceable during a drug's market exclusivity period but become unenforceable after generic entry, due to the difficulty of observing and suing for 'off-label' use.
- The authors exploit delays between patent filing and drug approval, and differences in clinical trial lengths, to create variation in 'minimum patent exclusivity' unrelated to a drug's scientific potential.
Data
The paper uses FDA administrative records (drugs@FDA, Orange Book) for drug approvals, re-approvals, generic entry, and regulatory exclusivities. It also incorporates data from PubMed for scientific publications and multiple sources (NDA Pipeline, Pharmaprojects, Cortellis, ClinicalTrials.gov) for clinical trial activity, classifying diseases using ICD-10 codes.
Michael D. Frakes, Melissa F. Wasserman — Law and Economics
This paper constructs a novel database of patents for FDA-approved biologics to show how biopharmaceutical firms strategically use patent thicketing and evergreening to block biosimilar entry and extend market exclusivity.
Finance Application
- This research could inform asset pricing by creating a 'patent exclusivity factor' for pharmaceutical stocks, where firms with robust patent thickets and successful evergreening strategies (indicating longer periods of monopoly rents) might exhibit lower cash flow volatility and potentially higher valuations or different risk premia.
- In corporate finance, the findings suggest that traditional valuation models for biopharma firms should explicitly account for the quantity and strategic timing of secondary patents, as these significantly extend effective market exclusivity and thus future earnings potential, leading to more accurate intrinsic value assessments.
- This could also be relevant for M&A analysis, where the strength of a target's patent thicket could be a key determinant of acquisition premiums.
Intellectual PropertyPatentsPharmaceutical IndustryMarket StructureCompetitive StrategyFirm ValuationMonopoly RentsBarriers to EntryRegulatory RiskInnovation Economics
Core finding, identification, data
Core Finding
- Biopharmaceutical firms strategically engage in patent thicketing and evergreening by accumulating numerous secondary patents, often of questionable innovativeness and timed just before primary patent expiration, to significantly extend market exclusivity and delay biosimilar entry.
- This behavior intensified after the 2010 Biologics Price Competition and Innovation Act (BPCIA), demonstrating a direct response to competitive threats and resulting in biologics having substantially more patents and longer effective market lives than small-molecule drugs.
Identification Strategy
- The paper employs a difference-in-differences approach, exploiting the 2010 Biologics Price Competition and Innovation Act (BPCIA) as an exogenous shock to biosimilar competition.
- It compares patenting trends for biologics (treatment group) to small-molecule drugs (control group) before and after the BPCIA.
- Additionally, event-study analyses examine patenting activity around primary patent expiration dates to identify strategic timing.
Data
The authors construct a novel, comprehensive database of 11,595 drug-patent pairs for 515 FDA-approved therapeutic biologics, gathered from the Purple Book, litigation records, patent term extensions, and extensive hand-coded patent searches using Orbit Intelligence. They also use OECD data on EPO patent allowance rates to assess innovativeness.
Samuel Antill, Eleanor Jenke, Raymond Kluender — Law and Economics
This paper investigates how small filing fees act as a significant barrier to consumer bankruptcy for financially vulnerable households, even when substantial debt relief is available.
Finance Application
- This research has direct implications for household finance by demonstrating how seemingly small financial barriers can disproportionately exclude the most vulnerable from crucial financial safety nets, impacting their long-term financial health and access to credit.
- It suggests that liquidity constraints and behavioral frictions play a significant role in high-stakes financial decisions, which could be modeled in household portfolio choice or debt management research.
- For insurance, the findings highlight challenges in targeting efficiency for social insurance programs, informing the design of health, unemployment, or disability insurance to ensure take-up by those most in need.
- In asset pricing, the inability of vulnerable households to access debt relief could lead to higher aggregate default rates on consumer credit, influencing the pricing of consumer credit-backed securities and overall credit risk assessments.
Household FinanceConsumer BankruptcyDebt ReliefLiquidity ConstraintsBehavioral EconomicsRegression DiscontinuityPolicy EvaluationSocial InsuranceFinancial InclusionTargeting EfficiencyCredit Markets
Core finding, identification, data
Core Finding
- The $338 filing fee causally reduces bankruptcy filing rates by 7.8 percentage points (14%).
- Critically, "marginal filers" deterred by this fee are often more financially vulnerable with higher levels of debt, indicating that the fee worsens the targeting efficiency of debt relief by excluding those who stand to gain the most.
Identification Strategy
- The paper employs a fuzzy Regression Discontinuity (RD) design around the 150% Federal Poverty Level (FPL) income threshold, which determines eligibility for a fee waiver.
- An indicator for whether a user's reported income exceeds 150% of the FPL is used as an instrumental variable for fee waiver applications, allowing for causal inference on the impact of the filing fee.
Data
The study utilizes proprietary, de-identified data from 18,055 users of Upsolve, a nonprofit providing software for Chapter 7 bankruptcy filing forms, who completed their paperwork between September 2021 and May 2025. This is supplemented with data from the Federal Judicial Center (FJC) Integrated Database and the Survey of Consumer Finances (SCF).
David Abrams, Jonathan Choi — Law and Economics
This paper evaluates the ability of fine-tuned Large Language Models (LLMs) to accurately assess the legality of police stops and frisks based on narrative reports, comparing their performance to human experts and traditional machine learning models.
Finance Application
- In insurance, LLMs could analyze claim narratives (e.g., accident reports, property damage descriptions) to quickly assess policy compliance, identify potential fraud, or flag cases with high legal dispute risk, thereby streamlining claims processing and reducing legal costs.
- In asset pricing, LLMs could scan regulatory filings, legal opinions, or lawsuit documents for specific language patterns indicative of future legal liabilities or regulatory enforcement actions for firms, providing real-time risk signals that could impact stock or bond valuations.
- For household finance, LLMs could help consumers understand complex financial contracts like mortgages or insurance policies by highlighting legally problematic clauses or non-compliant terms, empowering better decision-making.
Large Language ModelsLegal AnalysisNatural Language ProcessingMachine LearningRisk AssessmentComplianceUnstructured DataInsurance ClaimsRegulatory RiskContract AnalysisFraud Detection
Core finding, identification, data
Core Finding
- Fine-tuned Llama 3 models can identify illegal police stops with 88% accuracy and are well-calibrated, achieving 95% accuracy on 74.6% of cases where the model is most confident.
- This performance significantly exceeds human RAs (77% accuracy) and other ML models, demonstrating LLMs' potential for efficient auditing and real-time guidance in legal analysis.
Identification Strategy
- The study's identification strategy involves training and evaluating LLMs against a 'ground truth' dataset of 41,332 police stops from 2014-2024, which were manually coded by expert attorneys for reasonable suspicion.
- The performance of LLMs is then benchmarked against this expert coding, as well as against other machine learning algorithms, law student RAs, and intercoder agreement rates between plaintiff and city attorneys.
Data
The paper uses a novel dataset of 41,332 attorney-coded police stops from 2014 to 2024 from a major U.S. city. This dataset includes police report narratives, outcomes (frisked, searched, arrested, contraband discovered), and expert attorney assessments of reasonable suspicion.
Felipe M. Gonçalves, Steven Mello, Emily K. Weisburst — Law and Economics
This paper develops a novel econometric strategy to correct for selection bias when studying racial disparities in police use of force, leveraging variation in officer enforcement intensity to identify the racial composition of the unobserved population at risk.
Finance Application
- The 'extremum-agent monotonicity' and 'identification at infinity' approach could be highly valuable in finance to correct for selection bias in observed data.
- For instance, in household finance, analyzing loan approval disparities could use 'high-intensity' loan officers (those with the highest approval rates) to identify the true racial composition of the 'eligible borrower' pool, thus correcting for selection bias in observed applications.
- Similarly, in market microstructure, the trading patterns of 'high-intensity' market makers could reveal the true distribution of potential order flow, allowing for selection-corrected estimates of liquidity or adverse selection costs.
Selection BiasRacial DisparitiesEconometricsIdentification at InfinityExtremum AgentsAdministrative DataLendingDiscriminationMarket MicrostructureHousehold FinanceInsurance
Core finding, identification, data
Core Finding
- The study finds that Black civilians comprise 56% of arrestees but only 49% of potential arrestees, indicating significant differential selection.
- Correcting for this bias doubles the measured racial disparity in police force rates, with about 70% of the corrected disparity unexplained by other demographic or incident characteristics, suggesting a substantial role for officer discrimination.
Identification Strategy
- The identification strategy, building on 'identification at infinity' ideas, uses the racial composition of individuals arrested by the 'most enforcing officers' as an estimate of the racial composition of the potentially-arrested sample.
- This relies on two key assumptions: exogeneity (officers in the same assignment encounter the same set of potential arrests) and extremum-agent monotonicity (a maximally-enforcing officer would arrest anyone arrested by any other officer).
Data
The paper uses detailed administrative data on arrests, force incidents, and officer work shifts from the Chicago (2012-2015) and Seattle (2019-2022) police departments.
Soonwoo Kwon, Jonathan Roth — Frontier Econometric Methods
This paper develops a robust econometric framework for testing whether a treatment's effect on an outcome is fully explained by a specific mechanism, and quantifies the importance of alternative mechanisms when the null is rejected.
Finance Application
- This methodology could be applied in household finance to understand how financial education programs (D) impact long-term wealth accumulation (Y).
- Is the effect solely mediated by increased savings rates (M), or do other behavioral changes (e.g., investment choices, debt management) play a role? The method can quantify the impact of these alternative channels for individuals whose savings rates didn't change.
- In corporate finance, it could test if mandatory ESG reporting (D) affects a firm's cost of capital (Y) solely through reduced carbon emissions (M), or if other mechanisms like improved governance or reputation are at play, even for firms with stable emissions.
EconometricsCausal InferenceMediation AnalysisInstrumental VariablesProgram EvaluationTreatment EffectsHousehold FinanceCorporate FinanceESG
Core finding, identification, data
Core Finding
- The paper introduces a robust econometric framework for testing the 'sharp null of full mediation,' which posits that a treatment's effect on an outcome (Y) is entirely channeled through a specific mechanism (M).
- Crucially, this framework avoids stringent assumptions on how M is assigned.
- When the null is rejected, the methodology provides sharp lower bounds on the proportion of individuals whose outcomes are affected by the treatment despite no change in M, thereby quantifying the importance of unobserved or alternative mechanisms.
Identification Strategy
- The key innovation is an equivalence result: under the sharp null and independence/monotonicity, the treatment (D) acts as a valid instrumental variable for the mediator (M) on the outcome (Y).
- Building on this, the paper extends existing IV validity tests to accommodate multi-valued/multi-dimensional mediators and relaxed monotonicity assumptions, yielding sharp testable implications.
- The framework is also adaptable to non-experimental settings like standard IV, conditional unconfoundedness, and difference-in-differences.
Data
The paper uses data from two empirical applications: a randomized controlled trial (RCT) by Bursztyn et al. (2020) in Saudi Arabia on information treatment, job-search service sign-up, and job applications; and an RCT by Baranov et al. (2020) on cognitive behavioral therapy, grandmother presence/relationship quality, and financial empowerment.
Iman Modarressi, Jann Spiess, Amar Venugopal — Frontier Econometric Methods
This paper introduces a machine-learning framework for causal inference on text-based outcomes, leveraging large language models (LLMs) to identify and validate systematic differences between groups of text documents.
Finance Application
- This framework offers significant arbitrage opportunities in finance by enabling causal inference on unstructured textual data.
- In asset pricing, it could identify specific linguistic themes in earnings call transcripts or news that causally drive stock price movements.
- For household finance, it could analyze text from financial advice platforms to determine the causal impact of different advice themes on savings behavior.
- In insurance, it could extract causal themes from claim descriptions that predict claim severity or fraud likelihood, improving risk assessment and pricing models.
Causal InferenceLarge Language ModelsText AnalysisMachine LearningEconometricsRandomized Controlled TrialsNatural Language ProcessingFinancial EconometricsQualitative DataHypothesis Generation
Core finding, identification, data
Core Finding
- The authors demonstrate a robust methodology that combines LLM-based hypothesis generation with sample splitting and human validation to uncover 'causal themes' that systematically differentiate text distributions.
- In an empirical application using academic abstracts, the approach achieved 93% completeness in capturing systematic differences, highlighting its ability to provide statistically valid and interpretable insights from complex textual data.
Identification Strategy
The core methodological innovation involves a three-step process: first, testing for differences between text distributions via a 'reverse prediction' task using LLMs and a permutation test; second, prompting LLMs to generate 'causal themes' and scoring scales from training data; and third, validating these themes on held-out data using human labeling and a bias-corrected estimator to ensure valid inference and quantify completeness.
Data
The empirical proof-of-concept utilizes 200 arXiv abstracts from the econometrics category, which were subjectively labeled by a research team member into two groups based on alignment with their research interests.
Aaron Pancost, Garrett Schaller — Frontier Econometric Methods
This paper introduces a meta-regression technique that uses multiple OLS and IV estimates to separately identify measurement error, omitted variable bias, and detect instrument invalidity or heterogeneous treatment effects.
Finance Application
- This meta-regression method could be applied to asset pricing to assess the extent of measurement error in accounting-based factors (e.g., investment, profitability, Tobin's q) and its impact on estimated factor premia or anomaly returns.
- In corporate finance, it offers a diagnostic tool to re-evaluate causal effects identified using IVs, by disentangling measurement error from omitted variable bias and scrutinizing instrument validity.
- For household finance, it could be used to test the robustness of causal estimates in studies relying on potentially noisy survey data for variables like wealth, income, or risk preferences.
Instrumental VariablesMeasurement ErrorOmitted Variable BiasMeta-RegressionEndogeneityHeterogeneous Treatment EffectsCausal InferenceCorporate FinanceAsset PricingHousehold Finance
Core finding, identification, data
Core Finding
- The meta-regression of OLS on IV estimates allows researchers to disentangle measurement error bias from omitted variable bias.
- Applying this to four published papers, the authors find that measurement error is often quantitatively more important than omitted variable bias, and in some cases, instruments appear invalid or heterogeneity renders IV estimates irrelevant, suggesting OLS estimates might be closer to the truth.
Identification Strategy
- The core methodological innovation is a meta-regression of OLS coefficients on IV coefficients, where the slope identifies proxy quality (inverse of measurement error) and the intercept identifies average omitted variable bias.
- This approach leverages multiple OLS-IV pairs for different dependent variables from the same study, allowing separate identification of biases that are otherwise indistinguishable from a single OLS-IV pair.
- It also extends to detect instrument invalidity (economic and measurement invalidity) and assess heterogeneous treatment effects.
Data
The paper applies its meta-regression to reported OLS and IV coefficients from four published economics papers: Schlenker and Walker (2015), Mian and Sufi (2014), Adelino, Ma and Robinson (2017), and Duranton, Morrow and Turner (2014).
Ashesh Rambachan, Rahul Singh, Davide Viviano — Frontier Econometric Methods
This paper develops a novel nonparametric method to rigorously estimate treatment effects using remotely sensed variables (RSVs) as outcomes, demonstrating that common practice is biased when RSVs are post-outcome variables.
Finance Application
- This method has significant applications in ESG investing and climate risk assessment, where RSVs (e.g., satellite images of deforestation, pollution, or flood damage) can serve as robust, objective measures of environmental impact or physical risk for real estate portfolios, infrastructure projects, or insurance underwriting.
- For household finance and microfinance, RSVs like nightlights or satellite imagery, which predict poverty, could be used to develop more accurate credit scoring models or assess repayment risk in data-scarce emerging markets, enabling better financial inclusion.
- The framework's robustness to complex ML models makes it ideal for integrating high-dimensional, unstructured data into financial decision-making.
Causal InferenceRemotely Sensed VariablesMachine LearningProgram EvaluationSatellite ImageryESGClimate RiskMicrofinanceCredit ScoringInsuranceData Fusion
Core finding, identification, data
Core Finding
- The common practice of using machine learning-predicted outcomes from remotely sensed variables (RSVs) in experiments is shown to be arbitrarily biased when the RSV is a post-outcome variable.
- The paper nonparametrically identifies the treatment effect by formalizing the intuition that the conditional distribution of the RSV given the outcome and treatment is stable across experimental and observational samples, requiring predictions of the outcome, treatment, and sample indicator from the RSV for efficient inference.
Identification Strategy
- The identification strategy combines an experimental sample (where outcomes are missing but RSVs are observed) with an observational sample (where outcomes and RSVs are observed, and treatment may be missing or deterministic).
- The key assumption is the 'stability' of the RSV's conditional distribution given covariates, treatment, and outcome across both samples.
- The method is robust to misspecification of the RSV-based predictions, allowing for valid inference even with complex deep learning algorithms.
Data
The paper uses semi-synthetic exercises calibrated to a field experiment in India evaluating an anti-poverty program (Smartcards). Remotely sensed variables include nighttime luminosity measures and high-dimensional, pre-trained embeddings of satellite images. It also re-analyzes a crop burning experiment in India using satellite-based spectral indices.
Filip Obradovic — Frontier Econometric Methods
This paper develops a novel two-step identification framework to estimate long-term average treatment effects by combining short-term experimental and long-term observational data, explicitly modeling temporal link functions and accommodating imperfect experimental compliance.
Finance Application
- This framework could be highly valuable in household finance and corporate finance.
- In household finance, researchers could combine short-term experimental data from financial literacy interventions (e.g., randomized workshops on budgeting or investing) with long-term observational data on household savings, debt accumulation, or investment performance.
- The 'temporal link functions' could model how immediate changes in financial knowledge translate into long-term financial well-being.
- Similarly, in corporate finance, one could combine short-term A/B tests on new investor relations strategies or ESG disclosures (experimental data) with long-term firm valuation, cost of capital, or stock return data (observational data), using temporal links to understand how immediate market reactions predict long-run firm performance.
- The ability to handle imperfect compliance is crucial, as many financial interventions or policy changes face take-up issues.
long-term effectstreatment effectscausal inferencepartial identificationexperimental dataobservational datatemporal linkshousehold financecorporate financefinancial literacyESGprogram evaluationimperfect compliance
Core finding, identification, data
Core Finding
The empirical illustration on Head Start participation finds lasting but smaller positive effects on educational attainment, employment, and reduced criminal involvement compared to sibling comparison studies, with estimated increases in high school graduation (2.4%) and decreases in grade repetition (1.2-5.3%), idleness (2.8-4.2%), and criminal involvement (1.3-3.9%).
Identification Strategy
- The paper introduces a novel two-step identification framework that computationally produces sharp bounds on long-term average treatment effects (LTEs).
- It relies on two treatment response assumptions: Latent Monotone Instrumental Variables (LIV), which posits that the mean of long-term potential outcomes is non-decreasing in short-term potential outcomes, and Treatment Invariance (TI), which states that the relationship between short-term and mean long-term potential outcomes is unaffected by treatment.
- This framework combines short-term experimental data (with imperfect compliance) and long-term observational data, using the experimental data to amplify the identifying power of these temporal link function restrictions.
Data
The paper uses data from the Head Start Impact Study (HSIS), a short-term experiment providing cognitive test scores, and the Child and Young Adult Supplement to the National Longitudinal Survey of Youth 1979 cohort (CNLSY), a long-term observational survey providing outcomes like educational attainment, employment, and criminal involvement.
Felipe M. Gonçalves, Steven Mello, Emily K. Weisburst — Economics of Crime
This paper develops a novel empirical strategy to correct for selection bias in studies of policing, specifically focusing on racial disparities in police use of force.
Finance Application
- The methodology for correcting selection bias due to discretionary agents could be highly valuable in finance.
- For instance, in mortgage lending, loan officers (discretionary agents) decide which applications to approve.
- Applying this method could help estimate true racial disparities in loan approval rates among all qualified applicants, not just those who are approved or even apply.
- Similarly, it could be used to analyze selection bias in venture capital funding decisions, insurance policy underwriting, or even financial advisor client selection, by identifying 'most active' agents and inferring the characteristics of the broader 'at-risk' population to uncover hidden discrimination or disparities in access to capital.
selection biasracial disparitiesdiscriminationpolice use of forceeconometricsidentification at infinitydiscretionary agentslendinginsuranceventure capital
Core finding, identification, data
Core Finding
- The study finds that while Black civilians comprise 56% of arrestees, they make up only about 49% of potential arrestees, indicating significant differential selection into the data.
- Correcting for this selection bias doubles the measured racial disparity in force rates, with Black civilians being 48% more likely to face force than non-Black civilians, and about 70% of this disparity is unexplained by other demographic and incident characteristics, suggesting a role for officer discrimination.
Identification Strategy
- The paper's identification strategy corrects for selection bias by estimating the racial composition of the 'potentially-selected sample' (individuals at risk of arrest).
- It leverages variation in enforcement intensity across police officers, identifying the racial composition of the target sample from the 'most enforcing officers' (those with the highest propensity to make arrests) under assumptions of exogeneity and 'extremum-agent monotonicity'.
Data
The paper uses detailed administrative data on arrests and force incidents from the Chicago (2012-2015) and Seattle (2019-2022) police departments, including officer work shifts, arrestee demographics, and incident information. It also references FBI's Uniform Crime Reports, ACS, and Fatal Encounters for broader context.
Kirill Borusyak, Mauricio M. Caceres Bravo, Peter Hull — Frontier Econometric Methods
This paper introduces a novel instrumental variable (IV) construction method, called 'recentered instruments,' for estimating demand systems, particularly robust to endogenous product characteristics.
Finance Application
- This methodology could be highly valuable in asset pricing and household finance for estimating demand for differentiated financial products.
- For instance, in asset pricing, one could estimate investor demand for mutual funds or ETFs, where fund characteristics (fees, ESG scores, past performance) are often endogenous.
- Cost shocks to fund providers (e.g., regulatory compliance costs, technology costs) could serve as recentered instruments.
- In household finance, estimating demand for insurance policies or mortgages, where contract terms are endogenous, could use shocks to insurer/lender capital costs or regulatory price caps as instruments.
Demand estimationInstrumental VariablesEndogeneityProduct characteristicsCost shocksAsset pricingHousehold financeInsuranceEconometrics
Core finding, identification, data
Core Finding
The proposed recentered instruments provide accurate and unbiased estimates of demand parameters (like price elasticities and substitution patterns) even when product characteristics are endogenous, outperforming conventional characteristic-based IVs which exhibit strong bias in such scenarios.
Identification Strategy
- The method constructs IVs from supply-side cost shocks and product characteristics, which are 'recentered' by removing their conditional expectation given only characteristics.
- This ensures the instruments are uncorrelated with unobserved demand shocks even if product characteristics are endogenous.
- The key assumption is that cost shocks are exogenous to demand shocks conditional on observed characteristics.
Data
The paper primarily uses Monte Carlo simulations based on the Gandhi and Houde (2020) model, with varying scenarios for exogenous and endogenous product characteristics, to evaluate the performance of different IV approaches.
David Abrams, Jonathan Choi — Economics of Crime
This paper evaluates the capability of Large Language Models (LLMs) to accurately assess the legality of police stops and frisks based on narrative reports, comparing their performance to human experts and traditional machine learning models.
Finance Application
- This methodology could be directly applied in insurance to assess legal risks associated with claims involving police interactions, such as property damage during stops or liability for alleged false arrests, informing claims adjustment and underwriting for municipal liability policies.
- In municipal bond markets, LLMs could analyze police department data and legal documents to quantify the risk of civil rights lawsuits and associated financial liabilities for cities, impacting bond ratings and pricing.
- More broadly, financial institutions could fine-tune LLMs on their vast repositories of legal and regulatory documents to automate compliance checks, contract review, and legal risk assessments in areas like M&A due diligence or anti-money laundering, significantly reducing legal costs and improving operational efficiency.
Large Language ModelsLegal AnalysisPolice StopsRisk AssessmentNatural Language ProcessingMachine LearningInsuranceMunicipal BondsComplianceLitigation RiskLegal Tech
Core finding, identification, data
Core Finding
- A fine-tuned Llama 3 model can identify illegal police stops with 88% accuracy and illegal frisks with 77% accuracy, significantly outperforming human RAs and other ML models.
- The LLM is well-calibrated and can classify a substantial portion of cases (75% of stops) with 95% accuracy, demonstrating its potential for efficient auditing and real-time legal guidance in law enforcement.
Identification Strategy
- The study's identification strategy involves training and evaluating various LLMs and traditional ML models against a 'ground truth' derived from 41,332 police stop narratives that were attorney-coded for reasonable suspicion.
- Performance is benchmarked against multiple human experts (police officers, law students, and inter-attorney agreement rates), and the use of 'narrative-only' inputs helps isolate the LLM's legal reasoning capabilities.
Data
The paper utilizes a novel dataset of 41,332 attorney-coded police stops from 2014 to 2024 from a major U.S. metropolitan police department, which includes police report narratives and expert legal assessments of reasonable suspicion.
Natalia Emanuel, Pim Welle, Valentin Bolotnyy — Economics of Crime
This paper uses quasi-random assignment of physicians to evaluate the causal impact of involuntary hospitalization on individuals experiencing mental health crises, finding it significantly increases violent crime charges, suicide/overdose deaths, and causes employment/housing disruptions.
Finance Application
- The findings on earnings loss, housing disruption, and increased mortality/crime risk following involuntary hospitalization have significant implications for household finance and insurance.
- In household finance, this could inform research on how mental health crises contribute to financial distress, debt accumulation, and wealth erosion, and how financial institutions could better assess and manage these risks.
- For insurance, the quantified increase in suicide/overdose and violent crime risk directly impacts life, health, and disability insurance pricing, potentially leading to new models for underwriting mental health-related risks or designing targeted interventions for policyholders post-crisis.
Mental HealthInvoluntary HospitalizationCausal InferenceJudge IVHousehold FinanceInsuranceMortality RiskEmploymentHomelessnessCrimeAdministrative Data
Core finding, identification, data
Core Finding
- For individuals whose cases are "judgment calls," involuntary hospitalization nearly doubles the probability of dying by suicide or overdose and being charged with a violent crime within three months after evaluation.
- These adverse effects are partly driven by significant decreases in employment and earnings, and increased homelessness.
Identification Strategy
- The study employs a "judge IV" instrumental variable approach, leveraging the quasi-random assignment of individuals experiencing mental health crises to different emergency department physicians.
- The instrument is a physician's residualized tendency to hospitalize, which is shown to be conditionally random and affects hospitalization decisions but not outcomes directly.
Data
The paper utilizes comprehensive administrative data from Allegheny County, Pennsylvania, including records on involuntary hospitalizations, Medicaid claims, Homeless Management Information System (HMIS) data, state unemployment insurance records, and National Plan & Provider Enumeration System (NPPES) data for physician characteristics.
Kaiwen Leong, Huailu Li, Linh T. Tô — Economics of Crime
This paper demonstrates that publicly revealing actual peer attitudes can correct misperceived social norms, leading to a significant reduction in institutional offenses within a youth correctional facility.
Finance Application
- The finding that misperceived social norms drive behavior and can be corrected by public signals has strong implications for household finance and behavioral asset pricing.
- In household finance, individuals might undersave or overspend due to misperceptions about their peers' financial prudence; a public campaign revealing actual savings rates or debt aversion could correct these beliefs and encourage better financial habits.
- In asset pricing, herd behavior or the formation of bubbles/crashes could be influenced by misperceived market sentiment or the prevalence of certain investment strategies (e.g., ESG, meme stocks); public, aggregated data on actual investor beliefs or portfolio allocations could serve as a common knowledge signal to coordinate behavior, potentially stabilizing markets or accelerating the adoption of new, efficient investment norms among retail investors.
social normsmisperceived normscommon knowledgebehavioral financehousehold financeasset pricingpeer effectsinformation interventioncoordination gamesretail investorsESG investingmarket sentimentbehavioral economics
Core finding, identification, data
Core Finding
- Youths in a Reformative Training Center privately oppose misbehavior but substantially underestimate their peers' support for positive behavior.
- A randomized controlled trial publicly disclosing actual peer attitudes narrowed this belief gap, significantly reducing institutional offenses by 40-50% for up to six months.
- The effects were more pronounced for less popular and more educated youths, and faded with institutional turnover, underscoring the dynamic and fragile nature of social norms sustained by common knowledge.
Identification Strategy
- The study employs a population-level randomized controlled trial within a youth correctional facility, where youths were randomly assigned at the community group level to a control group or one of two treatment groups (whiteboard or paper announcement).
- The intervention publicly disclosed aggregated survey results on peer attitudes, allowing for the causal identification of norm-correcting information and the formation of common knowledge.
- Administrative offense data and a two-phase survey design (pre- and post-intervention) further strengthen the causal inference.
Data
The paper uses detailed surveys from 327 youth offenders in Singapore's Reformative Training Center (RTC) between 2020 and 2023, capturing private attitudes, beliefs about peer preferences, confidence in beliefs, social networks, cognitive ability, demographics, and psychological traits. It also utilizes administrative offense data from RTC (2020-2023) and inmates' admission/release dates.
Romaine A. Campbell, Logan M. Lee — Economics of Crime
This paper investigates the causal impact of prison education on post-release outcomes, finding that it unexpectedly increases reincarceration due to institutional responses and heightened supervision, while also boosting employment.
Finance Application
- This paper offers a rich framework for household finance and insurance research.
- Financial institutions often use 'positive signals' (e.g., financial literacy course completion, participation in debt management programs, or even certain educational achievements) to assess risk or offer products.
- This research suggests investigating the *unintended consequences* of these signals, particularly how institutional agents (e.g., loan officers, insurance underwriters, financial advisors) respond to them.
- For example, does completing a financial literacy course lead to higher credit limits or lower insurance premiums, which then, due to increased monitoring or behavioral changes, result in a higher *detection rate* of defaults or claims, even if underlying financial health improves? The heterogeneity by race also suggests exploring biases in how financial signals are interpreted across demographic groups.
EducationPrisonRecidivismEmploymentUnintended ConsequencesBehavioral EconomicsInstitutional BehaviorSignalsHousehold FinanceInsuranceInstrumental Variables
Core finding, identification, data
Core Finding
- The core finding is that participation in prison education, particularly postsecondary courses, increases the likelihood of reincarceration for technical violations (revocations) but not new crimes, especially among white inmates.
- This occurs because education serves as a 'positive signal' to case managers, leading to increased assignment to work release, which entails more intensive post-release supervision and thus a higher probability of detecting violations.
- Despite this, prison education also significantly increases post-release employment.
Identification Strategy
- The paper employs an instrumental variable (IV) approach.
- The instrument is an 'opportunity score' calculated as the number of courses that started in a prisoner's facility during their incarceration.
- This exploits quasi-random variation in course access driven by the interaction of academic calendar-based course start times and individual prison entry/transfer timing, conditional on primary prison, time served, and release year fixed effects.
Data
The study uses a unique, comprehensive individual-level dataset from Iowa, combining administrative data from the Iowa Department of Corrections (IDOC), Iowa Department of Education (IDOE), Iowa Workforce Development (IowaWORKS), and Grinnell College. This data tracks individuals released from Iowa prisons between 2014 and 2018, including their incarceration spells, course-taking history, misconduct records, and post-release employment and education outcomes.
James D. Macek — Urban Economics
This paper examines the impact of minimum lot size regulations on housing affordability, welfare, inequality, and income segregation across US cities and neighborhoods.
Finance Application
- The paper's findings on significant capital losses for landowners (17% of land values) due to deregulation directly inform real estate asset pricing and household wealth.
- This regulatory risk could be priced into REITs, real estate development loans, and mortgage-backed securities, especially in highly regulated markets.
- The impact on household wealth accumulation and consumption patterns across different income and ownership groups (renters vs. homeowners) could be studied in household finance, examining how regulatory changes affect mortgage demand, default risk, and overall portfolio allocation.
Housing RegulationReal EstateHousehold FinanceAsset PricingUrban EconomicsWealth InequalityLand ValuesPolicy ImpactMortgage Markets
Core finding, identification, data
Core Finding
- Minimum lot size regulations are found to be most expensive in low-density neighborhoods of productive cities, explaining income sorting.
- Deregulation leads to significant progressive welfare gains for renters (9% of income) but substantial capital losses for landowners (17% of land values), with limited aggregate productivity gains due to out-migration of affluent households.
- Cities lack incentives to unilaterally deregulate due to concerns over deteriorating amenity values.
Identification Strategy
- The causal effect of neighborhood affluence on amenities is identified using a 'donut identification design' based on terrain slopes, where the income of a neighborhood is instrumented with the slopes of other neighborhoods within a specific distance band.
- Minimum lot sizes are detected by identifying the mode of observed lot size distributions within geographically constructed zoning districts.
Data
The study uses CoreLogic property assessments and transactions, American Community Survey (ACS) data for household income distributions and demographics, the National Neighborhood Data Archive (NaNDA) for neighborhood data, and the USGS EDNA database for terrain slopes.
Manuel Alcaino, Raimundo Undurraga — Economics of Crime
This paper evaluates a low-cost, scalable centralized dropout monitoring system in Chile, finding it significantly increases student re-enrollment and grade completion, and substantially reduces juvenile crime rates.
Finance Application
- This research has direct implications for household finance and insurance.
- In household finance, the intervention's impact on human capital (education, crime reduction) could be linked to long-term wealth accumulation, savings behavior, and debt management, especially for vulnerable households.
- For insurance, reduced juvenile crime and increased educational attainment directly lower future risks (e.g., unemployment, incarceration), informing more accurate pricing for life, disability, or property/casualty insurance products.
- The high benefit-cost ratio also suggests potential for social impact bonds or ESG investing, where financial instruments could fund such interventions with returns tied to measurable social outcomes like reduced crime rates.
Human CapitalCrimeEducationHousehold FinanceInsuranceRisk ManagementSocial Impact BondsPublic PolicyRegression Discontinuity
Core finding, identification, data
Core Finding
- A centralized dropout monitoring system in Chile increased re-enrollment by 32% and grade completion by a similar amount.
- Two years later, students listed in the report had a 79% lower crime rate compared to non-listed counterparts, primarily due to reductions in non-severe crimes.
- This intervention yields a benefit-cost ratio of 6:1 for crime prevention, generating annual savings equivalent to US$9 million.
Identification Strategy
- The study employs a Regression Discontinuity (RD) design, leveraging a predetermined cutoff date (May 30th, 2022) for eligibility to be included in the dropout report.
- It compares students who disenrolled just before and just after this cutoff, focusing on 'two-sided schools' (those with dropouts on both sides of the cutoff) to ensure internal validity against endogenous reporting practices.
Data
The paper uses administrative data from Chile's Ministry of Education (MINEDUC) including student-level enrollment, disenrollment, graduation records, grades, and demographics. It also incorporates administrative data from SENAME (Chile's Youth National Protection Program) on criminal records for 14-17-year-old juveniles, covering the period 2022-2024.
Ingrid Ellen, Daniel Hartley, Jeffrey Lin, Wei You — Urban Economics
This paper studies the unintended consequences of Fair Access to Insurance Requirements (FAIR) plans on urban disinvestment and neighborhood decline in US cities during the mid-20th century.
Finance Application
- The paper's findings on how poorly designed insurance policies can induce moral hazard and lead to urban disinvestment have significant implications for real estate asset pricing and household finance.
- Researchers could investigate how the presence and design of government-backed insurance programs (e.g., flood insurance, wildfire insurance, or even mortgage insurance) affect long-term real estate valuations, REIT performance, and municipal bond yields in vulnerable areas.
- For household finance, the identified landlord abandonment behavior due to over-insurance could inform models of strategic default or property maintenance decisions by homeowners under various insurance schemes, particularly when property values decline below replacement costs.
- This framework could also be applied to analyze the unintended consequences of climate risk insurance policies.
InsuranceMoral HazardUrban EconomicsReal EstateDisinvestmentPublic PolicyProperty ValuesAsset PricingHousehold FinanceHistorical DataTriple Difference
Core finding, identification, data
Core Finding
- FAIR plans, designed to address insurance redlining in urban neighborhoods, inadvertently led to significant housing disinvestment, accelerated declines in neighborhood population and income, and increased the Black population share.
- This was driven by problematic features like prohibitions on considering environmental hazards, diluted underwriting incentives, and payouts exceeding market values, which created moral hazard and incentivized property abandonment or arson.
Identification Strategy
- The authors use a triple-difference design.
- The first difference compares outcomes before and after FAIR plan authorization (1968).
- The second distinguishes between neighborhoods with and without likely FAIR access, identified by measuring the withdrawal of private property insurance establishments from central neighborhoods using purpose-digitized city directories (1940–1967).
- The third compares this within-city neighborhood contrast between states that launched FAIR plans early (by 1970) and states that did not.
Data
The paper uses a balanced panel of consistent-boundary census tracts from 1950 through 1990 for 26 major US cities. Key data sources include purpose-digitized city directories from 1940 and 1967 to measure private insurer withdrawal, and fire incident data from National Fire Protection Association (NFPA) reports, US Department of Justice (DOJ) surveys, and the National Fire Incident Reporting System (NFIRS).
Marcella Alsan, Joshua Schwartzstein, Stefanie Stantcheva — Economics of Crime
This paper investigates how beliefs about safety, benefits, and harms of firearms, and the malleability of these beliefs, influence gun ownership and policy preferences among owners and non-owners.
Finance Application
- The core framework of a 'safety-possibilities frontier' and the malleability of beliefs about benefits and harms, particularly in response to information and endorsement, has direct applications in household finance and insurance.
- In household finance, this could model how investors perceive the risk-return trade-off of different asset classes or financial products.
- Information campaigns about the true costs (e.g., fees, behavioral biases) or benefits (e.g., diversification, long-term growth) of financial decisions, especially if endorsed by trusted sources, could shift perceived frontiers and influence investment choices.
- For insurance, the framework can explain demand for various policies (e.g., life, health, property) by understanding how individuals weigh perceived risks and protective benefits, and how targeted information about specific risks (e.g., climate change, cyber threats) or policy effectiveness could alter insurance uptake.
Behavioral EconomicsRisk PerceptionInformation AsymmetryBelief FormationDecision Making Under UncertaintyExperimental EconomicsHousehold BehaviorMalleability of BeliefsSafety-Possibilities Frontier
Core finding, identification, data
Core Finding
- The paper finds that while both gun owners and non-owners share a common objective of safety, they hold significantly different beliefs regarding the benefits and harms of lethal firearms.
- Gun owners tend to emphasize protective benefits and downplay risks, whereas non-owners focus on private and social harms.
- Crucially, these beliefs are malleable: randomized information treatments about the private costs of lethal firearms or the existence and efficacy of non-lethal alternatives (especially when endorsed) effectively shift gun owners' perceptions and policy preferences towards safer practices and greater openness to non-lethal options.
Identification Strategy
- The study employs a large-scale online survey with a randomized controlled trial design.
- Participants, comprising both lethal firearm owners (LFAO) and non-owners (NO), are randomly assigned to one of several treatment groups: a control group, a group receiving information on the private costs of lethal firearm ownership, a group receiving information about a specific non-lethal firearm alternative (Byrna), or a group receiving both information and endorsement for the non-lethal alternative.
- The causal effects are identified by comparing the post-treatment beliefs, attitudes, and real-stakes policy preferences across these randomized groups.
Data
The paper uses an original large-scale survey of lethal firearm owners and non-owners, recruited from Prolific with quota sampling to mirror U.S. population gun-ownership status. The survey collects data on preferences, needs, values, beliefs, backgrounds, behaviors, and policy views. It also incorporates FBI Uniform Crime Reporting (UCR) data for county-level violent crime incidence.
Clare A. Balboni, Gharad T. Bryan, Melanie Morten, Caylee O'Connor — Urban Economics
This paper develops a framework to distinguish between place-based and people-based causal effects of urban infrastructure by accounting for population mobility and selection, finding that place-based estimates often overstate individual-level impacts due to gentrification.
Finance Application
- The paper's findings on gentrification and selection are highly relevant for household finance, particularly in understanding wealth inequality and financial vulnerability.
- Infrastructure projects, by attracting wealthier residents, can displace lower-income households, impacting their savings, debt, and access to financial services.
- For real estate asset pricing, the observed increase in rental rates in treated areas suggests a direct impact on property values, but the 'place-based' versus 'people-based' distinction is crucial: investors focusing solely on aggregate property value increases might misprice assets if the underlying economic well-being of long-term residents does not improve.
- This could lead to mispricing in REITs or other real estate investment vehicles if the changing demographic risk is not adequately factored in.
Urban economicsInfrastructureGentrificationPopulation mobilitySelection effectsPlace-based effectsHousehold financeReal estateAsset pricingWealth inequalityUrban developmentTransportationPanel dataQuasi-experimentFinancial vulnerability
Core finding, identification, data
Core Finding
- While place-based estimates show strong positive impacts of Bus Rapid Transit (BRT) on labor force participation and household income, these effects are significantly weaker and often statistically insignificant for the actual residents.
- This divergence is driven by positive selection, where new residents (arrivers) in BRT-affected areas have higher education and baseline income than those who leave (exiters), indicating that gentrification explains much of the observed place-based gains.
Identification Strategy
- The study employs a difference-in-differences (DiD) framework, leveraging novel panel data that tracks both structures and individuals over time.
- To address spatial endogeneity in BRT placement, the treatment variable (reduction in log travel time) is demeaned by the expected reduction across all planned BRT phases, following Borusyak and Hull (2023), under a conditional parallel trends assumption.
Data
The paper uses bespoke panel data collected from 1,750 households in Dar es Salaam, Tanzania, surveyed at baseline (pre-BRT in 2016) and tracked for a follow-up three years later. It includes detailed pre- and post-BRT travel time data and incorporates retrospective information for new residents, complemented by World Bank LSMS data for pre-trends analysis.
Ghizlen Ouasbaa, Albert Solé-Ollé, Elisabet Viladecans-Marsal — Urban Economics
This paper uses a regression discontinuity design to causally estimate the impact of electing real estate developers to city councils on local housing supply in California.
Finance Application
- This research offers a novel mechanism for predicting local housing supply, which is crucial for real estate asset pricing.
- The election of a developer could serve as a predictive signal for future housing stock growth, impacting the valuation of residential REITs, private real estate funds, and mortgage-backed securities tied to specific metropolitan areas.
- For household finance, increased housing supply could affect local housing price appreciation and rental growth, influencing homeowner equity, mortgage demand, and household wealth accumulation.
- Furthermore, changes in development activity could impact municipal bond credit risk and pricing, as local government revenues are often linked to property development and taxes.
Political EconomyReal EstateHousing SupplyLocal GovernmentRegression Discontinuity DesignAsset PricingHousehold FinanceMunicipal FinanceREITsUrban Economics
Core finding, identification, data
Core Finding
- Electing a real estate developer to a city council leads to a 68% increase in approved housing units during their term, particularly for multifamily projects.
- This effect is primarily driven by discretionary zoning approvals and does not result in lasting regulatory reforms, suggesting developers act as effective deal-makers rather than broad reformers.
Identification Strategy
- The study employs a Regression Discontinuity Design (RDD) by analyzing close elections where a real estate developer candidate narrowly won or lost against a non-developer candidate.
- This approach leverages the quasi-random assignment of a developer to office around the vote share cutoff to identify the causal impact on housing supply.
Data
The paper uses candidate occupation data from the California Election Data Archive (CEDA) for over 30,000 city council candidates (1995-2017), supplemented by external sources like LinkedIn. Housing supply data comes from the U.S. Census Building Permits Survey (1990-2020), and council voting data on housing-related issues is extracted from city council minutes using web scraping and AI tools.
Daniel E. Gold, Lu Han, Christopher Timmins — Urban Economics
This paper uncovers a novel form of housing discrimination, 'selective unadvertising,' where landlords withhold available rental units from public listings to disproportionately benefit white renters, particularly in desirable neighborhoods.
Finance Application
- This research offers significant insights for household finance by demonstrating how information asymmetry in housing markets can exacerbate racial wealth gaps and financial vulnerability, particularly for minority households steered into less desirable areas.
- For real estate asset pricing, the existence of a substantial 'off-market' segment, selectively allocated, implies that publicly available data may not reflect the true supply-demand dynamics or intrinsic value of properties, potentially leading to mispricing or inefficient capital allocation for REITs and other real estate investors.
- In insurance, the steering of minority groups into neighborhoods with higher environmental risks or lower amenities could lead to differential risk exposures and, consequently, higher insurance premiums or reduced access to coverage for these populations.
Housing DiscriminationRental MarketsRacial BiasSelective AdvertisingResidential SegregationHousehold FinanceReal EstateInformation AsymmetryWealth InequalityInsurance Risk
Core finding, identification, data
Core Finding
- Black renters are significantly more likely to occupy publicly listed units than comparable white renters, with this disparity being especially pronounced in neighborhoods offering better amenities, environmental quality, and upward mobility.
- This selective unadvertising disproportionately affects lower-income renters and families with children, reinforcing residential segregation and unequal access to neighborhood resources.
Identification Strategy
- The study employs a novel data-driven approach by merging unit-level rental listings from an online platform (Dwellsy) with detailed tenant turnover records (InfoUSA) across 27 U.S. metro areas.
- It uses multinomial logit models to compare the racial composition of occupants in listed versus 'hidden' (unadvertised but turned over) units, controlling for building characteristics, neighborhood attributes, and manager-specific factors through various fixed effects (manager, tract, building).
Data
The paper uses Dwellsy rental listings (2021-2024), InfoUSA Residential Historical Files (2006-2024) for individual and household characteristics, and tract-level sociodemographic data from the American Community Survey (ACS). Supplemental data from Census, rental registration records (Seattle), CoStar database, and web-scraped listings are used for validation.
Konhee Chang — Urban Economics
This paper studies how rental housing supply, driven by large-scale corporate landlords, shapes residential segregation and spatial inequality in American suburbs.
Finance Application
- The findings on geographic scale economies and acquisition premiums for corporate SFR landlords directly inform the valuation and investment strategies of SFR REITs and other institutional real estate investors, suggesting optimal portfolio concentration.
- The 'disamenity' effect of renters on incumbent homeowners could be incorporated into hedonic pricing models for real estate, affecting property valuations and thus the underlying collateral value for real estate-backed securities (e.g., CMBS).
- In household finance, the paper's emphasis on financial constraints and the impact on homeownership accessibility has direct implications for mortgage demand, underwriting standards, and the risk profiles of mortgage portfolios, especially for different wealth and racial groups.
Core finding, identification, data
Core Finding
- The entry of large-scale corporate single-family rental (SFR) landlords reduces residential segregation by enabling lower-income, disproportionately non-White renters to move into high-amenity suburban neighborhoods where they previously could not afford to own.
- This reallocation is driven by landlords' scale economies and conversion of owner-occupied homes to rentals, leading to lower rents but higher home prices, and causing incumbent homeowners to move out due to perceived disamenities.
- While progressive for low-wealth renters, it negatively impacts median households priced out of homeownership and high-wealth households due to amenity changes.
Identification Strategy
- The paper employs property-level event studies, leveraging variation in acquisition timing by corporate SFR landlords, controlling for neighborhood trends and property fixed effects.
- For incumbent household responses, it exploits ex-ante proximity to properties acquired by SFR landlords.
- Local scale economies are identified using a repeat-sales design to measure the acquisition premium paid by corporate landlords as their local scale expands.
- Finally, GMM with exposure to corporate SFR landlords as an instrument identifies elasticities for endogenous amenities and rental supply in a quantitative spatial equilibrium model.
Data
The paper constructs a novel property-level dataset combining housing deeds, assessor records, rental listings (MLS), and address histories (Data Axle) for 23 corporate SFR landlords. It also uses TREPP commercial mortgage data for operating costs, ACS for neighborhood characteristics, and SCF for household wealth distribution.
Andrew Garin, Ethan Jenkins, Evan E. Mast, Bryan A. Stuart — Urban Economics
This paper uses administrative data to show that individuals in high-poverty neighborhoods experience significant earnings growth, which often leads them to migrate to wealthier areas, contributing to the persistence of poverty in their original neighborhoods.
Finance Application
- This paper's findings offer significant arbitrage opportunities for household finance and real estate asset pricing.
- The causal link between idiosyncratic earnings shocks and housing location/value changes can inform models of household portfolio choice, particularly housing investment and mortgage demand, by incorporating endogenous mobility and neighborhood sorting.
- For asset pricing, understanding how earnings growth drives demand for housing in specific neighborhoods could refine local real estate market models, impacting valuations of REITs, mortgage-backed securities, and local bank portfolios.
- Furthermore, the persistence of neighborhood poverty due to selective migration could be a factor in municipal bond pricing or local business investment decisions.
MigrationEarnings ShocksNeighborhoodsPovertyHousehold FinanceReal EstateHousing MarketsWealth AccumulationLocal Economics
Core finding, identification, data
Core Finding
- The paper finds that 36.5% of individuals in high-poverty neighborhoods move to less poor areas within eight years, and their earnings growth rates are similar to those in richer neighborhoods.
- Using idiosyncratic firm-specific pay shocks, they causally demonstrate that higher earnings drive migration to better neighborhoods, with elasticities of tract median income and housing values to earnings shocks being 0.175 and 0.212, respectively.
- This selective out-migration of upwardly mobile residents is a key mechanism explaining why poor neighborhoods remain poor.
Identification Strategy
- The causal effect of earnings changes on neighborhood choice is identified by isolating variation in idiosyncratic, firm-specific pay shocks.
- This is achieved by constructing a 'pay shock' variable based on the average percent change in earnings among a holdout sample of an individual's coworkers, following methods from Koustas (2018) and Ganong et al. (2020).
- This design allows for controlling for broader neighborhood changes and individual/firm characteristics.
Data
The study utilizes newly available administrative data from the Census Bureau's Master Address File-Auxiliary Reference File (MAFARF) for longitudinal address information, the Longitudinal Employer-Household Dynamics (LEHD) database for earnings data, and the American Community Survey (ACS) for individual/household demographics and neighborhood characteristics.
Hadar Avivi, Tslil Aloni — Urban Economics
This paper studies how childhood residential location affects adult income for native-born and immigrant children in Israel, highlighting significant heterogeneity in location effects across these groups and proposing robust targeting policies.
Finance Application
- This research offers crucial insights for household finance and insurance.
- Understanding heterogeneous location effects on adult income can inform optimal housing investment and mortgage lending strategies, as families' long-term earnings potential is tied to childhood residence.
- Financial advisors could leverage this to provide personalized advice on residential choices for maximizing human capital and lifetime wealth.
- For insurance, these insights could refine risk assessment for life or long-term care policies, potentially leading to location-adjusted premiums or the development of new "human capital insurance" products that mitigate risks associated with suboptimal childhood environments.
human capitalresidential mobilityincome inequalityimmigrant integrationhousehold financereal estatesocial mobilityrisk assessmentpolicy evaluation
Core finding, identification, data
Core Finding
- Childhood residential location effects on adult income vary substantially between native-born Israeli children and immigrants from the former Soviet Union.
- For low-income families, places that boost income for one group do not necessarily benefit the other, leading to a near-zero correlation in location effects.
- This heterogeneity implies that unified neighborhood recommendation policies, like those in "moving to opportunity" programs, can yield inferior outcomes for minority groups.
Identification Strategy
- Causal location effects are identified by exploiting variations in children's exposure time to different cities during childhood, due to household moves at different ages.
- For immigrants, this combines variations in moves within Israel with the age at which children migrated from the former Soviet Union, assuming that for families with the same location choices, the child's age at arrival is unrelated to unobserved factors affecting outcomes.
Data
The paper uses comprehensive administrative data from the Israeli Central Bureau of Statistics (CBS), covering all registered Israeli citizens born between 1950-1995 and their parents. This includes tax records (1995-2019), education records, civil registry records (demographics, family links, annual location of residence 1999-2019 and 1995), and the 1995 census.
Matthew Freedman, Noah A. Kouchekinia, David Neumark — Urban Economics
This paper evaluates the longer-run labor market impacts of Opportunity Zones (OZs), finding that while OZs increase job creation within zones, these jobs primarily benefit higher-income residents from outside the zones and are largely offset by job declines in nearby low-income tracts.
Finance Application
- The finding that OZ investments primarily benefit higher-income outsiders and lead to job displacement in nearby low-income areas has significant implications for real estate finance and ESG investing.
- Real estate investors and developers targeting OZs might be mispricing the long-term social and economic sustainability of their projects if they assume broad community uplift, potentially leading to overvaluation or misaligned impact goals.
- For household finance, the limited impact on OZ residents' earnings and poverty suggests that financial institutions lending in these areas should adjust credit risk models, as the underlying economic conditions for local residents may not be improving as expected.
- This also highlights a potential 'impact washing' risk for ESG funds promoting OZ investments as socially beneficial.
Opportunity ZonesPlace-Based PoliciesReal EstateEmploymentPovertyGentrificationSpillover EffectsTax IncentivesCausal InferenceESG InvestingHousehold FinanceAsset PricingLocal Economic Development
Core finding, identification, data
Core Finding
- Opportunity Zone designation increases job creation among businesses within zones, but these newly created jobs are largely taken by residents of higher-income census tracts rather than residents of the OZ or other non-OZ low-income tracts.
- Consequently, there are limited impacts on OZ resident employment rates, earnings, or poverty rates.
- Moreover, newly created jobs in zones are nearly offset one-to-one by declines in nearby low-income tracts, effectively undoing redistributive benefits.
Identification Strategy
- The study employs an event-study framework, comparing designated OZs to eligible but not designated tracts.
- To address violations of the parallel trends assumption, it uses Inverse Probability Weighting (IPW) and doubly robust regression adjustment, which weight control tracts to match the pre-treatment economic trajectories of treated tracts.
- The models include tract fixed effects and year fixed effects to control for unobserved heterogeneity and common time trends.
Data
The paper utilizes American Community Survey (ACS) data for 2013-2023 for resident outcomes and LEHD Origin-Destination Employment Statistics (LODES) for 2013-2022 for workplace jobs and commuting flows. Data on OZ designation comes from the Community Development Financial Institutions (CDFI) Fund.
Patrick W. Farrell, Matthew Easton — Urban Economics
This paper proposes a theoretical explanation for the observed power law distribution of city populations, showing that lognormal distributions arise naturally from quantitative spatial equilibrium models due to lognormally distributed locational fundamentals and market access.
Finance Application
- This framework provides a robust method for modeling the spatial distribution of economic activity, which is highly relevant for finance.
- In real estate, the model could be adapted to predict the lognormal distribution of commercial or residential property values and rents across regions, driven by local amenities (fundamentals) and access to economic centers (market access), informing REIT investment strategies and mortgage-backed security risk.
- For asset pricing, understanding regional economic growth patterns, derived from these spatial equilibrium dynamics, could help explain cross-sectional differences in local equity returns or private equity valuations for geographically concentrated businesses.
- In insurance, the lognormal distribution of population and economic activity could significantly improve catastrophe risk modeling by providing a more robust spatial distribution of insured values, enhancing pricing for property & casualty insurance and reinsurance contracts, especially for tail events.
spatial economicscity size distributionlognormal distributionpower lawreal estateasset pricingregional economicsinsurance riskgeographic factorsmarket accesseconomic geography
Core finding, identification, data
Core Finding
- The paper demonstrates that in quantitative spatial equilibrium models, a location's population is the product of its locational fundamentals (qualities) and its market access (trade with other locations).
- When both these terms are lognormally distributed (due to random variation in geographic attributes and trade costs, respectively), the resulting population distribution is also lognormal, which naturally approximates a power law in its upper tail, explaining the observed regularity in city sizes.
Identification Strategy
- The authors validate their theoretical predictions by inverting their quantitative spatial equilibrium model using U.S. county-level data (2020 Census, ACS, CFS, TIGER/Line maps) to empirically recover and show that both locational fundamentals and market access terms are lognormally distributed.
- They further simulate the model with empirically relevant parameters to demonstrate its robustness and ability to match observed city size distributions and their sensitivity to changes in agglomeration benefits, congestion, and transportation costs.
Data
The paper uses U.S. county-level data from the 2020 Decennial Census for population, 2020 American Community Survey (ACS) for incomes/wages, 2017 Commodity Flow Survey (CFS) for trade flows, and TIGER/Line maps for transportation networks. It also incorporates gridded geographic data from Henderson et al. (2018) for attributes like ruggedness, elevation, temperature, and land suitability.
Anaïs Fabre — Urban Economics
This paper quantifies how the uneven spatial distribution of higher education options and student mobility frictions contribute to spatial inequalities in educational attainment and skill sorting in France.
Finance Application
- The findings on mobility frictions and brain drain directly impact lifetime earnings and wealth accumulation for households in 'education deserts.' This could inform research on optimal student loan design (e.g., income-contingent repayment tied to location), the effectiveness of place-based financial aid, or how regional human capital disparities contribute to household wealth gaps and intergenerational mobility of financial outcomes.
- Regional skill sorting and brain drain could also affect local real estate markets and the valuation of local businesses, influencing regional risk premia or the pricing of geographically concentrated assets.
Human CapitalSpatial EconomicsEducationLabor MarketsMigrationHousehold FinanceRegional InequalityStructural ModelsTwo-Sided MatchingAsset Pricing
Core finding, identification, data
Core Finding
- The paper demonstrates that the within-country spatial distribution of colleges significantly contributes to spatial inequalities.
- Higher education options are unevenly distributed, and student demand is highly sensitive to geographic proximity, leading to gaps in educational attainment and spatial skill sorting.
- Mobility frictions explain one-third of regional gaps in educational attainment, but eliminating them creates a trade-off: it benefits students from low-opportunity areas but accelerates brain drain to education hubs, magnifying regional inequalities.
Identification Strategy
- The paper builds a dynamic model combining a two-sided matching model of the higher education market with a model of location decisions at entry on the labor market.
- It identifies students' and programs' preferences from data on choices and equilibrium outcomes without parametric assumptions on payoffs, leveraging local policy variations (moving subsidies, removal of geographic admission priorities) as quasi-experiments.
Data
The study uses comprehensive administrative data on the universe of college applicants and programs in France, tracking students' educational paths until exit from the higher education system. It also incorporates a survey of individuals' home, study, and work locations, along with their wages, after they enter the labor market.
Gabriel Ahlfeldt, Ismir Mulalic, Caterina Soto Vieira, Daniel M. Sturm — Urban Economics
This paper uses individual-level panel data for Denmark to document how location choices and residential floor space consumption evolve over the life cycle and in response to life events, and quantifies the role of amenity preferences and demographic changes in spatial sorting.
Finance Application
- The findings on how life events and amenity preferences drive spatial sorting directly inform models of local housing demand and price dynamics, crucial for real estate asset pricing.
- Investors in REITs or local real estate could forecast demand shifts in urban sub-markets based on demographic trends (e.g., aging populations shifting demand from central to suburban areas).
- In household finance, the paper highlights how life events (cohabitation, childbirth, separation) significantly alter housing consumption and location choices, which are major financial decisions impacting housing equity, mortgage demand, and overall wealth accumulation.
- For insurance, the finding that widowhood substantially increases floor space consumption could inform life insurance product design, highlighting a specific financial need for surviving spouses to maintain living standards.
Urban EconomicsSpatial SortingLife CycleDemographicsHousing MarketsAmenitiesReal EstateHousehold FinanceEvent StudiesQuantitative ModelsDenmark
Core finding, identification, data
Core Finding
- The main mechanism driving spatial sorting within cities is heterogeneity in preferences for local amenities, with young singles preferring central consumption amenities and families/seniors valuing natural amenities in suburban areas.
- While individual demographic changes (population aging, falling fertility, increasing singlehood) have substantial impacts on urban geography, their combined effect on cities is found to be substantially more muted due to counteracting forces.
Identification Strategy
- The paper uses newly constructed individual-level panel data for Denmark and employs event-study specifications, leveraging the imputation estimator proposed by Borusyak et al. (2024) and Liu et al. (2024).
- This method disentangles sorting by age and family status from correlated individual effects and allows for joint estimation of the effects of various life events (cohabitation, childbirth, separation, empty nesting, retirement, death of spouse) on location choices and floor space consumption, controlling for individual fixed effects.
- A quantitative urban model is then estimated to invert for structural parameters like amenity valuations.
Data
The paper uses newly constructed individual-level panel data covering the entire population of Denmark from 1986 to 2019. This dataset includes residence and workplace locations (100x100 meter grid cells), income, residential floor space consumption, property prices, and detailed demographic information (age, education, marital status, children). Travel times between locations are also computed.
Grace Ortuzar, David C. Phillips, James X. Sullivan — Urban Economics
This paper presents a randomized controlled trial evaluating the effectiveness of time-limited rental subsidies in reducing homelessness among single, homeless adults in Santa Clara County, CA.
Finance Application
- This research demonstrates that targeted, temporary financial assistance can significantly reduce homelessness and improve housing stability, with effects persisting beyond the subsidy period.
- This insight is highly relevant for household finance, suggesting that such interventions can improve creditworthiness and reduce default risk for vulnerable populations, impacting consumer lending portfolios.
- For asset pricing, reduced homelessness could signal healthier local rental markets and more stable property values, affecting the risk and return of real estate investment trusts (REITs) and mortgage-backed securities (MBS).
- In insurance, it highlights the potential for new social or micro-insurance products designed to mitigate housing instability risks, and could indirectly lower health and life insurance claims by reducing the severe health consequences of homelessness.
HomelessnessRental SubsidiesRandomized Controlled TrialHousehold FinanceCredit RiskHousing MarketsSocial InsuranceFinancial StabilityReal EstateSocial Impact
Core finding, identification, data
Core Finding
- The study finds that temporary rental subsidies significantly reduce the incidence of homelessness by one-third and shelter days by two-thirds while active.
- Preliminary sub-sample results suggest these effects persist even after the subsidy ends, indicating that financial constraints are a primary driver of homelessness for a significant fraction of the population.
Identification Strategy
- The study employs a randomized controlled trial (RCT) design, randomly assigning eligible single, homeless adults to either a treatment group (offered temporary rental subsidies and case management) or a control group (usual care).
- This random assignment ensures a causal estimate of the intervention's impact by creating statistically equivalent groups, allowing for direct comparison of outcomes.
Data
The primary data source is Santa Clara County's Homeless Management Information System (HMIS), which provides client-level data on baseline characteristics, program enrollment outcomes (including emergency shelter, street outreach, and housing programs), and financial assistance received.
Martin Koenen, Drew M. Johnston — Urban Economics
This paper uses Facebook data to show that social ties are spatially concentrated and causally influence individuals' residential choices, explaining why people, especially the less-educated, are less responsive to economic opportunities and shocks.
Finance Application
- The paper's findings have significant implications for household finance and local asset pricing.
- First, the strong influence of social networks on residential inertia suggests that local economic shocks (e.g., industry downturns) could have prolonged effects on local housing markets and the value of locally concentrated assets, as individuals are less likely to move even in the face of declining economic opportunities.
- This could lead to slower price discovery or higher risk premia for real estate in areas with strong social ties.
- Second, the differential responsiveness to shocks across education levels, driven by network concentration, implies that less-educated households might be more exposed to local economic risks, potentially affecting their savings behavior, debt accumulation, and demand for formal insurance products (e.g., unemployment or mortgage insurance), as informal network support might substitute for formal mechanisms.
social networksmigrationresidential choicespatial equilibriumhousehold financeasset pricinglocal marketsreal estatelabor mobilityeconomic shocksinsurancedemographics
Core finding, identification, data
Core Finding
- The study finds that social ties are highly concentrated locally (78% of friends within 100 miles) and causally increase an individual's likelihood of living in a given commuting zone by 0.3 percentage points for each additional friend, an effect comparable to a $470 annual wage increase.
- Incorporating these network effects into a spatial equilibrium model significantly reduces unexplained residuals and explains why migration is inelastic to wages, particularly for less-educated individuals with more concentrated networks.
Identification Strategy
- The causal effect of social networks on migration is identified by exploiting plausibly exogenous variation in the timing of friends' moves around an individual's college graduation.
- The key assumption, termed 'symmetric bias,' posits that unobservable factors influencing friends' moves and location decisions are symmetrically correlated for friends moving before versus after an individual's graduation, allowing the difference in predictive power to isolate the causal effect.
Data
The paper uses de-identified, individual-level data from Facebook users in the United States from 2012 to 2023, which includes individuals' monthly location, demographics, socioeconomic status, and social connections (friends). It also uses data from the Quarterly Census of Employment and Wages for labor market information and a custom survey.
Pablo A. Valenzuela-Casasempere — Urban Economics
This paper studies the long-run effects of forced displacement due to the construction of the U.S. Interstate Highway System on individuals' mortality, mobility, and socioeconomic outcomes.
Finance Application
- The findings have direct implications for **insurance** and **household finance**.
- For life insurers, the quantified reduction in longevity for displaced populations could be integrated into actuarial models to refine premium pricing and assess long-term liabilities, especially for cohorts affected by large-scale urban development.
- In household finance, the observed decline in wealth accumulation and relocation to lower-SES areas for displaced individuals highlights increased credit risk, potentially affecting mortgage lending, consumer credit, and default rates in historically impacted communities.
- This could inform the development of more nuanced credit scoring models and targeted financial products for vulnerable populations, or be used in **real estate asset pricing** to evaluate the long-term social and economic externalities of infrastructure projects on property values and local market stability.
Household FinanceInsuranceMortality RiskWealth InequalityReal EstateUrban EconomicsInfrastructureSocial CostsDisplacementLongevityCredit Markets
Core finding, identification, data
Core Finding
- Displaced individuals die approximately three months younger, are more likely to leave their original neighborhoods, and relocate to areas with lower socioeconomic characteristics, leading to reduced long-term wealth accumulation.
- These negative effects are highly localized (within 100 meters of the highway) and disproportionately impacted racial and socioeconomic minorities, with destination neighborhoods explaining 30% of the displacement-mortality effect.
Identification Strategy
- The study identifies displaced individuals by geocoding 1940 census addresses and linking them to detailed construction data of the Interstate Highway System (1956-1960).
- It employs three complementary strategies: comparing affected individuals to unaffected neighbors (100-200 meters away), matching affected individuals to controls far from highways, and using planned highway maps as a placebo control group, while controlling for various individual and city-level fixed effects.
Data
The paper uses geocoded address-level data from the full-count 1940 U.S. Census, linked to administrative mortality records from 1995-2005. It also incorporates highway construction data (PR-511, OpenStreetMap, Federal Engineering Maps), 1950 census data for neighborhood characteristics, municipal tax records for property values, and Facebook data for social connection measures.
Paul Gertler, Marco Gonzalez-Navarro, Raimundo Undurraga, Joaquin A. Urrego — Urban Economics
This paper empirically compares the direct impacts and spatial spillover effects of two major slum renewal policies—in-situ upgrading and population relocation—on slum characteristics, residents' socioeconomic profiles, and adjacent formal neighborhoods in Chile using a comprehensive, long-term panel dataset.
Finance Application
- This research offers several avenues for finance.
- In asset pricing, the differential impact of urban renewal policies on local housing investment, property values, and crime rates could be used to model and predict real estate returns and risk premia for urban properties, especially for REITs or private equity funds investing in urban development.
- For household finance, the findings on improved housing quality and socioeconomic status from in-situ upgrading could inform studies on household wealth accumulation, mortgage access, and default risk in developing urban areas.
- In insurance, the significant reduction in property and violent crimes in upgraded neighborhoods provides direct evidence for re-evaluating and optimizing property and casualty insurance pricing and risk assessment models for urban areas undergoing similar development.
Urban EconomicsReal Estate FinanceHousehold FinanceInsuranceSpatial SpilloversPolicy EvaluationCausal InferenceSynthetic Difference-in-DifferencesSatellite ImageryMachine LearningCrimeHousing QualitySocioeconomic Status
Core finding, identification, data
Core Finding
- The study finds that in-situ upgrading significantly improves housing quality and residents' socioeconomic status, and generates positive spillovers in adjacent neighborhoods, reducing crime and increasing formal housing investment.
- In contrast, population relocation primarily reduces slum population and land use for residential purposes but shows no significant improvements in housing quality, socioeconomic status, or positive spillovers in nearby areas.
- Furthermore, in-situ upgrading is found to be more cost-effective.
Identification Strategy
- The paper primarily uses a Synthetic Difference-in-Differences (SDiD) approach (Arkhangelsky et al., 2021; Clarke et al., 2023) to estimate causal effects, accounting for staggered treatment timing and non-random assignment by constructing synthetic control groups from never-treated slums.
- For outcomes observed only in census years (socioeconomic variables), a Two-Way Fixed Effects (TWFE) model is employed.
Data
The study utilizes a unique slum-level panel dataset spanning over 20 years (2000-2021) for the universe of slums in Chile. This dataset integrates Google Earth satellite imagery (processed with CNN algorithms for building footprints), population census data (2002, 2017), administrative records (MINVU), construction permits, property tax data (Chilean Internal Revenue Service), and geocoded crime reports (Subsecretaría de Prevención del Delito).
Andrew C. Johnston, Nolan G. Pope, Maggie R. Jones — Children and Families
This paper uses linked tax and Census records for over 5 million children to examine how parental divorce affects family arrangements and children's long-term adult outcomes.
Finance Application
- The paper's findings have direct implications for household finance, insurance, and real estate.
- In household finance, the significant income drop and its long-term effects on children's earnings can inform models of consumption, savings, and debt management around major life events, potentially revealing new risks for mortgage or credit default.
- For insurance, the increased child mortality (35-55%) and teen birth rates (63%) directly impact life and health insurance pricing, suggesting opportunities for insurers to refine risk assessments and develop tailored products for individuals from divorced families.
- In real estate, the relocation to poorer neighborhoods post-divorce affects housing demand and property values, which could be integrated into real estate investment strategies or analyses of mortgage-backed securities.
DivorceFamily StructureHousehold IncomeChild OutcomesMortalityTeen BirthsIncarcerationCollege ResidencyNeighborhood QualityParental ProximityHousehold FinanceInsuranceReal EstateLife EventsHuman Capital
Core finding, identification, data
Core Finding
- Parental divorce significantly reduces children's adult earnings and college residence while increasing incarceration, mortality, and teen births.
- These detrimental effects are substantially mediated by changes in household income, neighborhood quality, and parent proximity, which collectively explain 25 to 60 percent of the observed impacts.
Identification Strategy
- The study employs a within-family design, comparing siblings with different lengths of exposure to the same parental divorce.
- This approach identifies age-specific divorce effects by controlling for shared family fixed effects, initial environment, and inherited endowments, thus addressing selection bias.
Data
The paper utilizes linked U.S. federal tax records, Social Security Administration data, and Census Bureau records for over 5 million children born between 1988 and 1993, tracking marital histories, household characteristics, and adult outcomes through 2018.
John Eric Humphries, Christopher Neilson, Xiaoyang Ye, Seth D. Zimmerman — Children and Families
This paper uses admissions lotteries for an extended-day universal pre-kindergarten (UPK) program in New Haven, Connecticut, to evaluate its impact on parental earnings, labor supply, and children's educational outcomes.
Finance Application
- The paper's findings on sustained parental earnings gains and reduced career disruption due to childcare access have significant implications for household finance.
- Researchers could investigate how this increased income stability and human capital accumulation affects household portfolio choices, such as equity risk-taking, or demand for financial products like life and disability insurance.
- Furthermore, the reduced career disruption could lower credit risk for lenders, impacting mortgage underwriting standards or consumer loan pricing in areas with robust UPK programs.
human capitallabor supplyhousehold financepublic policychildcareearningsnatural experimentcredit riskportfolio choice
Core finding, identification, data
Core Finding
- UPK enrollment significantly increases parental earnings by 21.7% during the pre-kindergarten years, with gains persisting for at least six years, primarily due to increased labor supply and reduced career disruption.
- These earnings gains dramatically reduce the net government costs of UPK by 90%, leading to a high marginal value of public funds (MVPF) of 10.1, even with limited effects on children's academic outcomes.
Identification Strategy
- The study leverages admissions lotteries for the UPK program as a natural experiment, providing exogenous variation in program access.
- This quasi-experimental design allows for instrumental variable (IV) estimation, where lottery assignment serves as the instrument for UPK enrollment, controlling for assignment propensities and demographics.
Data
The paper links administrative records of NHPS pre-kindergarten admissions lotteries (2003-2022) with state data on school enrollment and achievement (Connecticut State Department of Education), parent earnings and employment data (state Unemployment Insurance records, 1999-2022), and survey data from past applicants on childcare costs and hours worked.
Hamid Noghanibehambari, David Slusky, Hoa Vu — Children and Families
This paper investigates the multigenerational impacts of legalized abortion in the United States, finding that early-life exposure to this policy shift improves birth outcomes in the subsequent generation.
Finance Application
- This research offers a compelling framework for finance to study the long-term, intergenerational impacts of social policies on human capital and economic well-being.
- In household finance, researchers could investigate how variations in reproductive health access (e.g., pre- vs. post-Dobbs decision) influence household savings rates, investment in children's education, and retirement planning, given the documented improvements in health and education.
- For insurance markets, these findings could inform the pricing of health and life insurance products, as healthier populations imply lower long-term healthcare costs and potentially longer lifespans, affecting actuarial models.
- Furthermore, regional asset pricing could explore whether states with more permissive reproductive health policies, leading to a healthier and more productive workforce over time, exhibit different trends in local housing prices, municipal bond yields, or the valuation of local businesses.
reproductive health policyintergenerational effectshuman capitalbirth outcomeshousehold financehealth insuranceasset pricingpolicy uncertaintyeconomic growthregional economics
Core finding, identification, data
Core Finding
- The study finds that first-generation mothers' early-life exposure to legal abortion leads to a statistically significant increase in their children's birth weight (3.3 grams, or 0.1% relative to the mean) and reduces the likelihood of low birth weight by 1.6%.
- These improvements are primarily driven by enhanced educational attainment and increased prenatal care utilization among the first generation, rather than changes in maternal racial or age composition.
Identification Strategy
- The paper employs event study and two-way fixed effects models, leveraging the staggered implementation of abortion legalization across US states in the late 1960s and early 1970s as a quasi-natural experiment.
- It compares birth outcomes of mothers born in different years relative to the birth-state-specific year of abortion legalization, controlling for various maternal and state-level covariates, and using methods from Sun & Abraham (2021) for robustness.
Data
The primary data source is the restricted-access NCHS Natality Detail Files (1974-2017) for individual birth records. It also uses Census-ACS data (2000-2022) for disability outcomes, state-level disease inventory from the Tycho project, and state-level Aid to Families with Dependent Children (AFDC) and Fair Employment Practices Act (FEPA) data, as well as ASEC-CPS data for gender wage gap measures.
Liran Einav, Amy Finkelstein, Petra Persson — Children and Families
This paper uses Swedish administrative data and event studies to causally estimate the impact of having a child with Down syndrome on parental co-habitation, subsequent child-bearing, labor earnings, and total income, contrasting environments with and without prenatal testing.
Finance Application
- This paper offers direct insights for household finance and insurance research.
- The finding that generous disability allowances largely insure families against additional financial costs can be used to quantify the value of social insurance in mitigating financial distress from health shocks, informing models of household consumption-savings decisions under health risk.
- The observed selection in the 'prevalent-testing' environment provides a rich setting to study adverse selection in private insurance markets for rare health events, examining how information (prenatal testing) affects insurance demand and pricing.
- Furthermore, the impact on maternal labor supply and household income, even with allowances, could influence local housing markets and demand for specific goods/services, relevant for local asset pricing models.
health shockshousehold financeinsurancesocial welfarelabor supplyfamily structureevent studynatural experimentparental decisionsdisabilityprenatal testing
Core finding, identification, data
Core Finding
- Having a child with Down syndrome increases parental co-habitation and subsequent child-bearing.
- While it leads to a greater decrease in maternal earnings in a 'no-testing' environment, this effect reverses (smaller decline) in a 'prevalent-testing' environment, suggesting selection.
- Crucially, total income substantially increases due to generous disability allowances, largely insuring families against additional financial costs.
Identification Strategy
- The study employs an event-study design around birth, exploiting the 'no-testing' environment (births to mothers aged 34 or younger in 2005 or earlier) as a natural experiment where the occurrence of Down syndrome is random conditional on maternal age.
- This allows for a causal comparison of outcomes between families with and without a child with Down syndrome, controlling for maternal age and using fixed effects.
Data
The paper utilizes population-wide Swedish administrative data from 1990-2019, including Medical Birth Records, the Swedish Population Register, Statistics Sweden's longitudinal database of individuals (LISA) for income and demographics, and National Board of Health and Welfare records for death, patient visits, and prescription drug fills.
Adrienne Lucas, Patrick McEwan, David Torres Irribarra — Children and Families
This paper estimates the long-run and intergenerational effects of a scaled-up, government-run, education conditional cash transfer (CCT) program in Chile on indigenous adults and their children.
Finance Application
- The intergenerational effects on human capital and income stability could inform models of intergenerational wealth transfer and financial literacy transmission in household finance.
- Researchers could investigate how CCTs influence parental investment in children's financial education or early-life savings accounts, or how they alter risk preferences for financial investments among beneficiary households.
- For asset pricing, reduced ethnic inequality and increased human capital imply more stable and higher future labor income for a segment of the population, potentially affecting aggregate consumption growth and its volatility, which are key drivers in asset pricing models.
- This could be used to study the equity risk premium or the pricing of human capital-intensive firms in emerging markets.
- Insurers could analyze how CCTs affect the risk profiles and insurance needs of low-income, marginalized communities over the long run, potentially leading to new product designs or market expansion strategies.
Conditional Cash TransfersEducationHuman CapitalIntergenerational MobilityEthnic InequalityLabor EarningsHousehold FinanceAsset PricingEmerging MarketsSocial PolicyQuasi-Experiment
Core finding, identification, data
Core Finding
- The study finds that cohorts of indigenous adults with the greatest exposure to the Beca Indígena program completed 0.6 additional years of schooling, had 22% higher labor earnings, and increased hours worked by 10%, significantly reducing pre-treatment ethnic differences.
- Furthermore, mothers' exposure increased their children's early-grade test scores and reduced second-generation grant receipt, indicating positive intergenerational spillovers and a transformation in socioeconomic environments.
Identification Strategy
- The identification strategy leverages a quasi-experimental design based on variation in expected grant exposure across birth cohorts and indigenous status.
- It compares indigenous children born after 1974 (increasingly exposed to grants) with those born before 1975 (not exposed) and with never-treated non-indigenous children born in the same years and communes.
- The design uses granular fixed effects for combinations of birth cohorts and communes, and indigenous-status-by-birth-commune fixed effects, to control for unobserved variables and ensure a difference-in-differences estimation within birth communes.
Data
The study uses large samples of Chilean adults (born 1965-2000) and their children, pooling observations from multiple years of the CASEN household survey (Encuesta de Caracterización Socioeconómica Nacional) and Chile's national SIMCE tests (Sistema de Medición de la Calidad de Educación). Administrative data on grant awards and census microdata are also used to calculate expected grant exposure.
Kristin Forbes, Jongrim Ha, M. Ayhan Kose — International Finance and Macroeconomics Data Session
This paper analyzes how monetary policy tradeoffs between economic activity, inflation, and the price level have evolved over time and across countries, particularly focusing on the post-pandemic adjustment, by constructing a new database of 'rate cycles.'
Finance Application
- The identified 'rate cycles' and their characteristics (e.g., aggressive vs. gradual tightening, delayed liftoff) could serve as crucial state variables in dynamic asset pricing models, explaining variations in equity risk premia, bond yields, and currency returns across different monetary policy regimes.
- The finding that central bank credibility mitigates negative outcomes suggests that market perceptions of central bank independence and transparency could be priced into asset valuations.
- Furthermore, the emphasis on the price level's impact on households could inspire new research into inflation-linked financial products or insurance against price level risk, beyond just inflation rate risk, especially for long-term investors or pension funds.
monetary policycentral banksinterest ratesinflationprice leveleconomic activitysacrifice ratiorate cyclescredibilityasset pricingrisk managementmacroeconomics
Core finding, identification, data
Core Finding
- The post-pandemic tightening period exhibited unusually low 'Sacrifice Ratios' (output losses per inflation reduction) but a historically large increase in the price level, reflecting a combination of a delayed start to tightening, subsequent aggressive rate hikes, and strong central bank credibility.
- Delayed liftoff leads to larger price level increases and less attractive tradeoffs, while greater central bank credibility consistently yields better outcomes across all metrics without apparent tradeoffs.
Identification Strategy
- The paper identifies 'rate cycles' (easing and tightening phases) using a modified Bry and Boschan algorithm applied to policy interest rates and balance sheet programs for 24 advanced economies.
- It then uses principal component analysis to construct measures of 'Delayed Start' (timing of liftoff relative to macroeconomic conditions) and 'Aggressiveness of Rate Hikes,' alongside central bank credibility and oil shocks, to explain variations in Sacrifice Ratios, output gaps, inflation reduction, and price level changes through regressions and FRB/US model simulations.
Data
The study compiles a new monthly database of policy interest rates and QE/QT programs for 24 advanced economies from 1970-2024, drawing from BIS, Haver Analytics, FRED, OECD, and various NBER/CEPR papers. It also uses macroeconomic data (GDP, industrial production, employment, unemployment, CPI/PCE inflation) from OECD, Haver Analytics, and Ha, Kose, Ohnsorge, along with measures of central bank credibility and labor market flexibility.
John A. Mondragon — Capital Markets and the Economy
This paper quantifies the relative importance of price and rationing mechanisms in the mortgage market's response to credit supply shocks, particularly during the March 2020 market disruptions.
Finance Application
- The finding that rationing dominates price effects in credit supply adjustments has significant implications for asset pricing, particularly for mortgage-backed securities (MBS).
- Models of MBS prepayment and extension risk might be miscalibrated if they solely rely on interest rate elasticities, as rationing could decouple new mortgage supply from interest rate movements, leading to mispricing.
- In household finance, this framework could be applied to other consumer credit markets (e.g., auto loans, credit cards) to understand how non-price mechanisms affect credit availability during periods of uncertainty, impacting household consumption and investment decisions beyond what interest rates alone suggest.
credit rationingmortgage marketcredit supplynon-price mechanismshousehold financemonetary policy transmissionrisk managementfixed incomeMBScredit markets
Core finding, identification, data
Core Finding
- The paper finds that during the March 2020 market disruptions, jumbo mortgage borrowers faced significantly tighter credit supply conditions compared to conforming loan borrowers, leading to a 50% reduction in refinancing likelihood and a 4-6% reduction in borrowed amounts.
- A decomposition reveals that rationing (specifically, a sharp decline in cash-out refinances) accounted for two to three times as much of the decline in borrowing as price effects, challenging models that primarily rely on price clearing.
Identification Strategy
- The identification strategy leverages the discontinuity in credit supply conditions around the conforming loan limit (CLL).
- Loans above the CLL (jumbo loans) are privately held, while conforming loans (below CLL) are largely held by GSEs, creating differential credit and liquidity conditions.
- The paper uses a difference-in-differences approach, comparing refinancing behavior of borrowers with loans just above and below the CLL, and employs a 'donut' analysis to mitigate contamination from loans very close to the limit.
Data
The paper utilizes data from Equifax Credit Risks Insight Servicing McDash and Black Knight McDash Data, which merges mortgage performance and origination data with credit bureau data. This allows for observation of total mortgage balances, refinancing events, and new loan characteristics for borrowers.
Christoph Boehm, Andrei A. Levchenko, Nitya Pandalai-Nayar, Hiroshi Toma — International Trade & Investment
This paper derives closed-form expressions for steady-state gains from trade in dynamic models, highlighting the crucial role of short-run trade elasticities and the transition path.
Finance Application
- This framework's distinction between short-run and long-run trade elasticities, coupled with its dynamic transition paths for firm entry and trade flows, offers valuable insights for asset pricing and risk management.
- For asset pricing, it could inform models of cross-border investment and multinational firm valuation, as trade policy shocks (e.g., tariffs) would have differential impacts on firm profitability and capital flows depending on the elasticity horizon.
- For risk management, the gradual adjustment of trade flows and firm masses could be used to model supply chain resilience and price trade credit insurance or political risk insurance, especially for long-term contracts, by quantifying the duration and magnitude of economic adjustments to trade disruptions.
Dynamic Trade ModelsGains from TradeTrade ElasticitiesShort-run ElasticityLong-run ElasticityTransition DynamicsInternational FinanceAsset PricingTrade Policy UncertaintySupply Chain Risk
Core finding, identification, data
Core Finding
- The gains from trade in dynamic models are substantial and crucially depend on the short-run tariff elasticity; they can be arbitrarily large even with a high long-run tariff elasticity if the short-run elasticity is low.
- Accounting for the transition path has only a modest impact (7-12%) on the magnitude of these gains compared to steady-state comparisons.
Identification Strategy
- The paper's methodological innovation is to recover the structural parameters (long-run elasticity of trade with respect to iceberg costs and the exponent governing the mass of firms) by combining empirical estimates of both short-run and long-run tariff elasticities.
- This is because the short-run elasticity is a function of one parameter, while the long-run elasticity is a function of both, allowing for their separate identification.
Data
The study uses OECD Inter-Country Input Output tables (ICIO) for domestic absorption shares, TRAINS dataset for tariff data, BACI version of UN-COMTRADE for trade flows, World Bank for aggregate tariff revenues, and Penn World Tables for real GDP.
Gustavo De Souza, Ruben Gaetani, Martí Mestieri — International Trade & Investment
This paper demonstrates that trade liberalization, through tariff cuts, can paradoxically reduce technology diffusion in developing countries by incentivizing foreign firms to export directly rather than transfer technology.
Finance Application
- This research offers a novel perspective for asset pricing by suggesting that trade liberalization, often viewed as a positive catalyst, can have complex and even detrimental long-run effects on productivity and firm growth through reduced technology diffusion.
- This 'less diffusion' channel could lead to mispricing of long-term equity returns, particularly for firms in emerging markets undergoing trade reforms, if investors only focus on short-term trade volume increases.
- For corporate finance, it highlights how trade policy influences the strategic choice between direct foreign investment (FDI) and technology licensing, impacting capital allocation and R&D decisions of multinational corporations and their local partners.
- Furthermore, it introduces a 'technology diffusion risk' that could be incorporated into credit risk models or insurance products for firms heavily reliant on international knowledge transfer.
Trade PolicyTechnology DiffusionTariffsProductivity GrowthInternational TradePatent CitationsTechnology TransfersWelfare EconomicsAsset PricingCorporate FinanceInvestment DecisionsEconomic Growth
Core finding, identification, data
Core Finding
- Lower import tariffs in Brazil lead to a significant drop in technology diffusion, as foreign firms choose to export their products directly instead of engaging in technology transfers with local firms.
- This reduction in technology transfers, particularly know-how, slows the diffusion of foreign ideas and reduces long-run productivity growth, diminishing the overall welfare gains from trade liberalization by more than three-fourths compared to models that ignore this channel.
Identification Strategy
- The study uses an instrumental variable approach, following Boehm et al. (2023), to identify the causal effect of tariffs on technology diffusion.
- The instrument leverages the Most Favored Nation (MFN) tariff system, comparing changes in technology flows from small MFN partners (whose tariffs are primarily driven by factors related to large MFN partners) to those from preferential partners, thereby creating plausibly exogenous variation in tariffs.
Data
The paper utilizes comprehensive records of technology transfer contracts from the Brazilian patent office, patent citation data from the European Patent Office's Worldwide Patent Statistical Database (PATSTAT), firm ownership and FDI data from the Brazilian Firm Registry, and import tariff data from the World Bank Trade Analysis Information System.
Chang Liu, Kohei Takeda — Development of the American Economy
This paper investigates how universities drive structural transformation from manufacturing to services in local economies by acting as talent and innovation hubs, leading to higher growth in high-skilled services and increased skill premiums.
Finance Application
- The paper's insights could inform asset pricing by suggesting a 'university innovation factor' for local equity markets; firms located in CZs with top-ranked universities might exhibit higher growth and thus a premium in stock returns due to the continuous emergence of high-skilled tasks.
- In household finance, the persistent higher skill premium and dynamic labor markets in university regions imply different human capital risk profiles, influencing optimal savings, education investment, and insurance decisions for local households.
- For real estate finance, the findings suggest that commercial and residential property values in university-dense areas should show stronger appreciation and stability, driven by sustained demand from high-skilled labor and innovative businesses, which could be explored through geographically targeted REITs or private equity real estate funds.
Structural TransformationInnovationSkill PremiumEconomic GeographyHuman CapitalRegional DevelopmentLabor MarketsUniversitiesTask-Based ModelsAsset PricingHousehold FinanceReal Estate Finance
Core finding, identification, data
Core Finding
- The paper documents four key facts: (i) commuting zones (CZs) with universities, especially top-ranked ones, experience greater increases in service sector employment and establishments; (ii) regional differences in structural transformation are primarily driven by changes within tasks and skills; (iii) skill premiums are consistently higher in university-dense CZs; and (iv) new tasks disproportionately emerge in regions with universities.
- The theoretical model highlights that universities' innovation role, creating new tasks, is crucial for explaining both service sector growth and rising skill premiums in these regions.
Identification Strategy
- To establish a causal link between universities and local economic outcomes, the authors employ a quasi-experimental design.
- They leverage historical college site selections (1839-1954) as a natural experiment, comparing 'winning' counties (selected to host a university) with 'runner-up' counties from the same selection process.
- This approach helps to isolate the impact of university presence from confounding factors like initial economic conditions or local amenities.
Data
The study utilizes comprehensive regional data from the US, including County Business Patterns (CBP) for employment and establishments, American Community Survey (ACS) for individual-level wages, occupations, and skills, and the Integrated Postsecondary Education Data System (IPEDS) for university characteristics and rankings. It also incorporates 'new work' data from Autor et al. (2024) to track emerging occupation titles.
Heitor S. Pellegrina, Farid Farrokhi, Sebastian Sotelo, Elliot Kang — International Trade & Investment
This paper empirically documents and theoretically models deforestation in a dynamic world trade system, analyzing the impact of trade policy and population growth on global forest area and welfare.
Finance Application
- The paper's dynamic pricing of 'new land' (qi) and land rental rates (ri) based on future agricultural rents and a discount rate (ρ) could be directly applied to asset pricing models for real estate, timberland, or agricultural land, incorporating climate-related risks and carbon credit values.
- The quantification of CO2 emissions costs and welfare impacts from trade policies under different carbon prices offers a framework for assessing climate transition risks in investment portfolios exposed to agricultural commodities or land-intensive industries.
- Furthermore, the model's insights into global deforestation reallocation could inform ESG investing strategies and sovereign risk analysis, particularly for countries heavily reliant on agricultural exports and vulnerable to climate policy changes or supply chain disruptions.
deforestationinternational tradeclimate changeland usegeneral equilibrium modelasset pricingenvironmental economicsagricultural economicscarbon emissionsESG investingreal estate financesupply chain financesovereign riskdynamic modeling
Core finding, identification, data
Core Finding
- Global forest area decreased by 7.1% between 1990-2020, driven by agricultural expansion, comparative advantage in agriculture, and population growth.
- A dynamic general equilibrium model shows that multilateral reductions in agricultural trade costs can increase global forest area and lead to net welfare benefits, primarily by reallocating deforestation pressures across countries based on comparative advantage and structural change.
Identification Strategy
- For empirical patterns, the paper uses correlations and regressions.
- For causal inference regarding population growth and land use, it employs an instrumental variable (IV) approach.
- Domestic population growth is instrumented by the median age, crude birth rate, and average age for childbearing from the previous period, with similar instruments for trading partners.
- The identifying assumption is that past demographic structure predicts population growth for biological reasons, uncorrelated with current productivity shocks.
- The model's land-producing sector parameters are calibrated to match these reduced-form population impacts.
Data
The paper uses FAO Forest Resource Assessment (FAO-FRA) for forest area and carbon stock (1990-2020), FAO-STAT for agricultural production, land use, and international trade, UN-ILO for employment, and the GTAP database for economic accounts. Trade costs are calibrated using World Bank World Development Indicators (tariffs, import fees) and the CEPII Gravity Database (trade agreements, geography). UN population projections are used for future scenarios.
Richard Sylla — Development of the American Economy
This paper re-evaluates Charles Beard's economic interpretation of the US Constitution, arguing that slave ownership, not public debt, was the dominant economic interest shaping its provisions.
Finance Application
- This insight could be applied to asset pricing by examining how the political protection of a dominant asset class (e.g., real estate, specific industries, or even carbon credits today) affects its risk premium, liquidity, and long-term returns.
- In household finance, it could inspire research into how the concentration of wealth in specific asset types influences policy decisions that perpetuate wealth inequality or create 'protected' forms of capital.
- For institutional design, one could study how the asset holdings of founding shareholders or major stakeholders influence corporate charters and governance structures, similar to how slaveholders influenced the Constitution.
Historical FinancePolitical EconomyAsset ValuationInstitutional DesignWealth InequalitySlaveryUS Constitution
Core finding, identification, data
Core Finding
- The paper finds that the market value of slave property vastly exceeded the market value of public securities (national debt) at the time of the Philadelphia Convention in 1787, and continued to do so even after financial reforms.
- This suggests that the Constitution's provisions were primarily designed to protect slaveholders' interests, rather than merely being compromises.
Identification Strategy
- The paper's identification strategy involves a comparative historical economic analysis, presenting new evidence to quantify and compare the market values of slave property versus national debt during the founding era.
- It re-interprets historical institutional design by demonstrating the relative economic power of different asset classes.
Data
The paper uses historical data on the "market value of slave property" and the "market value of the national debt" (including assumed state debts) during the US founding era, specifically around 1787 and following subsequent financial reforms.
Christopher M. Meissner, Alexander Klein — Development of the American Economy
This paper examines the relationship between tariffs and labor productivity in US manufacturing from 1870 to 1909, concluding that tariffs generally reduced productivity.
Finance Application
- The findings on how tariffs affect industry productivity, firm size, and entry/exit dynamics could be highly relevant for asset pricing.
- Researchers could investigate how historical tariff changes impacted industry-specific stock returns or bond yields, considering the trade-off between protection and productivity.
- For household finance, the tariff-induced increases in output prices and shifts in employment could be linked to household consumption patterns, savings rates, or regional wealth accumulation.
- The paper's emphasis on lobbying and political economy also offers a unique historical setting to study the financial value of political connections for firms, examining whether firms in successfully protected industries experienced different access to capital or valuations.
TariffsProductivityEconomic HistoryManufacturingGilded AgeInstrumental VariablesTrade PolicyIndustry DynamicsPolitical EconomyFirm HeterogeneityAsset PricingHousehold Finance
Core finding, identification, data
Core Finding
- The main empirical finding is that tariffs reduced labor productivity in US manufacturing during the Gilded Age.
- While tariffs generally reduced the average size of establishments, they increased output prices, value-added, gross output, employment, and the number of establishments.
- The authors suggest that tariffs weakened import competition and encouraged the entry of smaller, less productive domestic firms, ultimately hindering the US from becoming a globally competitive manufacturer.
Identification Strategy
- The identification strategy employs an instrumental variable (IV) approach to address the endogeneity of tariff policy.
- It leverages the fact that many US tariff lines were specific tariffs, making actual tariff rates a function of exogenous foreign price shocks.
- The instrument is constructed using short-horizon changes in import unit values interacted with product-level reliance on specific tariffs, which predicts changes in the ad valorem equivalent of product-level tariffs, adapted from Greenland and Lopresti (2024).
Data
The paper uses a newly compiled industry-level database of tariffs from 1870-1900, digitized from the universe of tariff-lines and imports from official data (Foreign Commerce and Navigation of the United States - FCNUS). This is matched with state-level Census of Manufactures data for 1870, 1880, 1890, 1900, and 1909, covering over 80 dis-aggregated manufacturing industries.
Ran Abramitzky, Lena Greska, Santiago Pérez, Joseph Price, Carlo Schwarz, Fabian Waldinger — Development of the American Economy
This paper investigates how socio-economic background influences academic careers, from entry and discipline choice to productivity and recognition, using a large dataset of U.S. academics from 1900-1969.
Finance Application
- This research offers a powerful framework to analyze how socio-economic background influences career paths and recognition within finance.
- For instance, one could investigate if fund managers or financial analysts from lower SES backgrounds are more likely to pursue 'novel' investment strategies, potentially leading to higher variance in returns but also breakthrough performance, yet receive less recognition from institutional investors.
- This could also inform studies on diversity in corporate leadership or venture capital, examining if founders from less privileged backgrounds face higher hurdles in securing funding or achieving market recognition despite innovative ideas, impacting wealth creation and intergenerational mobility.
Socio-economic backgroundCareer progressionInnovationRecognitionTalent allocationIntergenerational mobilityHuman capitalLabor economicsSociology of scienceText analysisBig data
Core finding, identification, data
Core Finding
- Individuals from lower socio-economic backgrounds are severely underrepresented in academia, especially in elite institutions and humanities.
- While they are more likely to either not publish or have outstanding publication records and introduce more novel scientific concepts, they receive less recognition (fewer citations, Nobel nominations, and awards) compared to their higher SES peers, even after controlling for publication output.
Identification Strategy
- The study employs a comprehensive, linked dataset of U.S. academics (1900-1969) to establish correlational evidence.
- It measures parental socio-economic background using father's predicted income percentile from linked census records.
- Robustness checks include various SES proxies and controls for age, gender, cohort, discipline, and state.
- For discipline choice, text embeddings quantify semantic similarity between father's occupation and academic fields.
Data
The paper constructs a large individual-level dataset of 46,139 U.S. university academics (1900-1969) by linking faculty rosters (World of Academia Database) to full-count U.S. censuses (via Census Linking Project, Census Tree Project, IPUMS MLP), Clarivate Web of Science for publication/citation data, and Nobel Nomination Archive data.
Christian P. Hoeck, Tobias Renkin — Monetary Economics
This paper estimates firm-level supply curves using export demand shocks and integrates these micro-level findings into a New Keynesian model to derive the aggregate Phillips curve slope.
Finance Application
- The identified firm-level supply curve slope (price elasticity to output) could be a novel factor in cross-sectional stock returns.
- Firms with steeper supply curves might exhibit higher price volatility in response to demand shocks, leading to different risk premia or hedging demands in equity markets.
- This micro-level price sensitivity could also inform credit risk models, as firms with steeper supply curves might face higher earnings volatility and thus higher credit risk during demand fluctuations, impacting bond yields and credit default swap spreads.
- Furthermore, understanding the micro-foundations of inflation could lead to better pricing of inflation-linked securities or inflation hedges.
Phillips CurveFirm-levelSupply CurvesDemand ShocksMicro-macro LinkIdentificationShift-share InstrumentLocal ProjectionsPrice DynamicsOutput DynamicsNew Keynesian ModelExport ExposureAsset PricingCredit RiskInflation
Core finding, identification, data
Core Finding
- The paper finds that firm-level supply curves are steep, with a 1% increase in output raising prices by 0.5%.
- When integrated into a New Keynesian model, this 'capacity pressure' channel contributes significantly (0.032pp) to the Phillips curve slope, which is steeper than in textbook calibrations.
- However, with realistic (slightly counter-cyclical) real wages in Denmark, the overall Phillips curve remains rather flat (slope of 0.025).
Identification Strategy
- The identification strategy relies on plausibly exogenous firm-level demand shocks constructed using a shift-share instrument.
- This instrument combines heterogeneity in firms' exposure to different export destinations with fluctuations in aggregate import demand in those destinations over time.
- Local projections are used to estimate the dynamic response of prices and output to these shocks, controlling for aggregate supply shocks and inflation expectations with time-sector fixed effects.
Data
The paper uses Danish register data covering production, sales, and prices of manufacturing firms at the product and destination level, combined with macroeconomic data on countries' product-level imports and exports from the UN Comtrade database. It also uses self-reported capacity utilization data from the Danish Business Sentiment survey and Danish Producer Price Index (PPI) survey microdata.
David W. Berger, Nicholas Turner, Eric Zwick, Geoffrey Gee — Monetary Economics
This paper quantifies the impact of pandemic-era fiscal stimulus on household car purchases and auto inflation, differentiating between new and used car markets.
Finance Application
- The granular data and identification strategy could be applied in household finance to study how various financial shocks (e.g., interest rate changes, housing market downturns) affect durable goods consumption, household debt accumulation, and financial fragility across different income and demographic groups.
- For asset pricing, the paper's findings on supply constraints and demand shifts driving sectoral inflation offer a framework to model how macro shocks propagate to firm valuations and asset returns in durable goods industries, particularly the auto sector, considering the interplay between new and secondary markets.
- In insurance, the observed shifts in new vs. used car purchases and their price dynamics could inform models for auto insurance pricing, claims forecasting (especially for total losses), and risk pool management.
Household FinanceDurable GoodsFiscal StimulusConsumptionInflationAuto MarketSupply ChainsCredit ConstraintsMarginal Propensity to ConsumeAsset Pricing (Sectoral)Difference-in-Differences
Core finding, identification, data
Core Finding
- Pandemic fiscal stimulus increased total auto sales by 3.3% (5.5 million additional vehicles from 2020-2022), implying a 3-year marginal propensity to spend (MPX) on autos of 0.165 and a marginal propensity to consume (MPC) for all goods of 0.44.
- However, fiscal transfers alone explain less than 20% of the auto price surge; a combination of relaxed credit conditions, shifts in consumer preferences toward goods, and temporary supply constraints are necessary to match observed price dynamics.
Identification Strategy
- The study employs a difference-in-differences strategy, exploiting cross-sectional variation in pre-pandemic exposure to stimulus payments (based on Child Tax Credit eligibility by ZIP code and month) and the precise timing of payments.
- This allows for a within-CBSA analysis, controlling for broader macroeconomic shocks with time fixed effects.
Data
The paper utilizes administrative data on vehicle registrations (Experian), household transfers (IRS), loan-financed purchases (FRBNY/Equifax Consumer Credit Panel), COVID-related variables (CDC), mobility patterns (Opportunity Insights GPS), unemployment (BLS), mortgage refinances (McDash), PPP loans (SBA), and consumption shares (CEX).
Yannis Kastis, Hillary G. Vipond — Development of the American Economy
This paper demonstrates how the arrival of Jewish immigrant tailors, bringing distinct organizational practices, causally accelerated technology adoption and the transition to mass production in the English tailoring industry.
Finance Application
- This research offers a compelling framework for understanding how 'organizational capital' drives firm performance and technology adoption, which could be applied in asset pricing to identify a novel factor explaining cross-sectional stock returns.
- Firms with superior organizational practices, perhaps proxied by management quality scores or organizational structure data, might exhibit higher growth or resilience when faced with technological disruption, leading to an 'organizational premium.' In corporate finance, this insight could inform M&A strategies, emphasizing the value of integrating complementary organizational practices rather than just physical assets.
- For household finance, the observed displacement of self-employed generalists and the creation of new specialized roles could be used to model labor market risk and the demand for re-skilling insurance products in the face of automation.
organizational practicestechnology adoptionimmigrationlabor marketsfirm dynamicsproductivityhistorical economicsnatural experimentinstrumental variableshuman capital
Core finding, identification, data
Core Finding
- The arrival of Jewish immigrant tailors, who specialized in ready-to-wear production with a greater division of labor, significantly accelerated the adoption of the sewing machine and mass production in the English tailoring industry.
- This shift compelled native tailors to adopt similar organizational practices, leading to a 16% increase in the average size of native tailoring firms and a transition of native generalists into specialized roles or sewing machinists.
Identification Strategy
- The identification strategy employs a shift-share instrumental variable approach, exploiting the exogenous arrival of Jewish tailors fleeing pogroms in the Russian Empire.
- The instrument predicts district inflows of Russian tailors based on their 1851 settlement patterns (before sewing machines were in use and production was homogeneous), multiplied by the nationwide decennial inflow of Russian immigrants, thus addressing potential endogeneity from immigrants sorting into technologically advanced districts.
Data
The paper constructs a novel dataset on micro-occupations from digitized English census records (1851-1911), firm-level data from the British Business Census of Entrepreneurs, and immigrant occupational profiles from the Poor Jews' Temporary Shelter in London records.
Courtney Wiegand — Monetary Economics
This paper constructs high-frequency fiscal policy shocks from the US budget resolution process and analyzes their impact on various asset prices, including nominal yields, term premiums, convenience yields, and the stock market, considering the stance of monetary policy.
Finance Application
- The high-frequency fiscal shock measure could be applied to study the cross-sectional implications of fiscal policy on corporate bond spreads, credit default swaps, or specific firm-level characteristics like investment and leverage.
- Researchers could also use these shocks to analyze market liquidity and trading behavior around fiscal news, providing insights into information processing in financial markets.
- The findings on term premiums and convenience yields could inform dynamic portfolio allocation strategies for fixed-income investors, especially in environments with fiscal dominance or ZLB constraints.
fiscal policyasset pricinghigh-frequency shocksterm structurenominal yieldsreal yieldsbreakeven inflationterm premiumsconvenience yieldsstock marketmonetary policyzero lower boundevent studydeficit newsgovernment debtmacro-financeindustry portfolios
Core finding, identification, data
Core Finding
- Fiscal policy shocks significantly impact yields across the term structure, with a 1% increase in cumulative deficit-to-GDP over 5 years raising 2-year yields by 2 bps and 10-year yields by 2.3 bps, primarily driven by real yields (two-thirds) and breakeven inflation (one-third).
- Positive deficit news increases term premiums and reduces Treasury convenience yields, suggesting higher deficits make Treasuries riskier and less special.
- When monetary policy is constrained (ZLB), the cash flow effect dominates, leading to a positive stock market response, particularly in growth-sensitive industries.
Identification Strategy
- The paper constructs a novel measure of high-frequency fiscal policy shocks by tracking innovations in forward-looking deficit targets during the annual US budget resolution and reconciliation process.
- It identifies specific dates and magnitudes of deficit news by comparing deficit targets across consecutive legislative stages (President's budget, House/Senate resolutions, unified budget, reconciliation).
- These changes are discounted to present value and scaled by expected GDP.
- The shocks are validated as unpredictable by macroeconomic news or professional forecasts and shown to significantly influence investor expectations and predict realized deficits.
Data
The paper uses Congressional budget resolution documents, CBO cost estimates, and committee reports (1980-2022) to construct fiscal shocks. Financial market data include US nominal, real, and breakeven inflation yields (Gürkaynak et al.), S&P 500, T-Bill rates, term premium estimates (Adrian et al.), convenience yields, and 49 industry portfolios (Ken French Library). Macroeconomic data from FRED and various forecasts from the Survey of Professional Forecasters, Bloomberg, and Fed's Greenbook/Tealbook are also utilized.
Joseph Hoon, Chang Liu, Karsten Müller, Jonathan Payne, Zhongxi Zheng — Monetary Economics
This paper quantifies the heterogeneous state-level economic costs of financial crises in the U.S. from 1863-2022 using a novel macro-financial panel dataset and a comprehensive crisis chronology, finding that local financial distress is more frequent and costly than previously estimated.
Finance Application
- This research offers a rich historical panel dataset and a novel 'states-at-risk' measure that could be invaluable for finance.
- Asset pricing researchers could investigate how regional financial distress predicts cross-sectional returns of state-specific equity portfolios or municipal bonds, potentially constructing regional financial stress factors.
- Household finance scholars could analyze the impact of localized banking crises on household credit access, default rates, and wealth accumulation, especially for geographically concentrated portfolios.
- Insurance researchers could use the 'states-at-risk' measure to model regional systemic risk for property & casualty insurers or to price catastrophe bonds linked to regional economic downturns.
financial crisesregional economicsbanking historymacro-financestate-level dataoutput lossesbank failuresfinancial distress indicatorsasset pricinghousehold financerisk management
Core finding, identification, data
Core Finding
- U.S. financial crises lead to significant, heterogeneous state-level output losses (averaging 6%), which are predictable by local banking conditions like deposit contractions and bank failures.
- A novel local financial distress measure predicts 3% output losses, and the aggregate 'states-at-risk' fraction provides a superior predictor of national output downturns compared to binary crisis indicators.
Identification Strategy
- The study employs panel local projections on a newly constructed state-level macro-financial dataset (1863-2022).
- It develops a novel local financial distress chronology by combining a comprehensive survey of 25 U.S. crisis chronologies with statistical indicators of local banking troubles (bank failures, deposit/wholesale liability contractions), allowing for the identification of hundreds of regional distress events.
Data
The paper uses a newly constructed historical state-level macro-financial panel dataset for 48 U.S. states from 1863-2022, combining digitized bank balance sheets (from OCC, FDIC, FFIEC Call Reports) and 60 annual indicators of economic activity (from Hoon et al., 2025). It also incorporates a systematic survey of 25 major U.S. financial crisis chronologies and bank failure data.
Davide Debortoli, Jordi Galí — Impulse and Propagation Mechanisms
This paper empirically assesses the impact of wealth and income distribution on aggregate consumption dynamics.
Finance Application
- This finding has significant implications for asset pricing and household finance.
- In asset pricing, if aggregate consumption is insensitive to wealth/income distribution, models explaining the equity premium or other aggregate asset pricing puzzles using heterogeneous agents might need to re-evaluate the channels through which heterogeneity impacts aggregate markets.
- It suggests that while individual consumption paths differ, their aggregation might smooth out these differences for aggregate asset pricing.
- For household finance, this implies that households might be more effective at smoothing idiosyncratic income shocks than often assumed, or that the aggregate impact of such shocks on consumption is limited.
- This could inform research on household saving, debt, and portfolio choices, suggesting that policy interventions aimed at redistribution might have limited effects on overall consumption, but still significant welfare implications for specific groups.
MacroeconomicsConsumptionHeterogeneityIdiosyncratic RiskWealth DistributionIncome DistributionEuler EquationHANK ModelsTANK ModelsEmpirical Economics
Core finding, identification, data
Core Finding
- The study finds that wealth and income distribution statistics have, at most, a small quantitative impact on aggregate consumption, contrasting with the strong role played by current disposable income.
- This challenges the predictions of models with idiosyncratic income risk and incomplete markets (like HANK and TANK models) regarding the importance of distribution for aggregate consumption.
Identification Strategy
- The identification strategy involves using Granger causality tests and estimating consumption Euler equation models.
- These models are extended to explicitly incorporate wealth and income distribution statistics to determine their causal and quantitative impact on aggregate consumption, thereby testing the implications of heterogeneous agent models.
Data
The paper uses 'wealth and income distribution statistics' as key data inputs, alongside aggregate consumption and disposable income data. Specific datasets are not named in the abstract.
Luca Gagliardone, Mark Gertler, Simone Lenzu, Joris Tielens — Monetary Economics
This paper develops a unified framework to analyze micro and macro cost-price dynamics, demonstrating state-dependent pricing behavior that explains aggregate inflation in both stable periods and during inflation surges using firm-level data.
Finance Application
- The state-dependent and nonlinear cost passthrough dynamics are critical for asset pricing, especially in understanding inflation risk premia and the valuation of inflation-indexed securities like TIPs, as higher passthrough during large shocks implies greater inflation persistence.
- In corporate finance, these insights can refine earnings forecasts and firm valuations by accounting for how firms' profitability is nonlinearly affected by cost shocks.
- For risk management, the findings could improve the pricing and hedging strategies for inflation derivatives, highlighting the inadequacy of linear models during volatile periods.
InflationState-Dependent PricingMenu CostsPrice RigidityCost PassthroughMicrodataMacro-FinanceAsset PricingCorporate FinanceRisk Management
Core finding, identification, data
Core Finding
- The study finds strong microdata evidence for state-dependent pricing, where cost passthrough into inflation increases disproportionately during large aggregate shocks, while normal times exhibit constant price adjustment frequency.
- A generalized state-dependent pricing model accurately explains both low pre-pandemic inflation and the subsequent nonlinear surge, outperforming time-dependent models during large shocks.
Identification Strategy
- The paper's innovation lies in constructing an empirical measure of "price gaps" (deviation between listed and optimal price) from microdata on firms' prices and costs.
- It nonparametrically tests how firms' pricing strategies align with different models by characterizing the mapping between price gaps and price adjustments.
- A novel calibration strategy uses moments from the joint distribution of price changes and price gaps, and observable aggregate cost shocks are directly fed into the model.
Data
The research uses a rich micro-level dataset of administrative records for Belgian manufacturing firms from 1999 to 2023, including output prices, quantities, and production costs. Specific sources include PRODCOM data for firm-industry price indices, National Bank of Belgium Business Survey (NBB-BS) for price adjustment frequency, and firms' quarterly VAT and social security declarations for variable costs.
Silvia Miranda-Agrippino, Tsvetelina Nenova, Hélène Rey — Monetary Economics
This paper estimates a novel policy rule for the People's Bank of China (PBOC) to identify monetary policy shocks and analyze their domestic and international transmission, considering China's evolving policy framework and dual mandate.
Finance Application
- The paper's distinction between real (commodity markets, trade) and financial spillovers from Chinese monetary policy offers several finance applications.
- Asset pricing researchers could investigate whether global equity or bond markets, particularly those tied to commodity-exporting firms or countries, price in these 'real' Chinese shocks more significantly than direct financial contagion.
- For risk management, the identified liquidity shocks in commodity-exporting emerging markets suggest a need for tailored hedging instruments or sovereign debt structures to mitigate such real economy-driven capital flow volatility.
- In household finance, the pass-through of Chinese-induced commodity price changes to US consumer prices could be used to study how different household consumption baskets or income levels are affected by these external inflation shocks, impacting their savings and investment decisions.
Monetary PolicyChinaInternational SpilloversCommodity MarketsGlobal TradeFinancial FlowsExchange RatesPolicy RulesVAR ModelsAsset PricingRisk ManagementHousehold Finance
Core finding, identification, data
Core Finding
- Chinese monetary policy shocks transmit domestically via textbook channels (higher rates, lower credit, reduced output/trade).
- Internationally, these shocks primarily affect global commodity markets, leading to price falls, and global trade/production, with financial spillovers being second-order.
- The US economy is notably vulnerable, experiencing reduced exports, output, and consumer price inflation due to these spillovers.
Identification Strategy
- The identification strategy involves estimating a non-linear policy rule for the PBOC, which incorporates a novel composite monetary policy indicator (CMPI) and an explicit term for the renminbi's value against a currency basket.
- Monetary policy shocks are identified as the residuals from this rule.
- These shocks are then used as an instrumental variable in monthly Vector Autoregressive (VAR) models, normalized to induce a 1% year-over-year contraction in Chinese industrial production at its peak.
Data
The paper utilizes a proprietary Composite Monetary Policy Indicator (CMPI), monthly Chinese GDP, CPI, and exchange rate data (CFETS EER). It also includes Chinese capital flows, foreign exchange reserves, and various interest rates. Global data comprises global output and trade (excluding China/US), a global financial conditions index, several commodity price indices (CRB, S&P GSCI), and US economic variables (industrial production, imports, exports, import prices, PPI, CPI).
Al-Mahdi Ebsim, Chen Lian, Yueran Ma, Pablo Ottonello, Diego J. Perez — Impulse and Propagation Mechanisms
This paper develops a model with sophisticated corporate borrowing constraints (covenants) that allow for creditor control to reduce managerial agency frictions, and analyzes its micro and macroeconomic implications.
Finance Application
- This research offers several avenues for finance applications.
- In corporate finance, the mechanism of creditor control reducing agency frictions could be applied to study the performance and valuation of firms undergoing private equity buyouts or in distressed debt situations, where debt covenants are actively renegotiated.
- For asset pricing, the muted financial acceleration under sophisticated constraints suggests that firms with such debt structures might exhibit lower cash flow volatility and thus potentially lower equity risk premiums, which could be tested by classifying firms based on their covenant types and examining their stock returns.
- In household finance, one could explore if similar 'sophisticated' lending terms (e.g., dynamic credit card limits or mortgage covenants tied to income) exist and how they affect household consumption and default behavior during economic downturns.
Corporate FinanceDebt CovenantsAgency CostsMacroeconomicsFinancial FrictionsFirm DynamicsCredit CyclesAsset PricingPrivate EquityDistressed Debt
Core finding, identification, data
Core Finding
- The model demonstrates that sophisticated borrowing constraints, enforced by creditor control upon violation, improve firm performance (earnings and capital) by mitigating managerial agency frictions, contrasting with traditional hard constraints that indiscriminately cut credit.
- At the macroeconomic level, these sophisticated constraints lead to muted financial acceleration and milder aggregate impacts from shocks, aligning with the observed resilience of U.S. nonfinancial firms during recessions.
Identification Strategy
- The paper's identification strategy involves calibrating a quantitative model to match observed micro-level empirical regularities of firm behavior around covenant violations, specifically the dynamics of capital and earnings.
- It then uses this calibrated model to conduct counterfactual simulations, comparing the macroeconomic impacts of shocks under sophisticated versus hard borrowing constraints, and under different monetary policy regimes.
Data
The paper calibrates its model to micro-level evidence on firm behavior around covenant violations, drawing on empirical findings from studies such as Roberts & Sufi (2009), Nini, Smith, & Sufi (2012), and Lian & Ma (2021). It also references the resilience of U.S. nonfinancial firms during recessions, as documented by Gertler & Gilchrist (2018).
Kai Christoffel, Matyas Farkas — Impulse and Propagation Mechanisms
This paper develops a monetary policy framework using a regime-switching DSGE model to quantify and manage the risks of inflation expectation de-anchoring, deriving optimal policy responses and applying it to the euro area.
Finance Application
- This framework could be applied to asset pricing by modeling inflation risk premia in nominal bonds, where a higher probability of de-anchoring (as quantified by the model) could lead to increased demand for inflation-indexed securities or higher inflation risk premia in bond yields.
- In household finance, the model's insights into de-anchored expectations could explain shifts in household savings behavior, such as increased allocation to real assets or inflation hedges, particularly during periods of perceived central bank credibility loss.
- For insurance, the quantified de-anchoring risk could inform the pricing of long-duration liabilities (e.g., annuities) and asset-liability management strategies, as unexpected shifts in inflation regimes directly impact the real value of future payouts and investment returns.
Inflation ExpectationsRegime SwitchingMonetary PolicyCentral Bank CredibilityDSGE ModelsInflation RiskBond MarketsHousehold BehaviorInsurance LiabilitiesMacro-Finance
Core finding, identification, data
Core Finding
- The paper finds that optimal monetary policy should be more aggressive in responding to inflation deviations when there are risks of de-anchoring, especially if central bank actions influence its credibility.
- Empirically, it identifies a de-anchoring episode in the euro area between 2011-2015 but notes that the post-COVID inflation surge did not lead to de-anchoring due to a strong central bank response.
- It also shows that a 'looking-through' strategy for supply-side inflation shocks can increase de-anchoring risks.
Identification Strategy
- The paper uses a regime-switching DSGE model, specifically a New Keynesian 3-equation model augmented with a Markov switching process for the perceived inflation target.
- It employs a Kim (1994) filter and an Expectation Maximization algorithm to estimate regime probabilities and parameters, and uses conditional stochastic simulations to quantify de-anchoring risks in real-time.
- The core idea is that agents' perceived inflation target can switch between an anchored and a de-anchored regime, with the central bank's credibility and policy actions influencing these transition probabilities.
Data
The paper uses data from the ECB's Survey of Professional Forecasters (SPF) for euro area long-term inflation expectations and realized HICP inflation. It also leverages ECB macroeconomic forecasts for real-time policy analysis and counterfactual simulations.
Luca Fornaro, Christoph Große Steffen — International Finance & Macroeconomics
This paper develops a theoretical model showing how monetary unions can experience endogenous financial fragmentation, where identical countries react differently to common shocks due to self-reinforcing loops and animal spirits.
Finance Application
- This framework offers insights into sovereign bond spreads and equity risk premia within monetary unions.
- Expectations of central bank anti-fragmentation policies could be priced into these assets, influencing their correlation and volatility across member states, especially during periods of stress.
- For household finance, the model's 'animal spirits' and capital flight mechanisms could explain how household portfolio choices and consumption decisions are affected by perceived fiscal fragility and fragmentation risk.
- Risk managers and insurers operating across a monetary union could use these insights to better price credit risk or underwrite policies in different member states, considering the potential for asymmetric economic outcomes and the stabilizing role of central bank interventions.
Monetary UnionsFinancial FragmentationCapital FlowsFiscal PolicyCentral BankSovereign DebtExpectationsSelf-Fulfilling CrisesAsset PricingRisk PremiaMacro-Finance
Core finding, identification, data
Core Finding
- The paper demonstrates that monetary unions can experience endogenous financial fragmentation, where identical countries react differently to common shocks, leading to capital flights in some and safe-haven inflows in others.
- Anti-fragmentation policies by the central bank, by counteracting private capital flows with public ones, can mitigate these effects and stabilize the union, often by crowding in private capital flows.
Identification Strategy
- The paper develops a theoretical two-country model of a monetary union with identical fundamentals.
- The key methodological innovation is demonstrating how self-reinforcing loops between capital flows, economic activity, and fiscal policy, driven by 'animal spirits' (expectations), can endogenously break symmetry and lead to financial fragmentation without exogenous asymmetric shocks.
Data
No primary data is used; the paper is purely theoretical, though it references empirical literature on wage rigidity and euro area financial integration.
Francis X. Diebold, Aaron Mora, Minchul Shin — Forecasting & Empirical Methods
This paper analyzes the properties of macroeconomic survey forecast averages, treating them as "portfolios" of forecasts, to understand diversification gains as the number of respondents grows.
Finance Application
- This framework could be applied to financial analyst forecasts for corporate earnings or stock returns, determining the optimal "crowd size" of analysts needed for robust consensus estimates.
- The equicorrelation model and signature plots could also assess diversification benefits in portfolios of expert opinions on asset price movements or credit risk, informing how many independent sources are needed to achieve stable risk assessments in insurance underwriting or investment management.
- This would help identify the point of diminishing returns for gathering more information.
wisdom of crowdsforecast combinationequicorrelationdiversificationportfolio theorysurvey datamacroeconomic forecastinganalyst forecastsrisk aggregation
Core finding, identification, data
Core Finding
- The study finds that an equicorrelation model provides a near-perfect fit for both growth and inflation forecast errors from the U.S.
- Survey of Professional Forecasters.
- Diversification gains are more significant for inflation forecasts but diminish rapidly for both, suggesting that a smaller number of forecasters than currently used might be adequate for optimal forecast accuracy.
Identification Strategy
- The paper's methodological innovation is the use of "crowd size signature plots," which summarize forecasting performance as a function of the number of averaged forecasts.
- An equicorrelation model is estimated by minimizing the divergence between these empirically derived signature plots and their analytical counterparts, using a "matching estimator" approach.
Data
The paper uses quarterly point forecasts for real output growth and GDP deflator inflation from the U.S. Survey of Professional Forecasters (SPF) for the period 1968Q4-2023Q2.
Aakash Kalyani, Serdar Ozkan — Forecasting & Empirical Methods
This paper develops a novel firm-level measure of marginal labor cost pressures using textual analysis of earnings calls, and investigates its pass-through to inflation and its impact on firm investment and productivity.
Finance Application
- This paper's firm-level labor cost pressure measure offers a powerful new variable for asset pricing, potentially explaining cross-sectional stock returns or informing factor models, especially given the observed heterogeneity across industries.
- In corporate finance, the findings on automation and R&D investment in response to labor costs could shed light on capital budgeting decisions, capital structure, and firm valuation, particularly for labor-intensive sectors.
- For risk management, this measure could be used to assess operational risk for firms and industries susceptible to labor market shocks, informing credit risk models or the pricing of business interruption insurance.
textual analysislabor costsinflationfirm-level dataearnings callsasset pricingcorporate financeinvestmentautomationproductivitymacroeconomicsrisk management
Core finding, identification, data
Core Finding
- The paper's novel firm-level labor cost pressure measure, derived from textual analysis, outperforms traditional aggregate indicators in forecasting inflation and shows a significant but heterogeneous pass-through to PPI inflation, highest in the service sector and near-zero in manufacturing.
- Firms respond to higher labor cost pressures by increasing investment and R&D, particularly in routine-manual task-intensive industries, leading to productivity gains through automation.
Identification Strategy
- The authors quantify firm-level labor cost pressures by leveraging a firm's cost minimization problem, projecting changes in intermediate input revenue shares onto the intensity of labor-related discussions from earnings calls.
- To establish causality for inflation pass-through, they use industry-level variation in PPI and Jorda projection specifications with time fixed effects to address simultaneity.
- For firm responses, they employ firm and year fixed effects, industry characteristics, and firm-specific controls to isolate the causal effect of labor cost pressures.
Data
The study uses over 250,000 earnings call transcripts from S&P Global for 6,237 US public companies (2002-2025), firm-level financial data from Compustat, and various aggregate macroeconomic indicators (e.g., ECI, PPI, PCE inflation).
Vito Cormun, Ryan Chahrour, Pierre De Leo, Pablo A. Guerrón-Quintana, Rosen Valchev — International Asset Pricing
This paper demonstrates that noisy news about future Total Factor Productivity (TFP) is a primary driver of U.S. dollar/G7 exchange rate fluctuations and explains several long-standing exchange rate puzzles.
Finance Application
- The methodology of disentangling 'true' anticipated fundamental shocks from 'expectational noise' could be applied to other financial markets to understand how asset prices react to different types of information.
- For instance, identifying noisy news about corporate earnings or inflation could help explain excess volatility and predictability in equity or bond markets.
- This framework could also inform the design of dynamic hedging strategies that differentiate between fundamental and sentiment-driven asset price movements, particularly in currency and fixed income markets, where the paper already shows significant effects on risk premia.
Exchange RatesTFP NewsExpectational NoiseAsset PricingUncovered Interest ParityBackus-Smith PuzzleExcess VolatilityRisk PremiaInternational FinanceMacroeconomicsVAR ModelsNews Shocks
Core finding, identification, data
Core Finding
- The study finds that over half of U.S. dollar/G7 exchange rate fluctuations are explained by variations in expected U.S. productivity, encompassing both correctly-anticipated changes and expectational 'noise'.
- This 'noisy news' is linked to medium-to-long-run TFP growth, leading to significant deviations from uncovered interest parity and generating puzzles such as predictable excess returns, low Backus-Smith correlations, and excess volatility.
Identification Strategy
- The paper employs a two-step identification strategy.
- First, a 'max-share' VAR approach (Uhlig, 2003) isolates a reduced-form shock that explains the largest share of real exchange rate variation.
- Second, using the method of Chahrour and Jurado (2021), two orthogonal disturbances to future U.S.
- TFP expectations are identified: 'technological disturbances' (partially anticipated changes to actual TFP) and 'expectational noise' (expectations that do not materialize in realized TFP), by imposing zero restrictions on the two-sided MA representation of TFP and a signal about future TFP.
Data
The analysis uses quarterly data from 1978:Q4-2008:Q2 (with an extended sample for robustness) including nominal exchange rates and Eurodollar interest rates (Datastream), CPI, real consumption, and investment (OECD), U.S. utilization-adjusted TFP (Fernald, 2012), U.S. R&D and trade balance (FRED), and MSCI equity prices/returns and 10-year government bond yields (Datastream, Global Financial Data).
Stephanie Schmitt-Grohé, Martín Uribe — Workshop on Methods and Applications for Dynamic Equilibrium Models
This paper estimates a dynamic general equilibrium model to disentangle and quantify the contributions of central bank information (CBI) and neo-Fisher effects (NFE) to inflation and output dynamics.
Finance Application
- This research offers several avenues for finance.
- First, understanding whether market reactions to monetary policy surprises are driven by NFE (long-term inflation expectations) or CBI (central bank's private signal about demand) is crucial for pricing fixed-income securities and inflation-linked products.
- Second, the finding that there's no central bank information advantage could be tested in financial markets by examining whether professional forecasters or high-frequency traders can predict central bank actions or economic outcomes better than what public information suggests.
- Finally, the model's ability to generate short-run inflation increases from rate hikes could inform strategies for hedging against unexpected inflation in equity and commodity markets, especially if NFE is the dominant channel.
monetary policycentral bank communicationinflationneo-fisher effectasset pricingfixed incomemacro-financeeconomic shocksinformation asymmetry
Core finding, identification, data
Core Finding
- The study finds that both neo-Fisher effects (driven by permanent monetary shocks) and central bank information (driven by preference shocks to which the central bank responds) are quantitatively important for explaining inflation and output changes.
- Specifically, NFE explains 20-30% of inflation variance, and removing the CBI channel significantly increases the preference shock's contribution to inflation and output variance.
- The paper also finds no evidence of an information advantage for the central bank.
Identification Strategy
- The paper identifies neo-Fisher effects through a permanent monetary shock and central bank information effects through a preference shock to which the central bank directly responds, modeled via a parameter (αξ) in the Taylor rule.
- The model is estimated using postwar U.S. data, and the effects are disentangled by comparing scenarios with and without the CBI channel (αξ=0) and by analyzing impulse responses under full versus imperfect information.
Data
The paper uses postwar U.S. data to estimate its dynamic general equilibrium model.
Qiushi Huang, Leonid Kogan, Dimitris Papanikolaou — International Asset Pricing
This paper documents a robust positive correlation between U.S. innovation and real dollar appreciation, rationalizing it with a general equilibrium model where innovation-driven productivity gains accrue disproportionately to entrepreneurs in incomplete markets.
Finance Application
- The model's mechanism of innovation-driven wealth redistribution and incomplete markets could inform new asset pricing factors, such as an 'innovation risk premium' that captures the exposure of specific asset classes (e.g., tech stocks, growth vs. value) to these displacement shocks.
- For household finance, the finding that innovation increases within-country inequality suggests research into how households, particularly those not directly involved in innovation, adjust their savings and portfolio choices to hedge against this specific form of 'creative destruction' risk.
- In insurance, the concept of unspanned innovation risk could inspire novel products that offer protection against the relative wealth decline for non-innovating segments of the population during periods of rapid technological advancement.
Exchange RatesTechnological InnovationIncomplete MarketsWealth RedistributionIncome InequalityCapital FlowsAsset PricingGrowth StocksValue StocksMacro-FinanceInternational FinanceRisk SharingCreative DestructionUIP Puzzle
Core finding, identification, data
Core Finding
- The paper finds a robust positive correlation between U.S. innovation (measured by patent value to market cap ratio) and real dollar appreciation, with a one-standard-deviation rise in innovation associated with approximately 3 to 4 log points of appreciation per year.
- This is explained by an incomplete markets model where innovation-driven productivity gains increase U.S. wealth, raise marginal utility for U.S. households, and attract foreign capital inflows into innovative U.S. firms, leading to dollar appreciation and increased inequality.
Identification Strategy
- The paper identifies the impact of innovation using annual measures of U.S. innovation (log ratio of total patent value to aggregate stock market capitalization) and a model-implied 'displacement shock' proxy (difference between aggregate market capitalization growth and market portfolio returns).
- They use panel regressions with country fixed effects and Newey-West standard errors to establish robust correlations, and VAR analysis to study impulse responses of key macroeconomic and financial variables to these shocks.
Data
The paper uses data from G-10/G-7 countries, including nominal exchange rates (IMF), CPI, household consumption, GDP, FDI, and portfolio equity flows (World Bank), top 1% income share and net wealth (World Inequality Database), three-month T-bills (Global Financial Data), equity index returns (MSCI from Datastream), patent data (Kogan et al., 2017), firm-level foreign institutional ownership (FactSet Lionshare), and firm fundamentals (Compustat).
Marco Del Negro, Ibrahima Diagne, Keshav Dogra, Pranay Gundam, Donggyu Lee, Brian M. Pacula — Workshop on Methods and Applications for Dynamic Equilibrium Models
This paper uses an estimated Heterogeneous Agent New Keynesian (HANK) model to analyze how the tradeoff between inflation stabilization and consumption volatility varies across different wealth levels in response to monetary policy and macroeconomic shocks.
Finance Application
- This research provides a robust framework for understanding how macroeconomic shocks and monetary policy differentially impact the financial well-being and investment behavior of households across the wealth spectrum.
- In asset pricing, it suggests that the risk premia for assets sensitive to demand shocks (e.g., growth stocks, financial sector equities) might be disproportionately borne by wealthy investors, leading to different pricing dynamics than assets sensitive to supply shocks (e.g., commodities, value stocks).
- For household finance, the findings highlight a need for financial products or insurance mechanisms tailored to protect low-wealth households against real wage erosion and unemployment during supply-side inflation, potentially informing the design of inflation-indexed savings or wage insurance products.
- Additionally, it could motivate research into how wealth-dependent exposure to different shock types influences household portfolio allocation decisions and the demand for various asset classes.
HANK modelsMonetary PolicyInflationInequalityWealth DistributionConsumption VolatilitySupply ShocksDemand ShocksAsset PricingHousehold FinanceLabor Market
Core finding, identification, data
Core Finding
- The rich experience a 'divine coincidence' where stabilizing inflation and consumption volatility go hand-in-hand, primarily because their consumption is driven by procyclical profits sensitive to demand shocks.
- In contrast, the poor face a significant tradeoff, as aggressive inflation stabilization is costly for them, increasing consumption volatility by further reducing real wages and increasing unemployment, especially during supply-side inflationary shocks.
Identification Strategy
- The study employs an estimated medium-scale HANK model, calibrated to match key features of household inequality from the 2007 Survey of Consumer Finances (SCF) and estimated using Bayesian methods with aggregate time-series data (FRED, BEA) from 1992-2019.
- The identification strategy involves simulating the model under varying monetary policy rules (specifically, the coefficient on inflation in a Taylor-type rule, φπ) and decomposing the welfare effects of monetary policy, supply, and demand shocks across different wealth quantiles.
Data
The paper uses micro data from the 2007 Survey of Consumer Finances (SCF) for model calibration, and aggregate time-series data from FRED and BEA (Q1 1992 to Q4 2019) for estimation, covering variables like output, consumption, inflation, interest rates, wages, unemployment, and corporate profits.
Mai C. Dao, Pierre-Olivier Gourinchas — International Asset Pricing
This paper develops a novel 'purified' measure of Covered Interest Parity (CIP) deviations in emerging markets using supranational bonds, free from local credit and liquidity risks, and analyzes its drivers.
Finance Application
- This purified CIP basis offers a cleaner measure of dollar funding stress and intermediation frictions in emerging markets, which could be a powerful new risk factor in EM asset pricing models.
- Researchers could investigate if this purified CIP predicts future returns of EM local currency bonds, currencies, or equities, or if it helps explain the cross-section of EM asset returns.
- It could also be used to evaluate the effectiveness of macroprudential policies or central bank swap lines in mitigating dollar funding pressures for EM firms and governments, thereby impacting their cost of capital and investment decisions.
Covered Interest ParityEmerging MarketsSupranational BondsDollar FundingFinancial FrictionsAsset PricingCurrency MarketsRisk PremiaIntermediation CapacityHedging Demand
Core finding, identification, data
Core Finding
- The 'purified' CIP basis in emerging markets (EMs), derived from supranational bonds, is significantly smaller and less volatile than conventional measures, and its behavior aligns with theoretical predictions.
- It correlates with global dollar funding costs, intermediary balance sheet capacity, and currency-specific dollar hedging demands, unlike naive CIP measures which are confounded by local credit risk.
Identification Strategy
- The paper's methodological innovation is the construction of a 'purified' CIP basis.
- This is achieved by using supranational bonds (e.g., EIB, IBRD) issued in both US dollars and local EM currencies, which are considered credit-risk-free.
- The z-spread method is applied to these bonds, and the resulting CIP deviations are further adjusted for differential liquidity premia using bid-ask spreads and fixed effects, thereby isolating the true arbitrage friction from credit and liquidity risks.
Data
The study utilizes Bloomberg data for supranational bonds (from issuers like EIB, IBRD, KfW) issued in eight major EM currencies and US dollars, along with spot/forward FX rates, interbank rates, and cross-currency basis swaps. It also incorporates data on dollar gaps (external asset/liability positions) from Benetrix et al. (2019) and Allen and Juvenal (2024), the Broad U.S. Dollar Index, and primary dealer leverage ratios from He et al. (2017).
Serdar Birinci, Fatih Karahan, Yusuf Mercan, Kurt See — Workshop on Methods and Applications for Dynamic Equilibrium Models
This paper develops a heterogeneous agent New Keynesian (HANK) model with a frictional labor market and on-the-job search (OJS) to quantitatively study the implications of employer-to-employer (EE) transitions for macroeconomic outcomes and optimal monetary policy.
Finance Application
- The paper's findings on how job mobility (EE flows) and OJS efficiency shocks drive inflation and monetary policy responses could be integrated into asset pricing models to better forecast bond yields and equity risk premia, especially by incorporating labor market tightness as a novel factor.
- For household finance, the heterogeneous agent model, with its detailed income risk and MPC heterogeneity, offers a framework to study how changes in job mobility and wage dynamics influence household saving, borrowing, and investment decisions across different wealth levels.
- In insurance, the endogenous income risk and human capital evolution could inform the design and pricing of new products like 'job mobility insurance' that protect against wage losses or facilitate transitions, going beyond traditional unemployment benefits.
HANK modellabor marketjob mobilityemployer-to-employer flowson-the-job searchmonetary policyinflationwage dynamicsheterogeneous agentsMPC heterogeneityasset pricinghousehold financeincome riskmacro-financeinsurance
Core finding, identification, data
Core Finding
- Employer-to-employer (EE) dynamics significantly influence inflation, with muted job mobility causing 0.42 percentage points lower inflation during the 2016–2019 recovery and elevated EE rates contributing 0.77 percentage points higher inflation during the 2021-2022 'Great Resignation'.
- Optimal monetary policy should respond strongly and positively to EE fluctuations, differentiating between episodes with similar unemployment but distinct EE dynamics.
- Crucially, accounting for market incompleteness (heterogeneous agents) substantially alters both macroeconomic outcomes and optimal monetary policy prescriptions regarding EE transitions.
Identification Strategy
- The paper uses a quantitative HANK model with a frictional labor market and on-the-job search, estimated by targeting empirical correlations and standard deviations of unemployment, EE rate, and inflation with output.
- It employs the sequence-space Jacobian (SSJ) method to solve and simulate the model under aggregate uncertainty, handling discretized worker distributions.
- Counterfactual simulations (e.g., fixing EE rates) and a decomposition exercise using a Directed Acyclic Graph (DAG) representation are used to quantify the impact and channels of OJS efficiency shocks on inflation during historical episodes.
Data
The study uses U.S. economy data from 1995-2024, including the Current Population Survey (CPS) for unemployment, Longitudinal Employer-Household Dynamics (LEHD) for job-switching rates and wage dynamics, BLS for PCE price index and Unit Labor Cost (ULC), U.S. Bureau of Economic Analysis for real GDP, JOLTS for vacancies, and the Survey of Income and Program Participation (SIPP) for retirement income.
Gadi Barlevy, Ines Xavier — Macro, Money and Financial Frictions
This paper develops a theoretical model of Ponzi schemes with asymmetric information to understand how they occur, what prevents them, and their welfare implications, focusing on the role of reputation and investor beliefs.
Finance Application
- The model's insights into how investor skepticism, reputation building, and detection probability affect the sustainability of fraudulent schemes could be applied to understanding asset bubbles or market manipulation, especially in less regulated or nascent markets like cryptocurrencies.
- In household finance, it offers a framework to study investor susceptibility to scams, the role of trust in investment decisions, and the comparative effectiveness of financial literacy versus regulatory oversight.
- For insurance, the dynamics of building trust through initial payouts before a scheme collapses could inform models of insurance fraud or fraudulent investment products.
Ponzi schemesasymmetric informationreputationfraudinvestor behaviorfinancial crimemarket manipulationgame theoryfinancial regulationhousehold finance
Core finding, identification, data
Core Finding
- Ponzi frauds can occur in equilibrium when investors are initially skeptical but there is a low probability of detection.
- The scheme operator builds trust over time by repaying early investors with funds from new investors, and the scheme's duration is inversely related to the promised return.
- The model suggests that enforcement is more effective than education in deterring Ponzi schemes, and that Ponzi equilibria can ex ante Pareto dominate no-trade equilibria, not because they are welfare-improving, but because they exploit the surplus created by genuine trade.
Identification Strategy
- This is a theoretical model, so it does not employ an empirical identification strategy.
- The methodological innovation lies in applying an asymmetric information framework to model Ponzi schemes, where the long-lived agent's type (genuine or imposter) is unknown to short-lived investors, and the imposter builds reputation over time by mimicking a genuine type.
Data
The paper is theoretical and does not use primary data. It references existing empirical literature and historical accounts of Ponzi schemes to motivate its framework and discuss the consistency of its predictions with observed patterns.
Tobias Adrian, Christopher Erceg, Marcin Kolasa, Jesper Lindé, Pawel Zabczyk — Macro, Money and Financial Frictions
This paper uses a DSGE model with segmented assets, behavioral discounting, and a nonlinear Phillips Curve to assess the macroeconomic and fiscal consequences of quantitative easing (QE) under different economic conditions and levels of uncertainty.
Finance Application
- The paper's detailed analysis of how QE affects term premiums, bond prices, and central bank profits/losses under various scenarios (deep vs. shallow liquidity traps, commitment, uncertainty) offers direct insights for asset pricing models, particularly for fixed income.
- This could inform the modeling of sovereign credit risk and the pricing of long-duration government bonds, especially concerning central bank balance sheet risks and duration exposure.
- For insurance companies, which hold significant long-term assets, the findings on interest rate risk and central bank losses in 'faster recovery' scenarios are crucial for assessing solvency and managing asset-liability mismatches under different monetary policy regimes.
Monetary PolicyQuantitative EasingCentral Bank Balance SheetGovernment DebtTerm PremiumLiquidity TrapDSGE ModelFiscal PolicyInflationMacroeconomicsBond MarketsRisk Management
Core finding, identification, data
Core Finding
- QE provides substantial macroeconomic benefits and improves the consolidated fiscal position in a deep liquidity trap, often with central bank profits.
- However, in a shallow liquidity trap, QE benefits are smaller, and there's a greater risk of overheating and central bank losses, especially with low initial term premiums, large QE size, or commitment-based forward guidance.
- QE is found to have significantly lower fiscal costs than conventional fiscal stimulus for a given output boost.
Identification Strategy
- The paper employs a DSGE model featuring bond market segmentation to allow QE to affect term premiums, behavioral discounting to mitigate the 'forward guidance puzzle,' and a nonlinear Phillips Curve to capture overheating risks.
- The model is calibrated to match empirical evidence on monetary transmission in the U.S. and Euro area, and stochastic simulations use shocks calibrated to match historical U.S. data volatility (1960-2019) to assess risks under uncertainty.
Data
The model's parameters are calibrated using empirical evidence on monetary transmission from the U.S. and Euro area, and targeted steady-state ratios are based on U.S. data. Stochastic simulations calibrate shock volatilities to match unconditional standard deviations and correlations of output growth, PCE inflation, nominal wage growth, and hours worked per capita from U.S. data for the 1960-2019 period.
Thierry Mayer, Isabelle Mejean, Mathias Thoenig — International Economics and Geopolitics
This paper develops a quantitative trade model embedding a diplomatic game of escalation to conflict to analyze how trade dependencies influence geopolitical disputes and the optimal degree of 'decoupling' strategies.
Finance Application
- This framework could be applied to asset pricing by modeling how geopolitical risk, particularly the 'fragmentation paradox,' impacts the equity risk premium for firms with varying supply chain exposures to rival nations.
- In household finance, it could inform how perceived changes in 'global safety' influence household portfolio decisions, such as allocations to domestic versus international assets or demand for hedging instruments.
- For insurance, the quantification of 'geoeconomic loss' and 'true cost of war' could be used to price political risk insurance or supply chain disruption policies, adjusting premiums based on trade policy changes and their effect on conflict probabilities.
GeopoliticsInternational TradeSupply ChainsConflict RiskTariffsDecouplingQuantitative ModelsAsset PricingHousehold FinancePolitical Risk InsuranceRisk Management
Core finding, identification, data
Core Finding
- The growing U.S. dependence on Chinese products over the past thirty years has increased the cost of geopolitical disputes with China for the US.
- While decoupling through tariffs may offer geopolitical benefits by reducing the cost of war and improving bargaining power, it can paradoxically raise the risk of escalation by weakening incentives for restraint, a phenomenon termed the 'fragmentation paradox'.
- The optimal US tariff on Chinese imports is 8% when global safety is high, but increases as global safety deteriorates.
Identification Strategy
- The paper's methodological innovation is embedding a diplomatic game of escalation, where leaders have private information about war costs, into a quantitative general equilibrium trade model with input-output linkages.
- This allows for the estimation of welfare-relevant geoeconomic factors (opportunity cost of war, peace-keeping costs, probability of de-escalation) and their evolution under various trade policies.
- The model is solved using exact hat algebra and calibrated to current data, with war damages and trade disruptions calibrated from historical large interstate wars.
Data
The model is calibrated using data from the Trade in Value Added Database (release 2023, 1995-2023) by the OECD. Elasticities are sourced from Hertel et al. (2007) and Ahmad and Schreiber (2024). War damage parameters, including productivity contraction, are based on estimates from Federle et al. (2024) using 150 years of large interstate war data, and trade disruption parameters are from Glick and Taylor (2010) based on WWI and WWII.
Sebastian A. Merkel, Ziang Li — Macro, Money and Financial Frictions
This paper develops a New Keynesian model with incomplete markets and nominal safe assets to show how uncertainty-induced flight to safety, combined with sticky prices, causes aggregate demand recessions and limits the effectiveness of conventional monetary policy.
Finance Application
- This paper's insights are highly relevant to asset pricing, particularly in understanding bond and equity market dynamics during periods of uncertainty.
- The mechanism of capital price overshooting due to sticky nominal safe asset values could explain excess equity market volatility and "liquidity spirals" during crises.
- It also highlights the importance of nominal safe asset supply and its interaction with monetary policy for asset valuations, suggesting that central bank actions on long-term bond purchases (QE) might be more effective through their impact on safe asset supply and duration rather than just interest rates.
- This could inform models of flight-to-liquidity and the pricing of safe assets.
New KeynesianFlight to SafetyUncertainty ShocksMonetary PolicySticky PricesSafe AssetsAsset PricingCapital Price OvershootingBond MarketsMacro-FinancePortfolio Choice
Core finding, identification, data
Core Finding
- A rise in uncertainty triggers a flight to safety, where investors reallocate portfolios from risky capital to nominal government bonds.
- Under sticky goods prices, the real value of these nominal safe assets cannot adjust flexibly, leading to capital price overshooting and aggregate demand recessions.
- Conventional monetary policy, acting through nominal interest rates, is largely ineffective in mitigating these recessions because it does not directly influence the portfolio reallocation between safe and risky assets.
Identification Strategy
- The paper uses a theoretical, analytically tractable continuous-time New Keynesian model augmented with incomplete markets and nominal safe assets.
- The key "shock" is an exogenous increase in time-varying idiosyncratic risk (uncertainty), which drives the "flight to safety" mechanism.
- The methodological innovation is combining the safe asset framework of Brunnermeier, Merkel and Sannikov (2024) with a standard New Keynesian supply side, allowing for analytical derivation of results without relying on numerical methods for core findings.
Data
The paper does not use empirical data for its main analysis. It relies on a numerical illustration for impulse response functions, with parameters calibrated to follow prior literature (Brunnermeier et al., 2024 for household side/volatility, Kaplan et al., 2018 for New Keynesian supply side).
Jaedo Choi, George Cui, Younghun Shim, Yongseok Shin — Economic Growth
This paper investigates how US multinational joint ventures in China facilitate technology transfer, intensify global competition, and ultimately impact US and Chinese welfare.
Finance Application
- The findings on how technology transfer and increased competition affect firm profitability, innovation, and market share could inform asset pricing models by identifying new risk factors related to global supply chain and intellectual property exposure.
- For corporate finance, the 'over-investment' insight suggests that firms may misallocate capital in JVs, leading to suboptimal long-term value creation, which could be analyzed through M&A performance or capital expenditure studies.
- In household finance, the negative impact on US real wages could explain regional disparities in household wealth accumulation or consumption patterns, particularly in sectors highly exposed to Chinese competition.
Technology TransferJoint VenturesGlobal CompetitionFirm PerformanceInnovationWelfareTradeFDIAsset PricingCorporate FinanceHousehold FinanceIntellectual Property RiskCross-sectional Returns
Core finding, identification, data
Core Finding
- Empirically, the paper establishes three facts: Chinese parent firms of joint ventures (JVs) grow larger, export more, and become technologically similar to their US partners; non-participating Chinese firms in industries with more JVs also grow larger and more technologically advanced; and US firms in these industries experience negative impacts on their size, exports, and innovation.
- A quantitative model shows that US leaders over-invest in JVs, leading to short-run gains but long-run losses for US welfare due to rising Chinese competition, while banning JVs would increase US welfare by 1.2% but reduce China's by 10.6%.
Identification Strategy
- For direct effects on Chinese partners, an event study design with propensity score matching is used, comparing firms that formed JVs to a control group.
- For indirect spillovers and competition effects, an instrumental variable (IV) strategy is employed, using joint venture investments from Japan and South Korea to India as an instrument.
- The identifying assumption is that the factors driving these investments are unrelated to those influencing US-China joint ventures.
Data
The study uses merged firm-level data for Chinese firms (National Bureau of Statistics balance sheets, patent data, Orbis ownership, Chinese Customs Trade Statistics) and US firms (Compustat, USPTO patent data). Sectoral data from NBER-CES and Comtrade, along with Indian firm data from Prowess and Orbis, are also utilized.
Andrea Cerrato, Francesco Filippucci — Economic Growth
This paper quantifies the long-term macroeconomic and welfare effects of Italy's large-scale regional development program (CasMez, 1950-1992) on its Southern regions, accounting for factor mobility and agglomeration economies.
Finance Application
- The findings on regional productivity, capital mobility, and crowding-out effects offer insights for asset pricing by suggesting that regional development policies could create differential growth opportunities and risk exposures for firms, potentially leading to regional asset pricing factors or influencing the cost of capital for firms with concentrated operations.
- For household finance, the welfare and labor mobility effects could inform models of regional household wealth accumulation, consumption patterns, and demand for financial services.
- In insurance, understanding how these policies alter regional economic stability and employment risks could help price regional-specific insurance products or assess demand for unemployment and business interruption insurance.
Regional DevelopmentIndustrial PolicyPublic CapitalAgglomeration EconomiesFactor MobilityCrowding-OutWelfare AnalysisCost-Benefit AnalysisEconomic GeographyMacroeconomicsGrowth ModelCapital Allocation
Core finding, identification, data
Core Finding
- The program boosted Southern Italy's value added by 19% (2011), leading to a 1.6% increase in national GDP.
- However, it caused a 2.7% decrease in value added in non-targeted regions due to crowding-out effects from factor reallocation, effectively halving the aggregate benefits.
- The benefit-to-cost ratio (BCR) was 1.34, indicating cost-effectiveness, but a counterfactual place-neutral program would have yielded larger welfare gains.
Identification Strategy
- The paper employs two identification strategies: a difference-in-differences (DiD) approach comparing municipalities within Industrial Development Areas (IDAs) to matched non-IDA Southern municipalities, and a long difference-in-discontinuities design exploiting the sharp border of CasMez's jurisdiction.
- These empirical estimates are then used to calibrate a two-region growth model featuring public capital, factor mobility, and agglomeration economies to quantify macroeconomic effects.
Data
The study uses administrative data from historical archives (Archivi dello Sviluppo Economico Territoriale - ASET) for CasMez-financed projects, decennial population Censuses for municipality-level demographic and labor market outcomes, province-to-province migration flows, and Istituto Tagliacarne data for province-level value added.
Shoumitro Chatterjee, Elisa Giannone, Tatjana Kleineberg, Kan Kuno, Luca Looser — Economic Growth
This paper examines how structural transformation, particularly the shift towards high-skill services, affects regional convergence and spatial inequality across a novel global dataset of subnational economic activity.
Finance Application
- The increasing spatial concentration of high-skill services and the resulting regional divergence could introduce new regional risk factors for asset pricing models, particularly for real estate, local equity, and private credit markets.
- For household finance, this trend implies diverging wealth accumulation and financial stability across regions, impacting mortgage markets, consumer lending, and demand for financial advice.
- Insurers could face heightened systemic risk from concentrated economic activity in high-skill hubs, necessitating new approaches to underwriting property, business interruption, and unemployment insurance for these specialized regional economies.
Regional economicsStructural transformationSpatial inequalityEconomic convergenceHigh-skill servicesAgglomeration effectsEconomic geographyAsset pricingHousehold financeInsuranceReal estateLocal economiesRisk factors
Core finding, identification, data
Core Finding
- Regional income convergence within countries has significantly slowed and stalled over time, a trend strongly associated with structural transformation towards high-skill private services.
- Employment in these high-skill services is substantially more geographically concentrated than in other sectors, suggesting that agglomeration effects in this sector, combined with non-homothetic preferences, create a self-reinforcing interplay between regional inequality, structural transformation, and aggregate growth, implying a trade-off between spatial equality and faster economic growth.
Identification Strategy
- The paper primarily establishes three novel empirical facts through descriptive analysis and regressions, controlling for country-specific heterogeneity and year fixed effects.
- It then uses a spatial equilibrium model with non-homothetic preferences and agglomeration effects to rationalize these facts and perform counterfactual simulations (e.g., eliminating agglomeration effects or income effects) to identify the underlying mechanisms driving the observed trends in regional convergence and structural transformation.
Data
The paper constructs a novel global dataset of regional GDPs and granular sectoral employment for over 1500 regions in more than 90 countries (1980-2019). This dataset integrates sources like Gennaioli et al. (2014), IPUMS, World Bank Global Labor Database, ARDECO, The Economist Intelligence Unit (city-level data), GGDC, ETD, and includes indicators such as years of schooling, trade agreements, road networks, and nightlight data for robustness.
Kristian Blickle, Xu Lu, Jian Li, Yiming Ma — Asset Pricing
This paper analyzes the time-varying nature of deposit flightiness, its determinants (like QE and interest rates), and its implications for financial stability, rationalizing these trends with a model featuring heterogeneous investors and switching costs.
Finance Application
- The concept of 'deposit flightiness' and its link to monetary policy could be integrated into asset pricing models to explain variations in bank stock returns or the pricing of bank-issued debt, potentially revealing a 'flightiness premium' or discount.
- In household finance, the granular depositor-level data and heterogeneity could inform models of household liquidity management and how different segments adjust their portfolios in response to interest rate changes and perceived bank stability.
- For insurance research, the amplified run risk due to flightier deposits and drastic rate hikes could be used to stress-test deposit insurance schemes (e.g., FDIC) and design more dynamic, risk-sensitive premiums or capital requirements for banks.
bankingfinancial stabilitymonetary policyquantitative easinginterest ratesdeposit flightinessbank runsheterogeneous investorsasset pricinghousehold financerisk management
Core finding, identification, data
Core Finding
- Empirically, deposit flightiness, measured by deposit flow sensitivity to rates, displays pronounced fluctuations over time, reaching unprecedentedly high levels after the Covid-19 crisis, coinciding with low interest rate environments and central bank reserve expansions.
- Theoretically, a model with heterogeneous investors and switching costs rationalizes this by showing that large deposit inflows (e.g., from QE) attract less convenience-valuing, more rate-sensitive depositors, making the marginal depositor flightier and increasing run risk, especially with drastic rate hikes.
Identification Strategy
- To measure deposit flightiness, the paper estimates the sensitivity of bank-level deposit flows to deposit rates using rolling window regressions.
- Deposit rates are instrumented using supply-side instruments, specifically per unit asset fixed costs and salary expenses, assuming these affect deposit rates through the cost of producing deposits rather than depositors' demand.
Data
The paper uses Call Reports data for bank-level characteristics, regulatory data from the Complex Institution Liquidity Monitoring Report (FR2052) for deposits by counterparty and account type, and novel transaction-level data from a financial data processor for over 1,400 U.S. depository institutions (2015-2022) to track fund movements. Aggregate data like the Fed funds rate and outstanding reserves are sourced from FRED.
Matthew Turner, David N. Weil — Economic Growth
This paper quantitatively assesses the contribution of big cities to US economic growth by examining static agglomeration effects on production and dynamic effects on invention productivity, using counterfactual scenarios of limited city sizes.
Finance Application
- The paper's quantification of agglomeration effects (σA for output, σB for patents) provides novel city-level economic growth and innovation metrics.
- In asset pricing, these parameters could be used to construct 'urban growth factors' to explain cross-sectional variation in local real estate returns or regional equity valuations.
- For household finance, the differential income growth and human capital accumulation in larger cities could explain variations in household savings rates, portfolio allocations, or mortgage default probabilities across metropolitan areas.
- Insurers could leverage these metrics to better price commercial property and casualty insurance, accounting for the concentrated value-at-risk and innovation-driven obsolescence in highly agglomerated urban centers.
Urban economicsEconomic growthAgglomeration economiesProductivityInnovationPatentsCity sizeCounterfactual analysisHuman capitalTFP
Core finding, identification, data
Core Finding
The effects of restricting city sizes (e.g., to one million people since 1900) on US output by 2010 are surprisingly small, resulting in only an 8% lower output than the observed value under baseline parameters, suggesting urban scale effects are not the primary engine of economic growth.
Identification Strategy
- The paper employs a counterfactual analysis, similar to Fogel's (1964) work on railroads, to estimate the impact of hypothetical city size limitations.
- It combines existing literature estimates of static agglomeration effects on output (σA) and dynamic effects on patent production (σB) with MSA-level patent and population data since 1900, integrating these into a simple growth model to simulate aggregate output and technological progress under various city size caps.
Data
The study uses MSA-level patent and population data since 1900 (from Forstall and NBER 1995, and the CUSP patent database by Berkes 2018) and county-level output data from the BEA (US-DOC/BEA/RD 2023).
Craig A. Chikis, Benny Kleinman, Marta Prato — Economic Growth
This paper develops an endogenous growth model with multi-market innovative firms and local knowledge spillovers, empirically estimates it for the U.S., and evaluates policies for spatial expansion.
Finance Application
- This research offers several avenues for finance.
- Asset pricing could investigate whether firms with broader geographic R&D scope or those expanding into new markets exhibit different risk-adjusted returns, potentially reflecting mispricing of knowledge spillover benefits.
- The regional inequality aspect suggests that investors might find alpha by identifying under-invested regions with high innovation potential, or by analyzing how regional policy changes (like expansion subsidies) affect local firm valuations and credit risk.
- Furthermore, the model's parameters (e.g., β, η) could be used to construct novel firm-level or regional innovation metrics that predict future growth or investment opportunities.
InnovationFirm GeographyKnowledge SpilloversEconomic GrowthR&D PolicyRegional DisparitiesAsset PricingCorporate FinanceRegional EconomicsPatent DataEvent StudyInstrumental Variables
Core finding, identification, data
Core Finding
- The paper finds that U.S. innovative firms operate in too few markets relative to the social optimum due to uninternalized knowledge spillovers, which increase with firm size but with diminishing returns.
- Policies promoting broader spatial scope yield larger welfare gains and can reduce regional inequality compared to traditional R&D subsidies.
Identification Strategy
- The empirical analysis uses an event-study design, leveraging the staggered expansion of firms into new markets.
- It employs a within-firm matching approach to compare citations in treated versus non-treated commuting zones for the same firm.
- For estimating spillover and innovation elasticities (α and η), an instrumental variable strategy using an immigration instrument (Burchardi et al., 2020) is used to address endogeneity and measurement error.
Data
The paper uses firm-level data from Dun & Bradstreet for R&D facility locations and cross-firm patent citations from USPTO's PatentsView. It also incorporates county-level economic activity data from Business Dynamics Statistics and County Business Patterns, and LMA-level data for estimation.
Niels Joachim Gormsen, Eben Lazarus — Asset Pricing
This paper decomposes real interest rate changes into pure discounting, expected growth, and uncertainty components, and analyzes their differential impact on equity valuations.
Finance Application
- The paper's 'pure discounting' term provides a theoretically sound counterfactual for a duration-matched risk-free asset, which could be used to re-evaluate the equity premium puzzle for other asset classes like real estate or private equity.
- The framework can also be applied in household finance to quantify how much of observed wealth inequality is due to 'true' economic gains versus mechanical interest rate effects.
- Furthermore, the decomposition can help analyze how monetary policy shocks transmit to different asset classes and investor behaviors by identifying the active channels (pure discounting, growth, or uncertainty).
asset pricinginterest ratesequity valuationdurationrisk premiamonetary policywealth inequalitymacro-financestructural decompositionforecastingcross-section of stocks
Core finding, identification, data
Core Finding
- The paper finds that only the 'pure discounting' component of real interest rates transmits one-for-one to equity valuations, explaining over 80% of cross-country changes in stock valuations since 1990.
- Growth and risk shocks have little or negative transmission to equities, highlighting a 'stock-yield disconnect' that is resolved by isolating the pure discounting term.
Identification Strategy
- The methodological innovation is a theoretical decomposition of real interest rates into three structural components: pure discounting (time preference), expected growth, and uncertainty (risk/entropy).
- Empirically, these components are estimated using long-term professional forecasts (for real rates and growth) and option-implied volatility (for uncertainty), backing out the pure discounting term as a residual.
- This decomposition allows them to isolate the specific drivers of interest rate changes and their distinct effects on equity.
Data
The paper uses an international panel of long-term professional forecasts for interest rates, inflation, and growth rates from Consensus Economics (1990-2023 for G7, later for others). It also uses option prices from OptionMetrics to proxy for uncertainty (VIX squared) and value-weighted equity prices/earnings from CRSP and Compustat.
Aniket Baksy, Daniele Caratelli, Niklas Engbom — Economic Fluctuations and Growth Program Meeting
This paper documents a significant long-term decline in upward job mobility in the U.S. labor market, primarily driven by increased mismatch, greater employer concentration, and reduced search intensity by employed workers, leading to substantial real wage stagnation.
Finance Application
- The documented decline in upward job mobility and wage stagnation has direct implications for household finance, affecting savings rates, debt accumulation, and retirement planning, as households face lower and more uncertain income growth.
- In asset pricing, reduced labor market fluidity and wage growth could impact aggregate consumption, influencing consumption-based asset pricing models, and potentially creating cross-sectional differences in equity returns for firms in industries with varying employer concentration or labor market mismatch.
- For insurance, increased income risk due to stagnant wages and reduced mobility could drive demand for income protection products like unemployment or disability insurance, requiring insurers to re-evaluate pricing models based on these structural labor market changes.
labor economicsjob mobilitywage stagnationemployer concentrationlabor market mismatchhousehold incomeconsumptionasset pricingrisk managementunemployment insurancestructural model
Core finding, identification, data
Core Finding
- Upward job mobility in the U.S. has fallen by approximately 40-60% between the 1980s and 2010s.
- This decline is attributed to three structural factors: greater mismatch between open jobs and searching workers, increased employer concentration limiting job shopping, and less active job search by employed individuals.
- The combined effect of these changes is a 4 percentage point lower real wage, accounting for about 40% of the fall in the aggregate labor share.
Identification Strategy
- The paper uses a structural job ladder model with two versions: a stylized partial equilibrium model and a richer full model.
- The full model identifies structural factors through a three-step estimation process: Step I directly estimates job finding rates, vacancy rates, and a matching wedge from data; Step II estimates eight key parameters (including job-to-job offer rates, wage dynamics, and separation rates) by matching moments of wage and offer distributions, and stayers/losers distributions; Step III identifies employer concentration by analyzing the covariation between job mobility and firm size in a panel of U.S. states.
Data
The paper uses cross-sectional data to estimate wage and offer distributions, and a panel of U.S. states to estimate employer concentration. It also uses the National Longitudinal Survey of Youth (NLSY) for validation and to provide supporting evidence on individual wage and employment dynamics.
Lautaro Chittaro, Monika Piazzesi, Marcelo J. Sena, Martin Schneider — Economic Fluctuations and Growth Program Meeting
This paper establishes a theoretical equivalence between carbon taxes and firm-specific return schedules based on emission intensity, and quantitatively assesses the effectiveness of different green investor preferences on emission reductions in a general equilibrium model.
Finance Application
- This paper offers a powerful framework for asset pricing to theoretically benchmark and interpret observed ESG or pollution premia, linking them directly to carbon pricing.
- For household finance and asset management, the heterogeneous investor model provides a quantitative tool to evaluate the real-world impact of different sustainable investing strategies, such as 'negative screening' (nonpecuniary costs) versus 'impact investing' (nonpecuniary benefits).
- It suggests that strategies focused on subsidizing clean capital in key sectors (like electricity) may be more effective in driving emission reductions than those punishing dirty assets, especially in markets with diverse investor preferences.
- This could inform the design of ESG funds and institutional investor mandates aiming for tangible environmental outcomes.
Carbon TaxESG InvestingAsset PricingHeterogeneous InvestorsPollution PremiaClimate FinanceGeneral EquilibriumCorporate FinanceSustainable InvestingEnergy Economics
Core finding, identification, data
Core Finding
- The paper demonstrates that a carbon tax on energy users provides the same incentives as a replicating return schedule tied to firms' emission intensities.
- Analyzing pollution premia through this lens, it concludes that current market incentives for emission reduction are modest.
- While replicating a serious carbon tax requires high returns for highly polluting firms, this is unsustainable with heterogeneous investors unless nearly all perceive large nonpecuniary costs.
- However, substantial emission reductions can be achieved if even a small share of investors perceive nonpecuniary benefits from owning clean electricity capital, effectively subsidizing green assets.
Identification Strategy
- The core methodological innovation is establishing a theoretical equivalence between a carbon tax and a capital tax schedule (or investor preferences) that depends linearly on a firm's emission intensity (scope 1 emissions relative to enterprise value).
- This equivalence is then applied within a multisector general equilibrium growth model, calibrated with sector-level and firm-level data, to quantitatively assess the impact of different carbon tax rates and heterogeneous investor preferences (nonpecuniary costs vs. benefits, with short-sale constraints) on equilibrium returns and emission reductions.
Data
The paper primarily uses the GTAP 11 database (Global Trade Analysis Project) for sector-level costs and emissions data. For firm-level analysis, it integrates S&P Global Trucost data on electricity production and emissions by fuel type with Compustat Fundamentals for enterprise value. Depreciation rates are sourced from the Fixed Asset Tables of the Bureau of Economic Analysis.
Simon Mongey, Michael E. Waugh — Economic Fluctuations and Growth Program Meeting
This paper develops a dynamic general equilibrium model where household wealth endogenously determines price elasticities of demand, leading to sorting of households across firms and differential markups.
Finance Application
- The finding that household wealth significantly impacts firm markups and aggregate inflation, especially in response to fiscal transfers, has direct implications for asset pricing.
- Wealth shocks (e.g., from housing or equity market booms/busts) could propagate to corporate profitability and equity valuations, suggesting a need for investors to consider wealth distribution effects when forecasting earnings or inflation.
- In household finance, the model's mechanism where wealthier households are less price-sensitive and sort into higher-markup goods could explain consumption patterns and demand for luxury assets or premium financial services, informing models of consumption-portfolio decisions under heterogeneous price elasticities.
MacroeconomicsIndustrial OrganizationHousehold HeterogeneityMarkupsPrice ElasticityWealthFiscal PolicyConsumptionFirm PricingGeneral EquilibriumAsset Pricing ImplicationsHousehold Finance
Core finding, identification, data
Core Finding
- The model finds that household heterogeneity is the dominant source of markup variation across firms, accounting for 58.5% of the variation.
- Furthermore, a one-time fiscal transfer of one percent of GDP to households leads to a 0.3 percentage point increase in the aggregate markup, driven by reduced household price sensitivity and re-sorting across firms.
Identification Strategy
- The model's parameters are calibrated by replicating key empirical relationships from existing literature.
- This includes matching the negative relationship between household income and demand elasticities (Auer et al., 2024), household sorting into expensive varieties (Jaimovich et al., 2019), and the cross-sectional relationship between markups and sales shares (Edmond et al., 2023).
- The fiscal transfer shock experiment mimics wealth-induced demand shocks by varying homeownership rates and house price changes across simulated regions, consistent with Stroebel and Vavra (2019).
Data
The paper uses various microdata sources for calibration and validation, including PSID data for income processes, Survey of Consumer Finances for household assets, NAICS 5 industry-level data for firm concentration, Swiss microdata on household purchases (Auer et al., 2024), consumer packaged goods data from the Kilts-Nielsen dataset (Jaimovich et al., 2019), and data from Stroebel and Vavra (2019) on homeownership rates and house price changes.
Fernando E. Alvarez, Francisco J. Buera, Nicholas Trachter — Macroeconomics Within and Across Borders
This paper develops a dynamic general equilibrium model to analyze inefficiencies in technology adoption and design optimal industrial policies, particularly in the presence of complementarities.
Finance Application
- The concept of 'technology adoption traps' and 'Big Push' policies could be applied to asset pricing to explain long-term regime shifts in equity market returns or persistent cross-sectional differences in firm valuations based on their technological dynamism.
- For corporate finance, the fixed costs and non-convexities of technology adoption directly relate to real options theory, where industrial policy acts as a subsidy to the option value of investing in new technologies, influencing capital expenditure and M&A activity.
- In credit markets, economies or sectors stuck in low-adoption traps could exhibit higher systemic risk and default probabilities, impacting bond spreads and bank lending decisions.
Technology AdoptionIndustrial PolicyEconomic GrowthGeneral EquilibriumMarket InefficienciesMultiple EquilibriaAsset PricingCorporate InvestmentReal OptionsCapital Allocation
Core finding, identification, data
Core Finding
- The paper identifies both static (misallocation of intermediate goods) and dynamic (firms undervalue technology adoption due to fixed costs and market power) inefficiencies in a dynamic economy.
- Optimal industrial policy involves subsidies to correct these, and in the presence of strong complementarities, can lead to multiple equilibria, including 'traps' where economies get stuck in low-adoption states.
- Policy can facilitate a 'Big Push' out of these traps.
Identification Strategy
- The paper is theoretical, building a dynamic general equilibrium model with monopolistic competition, fixed costs for technology adoption (non-convexities), market power, and firm heterogeneity.
- The methodological innovation lies in integrating these elements to analyze optimal industrial policy and the conditions for multiple equilibria in technology adoption, using both a 'growing frontier' (vintage capital) and 'static frontier' (neoclassical growth) framework.
Data
This paper is purely theoretical and does not use any empirical data.
John C. Haltiwanger, Ron S. Jarmin, Robert Rodriguez, Matthew D. Shapiro — Conference on Research in Income and Wealth
This paper develops a quality-adjusted unit value index (QUVI) to decompose observed price changes into pure inflation, quality improvements, and product mix effects, addressing the limitations of standard unit value indices.
Finance Application
- This framework offers a powerful tool for financial economists to construct more accurate inflation measures for specific asset classes or sectors where product quality evolves rapidly, such as technology, healthcare, or even real estate.
- For asset pricing, quality-adjusted price indices for the underlying goods and services produced by publicly traded firms could refine revenue growth forecasts and improve valuation models, especially for growth stocks.
- In household finance, a better understanding of quality-adjusted consumption costs could lead to more precise estimates of real household wealth and consumption patterns.
- For insurance, this method could enhance the accuracy of replacement cost estimation for rapidly depreciating or technologically advancing assets, leading to more robust underwriting and claims management.
inflationquality adjustmenthedonic pricingprice indicesmarket frictionsproduct turnoverbig dataconsumer techeconomic measurement
Core finding, identification, data
Core Finding
- The paper demonstrates that standard Unit Value Indices (UVIs) significantly overstate inflation by conflating pure price changes with improvements in product quality and shifts in product mix.
- By constructing Quality-Adjusted Unit Value Indices (QUVIs) using hedonic methods for notebook computers, the authors show that actual price growth is substantially lower, with quality improvements and product mix changes accounting for a large positive component that masks underlying price declines.
- For example, UVI shows 0.51% quarterly growth, while QUVI shows -2.74% after quality adjustment.
Identification Strategy
- The methodological innovation involves developing a framework for Quality-Adjusted Unit Value Indices (QUVIs) that explicitly incorporates hedonic adjustments for product characteristics.
- It proposes a bounding approach using hybrid hedonics to account for product entry and exit, and an exact price index approach that introduces "market frictions" (z_kt) to relax the strict quality-adjusted law of one price.
- Acknowledging an inherent identification problem in separating quality (λ) from market frictions (z), the paper explores polar cases and references prior work (Byrne et al., 2017) that addresses this through the life-cycle evolution of price dispersion.
Data
The paper uses Point of Sales (POS) data from Circana (formerly NPD Group) covering consumer tech goods, specifically notebook computers, from 2017Q1 to 2020Q4. The data is at the item (SKU) level and includes detailed product attributes such as battery life, disk space, RAM, and screen size, which are crucial for hedonic quality adjustments.
Guido Menzio, Saverio Spinella — Micro Data and Macro Models
This paper develops a general equilibrium incomplete-market model with search frictions in financial markets and endogenous financial human capital to explain persistent heterogeneity in individual rates of return on wealth.
Finance Application
- This framework offers a powerful lens for household finance, directly linking financial literacy (as financial human capital) to investment outcomes and wealth accumulation, suggesting targeted financial education policies could mitigate wealth inequality.
- In asset pricing, the search friction mechanism could explain persistent cross-sectional return differences for seemingly similar assets in less liquid or over-the-counter markets, where investor sophistication (financial human capital) dictates access to better opportunities and thus higher effective returns, challenging the Law of One Price.
- It could also inform market microstructure models by quantifying how search costs and heterogeneous investor knowledge affect bid-ask spreads and market efficiency.
Heterogeneous ReturnsWealth InequalityFinancial Human CapitalSearch FrictionsIncomplete MarketsMonetary PolicyHousehold FinanceAsset PricingFinancial LiteracyMarket Efficiency
Core finding, identification, data
Core Finding
- The paper finds that search frictions in financial markets, combined with endogenous financial human capital, lead to a dispersion in rates of return offered by firms and earned by households.
- Households with higher financial human capital earn systematically higher returns and tend to be richer, contributing significantly to wealth inequality.
- The calibrated model successfully reproduces the observed extent of residual dispersion in returns to wealth across individuals.
Identification Strategy
- The paper's identification strategy is based on a micro-founded general equilibrium model where heterogeneous financial human capital (accumulated through costly investment) and search frictions in capital markets endogenously generate heterogeneous returns to wealth.
- The model is quantitatively calibrated to match specific empirical moments from Norwegian data, including the distribution of individual fixed-effects in returns to wealth, the fraction of wealth held in cash by different households, and wealth persistence.
Data
The paper uses empirical evidence from Fagereng, Guiso, Malacrino and Pistaferri (2020) and Halvorsen, Hubmer, Ozkan and Salgado (2024), primarily based on Norwegian data. These data provide moments on the distribution of returns to net worth, wealth persistence, and money holdings across wealth percentiles for calibration.
Rodolfo G. Campos, Jesús Fernández-Villaverde, Galo Nuño, Peter Paz — Micro Data and Macro Models
This paper analyzes the interaction between monetary and fiscal policy in a heterogeneous-agent New Keynesian (HANK) model, showing how public debt influences the natural interest rate and the central bank's policy response.
Finance Application
- The paper's findings on public debt influencing natural rates and monetary policy have direct implications for bond pricing and asset allocation.
- Financial institutions could develop dynamic bond portfolio strategies that account for fiscal policy-driven shifts in the natural rate, potentially exploiting mispricings arising from central bank reaction lags.
- In household finance, the HANK model's insights into heterogeneous agent responses to fiscal shocks could inform models of household savings, debt accumulation, and investment decisions, especially how different wealth segments react to changes in real interest rates and inflation expectations.
- This could lead to better predictions of consumer spending and demand for various asset classes.
HANK modelsmonetary policyfiscal policynatural interest ratepublic debtinflationheterogeneous agentsasset pricingbond marketshousehold financepolicy timingyield curveterm premia
Core finding, identification, data
Core Finding
- In heterogeneous-agent New Keynesian (HANK) models, the stock of public debt significantly influences the natural interest rate, necessitating central bank adjustments to its monetary policy rule to maintain inflation targets.
- The paper demonstrates that policy timing is crucial, with early central bank responses anchoring inflation expectations and reducing welfare costs, while delayed responses lead to greater inflation volatility.
Identification Strategy
- The paper employs a calibrated HANK model to theoretically analyze the long-run and short-run effects of debt-financed fiscal expansions on natural rates and inflation under various monetary policy rules and timings.
- Empirically, it quantifies the natural rate's response to public debt changes using local projections (LP) and structural vector autoregression (SVAR) models on U.S. data, identifying the impact via lagged debt-to-GDP ratios and a Cholesky decomposition.
Data
The study uses U.S. economic data for model calibration, 2019 SCF data for household characteristics, and market-based measures for empirical analysis. These market data include daily 5-year 5-year (5y5y) forward nominal yields, 5y5y inflation-linked swaps (ILS), and 5y5y TIPS yields, covering the period from 1967:Q1 to 2023:Q2.
Roberto Colarieti, Pierfrancesco Mei, Stefanie Stantcheva — Micro Data and Macro Models
This paper uses large-scale survey data to quantify household spending, saving, and borrowing responses to income shocks (MPCs and MPDs) and to understand the underlying motivations and decision-making processes, identifying four distinct household types.
Finance Application
- The identification of distinct household types (Strongly Constrained, Precautionary, Quasi-Smoothers, Spenders) based on their underlying motivations, rather than just observed financial characteristics, offers a rich framework for refining heterogeneous agent models in household finance.
- This could lead to more accurate predictions of how different segments of the population respond to changes in interest rates, credit availability, or fiscal policy, impacting their demand for specific asset classes (e.g., demand for safe assets by Precautionary types) or their propensity to default on debt.
- For insurance research, understanding the specific reasons why 'Strongly Constrained' households rely on credit for emergencies or 'Precautionary' households build buffers could inform the design of targeted insurance products or financial literacy interventions that better align with their behavioral biases and constraints.
household financeconsumptionsavingborrowingincome shocksmarginal propensity to consumemarginal propensity to deleveragesurvey dataheterogeneous agentsmachine learningbehavioral economicsfinancial decision-makingliquidity puzzleco-holding puzzleasset pricing implicationsinsurance demand
Core finding, identification, data
Core Finding
- The paper finds that intertemporal marginal propensities to consume (iMPCs) are significantly higher immediately following an income shock and diminish over time, while marginal propensities to deleverage (iMPDs) play a critical role and exhibit substantial cross-sectional heterogeneity.
- Four distinct household types (Strongly Constrained, Precautionary, Quasi-Smoothers, and Spenders) are identified based on their reasoning for financial decisions, which are not solely explained by socioeconomic factors but also by psychological factors, past experiences, and expectations.
- These types offer nuanced explanations for puzzles such as why constrained households may have lower iMPCs and liquid households may have high iMPCs.
Identification Strategy
- The paper employs a novel survey methodology that elicits household responses to hypothetical income shocks of varying sizes, signs, and timings, allowing for quantitative estimation of iMPCs and iMPDs.
- A key methodological innovation is 'cross-validation,' where survey responses to hypothetical scenarios are compared against observed behaviors from past studies (e.g., responses to unemployment shocks, tax refunds, mortgage payment changes, and economic impact payments) to validate survey reliability.
- Furthermore, Latent Class Analysis (LCA), a machine learning algorithm, is used to classify households into distinct types based on their stated reasons for financial actions.
Data
The study utilizes new large-scale survey data from representative samples of the U.S. working-age population (a primary sample of 2923 observations and a previous wave of 1293 respondents), supplemented by two cross-validation surveys. The survey collects comprehensive information on respondents' socioeconomic background, financial decision-making processes, perceived hurdles and concerns, usual spending and saving behaviors, and detailed assets and liabilities (including mortgages, student loans, credit card debts, real estate, and various financial assets).
Hassan Afrouzi, Andrés Blanco, Andrés Drenik, Erik Hurst — Micro Data and Macro Models
This paper develops a macro-labor search model with nominal wage rigidities to analyze how unexpected inflation affects labor market flows, real wages, and worker welfare, explaining observed patterns during recent inflationary periods.
Finance Application
- The model's finding that unexpected inflation transfers resources from workers to firms (increasing corporate profits) could inform equity valuation models, suggesting that firms with greater wage stickiness or market power might exhibit higher profitability and thus higher stock returns or lower cost of equity during unexpected inflation.
- For household finance, the documented worker welfare losses and costly job search/renegotiation due to inflation directly impact household consumption, savings, and debt decisions, potentially increasing demand for inflation-indexed financial products or income protection insurance.
InflationLabor MarketsWage RigiditiesJob SearchVacanciesBeveridge CurveWorker WelfareCorporate ProfitsAsset PricingHousehold FinanceMacro-Labor
Core finding, identification, data
Core Finding
- Unexpected inflation, due to nominal wage rigidities, incentivizes workers to engage in costly job-to-job transitions and wage renegotiations, leading to higher vacancies, a seemingly 'tight' labor market, and an upward shift in the Beveridge curve.
- This simultaneously causes real wage declines and significant worker welfare losses, while increasing firm profits, a pattern observed in recent and historical U.S. inflationary periods.
Identification Strategy
- The paper's primary identification strategy involves a calibrated macro-labor model that simulates the effects of an unexpected temporary inflation shock (both a one-time shock and a a series of shocks replicating historical inflation) on labor market dynamics, holding other labor market shocks constant.
- It then validates these model predictions against historical U.S. data, showing that prior periods of high inflation systematically exhibited similar patterns.
Data
The study uses data from the Job Openings and Labor Turnover Survey (JOLTS), Current Population Survey (CPS), Atlanta Fed Wage Tracker Index, ADP Pay Insights, Barnichon (2010)'s unified vacancy series, Longitudinal Employer-Household Dynamics (LEHD), and FRED (Federal Reserve Economic Database).
Gregor Jarosch, Laura Pilossoph, Anthony Swaminathan — Micro Data and Macro Models
This paper quantifies the welfare costs arising from discrepancies between actual and desired working hours in Germany, using survey data and a labor supply model to evaluate the impact of a shorter workweek.
Finance Application
- The pervasive overwork and its associated welfare losses could directly impact household financial decisions, as time-constrained individuals may have less capacity for financial planning, education, or active investment management, potentially leading to suboptimal portfolio choices or higher demand for passive investment products.
- From an asset pricing perspective, if a shorter workweek policy enhances worker well-being and productivity, it could influence firm profitability, labor costs, and ultimately stock valuations, requiring investors to price in the potential for such labor market reforms.
- Furthermore, the identified 'hours mismatch' represents a form of human capital risk, suggesting a potential market for insurance products that mitigate the utility loss from working undesired hours or facilitate flexible work arrangements.
labor economicshousehold financelabor supplywork-life balancewelfare economicspolicy analysisasset pricinghuman capitalproductivitytime use
Core finding, identification, data
Core Finding
- More than two-thirds of full-time German workers are overworked, desiring to work fewer hours at their current wage, leading to welfare losses equivalent to 1.2% of GDP.
- The optimal workweek length in Germany is estimated to be 37 hours, which could yield welfare gains of 0.5-1% of GDP, with larger gains for women, college-educated, white-collar, middle-aged, and high-income workers.
Identification Strategy
- The paper uses self-reported survey data on desired versus usual hours, combined with a static labor supply model (MaCurdy, 1981) to calculate individual willingness-to-pay (WTP) for working desired hours.
- It then conducts counterfactual policy analysis by simulating the welfare impact of different weekly hours caps, considering various assumptions on how these caps affect different worker groups and the empirical wage-hours relationship, and integrating macro-level considerations like capital returns and tax base effects.
Data
The study primarily uses the German Socioeconomic Panel (SOEP) data from 1985 to 2021, supplemented by the British Household Panel Survey (BHPS), US Current Population Survey (CPS) data from 1985 and 2001, and a custom survey fielded in 2024 as part of the Real-Time Population Survey (RPS).
Gregor Jarosch, Laura Pilossoph, Anthony Swaminathan — The Micro and Macro Perspectives of the Aggregate Labor Market
This paper quantifies the welfare costs of discrepancies between actual and desired work hours using survey data and a structural labor supply model, and evaluates the welfare gains from a shorter workweek in Germany.
Finance Application
- The quantified welfare costs of labor supply mismatch could be directly imported into household finance models to analyze how hours constraints affect consumption-saving decisions, retirement planning, and demand for financial products.
- For asset pricing, aggregate welfare gains/losses from labor market policies could influence equity risk premia or the valuation of firms, particularly those in labor-intensive sectors, by altering aggregate productivity and consumption growth.
- The observed heterogeneity in WTP across demographic groups could also inform models of human capital pricing and cross-sectional asset returns.
Labor SupplyWork HoursWelfare EconomicsHousehold FinanceAsset PricingMacroeconomicsSurvey DataGermanyLabor Market FrictionsConsumption-Saving
Core finding, identification, data
Core Finding
- More than two-thirds of full-time German workers are overworked, desiring to work less at their current wage.
- The optimal workweek in Germany is found to be approximately 37 hours, which could yield welfare gains of 0.5-1% of GDP, with larger gains for women, college-educated, white-collar, middle-aged, and high-income workers.
- These gains are robust to various assumptions, including the empirical wage-hours relationship and macroeconomic losses.
Identification Strategy
- The paper identifies welfare costs by combining self-reported desired and usual work hours from survey data with a static labor supply model (MaCurdy, 1981) to calculate individual willingness-to-pay (WTP) for working desired hours.
- Counterfactual policy analysis then assesses aggregate and distributional welfare changes from tightening weekly hours caps, accounting for the empirical wage-hours schedule and macroeconomic losses.
Data
The study primarily uses the German Socioeconomic Panel (SOEP) from 1985-2021, supplemented by the British Household Panel Survey (BHPS), US Current Population Survey (CPS) supplements, and a custom Real-Time Population Survey (RPS).
Xiaonan Ma — The Micro and Macro Perspectives of the Aggregate Labor Market
This paper investigates why workers accept earnings cuts when changing jobs, finding that many do so strategically to access "stepping-stone" employers that offer better future career prospects, a mechanism formalized through a search model with vectorized offer arrival rates.
Finance Application
- The strategic acceptance of initial earnings cuts for future career growth has direct implications for household financial planning, suggesting that households might optimally choose to under-save or take on more debt in the short term, anticipating higher future earnings from "stepping-stone" job transitions.
- For asset pricing, firms identified as "stepping-stone" employers might exhibit different labor costs, human capital dynamics, or turnover rates, which could be priced into their equity, potentially leading to specific risk premia or valuation anomalies.
- The prevalence of ECUTs also highlights a form of earnings volatility that could inform the design of novel insurance products, such as career transition insurance, or influence the optimal structure of unemployment benefits.
labor economicsjob mobilityearnings dynamicshuman capitalsearch theoryhousehold financeasset pricingrisk managementcareer developmentadministrative datasurvey datastructural model
Core finding, identification, data
Core Finding
- A significant portion of U.S. job transitions (38% of all, 36% of EE) involve earnings cuts (ECUTs), even for those motivated by pecuniary reasons.
- Workers who accept ECUTs for pecuniary reasons, particularly when moving to "stepping-stone" employers (firms offering higher quality future job offers), experience significantly higher future earnings growth (6 percentage points higher) and increased subsequent mobility.
- The model suggests that 48% of all transitions and 52% of ECUTs are driven by stepping-stone motivations, primarily due to the quality, rather than quantity, of future job offers.
Identification Strategy
- The paper identifies the drivers of ECUTs by linking self-reported transition motivations from the National Survey of College Graduates (NSCG) with administrative earnings data from the Longitudinal Employer-Household Dynamics (LEHD).
- It uses OLS and 2SLS regressions to establish correlations between motivations, earnings dynamics, and subsequent mobility.
- The core identification for "stepping-stone" employers comes from a structural search model that formalizes employer heterogeneity in vectorized offer arrival rates (distinguishing offer quantity and quality from different employer groups), calibrated to match observed labor market moments and then used for counterfactual analysis to isolate the impact of offer quality.
Data
The study primarily uses linked administrative-survey data from the U.S. Census Bureau's Longitudinal Employer-Household Dynamics (LEHD) program and the National Survey of College Graduates (NSCG). The LEHD provides quarterly earnings and employment records, while the NSCG provides self-reported motivations for job transitions, covering 28 U.S. states from 2010 to 2019.
Ayse Imrohoroglu, Kai Zhao — Inequality and Macroeconomics
This paper develops a dynamic general equilibrium model of homelessness, calibrated to U.S. data, to evaluate the effectiveness and welfare implications of various housing and poverty reduction policies.
Finance Application
- The model's insights into housing market dynamics under policy interventions could inform asset pricing models for real estate, particularly for affordable housing REITs or local housing market indices, by quantifying how subsidies and supply elasticities affect rental prices and demand.
- In household finance, the detailed modeling of household budget constraints, saving behavior, and default decisions under income and health shocks provides a framework to study the impact of various social insurance programs (e.g., health insurance, unemployment benefits) on household balance sheets, consumption smoothing, and housing stability.
- The welfare analysis of different subsidy types could also guide the design of optimal emergency savings or credit products for vulnerable populations, considering their impact on both housing outcomes and overall financial well-being.
HomelessnessHousing PolicyGeneral EquilibriumHousehold FinanceHousing MarketIncome ShocksHealth ShocksWelfare AnalysisHousing VouchersRent SubsidiesPoverty ReductionReal EstateSocial InsuranceDynamic Stochastic General Equilibrium
Core finding, identification, data
Core Finding
- The existing housing voucher program significantly reduces homelessness, with general equilibrium (GE) effects amplifying this reduction (50% increase in homelessness if removed).
- Expanding the program's reach is more effective than increasing voucher amounts.
- While housing policies are better at reducing homelessness than general poverty reduction tools like cash subsidies, the latter generate larger welfare gains (0.91% CEV for cash vs. 0.25% for vouchers), highlighting a trade-off between targeting homelessness and broader welfare.
- GE effects are crucial, as they can lead to different conclusions about policy effectiveness compared to partial equilibrium analyses.
Identification Strategy
- The paper's methodological innovation is the development of a dynamic general equilibrium (GE) model of homelessness, which explicitly accounts for market adjustments (e.g., housing prices and rents) in response to policies.
- This contrasts with existing individual-level randomized controlled trials (RCTs) that primarily capture partial equilibrium (PE) effects.
- The model incorporates idiosyncratic income and health shocks to simulate diverse homelessness experiences and compares PE and GE outcomes to highlight the importance of market-wide effects.
Data
The model is calibrated and validated using U.S. data from sources such as the 1996 National Survey of Homeless Assistance Providers and Clients (NSHAPC), 2010 Decennial Census, 2006-2016 American Community Survey (ACS), 2013 Survey of Consumer Finances (SCF), U.S. Census Bureau American Housing Survey 2017, PSID data, and HUD data.
Aditya Soenarjo — The Micro and Macro Perspectives of the Aggregate Labor Market
This paper examines how displaced workers' liquidity positions influence their choice of re-employment sector, and how this dynamic impacts economic fluctuations and the effectiveness of unemployment insurance policies.
Finance Application
- This research offers several avenues for finance.
- In household finance, the link between liquidity and labor reallocation can inform models of optimal household portfolio choice, especially regarding liquid asset holdings and demand for credit during unemployment spells.
- For asset pricing, understanding how labor market frictions and liquidity affect sectoral labor reallocation could help explain cross-sectional differences in asset returns and risk premia across industries, particularly during periods of structural change or uneven sectoral shocks.
- In insurance, the findings highlight the role of unemployment insurance as a liquidity provider that facilitates labor market transitions, suggesting potential for new private insurance products designed to smooth income and support career transitions during sectoral shifts, thereby impacting demand for traditional life and disability insurance.
Labor EconomicsLiquidityUnemployment InsuranceLabor ReallocationSectoral ShocksHeterogeneous AgentsIncomplete MarketsRegression Kink DesignHousehold FinanceAsset PricingRisk Management
Core finding, identification, data
Core Finding
- Empirically, using administrative data from Washington state, a marginal increase in liquidity (e.g., a $10 increase in weekly benefits) significantly increases the propensity for displaced workers to switch industries by about half a percentage point.
- While industry switchers initially experience 10 percentage points lower earnings, this gap closes within 8 quarters.
- The model predicts that more generous unemployment insurance fosters greater labor reallocation, which can lead to less severe recessions when sectoral shocks are uneven.
Identification Strategy
- The paper employs a Regression Kink Design (RKD) to identify the causal effect of liquidity on labor reallocation.
- It exploits the kink in the unemployment insurance schedule, where weekly benefits are capped.
- Workers just above and below this kink are quasi-exogenously assigned different benefit levels based on their Highest Quarterly Wage (HQW), allowing for a local causal inference on the impact of liquidity.
Data
The paper uses administrative data from the Continuous Wage and Benefit Histories (CWBH) Project in the United States, specifically focusing on Washington state data from 1979-1983. This dataset includes information on unemployment spells, weekly benefits, and matched employer-employee data, allowing the author to track workers' industries before and after unemployment.
Huiyu Li, Chen Lian, Yueran Ma, Emily Martell — Macroeconomics and Productivity
This paper documents new empirical facts linking firms' markups to borrowing constraints, explains these relationships using a Kimball demand model, and analyzes the implications for allocative efficiency and TFP losses.
Finance Application
- The two-way feedback between markups and borrowing constraints, particularly the role of earnings-based borrowing, could be integrated into asset pricing models to explain cross-sectional return anomalies related to financial constraints or industry characteristics.
- Firms in industries with higher reliance on earnings-based borrowing might exhibit different sensitivities to macroeconomic shocks, impacting their systematic risk.
- In corporate finance, this framework could inform optimal capital structure decisions, as firms might strategically manage markups to relax debt capacity.
- For insurance, the higher markup dispersion and earnings volatility in industries relying on earnings-based borrowing could influence the demand for and pricing of corporate insurance products like business interruption or credit risk insurance.
Borrowing ConstraintsMarkupsMisallocationTFPFinancial FrictionsEarnings-based BorrowingAsset-based BorrowingFirm HeterogeneityCorporate FinanceAsset PricingIndustry Dynamics
Core finding, identification, data
Core Finding
- Less constrained firms within an industry have higher markups, especially in industries where assets are difficult to borrow against and firms rely more on earnings to borrow.
- Markup dispersion is also higher in these industries that rely more on earnings to borrow.
- These relationships are explained by a two-way feedback: looser constraints allow firms to attain larger market shares and charge higher markups, and higher markups relax earnings-based borrowing constraints, which in turn lowers TFP losses from markup dispersion.
Identification Strategy
- The paper primarily uses a structural Kimball demand model, augmented with asset-based and earnings-based borrowing constraints, to quantitatively explain observed empirical facts.
- It identifies relationships through cross-sectional and industry-level variations in firm characteristics (cash holdings, asset liquidation values) and their correlation with markups.
- Robustness checks use alternative measures of constraint tightness, such as the distance to violating financial covenants and the capital wedge.
Data
The paper uses U.S. Compustat data, Census of Manufacturing data, CapitalIQ data, DealScan loans data, Census Business Dynamics Statistics (BDS) database, and Quarterly Financial Report (QFR).
Ursula Berresheim, David Koll — Inequality and Macroeconomics
This paper investigates how overoptimistic divorce expectations influence household decisions, leading to higher specialization, lower savings, and adverse welfare outcomes, particularly for lower-wage spouses.
Finance Application
- This research offers significant insights for household finance, particularly regarding portfolio choice and retirement planning.
- Overoptimistic households may under-save or take on excessive risk, assuming a stable dual-income future, leading to severe financial distress upon divorce.
- This could be modeled by incorporating a 'divorce-contingent' utility function or a behavioral bias in risk aversion, influencing demand for specific asset classes or the uptake of insurance products designed to mitigate divorce risk, such as 'human capital insurance' or 'income protection' policies.
Household FinanceBehavioral EconomicsSubjective ExpectationsSavingsHuman CapitalDivorce RiskLife-Cycle ModelSpecializationGender InequalityInsurance
Core finding, identification, data
Core Finding
- Overoptimistic divorce expectations lead couples to exhibit higher within-couple specialization and accumulate significantly lower wealth and human capital compared to rational couples.
- This behavior is particularly detrimental for the lower-wage spouse, resulting in lower assets and human capital upon divorce and contributing to higher poverty rates among divorced mothers.
Identification Strategy
- The study employs a structural household life-cycle model with endogenous human capital and asset accumulation.
- It quantifies divorce expectation bias using a novel survey and a Random Forest algorithm trained on PSID data, comparing individuals' subjective divorce probabilities to their predicted objective probabilities to identify overoptimistic agents.
Data
The paper uses the Panel Study of Income Dynamics (PSID) for 1999-2019 and the American Time Use Survey (ATUS-CPS) for 2003-2019 to calibrate its model. It also conducted its own survey on subjective divorce expectations among married individuals in the U.S.
Fatih Guvenen, Serdar Ozkan, Sergio Ocampo-Diaz — Inequality and Macroeconomics
This paper conducts a "horse race" among three prominent theoretical frameworks to determine which best explains the observed features of top-end wealth inequality, including Pareto tails and the prevalence of self-made billionaires.
Finance Application
- The finding that return heterogeneity, especially from entrepreneurial activity, is the primary driver of top wealth inequality has significant implications for household finance and asset pricing.
- It suggests that models of household portfolio choice should explicitly incorporate heterogeneous investment opportunities and entrepreneurial risk-taking, rather than solely focusing on labor income shocks.
- For asset pricing, understanding the investment behavior of these high-return, high-wealth individuals could shed light on the demand for private equity, venture capital, and other illiquid assets, potentially influencing their pricing and market liquidity.
- Furthermore, the distinct income processes analyzed could inform the design and demand for specialized insurance products tailored to entrepreneurial risks.
wealth inequalityPareto tailpower law modelsreturn heterogeneityentrepreneurial returnsincome riskhousehold financeasset pricingsavingsportfolio choicelife cycle models
Core finding, identification, data
Core Finding
- The study reveals that models relying on 'awesome-state' income shocks or empirically grounded non-Gaussian income processes alone fail to reproduce key features of top wealth inequality, such as the Pareto tail and the existence of 'hundred-millionaires,' and often generate counterfactual demographic implications.
- In contrast, frameworks incorporating heterogeneity in rates of return, particularly those modeling endogenous entrepreneurial returns, successfully match the shape of the wealth distribution's right tail, the concentration of wealth at the top, and the rapid life-cycle wealth growth observed among the super-rich.
Identification Strategy
- The paper employs a comparative evaluation strategy, pitting three distinct theoretical frameworks against each other.
- Each framework (infinite-horizon Aiyagari-style with awesome-state shocks, lifecycle Aiyagari-style with empirically estimated non-linear income processes, and lifecycle Power-law models with return heterogeneity) is calibrated to match specific moments and then assessed on its ability to reproduce a broader set of salient features of top wealth inequality, including the Pareto index, the fraction of self-made billionaires, and life-cycle wealth dynamics.
Data
The paper utilizes US wealth distribution data from various sources (Vermeulen, 2018; Smith, Zidar, and Zwick, 2023), Forbes' billionaires list, administrative data on earnings dynamics (Guvenen et al., 2021), and demographic data from the Social Security Administration (Bell and Miller, 2002).
Joseph H. Pedtke — Inequality and Macroeconomics
This paper develops a quantitative dynastic model to assess how trends in family structure, education prices, and policy reforms contribute to rising inequality and declining intergenerational mobility in the United States.
Finance Application
- This research offers crucial insights for household finance by demonstrating how family structure and policy significantly shape intergenerational wealth and human capital transmission, influencing educational and career outcomes.
- This could inform models of household saving, investment in education (e.g., 529 plans), and life-cycle portfolio choices, particularly for single-parent households facing unique constraints.
- In asset pricing, understanding the long-term evolution of human capital and income inequality could impact the demand for education-related assets or influence risk premia associated with human capital shocks across different demographic groups.
- Insurers could leverage these findings to better price long-term care, disability, or life insurance products, considering the persistent effects of early-life family structure on adult income and health outcomes.
intergenerational mobilityinequalityfamily structurehuman capitaleducationpublic policyhousehold financeasset pricinglabor economicsquantitative macro
Core finding, identification, data
Core Finding
- The increase in single-parent families is a major contributor to both growing inequality (accounting for 50-64% of the increase in the Gini index) and falling intergenerational mobility (32% of the increase in the IGE), primarily by limiting resources and human capital investments for children, which perpetuates single parenthood across generations.
- Skill-biased technological change (SBTC) is the largest driver of inequality and persistence, while rising education prices have minimal effects.
- Policy changes to taxes and transfers had mixed impacts, mitigating inequality and persistence when family structure was fixed, but potentially increasing single parenthood and thus inequality when family structure was endogenous.
Identification Strategy
- The paper develops a dynastic model of human capital investment, calibrated to match key empirical moments of family structure, parental investments, and the income distribution in the U.S. over three historical periods (1947-1970, 1971-1994, 1995-2018).
- It then uses counterfactual simulations, where specific driving forces (e.g., single parenthood rates, SBTC, education prices, tax and transfer policies) are exogenously varied, to quantify their causal contributions to inequality and intergenerational mobility.
Data
The study utilizes a variety of U.S. micro and aggregate data sources, including the Panel Study of Income Dynamics Child Development Supplement (PSID-CDS), American Heritage Time Use Study (AHTUS), Consumer Expenditure Survey (CEX), Current Population Survey Annual Social and Economic Supplement (CPS-ASEC), Congressional Budget Office (CBO) data on taxes and transfers, Department of Education Common Core of Data (CCD), and US Census Small Area Income and Poverty Estimates (SAIPE).
Nicolas Longuet-Marx — Political Economy
This paper develops and estimates a political equilibrium model to disentangle demand (voters) and supply (politicians) factors driving political realignment, specifically focusing on blue-collar voters in US House elections from 2000-2020.
Finance Application
- This research offers a rich framework for understanding how political polarization, especially on cultural versus economic issues, could translate into financial market outcomes.
- In asset pricing, the differential polarization could be a novel factor explaining cross-sectional returns, with firms exposed to 'cultural' policy risk (e.g., ESG, gun control, reproductive rights) facing different premia than those exposed to 'economic' policy risk (e.g., taxes, trade).
- For household finance, the paper's findings on how voter preferences evolve and interact with party platforms could explain heterogeneous household investment decisions, savings rates, or debt choices based on their political alignment and perceived policy shifts.
- Insurers could use this granular understanding of political realignment to better model regulatory risk and demand for specific products (e.g., health insurance, climate risk policies) in politically diverse regions.
Political EconomyVoter PreferencesPolitical RealignmentParty PolarizationCultural IssuesEconomic IssuesUS House ElectionsDemand-Side FactorsSupply-Side FactorsIdentificationSpatial DiscontinuityPrecinct-level DataCampaign WebsitesText AnalysisIdeology
Core finding, identification, data
Core Finding
- The primary driver of voter realignment by education (2000-2020) is supply-side factors, particularly parties' stronger polarization on cultural issues compared to economic issues.
- Shifts in voter preferences, such as increasing blue-collar support for progressive economic policies, actually mitigated this realignment, implying it would have been even more pronounced otherwise.
Identification Strategy
- The paper identifies voter preferences by exploiting congressional districts' border discontinuities.
- It matches contiguous precincts across district borders, assuming that differences in candidate positions between these precincts can be treated as random after accounting for time-invariant precinct characteristics, thus addressing endogeneity in candidate ideological positioning.
Data
The paper uses a new panel dataset of 1.3 million precinct-level election results (2000-2020), candidate ideological positioning estimated from a multimodal text-and-survey model using campaign websites, and granular census demographics (IPUMS) at the block-group level, including unionization and religious affiliation data.
Susan Helper, Resem Makan, Daniel W. Shoag — Innovation
This paper investigates the long-term local innovation and economic impacts of federal investments in national laboratories, often established in rural areas, using historical data and quasi-experimental methods.
Finance Application
- The findings on long-term, localized economic growth and innovation spillovers from federal R&D investments could inform real estate investment strategies, particularly for private equity funds or REITs targeting regions near new government-funded research hubs.
- Identifying areas with planned federal science investments could predict future demand for commercial and residential properties, leading to potential alpha.
- Furthermore, the observed 'multiplier effect' on retail sales and household income suggests opportunities for private credit or venture capital in local businesses and SMEs within these emerging innovation ecosystems, as well as for insurance companies assessing regional economic stability for underwriting or municipal bond investments.
Regional EconomicsInnovationGovernment SpendingEconomic GrowthKnowledge SpilloversSynthetic ControlPlace-Based PolicyReal EstatePrivate EquityVenture CapitalLocal DevelopmentLong-Term InvestmentHistorical Data
Core finding, identification, data
Core Finding
- The establishment of national laboratories led to significant and sustained increases in local patenting activity, retail sales, and household income in host counties, even in regions that previously lacked innovation infrastructure.
- These benefits, including knowledge spillovers and economic multipliers, accrued to both existing and new residents, suggesting that federal science funding can catalyze broad-based regional development.
Identification Strategy
- The study primarily employs the synthetic control method, comparing counties that received labs to synthetic counterfactuals constructed from a donor pool of similar counties within the same state, using lagged outcome variables.
- It also utilizes identified 'runner-up' locations (counties considered but not chosen for a lab) as a control group and a difference-in-differences approach for individual-level wage analysis.
- The siting decisions for these labs were often driven by exogenous geopolitical, military, or logistical concerns, rather than pre-existing innovation potential.
Data
The paper uses historical patent data from the HistPat database (1836 onwards) and Patstat for classifications and citations. It introduces a newly digitized county-level dataset of retail sales and buying income from Sales Management's Survey of Buying Power (1936-1970), and individual-linked Census data (1940-1950) for wage analysis. Archival documents provide details on lab operating costs and site selection.
Rebekah Dix, Todd Lensman — Innovation
This paper analyzes the incentives for innovation in cancer drug combination therapies, revealing how market externalities and missing property rights lead to underinvestment by private firms.
Finance Application
- This research offers a valuable framework for understanding innovation in complex, multi-component products, which is highly relevant for asset pricing and household finance.
- In asset pricing, the 'market expansion externality' could explain why technology firms might underinvest in developing open standards or interoperable software components, even if it expands the total market, because the benefits accrue to other firms in the ecosystem (e.g., app developers, hardware manufacturers).
- This could lead to mispricing of platform stocks if the market doesn't fully account for these unappropriated spillovers.
- In household finance, the 'missing property rights' could explain underinvestment in user-friendly interfaces or educational tools for complex financial products (e.g., bundled insurance, robo-advisors), as the benefits of increased adoption might not be fully captured by the innovator, affecting consumer welfare and product market dynamics.
InnovationIndustrial OrganizationHealthcare EconomicsMarket ExternalitiesProperty RightsDynamic ModelsStructural EstimationPharmaceutical IndustryAsset PricingHousehold FinancePlatform Economics
Core finding, identification, data
Core Finding
- Private firms significantly underinvest in cancer drug combination therapies due to positive market expansion externalities that benefit other firms and missing property rights that limit their ability to appropriate value.
- This leads to firms trialing fewer combinations, preferring those with their own drugs, and delaying trials until generic entry.
- The study quantifies these externalities and demonstrates that targeted public innovation policies can substantially increase welfare by redirecting investment towards socially beneficial combinations.
Identification Strategy
- The paper employs a structural dynamic discrete-choice model of innovation decisions.
- It first estimates a demand system for cancer drug regimens (bundles of drugs) using microdata on patient usage and prices, and a Nash bargaining model for drug price setting to recover marginal costs and firm profits.
- These estimates are then used as inputs into the dynamic model, which recovers innovation fixed costs, success probabilities, and the public innovator's objective function by matching observed trialing decisions.
- The model uses a partially oblivious equilibrium approach and sieve value function approximation for computational tractability.
Data
The paper utilizes clinical trial data from ClinicalTrials.gov (1990-2022), drug characteristics and patent information from GlobalData and Drugs@FDA, cancer treatment guidelines (Chu and DeVita, 2019), and microdata on drug usage and prices from Medicare (1998-2019) and Marketscan Commercial Claims and Encounters (1996-2013). It also uses laboratory test data from NCI ALMANAC for drug efficacy measures.
Samuel Goldberg, Tai Lam — Digital Economics and Artificial Intelligence
This paper examines the causal impact of generative AI (GenAI) on the production, consumption, and market structure of creative goods in an online marketplace.
Finance Application
- This research offers insights for asset pricing by demonstrating how a disruptive technology like GenAI can alter industry structure, firm profitability, and the value of creative assets.
- For instance, it suggests that companies heavily reliant on traditional creative labor may face declining valuations, while those effectively integrating GenAI could see market expansion and increased sales.
- In household finance, the observed crowd-out of non-GenAI artists and reduced sales rates highlight the increasing income volatility and career risk for creative professionals, prompting questions about their savings, investment in new skills, and demand for unemployment or income protection insurance.
- The legal and copyright risks associated with GenAI also create new opportunities for specialized insurance products covering IP infringement and AI-generated content liability.
Generative AICreative IndustriesMarketplace EconomicsTechnology AdoptionCopyrightMarket EquilibriumProduct QualityProduct VarietyCrowd-outAsset ValuationIntellectual PropertyLabor MarketsRisk Management
Core finding, identification, data
Core Finding
- Generative AI acts as a substitute for non-GenAI products, leading to crowd-out of traditional content production and exit of non-GenAI artists.
- However, substantial GenAI firm entry increases overall product quality and variety, expands total sales, and benefits consumers, particularly in smaller niche markets, implying a net benefit to the platform despite adverse effects on non-GenAI producers.
Identification Strategy
- The study employs a difference-in-differences design, leveraging a platform policy change in December 2022 that allowed GenAI entry into non-branded markets (treated group) but prohibited it in branded markets (control group) due to legal concerns.
- This quasi-experimental setup allows for causal inference on GenAI's impact, with robustness checks including event studies and matching estimators.
Data
The paper uses granular, image-level data from a large online stock image marketplace, including information on authors, publication dates, sales status, tags, and a critical flag indicating GenAI production. It also utilizes 1,046-dimension image embeddings from Google's Vertex AI to derive measures of product quality, similarity, and variety.
Anders Humlum, Emilie Vestergaard — Digital Economics and Artificial Intelligence
This paper examines the early labor market impacts of AI chatbots using large-scale adoption surveys linked to Danish administrative data, finding minimal effects on earnings and hours despite rapid adoption and modest productivity gains.
Finance Application
- The paper's finding of minimal labor market effects from AI chatbots, despite widespread adoption, offers a crucial counter-narrative to prevalent "imminent disruption" expectations in financial markets.
- This insight could be imported into asset pricing to re-evaluate the valuation of AI-exposed firms, suggesting that current market premiums for AI-driven productivity gains might be overstated if labor market impacts are indeed muted and follow a "productivity J-curve." In household finance, the stability of earnings and hours implies that AI's early impact on household income streams is limited, affecting models of consumption, savings, and debt accumulation for workers in exposed occupations.
- For insurance, the observed task restructuring without net employment changes could inform new occupational risk profiles, allowing for more precise underwriting of income protection or professional liability policies based on evolving job content.
Artificial IntelligenceLabor MarketsEarningsHours WorkedProductivityDifference-in-DifferencesAdministrative DataSurveysOccupational MobilityTask RestructuringFirm ValuationHousehold FinanceRisk ManagementTechnology Adoption
Core finding, identification, data
Core Finding
- Despite rapid adoption and substantial employer investments in AI chatbots, the paper finds minimal average impact on workers' earnings and hours one and a half years after ChatGPT's launch, with confidence intervals ruling out effects larger than 1%.
- While occupational switching and task restructuring occur, these shifts do not result in net changes to hours or earnings, explained by modest productivity gains (3% time savings) and the emergence of integration/oversight tasks.
Identification Strategy
- The study employs a difference-in-differences framework, comparing adopters and non-adopters before and after ChatGPT's launch in November 2022.
- It uses a dynamic diff-in-diff design and analyzes heterogeneity across occupations and employer initiatives, with robustness checks for pre-trends and observable characteristics.
Data
The paper uses two large-scale surveys (late 2023 and 2024) covering approximately 25,000 workers from 7,000 Danish workplaces across 11 exposed occupations. These survey responses are linked to comprehensive Danish administrative data, including the E-Income Register (earnings, hours, occupation), Population Register (demographics), Personal Wealth Register, and Firm Statistics Register.
Aarushi Kalra — Digital Economics and Artificial Intelligence
This paper conducts a large-scale randomized experiment on an Indian social media platform to study how disabling personalization algorithms affects user engagement with "toxic" content and overall platform usage.
Finance Application
- This research offers several insights for finance.
- The finding that users with strong preferences for "toxic" content (e.g., misinformation, pump-and-dump tips) will seek it out on other platforms if their primary platform curates it away, suggests that content moderation on a single trading app or investment forum might be ineffective without cross-platform regulation.
- Financial platforms (e.g., Robinhood, eToro) using personalization algorithms could inadvertently reinforce risky trading behaviors or exposure to financial misinformation, leading to "echo chambers" of speculative assets.
- The inelasticity of user behavior implies that simply changing the feed algorithm might not deter investors from seeking out high-risk, high-reward (or misinformed) strategies, potentially leading them to less regulated platforms or dark pools for such "toxic" financial content.
Social MediaAlgorithmsPersonalizationToxic ContentUser BehaviorRandomized ExperimentPlatform EconomicsMisinformationContent ModerationBehavioral FinanceMarket MicrostructureFinancial RegulationFinTechCross-Platform EffectsInvestor Behavior
Core finding, identification, data
Core Finding
- Randomizing content reduces exposure to toxic posts by 27% but causes a 35% drop in overall platform usage.
- Critically, users with a high baseline interest in toxic content are less responsive to content changes, increasing their propensity to share toxic posts they do see and actively seeking out such content on other platforms, demonstrating the immalleability of user preferences and the limitations of content moderation.
Identification Strategy
- The study employs an individually randomized controlled trial with 8 million users on a prominent TikTok-like platform in India.
- The identification strategy involves replacing the platform's standard feed-ranking personalization algorithm with a random content delivery mechanism for a treatment group, allowing for a causal estimation of algorithmic effects on user behavior.
Data
The paper uses administrative data from TipTop, an Indian social media platform with 200 million users, covering user engagement (views, shares, time spent, logins) and post characteristics (genre, toxicity scores via Google's Perspective API). It also includes survey data from over 8,000 users to capture off-platform behavior and political attitudes.
Verina F. Que — Digital Economics and Artificial Intelligence
This paper investigates whether consumers' past privacy choices on a digital platform influence their subsequent privacy decisions, demonstrating structural state dependence and its implications for platform strategy.
Finance Application
- This research on state-dependent privacy choices has significant implications for household finance and fintech.
- For instance, an urgent financial need (analogous to the low-battery shock, e.g., needing a quick loan or urgent payment) might compel users to adopt a new financial app and share data.
- This initial 'forced' data sharing could then lead to state dependence, making them more likely to share data with other financial service providers (e.g., for credit scoring, personalized insurance, or investment advice), even if their underlying privacy preferences are strong.
- This could explain patterns in fintech adoption, cross-selling effectiveness, and the impact of 'open banking' initiatives on consumer data sharing behavior across different financial platforms.
- The LLM-based categorization of 'preference strength' could also be applied to financial products to study how state dependence varies between 'necessity' (e.g., mortgages) versus 'discretionary' (e.g., speculative investment apps) financial decisions.
privacystate dependenceconsumer behaviordigital platformsfintechhousehold financedata sharingnatural experimentLLMsbehavioral economics
Core finding, identification, data
Core Finding
- Past privacy choices significantly affect consumers' current privacy choices, with an acceptance of data terms reducing the probability of rejecting the next request by approximately 15%.
- This effect decays over time, is larger when preferences for the app are weak, and for users with weak privacy preferences, suggesting temporary state dependence and externalities across apps within a platform ecosystem.
Identification Strategy
- The study leverages a quasi-experiment on Alipay, a major digital platform, where consumers are highly likely to accept data requests from 'portable power bank apps' due to a low-battery emergency.
- This 'mandatory acceptance' shock exogenously increases the likelihood of a 'yes' decision, allowing the researchers to identify structural state dependence by observing its impact on subsequent data-consent decisions, separating it from persistent heterogeneity.
Data
The paper uses proprietary individual-choice-level panel data from Alipay, covering 358,004 randomly sampled users from March 2020 to September 2022. It also incorporates service description data from Qichacha, categorized using Large Language Models (LLMs), for app classification.
Rafael Berriel, Ravi Jagadeesan — Macro Public Finance
This paper develops a dynamic mechanism design model to analyze optimal capital and consumption taxation when consumption taxes are imperfectly enforceable due to business owners disguising personal consumption as business expenses.
Finance Application
- The model's mechanism of disguised consumption by owner-managers directly applies to corporate finance, explaining how tax policy influences private benefits extraction, corporate investment, and capital structure in closely-held firms.
- In household finance, it could shed light on tax planning strategies of wealthy entrepreneurs, their effective tax rates, and how these affect their savings and portfolio decisions.
- For asset pricing, the paper's findings on optimal capital taxation and intertemporal distortions could be integrated into models to understand how tax policy affects the cost of capital, asset returns, and wealth distribution across different investor types.
Optimal TaxationCapital TaxationConsumption TaxationDynamic Mechanism DesignWealth InequalityTax AvoidanceEntrepreneurshipCorporate FinanceHousehold FinanceAsset Pricing
Core finding, identification, data
Core Finding
- In contrast to canonical models that suggest zero capital taxation with observable consumption, this paper finds that imperfect consumption tax enforceability makes sustained intertemporal distortions and capital taxation optimal.
- The optimal policy balances revenue generation from capital taxes with their negative impact on wages, and can be implemented via time-varying linear taxes/subsidies on wealth, capital income, and consumption.
- When it is easier to disguise consumption, the optimal consumption tax is lower, and the wealth tax is higher.
Identification Strategy
- The paper uses a dynamic mechanism design framework in general equilibrium, leveraging stochastic calculus methods to achieve tractability.
- The key methodological innovation is modeling the imperfection of consumption taxes by allowing business owners to disguise personal consumption as business expenses, which are observably indistinguishable from productive intermediate spending, thus creating a moral hazard problem for the planner.
Data
The paper is theoretical, but for numerical examples, it uses standard parameter values from the literature, including capitalist discount rates, risk aversion, and elasticity of intertemporal substitution. It also cites Moskowitz and Vissing-Jørgensen (2002) for the volatility of profits and references empirical work on wealth inequality and tax avoidance using administrative data (e.g., Portuguese data by Leite (2024)).
Yihao Yuan — Digital Economics and Artificial Intelligence
This paper develops and estimates a structural model of the video streaming market to analyze the welfare implications of exclusive contracts for streaming services, studios, and consumers.
Finance Application
- The model's framework for valuing "content" (titles) based on consumer willingness-to-pay and its impact on subscription demand could be adapted to value intangible assets like brand equity or proprietary financial algorithms for financial firms.
- The multi-homing behavior and demand for bundles are highly relevant to household finance, analyzing how exclusive features (e.g., premium credit card benefits, exclusive investment opportunities) drive consumer choice and willingness-to-pay for bundled financial services.
- In insurance, the model could quantify the impact of exclusive distribution agreements between carriers and agents on premiums, product availability, and consumer welfare.
Structural ModelBargaining PowerExclusive ContractsSubscription EconomyConsumer DemandMulti-homingVertical IntegrationIntangible AssetsFinancial ProductsBundlingMarket StructureWelfare Analysis
Core finding, identification, data
Core Finding
- Exclusive contracts benefit small streaming services (like Hulu) and small studios by enabling content differentiation and improving bargaining leverage, while larger services (Netflix, Amazon Prime) and "Big Five" studios see minimal or negative effects.
- Consumers are harmed in the short run by reduced title distribution and higher prices, but long-run benefits may arise from stimulated content production and service entry.
Identification Strategy
- The demand model identifies price elasticities from geographical variation in subscription tax rates and content utility from household size and subscription demand.
- The supply model identifies studio bargaining power by comparing observed distribution networks to those maximizing bilateral surplus, and the internalization parameter for vertically integrated studios by their licensing patterns (e.g., Disney to Hulu).
Data
The paper uses Nielsen's title rating data for weekly viewing time by demographics, Reelgood data for title characteristics and distribution across streaming services, Nielsen Household Universe Estimates for monthly market shares of service bundles, and subscription prices and tax rates from various sources.
Mark A. Aguiar, Benjamin Moll, Florian Scheuer — Macro Public Finance
This paper develops a theory of optimal capital gains taxation that explicitly models asset price movements driven by both cash flow and discount rate changes.
Finance Application
- This framework could be used to model how different tax regimes (e.g., realization-based vs. accrual-based capital gains taxes, or wealth taxes) impact investor behavior and asset prices, especially for long-duration assets or assets with significant intangible value where cash flows are hard to pin down.
- It could also inform research on the impact of tax policy on market liquidity and trading activity, given its emphasis on 'realized trades.' For household finance, it highlights how tax policy can redistribute wealth gains/losses from asset price changes, which is critical for understanding wealth inequality dynamics and designing effective social safety nets or wealth transfer policies.
Optimal taxationCapital gainsWealth taxationAsset pricingDiscount ratesCash flowsPublic financeHousehold financeRedistributionTax policyMirrlees problemSlutsky compensationRealized trades
Core finding, identification, data
Core Finding
- The central theoretical finding is that optimal redistributive taxation should target realized trades (sales and purchases) when asset price changes are not solely due to cash flow changes (i.e., when discount rates play a role).
- This contrasts with the Haig-Simons concept of comprehensive income (which includes unrealized gains) and recent proposals for wealth or accrual-based capital gains taxes.
- A combination of realization-based capital gains and dividend taxes is shown to be a robust implementation.
Identification Strategy
- The paper develops a theoretical model of optimal redistributive taxation in a small open economy with asset price movements driven by both cash flow and discount rate changes.
- It uses 'Slutsky compensation' as a conceptual tool to derive first-best lump-sum taxes, showing how they should respond to price changes based on whether investors are buyers or sellers.
- This framework is then extended to second-best settings with distortive taxes (e.g., asset sales tax, wealth tax) and general equilibrium, demonstrating the robustness of the realized trades principle.
Data
The paper is primarily theoretical and does not use empirical data. It references existing empirical literature on asset price fluctuations to motivate its theoretical distinction between cash flow and discount rate changes.
Eugenio Miravete — Industrial Organization
This paper empirically evaluates the output and welfare effects of third-degree price discrimination (3DPD) using supermarket scanner data, focusing on demand curvature conditions.
Finance Application
- The finding that price discrimination often reduces overall output and welfare, and that output is a poor proxy for welfare, has significant implications for corporate valuation and investment strategies.
- Financial analysts could incorporate these insights into their models for evaluating firms' long-term profitability and sustainability, especially for companies operating in segmented markets or facing potential antitrust scrutiny.
- For private equity or M&A, understanding the true welfare effects of a target firm's pricing strategies could inform due diligence and post-acquisition value creation plans, particularly when assessing regulatory risk or market expansion potential.
- This framework could also be used to analyze the impact of regulatory changes on firms' pricing power and its subsequent effect on their stock performance and credit risk.
Price DiscriminationWelfare EconomicsIndustrial OrganizationDemand EstimationNonparametric MethodsMarket StructureAntitrust PolicyCorporate StrategyConsumer SurplusOutput EffectsEmpirical EconomicsRetail MarketsScanner Data
Core finding, identification, data
Core Finding
- The paper finds that third-degree price discrimination (3DPD) generally decreases output and welfare relative to uniform pricing, contrary to common theoretical assumptions and the views of legal scholars like Robert Bork.
- Specifically, output is predicted to increase for only 26% of product-chains, and welfare for only 19%, while welfare is predicted to decrease for 76% of product-chains.
- The study also shows that using output as a proxy for welfare overstates potential gains and understates potential damages of 3DPD.
Identification Strategy
- The study employs a nonparametric approach to estimate demand curvature conditions for thousands of store-product combinations by fitting H-degree Stone-Weierstrass polynomial approximations.
- To address price endogeneity, it uses Hausman instruments derived from the average prices of the same product in other geographic markets within the same chain.
- This flexible specification is crucial for allowing demand curvature heterogeneity across local markets, which is key to evaluating the true effects of 3DPD.
Data
The paper utilizes weekly sales data from the IRI Marketing Data Set spanning 2008-2011. This dataset includes information for over 3,000 UPCs across ten retail product categories, collected from nearly one thousand stores belonging to seventy-one supermarket chains in fifty U.S. metropolitan areas.
Samuel M. Altmann — Industrial Organization
This paper structurally estimates food banks' dynamic demand and storage costs for various food types under Feeding America's auction-based 'Choice System' to evaluate welfare gains compared to alternative allocation mechanisms.
Finance Application
- The methodology for identifying and estimating unobserved inventory in a dynamic auction setting could be applied to corporate finance or asset pricing.
- For instance, it could model how firms bid for inputs (e.g., commodities, components) when their internal inventory levels are unobservable to researchers, impacting production costs and ultimately firm value.
- In household finance, it could analyze how households bid for durable goods or housing, where their 'inventory' of existing assets influences demand, even if unobserved.
- The welfare analysis of different allocation mechanisms could inform market design for illiquid assets or distressed debt, where sequential vs. batch auctions might yield different outcomes.
dynamic auctionsstructural estimationunobserved statesinventory managementmarket designwelfare analysisstorable goodsnon-profit economicseconometrics
Core finding, identification, data
Core Finding
- The 'Choice System' increases welfare by 36% compared to the 'Old System', equivalent to an additional 84 tons of food per day.
- This gain primarily arises from 'batching' (simultaneous allocation of multiple items) rather than 'signaling' (bidding intensity), with 55% of the welfare improvement attributed to batching and 24% to signaling.
Identification Strategy
- The model identifies unobserved food bank inventory (stocks) by exploiting observed variation in food banks' choice sets and winnings, which act as observed shifters of these unobserved stocks.
- A Gibbs Sampler with data augmentation draws unobserved stocks from their conditional posterior distribution, allowing for non-parametric identification of flow payoffs and stock transition processes.
Data
Proprietary bidding data from Feeding America's 'Choice System' (2014-2017) covering 26,617 auctions and 165 food banks, including winning/losing bids, food composition, and locations. Auxiliary data on food bank demographics, catchment areas, and food insecurity are also used.
Tianli Xia — Industrial Organization
This paper studies the welfare effects of Resale Price Maintenance (RPM) in the Chinese pharmaceutical industry using a quasi-natural experiment from an antitrust case.
Finance Application
- This research on vertical restraints and their welfare implications could be applied to household finance by studying how such practices affect consumer spending, savings, and debt decisions for specific product categories, especially those with significant household budget shares (e.g., groceries, electronics).
- In asset pricing, the paper's insights into manufacturer and retailer bargaining power, and how these affect margins and sales, could inform models of firm profitability and industry-specific risk factors, potentially predicting stock returns for firms in vertically integrated supply chains.
- Furthermore, the methodology for disentangling pro-competitive (double markup elimination) vs. anti-competitive (price coordination) effects could be used to analyze the impact of regulatory changes on firm value and market efficiency in other industries, such as financial services or insurance distribution, where vertical relationships are complex.
Resale Price MaintenanceVertical RestraintsAntitrustIndustrial OrganizationBargaining PowerDouble MarginalizationPrice CoordinationConsumer WelfareFirm ProfitabilitySupply ChainRegulatory RiskChina Market
Core finding, identification, data
Core Finding
- The paper finds that banning RPM led to a 5% increase in retail prices and a 3% drop in wholesale prices for RPM products, nearly doubling retail margins and decreasing quantities sold by 12%.
- This suggests RPM is pro-competitive in this setting by eliminating double markups, but it also facilitates price coordination, which drives retail prices higher than they would be under a two-part tariff.
- The structural model confirms RPM is overall welfare-improving, but consumer surplus gains would have been 77% higher without price coordination incentives.
Identification Strategy
- The identification strategy relies on a quasi-natural experiment from a high-profile antitrust case where the Chinese State Administration for Market Regulation (SAMR) banned RPM for five products from Yangtze River Pharma Group (YRPG).
- A difference-in-differences (DiD) approach is used, comparing RPM products to control products (non-YRPG products in unrelated categories) before and after the ban.
- Additional identification comes from a structural model using instrumental variables like the investigation shock, average wholesale prices in different regions, and the number of direct competitors to identify price coefficients and substitution patterns.
Data
The paper uses a novel monthly panel dataset of sales from over 80,000 retail pharmacies in 150 Chinese cities from 2018-2021, which includes retailer-specific wholesale prices, retail prices, and quantities sold. It also uses a consumer panel dataset tracking purchasing records in pharmacies in ten cities to capture substitution patterns.
Diego R. Känzig, Konstantinos Gavriilidis, Ramya Raghavan, James H. Stock — Macro Public Finance
This paper develops a novel measure of climate policy uncertainty and shows that higher uncertainty acts as a supply shock, decreasing output and emissions while raising prices, with heterogeneous firm-level impacts.
Finance Application
- The finding that climate policy uncertainty acts as a supply shock, distinct from other policy uncertainty, could inform the development of new climate-related risk factors in asset pricing models, particularly for 'brown' versus 'green' assets.
- The firm-level evidence of a 'green paradox' (increased brown investment, reduced green R&D) suggests potential mispricing or misallocation of capital that could be exploited by investors.
- Furthermore, the persistent decline in TFP and heterogeneous firm responses could impact long-term corporate valuations, credit risk, and the solvency of insurance companies exposed to climate transition risks, necessitating adjustments in risk management and portfolio construction strategies.
Climate policyUncertaintySupply shocksAsset pricingFirm investmentR&DGreen paradoxMacroeconomic effectsRisk factorsESGClimate transition riskMonetary policyTFPNarrative identification
Core finding, identification, data
Core Finding
- An increase in climate policy uncertainty (CPU) leads to a significant fall in industrial production, GDP, and investment, coupled with rising commodity and consumer prices, indicating CPU shocks transmit as supply shocks.
- At the firm level, CPU causes stronger declines in investment and R&D for highly exposed firms, but incentivizes fossil-related sectors to accelerate 'brown' projects, leading to a persistent fall in total factor productivity.
Identification Strategy
- The paper constructs a novel climate policy uncertainty index based on newspaper coverage and identifies exogenous uncertainty shifts using a narrative approach.
- It compiles 72 major U.S. climate policy events (1985-2019) driven by climate-related, political, or ideological considerations, which are then used as an external instrument in a vector autoregression (VAR) model and local projections to estimate causal effects.
Data
The study uses a climate policy uncertainty index derived from newspaper coverage (New York Times, Wall Street Journal, Washington Post), macroeconomic data from FRED (e.g., industrial production, unemployment, CPI, interest rates, GDP, CO2 emissions, TFP), and firm-level data from quarterly Compustat Fundamentals, complemented by climate change exposure measures from earnings conference call transcripts (Sautner et al., 2023) and BEA fixed assets data.
Jakub Kastl, Eric Richert, Jesper Rüdiger — Industrial Organization
This paper introduces a novel recursive algorithm for efficiently computing global equilibria in a large class of static games with private information, outperforming existing numerical methods in stability, speed, and accuracy.
Finance Application
- This algorithm could be applied to model strategic interactions in financial markets, such as optimal bidding strategies in Treasury auctions or IPOs with asymmetric information, to understand price discovery and market efficiency.
- In household finance, it could analyze complex decisions like optimal mortgage choice or insurance contract selection under private information (e.g., risk tolerance, income shocks) and budget constraints.
- For insurance, it could optimize contract design for various products, accounting for adverse selection or moral hazard with multi-dimensional consumer types and competitive insurers, leading to more realistic pricing and menu offerings.
game theoryequilibrium computationauctionsnonlinear pricingdynamic programmingprivate informationstructural estimationmarket designcomputational economicsindustrial organization
Core finding, identification, data
Core Finding
- The paper's core finding is a new recursive algorithm that solves static games with private information by approximating objective functions with piecewise constant functions and formulating agents' problems as sequential programs.
- This method finds global solutions more stably, faster, and with fewer model restrictions than traditional local optimization techniques like shooting or polynomial collocation, particularly in complex, asymmetric settings.
- Applications to highway procurement auctions and breakfast cereal markets demonstrate its superior performance and ability to handle intricate counterfactuals.
Identification Strategy
- The methodological innovation is a recursive algorithm that efficiently finds global solutions by decomposing the problem into sequential sub-problems.
- It relies on two key insights: (1) objective functions (e.g., bid functions) can be approximated by piecewise constant functions, and (2) games with 'local constraints' allow agents' optimization problems to be characterized as sequential programs.
- This approach acts as an 'intelligent grid search,' eliminating non-optimal paths early and leveraging dynamic programming to achieve computational gains and stability.
Data
The paper uses Nielsen Consumer LLC and NielsenIQ marketing databases for its breakfast cereal market application. For highway procurement auctions, it re-examines data from existing studies by Krasnokutskaya and Seim (2011) and Somaini (2020).
Felix Montag — Industrial Organization
This paper develops a structural model of demand and supply to quantify the trade-offs between consumer welfare and domestic employment effects of mergers, particularly in the context of foreign competition, using the U.S. appliance industry as a case study.
Finance Application
- The paper's quantification of consumer welfare losses and job preservation in mergers could inform asset pricing models by providing a framework to assess the 'social cost' or 'social benefit' of M&A, influencing regulatory approval and thus firm valuation.
- For household finance, the localized employment impacts and price changes could be linked to household financial resilience, debt, and savings behavior in affected regions.
- The explicit trade-off between consumer welfare and employment also offers a quantitative metric for ESG (Environmental, Social, and Governance) investing, allowing investors to evaluate the 'Social' component of corporate actions and their impact on stakeholders.
mergers and acquisitionsforeign competitionemploymentconsumer welfareindustrial organizationlabor economicsfirm valuationESG investinghousehold financesupply chain
Core finding, identification, data
Core Finding
- The Whirlpool-Maytag merger decreased consumer welfare by $300 million annually but preserved 825 domestic jobs compared to a hypothetical foreign acquisition.
- To offset the consumer harm, each preserved domestic job would need to be valued at more than $360,000 per year, and endogenous product portfolio adjustments exacerbate both consumer harm and employment losses.
Identification Strategy
- The paper employs a structural model of demand and supply with endogenous product portfolios and employment decisions.
- Demand parameters are identified using household-level income and price variation, second-choice characteristics, and an instrumental variable (IV) based on production locations and real exchange rates (RER) to address price endogeneity.
- Fixed cost bounds for product offerings are estimated via moment inequalities derived from the absence of profitable one-step deviations in firms' portfolio choices.
Data
The core data comes from TraQline, an annual survey of 600,000 U.S. households on appliance purchases (2005-2015). This is complemented by product-level manufacturing locations, brand-level repair rates, advertising expenditures, household income from IPUMS Current Population Survey (CPS), and plant-level output and employment data hand-collected from news reports and company archives.
Camilla Schneier — Industrial Organization
This paper examines the welfare implications of exclusive dealing contracts in the U.S. retail real estate sector, analyzing their impact on retailer entry, competition, prices, and consumer welfare, particularly in underserved areas.
Finance Application
- This research offers direct insights for real estate finance, particularly for commercial real estate (CRE) investment and REITs, by quantifying how specific contractual arrangements like exclusive dealing impact property values, tenant mix, and landlord profitability.
- For household finance, understanding how these contracts affect local retail access and prices can inform studies on household consumption, savings, and financial vulnerability in different neighborhoods, especially concerning access to essential goods.
- Corporate finance can leverage the findings on retailer strategic location choices and contract terms to analyze firm valuation, competitive advantages, and capital allocation strategies in the retail sector, while ESG investors might use the food desert implications to assess the social impact of retail companies' real estate practices.
Exclusive DealingRetail Real EstateCommercial Real EstateFood DesertsConsumer WelfareLocation ChoiceContract TheoryIndustrial OrganizationEmpirical IOReal Estate FinanceHousehold FinanceMarket CompetitionProperty ValuationREITsESG
Core finding, identification, data
Core Finding
- Exclusive dealing contracts, while potentially anti-competitive by limiting local entry and increasing prices, can stimulate retailer entry in underserved markets, thereby mitigating food deserts.
- A counterfactual ban on exclusive dealing would increase the percentage of people living in food deserts by 10-15 percentage points over 20 years, harming consumers in these under-resourced areas, though some households in other areas might benefit from increased competition.
Identification Strategy
- The paper employs a structural model combining household-level store choices (with price/distance sensitivity and complementarities) and a static entry game between retailers and landlords (with variable profits and information asymmetry).
- Demand parameters are identified using individual trips and consumer microdata, with price sensitivity identified via an instrumental variable (average prices in other markets) and distance sensitivity via within-zip-code variation.
- Information asymmetry parameters and fixed costs are identified by matching micro moments in retailer location choice and landlord problems.
Data
The study utilizes a novel database of all 'potential' retail locations (developed and planned), manually collected retail real estate contracts (including exclusive dealing terms), and Numerator data for household retail store choices and shopping behavior. Additional data sources include SNAP Retailer Locator Data, InfoGroup Historical Datafile, CompStak for lease characteristics, Cook County Clerk's Office Record of Deeds Search, and ACS/Census Demographic Data.
Christine Laudenbach, Elin Molin, Kasper Roszbach, Talina Sondershaus — Household Finance
This paper examines how a bank's proprietary Buy Now, Pay Later (BNPL) transaction data influences credit access, interest rates, and repayment behavior for regular bank loans, highlighting benefits for both lenders and borrowers.
Finance Application
- This research offers several avenues for finance.
- In household finance, the 'learning by doing' effect from small BNPL loans could be explored for other micro-financial products (e.g., micro-savings, small investments) and its impact on long-term financial literacy and self-control.
- For asset pricing, the paper's findings on the value of proprietary data and price discrimination could inform models for valuing fintech companies or traditional banks that successfully integrate alternative data sources, potentially revealing a 'data premium' in their equity.
- The enhanced credit risk assessment from BNPL data also has implications for pricing credit derivatives or securitized consumer debt, where such private information might not be fully reflected in public ratings.
household financefintechcredit riskinformation asymmetryprice discriminationfinancial inclusionbehavioral financeasset pricingbanking
Core finding, identification, data
Core Finding
- The study finds that internal BNPL customers are significantly more likely to be approved for bank loans and receive lower interest rates due to improved internal credit risk assessments based on their BNPL payment history.
- This private information allows the bank to price discriminate, earning profit margins while also fostering improved repayment behavior and lower default rates among these borrowers, suggesting a 'learning by doing' effect from BNPL use.
Identification Strategy
- The identification strategy compares 'internal' BNPL customers (those with recent, frequent BNPL transactions whose data is used for internal credit assessment) with 'external' customers (those with fewer or older BNPL transactions whose data is disregarded).
- The analysis controls for external credit scores using tight bins of fixed effects and includes daily time fixed effects to account for dynamic bank policies, effectively comparing observationally similar customers.
- Robustness checks include coarsened exact matching.
Data
The paper utilizes unique, proprietary data from a large Nordic financial service provider that operates as both a bank offering consumer loans and a BNPL provider. The dataset covers 2018-2022 and includes over one million unsecured bank loan applications, detailed BNPL transaction records, internal and external credit scores, loan offer terms (amounts, maturities, interest rates), and subsequent repayment behavior (late payments, defaults).
Guillermo Carranza, Aaron S. Goodman — Household Finance
This paper empirically investigates the effects of 401(k) vesting schedules on participant wealth accumulation, distributional equity, and employee retention, finding significant wealth losses for lower-income workers and no causal retention effect due to informational frictions.
Finance Application
- The findings on informational frictions preventing behavioral responses to vesting incentives are highly relevant for household finance, suggesting that financial literacy interventions could significantly impact retirement savings and investment decisions.
- For corporate finance, the observed lack of retention effect implies that firms' capital allocation to 401(k) vesting might not be achieving its intended labor efficiency benefits, potentially affecting firm valuation and optimal executive compensation design.
- This could also inform the design of portable retirement products or insurance solutions that mitigate wealth losses from forfeitures due to job mobility.
401(k)vesting schedulesretirement savingshousehold financeinformational frictionsfinancial literacyemployee retentionwealth inequalitylabor economicsdefined contribution planscorporate finance
Core finding, identification, data
Core Finding
- The study finds that 30% of 401(k) separations occur during vesting periods, leading to forfeitures concentrated among lower-income participants, which makes 401(k) compensation significantly more regressive.
- Despite firms' stated motivation for retention, there is no evidence that vesting schedules causally affect employee retention, primarily because a majority of participants are unaware of their plan's vesting rules.
Identification Strategy
- The paper employs two main identification strategies: a cross-plan difference-in-difference (DiD) approach comparing separation probabilities around vesting deadlines in plans with and without vesting requirements, and a within-plan discontinuity design exploiting quasi-random variation in vesting schedules for July vs.
- August hires at plans using the 'equivalency method' for service crediting.
Data
The study uses administrative recordkeeping data from Vanguard for 1,500 401(k) plans (2010-2022), covering 4.7 million separations, which includes hire/separation dates, vesting events, financial transactions, account balances, and participant demographics and income. It also incorporates data from a Vanguard survey of 1,018 defined-contribution plan participants on vesting knowledge and financial literacy.
Naser Hamdi, Ankit Kalda, Qianfan Wu — Household Finance
This paper uses judge leniency as an instrumental variable to causally estimate the long-term intergenerational effects of parental Chapter 13 bankruptcy protection on children's income, upward mobility, and homeownership.
Finance Application
- This research offers a robust framework for quantifying the long-term, intergenerational benefits of debt relief, which is highly relevant for household finance.
- The methodology could be applied to evaluate other forms of household debt interventions, such as student loan forgiveness or mortgage principal reductions, to understand their broader societal impact beyond immediate recipients.
- For asset pricing, the findings imply that policies affecting household financial stability can have long-lasting effects on future generations' wealth accumulation and demand for assets like housing, potentially influencing long-run asset prices.
- Insurers could use these insights to refine risk models for future generations, as improved financial stability and homeownership rates might correlate with lower default risks or altered demand for life and disability insurance products.
Household FinanceDebt ReliefBankruptcyIntergenerational MobilityIncomeHomeownershipInstrumental VariablesCausal InferenceConsumer CreditPolicy EvaluationWealth AccumulationEducation
Core finding, identification, data
Core Finding
- Parental Chapter 13 bankruptcy protection significantly increases children's lifetime income, intergenerational upward mobility, and homeownership.
- Children's annual income is $1,215 higher 10 years post-filing, growing to $18,831 by age 40.
- Upward mobility increases by 0.5-1.5% and homeownership by 4.2-5.0% 15-20 years post-filing.
- Quantitatively, each dollar of parental debt relief generates over two dollars in adjusted lifetime earnings for children.
- Key mechanisms are asset protection (especially housing), increased educational investments, and geographic stability, while neighborhood and experience effects are not strongly supported.
Identification Strategy
- The study employs a 2-stage least squares (2SLS) instrumental variables approach.
- The instrument is judge leniency in approving Chapter 13 bankruptcy filings, which varies across judges despite uniform federal law.
- The first stage uses this judge leniency (calculated as a judge's approval rate relative to the district's average, using a 'leave-one-out' method) to predict parental bankruptcy protection.
- The second stage then estimates the causal effect of this predicted protection on children's outcomes.
- Random assignment of judges to cases and limited judge-debtor interaction beyond confirmation support the exclusion restriction.
Data
The paper combines consumer bankruptcy filings from the Public Access to Court Electronic Records (PACER) system (1992-2009) with anonymized consumer credit histories and employment/earnings data from Equifax Inc. (2010-2022). Additional data sources include the IRS website, American Community Survey, and National Center for Education Statistics for ZIP Code characteristics and school graduation rates. Debt relief amounts are scraped from 'Final Report and Account' documents (2004-2009).
Eduard Boehm — Household Finance
This paper quantifies the consumer-welfare effects of intermediaries in the Chilean pension and annuity market, where retirees face complex product choices, adverse selection, and potentially biased advice.
Finance Application
- This framework could be applied to evaluate the impact of financial advisors in the US retail investment market, quantifying the welfare effects of advice in complex product selection (e.g., mutual funds, variable annuities, structured products) where fees and potential steering exist.
- It could also inform policy on disclosure requirements or fiduciary standards in the mortgage market, assessing how intermediaries' incentives affect household choices and loan terms.
- Furthermore, the rational inattention model could be used to explain persistent 'puzzles' in household portfolio allocation or insurance uptake, where consumers face high cognitive costs in evaluating complex options.
IntermediationFinancial AdviceChoice FrictionsRational InattentionAdverse SelectionAnnuitiesPensionsHousehold FinanceConsumer WelfareDynamic Models
Core finding, identification, data
Core Finding
- Intermediaries can improve welfare by eliminating choice frictions (valued at ~250 USD/year by retirees), but often steer customers towards annuities, even when Phased Withdrawal is optimal.
- Despite this steering, a ban on intermediation is consumer-welfare neutral because the gains from avoiding distortions are offset by increased decision costs and higher annuity prices due to exacerbated adverse selection, as the variety of annuity types allows for close substitutes.
Identification Strategy
- The model identifies preferences (mortality risks, bequest motives) by comparing choices within intermediation channels and across varying choice sets/prices.
- Intermediary distortions and information costs are identified by comparing choices across intermediation channels, leveraging geographic variation in intermediation demand (lagged share of intermediated individuals in a province) as an exogenous shifter, assuming information costs are uncorrelated with other unobservable preferences.
Data
The paper uses individual-level administrative data from the Chilean centralized offer exchange (SCOMP) covering 2004-2020, including demographics, savings, intermediation status, product choices, commissions, and mortality. It also incorporates data from the Social Protection Survey (SPS) and a choice-architecture experiment on retirees' financial literacy and preferences.
Ulrike Malmendier, Ryan Oprea, Laura Veldkamp — Household Finance
This paper introduces and rigorously tests the concept of "information resonance"—the extra weight people give to messages from those who feel like them—demonstrating its widespread nature and modeling its consequences for various economic phenomena.
Finance Application
- This research could be applied to household finance by explaining why individuals' investment decisions, savings behaviors, or insurance uptake are strongly influenced by their social networks and demographic peers, even when objective information is available.
- For asset pricing, information resonance could contribute to localized asset price anomalies or herd behavior, as investors disproportionately weight signals from similar individuals, leading to correlated trading patterns.
- In insurance, it could explain disparities in product adoption or risk perception across communities, driven by the resonant influence of local 'role models' or peer groups, potentially leading to suboptimal coverage decisions.
information diffusionsocial networkshomophilybehavioral economicsexperimentsdecision makingrole modelshousehold financeasset pricingsocial learningbias
Core finding, identification, data
Core Finding
- Through a controlled lab experiment, the paper finds that individuals are significantly more likely to follow advice from a stranger who shares a common belief, preference, or demographic characteristic, even when these characteristics are obviously irrelevant to the decision.
- This resonance is robust across diverse characteristics and decision domains, and it persists even when mechanisms like trust, relevance, or memory are minimized.
- The theoretical model shows how this behavioral bias can explain the power of role models, the influence of social media, and patterns of workplace discrimination.
Identification Strategy
- The primary identification strategy is a controlled lab experiment where subjects make various choices (e.g., charities, books, ETFs) based on recommendations from a stranger.
- Resonance is isolated by randomly assigning a single, often trivial, personal characteristic (a 'fun fact') about the recommender to each choice, while minimizing trust (anonymous recommenders), relevance (irrelevant characteristics), and memory (information salient at decision time) as alternative explanations.
Data
The paper uses data from a lab experiment involving 550 subjects on Prolific. For field evidence, it utilizes American Community Survey (ACS) data from 2005-2020, obtained via IPUMS, focusing on occupational choices and demographic characteristics.
Erina Ytsma — Science of Science Funding
This paper investigates how men and women in German academia respond differently to explicit performance incentives versus career-concern incentives, and the implications for output, innovation, and female representation.
Finance Application
- This research could inform the design of executive compensation packages in corporate finance, exploring if gender-specific incentive structures (e.g., stock options vs. long-term bonuses) differentially impact male and female executives' risk-taking, innovation, and ultimately firm performance and stock returns.
- In household finance, it could guide the development of gender-tailored incentives for retirement savings or investment products, considering whether explicit bonuses or long-term financial security appeals more to men versus women.
- For insurance, it could shed light on how different commission structures affect the sales performance and client engagement of male and female agents.
gender differencesincentivesperformance paycareer concernsnatural experimentlabor economicshuman capitalinnovationacademia
Core finding, identification, data
Core Finding
- Men exhibit a significant effort response (16-19%) only to career-concern incentives, while women show a significant effort response (36-40%) only to explicit performance incentives.
- Women increase highly cited work in response to performance pay, whereas men increase novel but medium-impact work.
- The study also suggests that productive women may be less likely to receive academic job offers despite their positive effort response, potentially hindering female representation and innovation.
Identification Strategy
- The paper employs a natural experiment by exploiting the introduction of performance pay in German academia.
- This allows for a causal identification of gender differences in the response to explicit performance incentives and career-concern incentives.
Data
The study utilizes a comprehensive data set that includes affiliations and publication records for the entire universe of German academics.
Lena Greska — Science of Science Funding
This paper investigates the trade-off between specialization and coordination in innovative teamwork, finding that generalist teams outperform specialist teams, especially in complex environments, with coordination costs as the key mechanism.
Finance Application
- This research offers valuable insights for structuring teams in various financial contexts.
- In asset management, it suggests that generalist investment teams, or those with broader expertise across asset classes or analytical methods, might outperform specialist teams, especially during periods of high market complexity or uncertainty.
- For financial product development, particularly for complex, novel instruments (e.g., structured products, new insurance policies for emerging risks), generalist teams could be more effective in navigating unforeseen interactions and coordinating diverse expertise.
- Furthermore, the impact of AI tools like ChatGPT on reducing coordination costs could inform how financial institutions integrate AI assistants to enhance team productivity and innovation.
TeamworkSpecializationCoordination CostsInnovationMachine LearningArtificial IntelligenceHuman CapitalOrganizational EconomicsProductivitySkill DiversityComplexity
Core finding, identification, data
Core Finding
- Generalist teams consistently produce higher quality innovation than specialist teams, particularly in high-complexity problems.
- This performance difference is primarily driven by higher coordination costs faced by specialist teams, which are mitigated by generalists' broader skill sets and by coordination-reducing technologies like ChatGPT.
Identification Strategy
- The study employs several empirical strategies, including controlling for observable heterogeneity and past performance, using user-fixed effects identified from individuals switching teams or between solo/team work, and an event study design based on team formation timing.
- A key quasi-natural experiment is the introduction of ChatGPT, which exogenously reduces coordination costs, allowing for a causal inference on its impact on team performance differences.
Data
The paper utilizes data from online machine learning competitions on Kaggle.com, encompassing user profiles, competition details, team structures, and code metadata. A novel measure of specialization is constructed based on the semantic diversity of a computer scientist's code, quantified using CodeT5+, a large language model.
Caroline Fry, Gauri Subramani — Science of Science Funding
This paper investigates how South-South fellowships for female scientists influence the research agendas of their non-migrating colleagues at home institutions, particularly focusing on local problem-solving.
Finance Application
- The insights on knowledge transfer and research agenda shifts due to mobility could be applied to financial analysts or fund managers.
- For instance, how does the mobility of a star analyst to a new region affect the research coverage and recommendations of their former colleagues, particularly regarding 'local' or 'shared' industry problems? Similarly, the movement of venture capitalists between emerging markets could influence investment patterns and innovation in financial products (e.g., fintech) in both their origin and destination markets, with geographic proximity potentially amplifying these effects on local financial challenges.
knowledge transferresearch mobilitygeographic proximityinnovationdeveloping countriesSouth-South migrationcollaborationinformation asymmetryanalyst researchventure capitalfinancial innovationlocal bias
Core finding, identification, data
Core Finding
- Non-migrating colleagues of successful South-South fellows increase their research focus on diseases highly prevalent in both their home and the fellow's host countries, leading to higher publication rates.
- This effect is stronger when home and host countries are geographically closer, driven by increased collaborations with host country scientists, especially when travel frictions are low.
Identification Strategy
- The study employs a differences-in-differences design, comparing publication patterns of non-migrants whose colleagues received an OWSD fellowship (treatment) with those whose colleagues applied but were unsuccessful (control).
- It leverages the timing of the fellowship and includes individual, year, and career age fixed effects.
- Robustness checks involve person × year × disease-level analysis and variations in fellowship duration.
Data
The paper uses data on female scientists applying for the Organization for Women in Science for the Developing World (OWSD) fellowship (1996-2016), Elsevier Scopus for publication records, the Global Burden of Disease database for country-specific disease prevalence, and the CEPII database for bilateral geographic distance, primary language, and flight time between countries.
Pierre Boutros, Eliana Diodati, Michele Pezzoni, Fabiana Visentin — Science of Science Funding
This paper analyzes how AI training during a PhD impacts the subsequent career paths and scientific productivity of STEM graduates in France, using a quasi-experimental approach.
Finance Application
- This research offers a robust framework to study the career trajectories of finance PhDs with specialized quantitative or AI/ML training.
- One could investigate if such training leads to a 'brain drain' from traditional asset management to quant funds or fintech startups, or if it correlates with higher alpha generation or risk management innovation.
- The disciplinary heterogeneity finding could be particularly relevant, examining if AI training impacts careers differently for finance PhDs focused on theoretical models versus those in financial engineering or computational finance.
AImachine learningPhD careerslabor markethuman capitalscientific productivitybrain draineconometricspropensity score matchingSTEMFranceinnovationpatentingacademic mobility
Core finding, identification, data
Core Finding
- AI training during a PhD does not increase the overall probability of pursuing a research career.
- However, AI-trained PhDs exhibit path dependence in publishing on AI topics and achieve higher citation impact.
- Interestingly, AI-trained Computer Science PhDs are less likely to enter private research organizations, while AI-trained PhDs in other STEM fields show increased patenting activity and international mobility, particularly to the US and China.
Identification Strategy
- The study employs Propensity Score Matching (PSM) in two steps to mitigate selection bias.
- First, it matches AI-trained PhD students with non-AI-trained students based on observable characteristics (supervisor traits, university location, defense year, discipline) to estimate the probability of being an AI student.
- Second, it applies PSM again to students who pursue a research career, comparing AI-trained and non-AI-trained individuals with similar propensity scores to isolate the causal effect of AI training on career outcomes and productivity.
Data
The paper uses a unique dataset of 35,492 French STEM PhD students who graduated between 2010 and 2018, identifying AI training through keywords in thesis titles and abstracts. Career outcomes and productivity metrics are derived from bibliometric data (OpenAlex), patent applications (EPO), and research grants (ANR).
Yagmur Yildiz, Diego Chavarro, Fabiana Visentin, Tommaso Ciarli — Science of Science Funding
This paper causally investigates whether post-doctoral mobility grants lead early career researchers to diversify their research topics.
Finance Application
- This research's methodology and findings could be applied to study the causal impact of early-career funding (e.g., seed grants, early-stage venture capital) on the diversification of research topics for finance academics or product development for fintech startups.
- An RDD could be used on funding thresholds for financial research grants to see if it encourages exploration of new asset classes or methodologies, or for startup accelerators to assess diversification of business models.
- This could inform optimal funding strategies to foster innovation in finance.
Research fundingCareer pathsDiversificationCausal inferenceRegression Discontinuity DesignAcademic mobilityInnovationEarly career researchersGrants
Core finding, identification, data
Core Finding
- Receiving a Marie Skłodowska-Curie Actions Individual Fellowship (MSCA-IF) positively impacts the breadth of a researcher's portfolio, increasing the share of publications with new topics (especially at the 90% new topics threshold) for 2-3 years after the grant.
- This effect suggests MSCA-IFs accelerate cognitive exploration but diminishes over time.
Identification Strategy
- The paper employs a fuzzy Regression Discontinuity Design (RDD) to establish causality.
- It compares 'similar' applicants who were just above (funded) and just below (not funded) the funding threshold, leveraging fine-grained evaluation scores and a two-stage equation model to account for non-compliance.
Data
The study combines MSCA-IF application data from the COmmon Research Data Warehouse (CORDA) for 2014-2019 under Horizon 2020, with publication data from the OpenAlex database. The final restricted sample includes 2,892 similar applicants (1,401 funded, 1,491 not funded).
Audrey Tiew — Environmental & Energy Economics
This paper examines the impact of the 2017 EU trade resolution on Malaysian palm oil markets, revealing how farmer heterogeneity leads to within-market leakage and evaluating alternative policies to limit deforestation.
Finance Application
- This paper's findings on 'within-market leakage' due to heterogeneous producer responses to environmental policies have direct implications for ESG investing and commodity markets.
- ESG funds or 'green' bonds targeting deforestation might misprice the true environmental impact and financial risks if they don't account for how policies shift production among heterogeneous firms (e.g., from smallholders to large estates).
- This could lead to mispricing of palm oil futures or the equity of food and agricultural companies, as the long-term supply stability and environmental performance are affected by these complex dynamics.
- For household finance, the differential impact on smallholders (e.g., lower scrap values, higher entry costs) highlights financial vulnerability, suggesting research into microfinance solutions or climate-resilient insurance products for small-scale farmers facing policy-induced income shocks.
environmental economicsindustrial organizationcommodity marketstrade policydeforestationESG investingsupply chainproducer heterogeneitypolitical economydynamic modelsmicrofinanceagricultural finance
Core finding, identification, data
Core Finding
- The 2017 EU trade restriction, intended to reduce deforestation, decreased smallholder land expansion but inadvertently incentivized large estate expansion, leading to significant within-market leakage.
- Large estates benefit from lower entry costs and higher scrap values, making them more resilient to negative demand shocks.
- While targeted taxes on large estates could be more environmentally effective, they face political opposition; conversely, producer quantity coordination (OPEC-like) emerges as a superior alternative, simultaneously reducing deforestation and increasing profits for all producer types, making it both effective and politically feasible.
Identification Strategy
- The paper employs a two-step estimation strategy.
- First, static supply and demand parameters are identified using instrumental variables: monthly rainfall in major soybean-producing countries (US, Brazil) for palm oil price in the supply equation, and regional Malaysian rainfall for palm oil price and US/Brazilian rainfall for soybean oil price in the demand equation.
- Second, dynamic parameters (entry/exit costs, scrap values) are estimated using a simulated method of moments, leveraging variation in land flows by farm type across regions and over time, with constant, marginal cost/rainfall supply shocks, and regional demand shocks serving as instruments.
Data
The study utilizes annual total palm oil land area by administrative region and farmer type from the Malaysian Ministry of Plantation Industries and Commodities (2006-2019), primary forest loss and carbon emissions data from Global Forest Watch, monthly Malaysian rainfall from the World Bank, world soybean oil prices from the IMF, and futures prices for crude palm oil and soybean oil from DataStream.
Shifrah Aron-Dine — Environmental & Energy Economics
This paper analyzes the distributional effects of natural disasters and the impact of post-disaster policies on migration, housing markets, and household welfare in Puerto Rico after Hurricane Maria using a structural model and novel data.
Finance Application
- The paper's insights into household undervaluation of risky, non-transferable rebuilding subsidies could inform the design of private disaster insurance products, highlighting the need for guaranteed and flexible payouts to increase uptake and welfare.
- For asset pricing, the general equilibrium effects on housing prices, rents, and migration patterns could be used to model and price local real estate investment trusts (REITs) or municipal bonds in disaster-prone areas, considering how different policy responses alter recovery trajectories and asset valuations.
- In household finance, the detailed analysis of mortgage defaults and the impact of policy design on household financial resilience could be extended to study optimal household portfolio choices and debt management strategies in the face of climate change risks, particularly for those with low home equity.
Household FinanceNatural DisastersMigrationHousing MarketsMortgage DefaultInsurancePublic PolicyGeneral EquilibriumWelfareAsset PricingReal EstateRisk Management
Core finding, identification, data
Core Finding
- Homeowners with property damage face significant welfare losses, but all households are affected by infrastructure destruction and general equilibrium price movements.
- Local infrastructure investment is found to be a cost-effective policy due to complementarities, while rebuilding subsidies for homeowners are not cost-effective because they are risky, non-transferable, and undervalued by recipients, especially if homes are foreclosed or sold.
- More flexible policies, like unconditional cash transfers or migration subsidies, could yield greater welfare improvements for similar costs.
Identification Strategy
- The paper leverages Puerto Rico's unique island geography and high-frequency airline passenger traffic data to measure temporary and permanent migration.
- It also uses an original survey of affected individuals to directly link property damage to subsequent household decisions like migration, mortgage default, and repair financing.
- A heterogeneous agent model is then calibrated to match these empirical patterns.
Data
The paper utilizes high-frequency airline passenger data from the Bureau of Transportation Statistics (BTS), an original survey of Puerto Rican households, and Census data (ACS, PRCS). It also incorporates FEMA Disaster Relief Fund reports, FHFA house price indices, and NOAA data on physical damages.
Christoph Semken, Amelie Michalke, Lennart Stein, Freek van Sambeek, Santiago Varela Seoane, Benjamin Oebel, Tobias Gaugler, Hunt Allcott — Environmental & Energy Economics
This paper develops a theoretical framework for optimal green retailing and empirically evaluates a "true cost" campaign by a German grocer, finding that consumer "warm glow" from donations reduced demand elasticity but also highlighting greenwashing concerns.
Finance Application
- The "warm glow" effect and greenwashing concerns could be applied to behavioral finance, examining if investors exhibit similar "warm glow" when investing in ESG funds, potentially accepting lower returns or higher fees, and how perceptions of greenwashing impact ESG fund flows or corporate bond yields.
- The paper's "Marginal Value of Profit Reduction" (MVPR) metric could be adapted for corporate finance to evaluate the financial efficiency of corporate sustainability investments, linking specific ESG initiatives to changes in firm valuation, cost of capital, or shareholder returns.
- Furthermore, the granular quantification of "uninternalized externalities" could inform the pricing of environmental liability insurance or climate-related risk bonds, allowing insurers to better assess and differentiate premiums based on a company's true environmental footprint.
ESGBehavioral FinanceCorporate FinanceGreenwashingConsumer BehaviorFirm ValuationRisk ManagementEnvironmental EconomicsPricingRetailSustainabilityWarm GlowMarginal Value of Profit ReductionInsurance
Core finding, identification, data
Core Finding
- A "true cost" campaign, where a grocer added estimated externality costs to product prices and donated proceeds to climate mitigation, resulted in a smaller-than-predicted demand decrease, suggesting a "warm glow" effect from donations.
- However, consumers grew skeptical and concerned about greenwashing.
- The paper introduces a "marginal value of profit reduction" (MVPR) metric to evaluate corporate sustainability strategies, finding that unilateral green retailing campaigns yield welfare gains significantly lower than optimal Pigouvian taxes.
Identification Strategy
- The study identifies the "warm glow" effect by comparing demand responses to the "true cost" campaign (price increase plus donations) with responses to standard short-term promotional price changes for the same products, using an event study design.
- For the structural model, price variation from retailer promotions and cross-sectional variation in consumer-retailer distances are used to identify elasticities and preferences.
Data
The paper utilizes item-level scanner data from a German grocer (Penny) from 2019-2023, household-level grocery transaction data from AiMark (GfK) for 2022, product characteristics from Open Food Facts, and a nationally representative panel survey (baseline and endline) on consumer beliefs and attitudes. Media articles related to the campaign were also collected and coded.
Amanda R. Kreider, Rachel M. Werner — Workshop on Aging
This paper investigates how US immigration enforcement, specifically the Secure Communities program, impacts the supply of home care workers and the access to home-based long-term care for older adults.
Finance Application
- This research offers direct insights for long-term care (LTC) insurance.
- Reduced formal home care supply and increased reliance on family caregivers could alter demand for LTC products, their pricing, and the types of benefits offered.
- For asset pricing, the profitability and risk of publicly traded home care agencies, nursing home operators, or healthcare REITs could be significantly impacted by labor supply shocks and shifts in care preferences.
- In household finance, the increased burden on family caregivers due to reduced formal care could affect household savings, debt, and intergenerational wealth transfers, especially for low-income families.
ImmigrationLabor SupplyLong-Term CareHealthcareMedicaidHousehold FinanceInsuranceAsset PricingNatural ExperimentDifference-in-DifferencesEvent Study
Core finding, identification, data
Core Finding
- The Secure Communities policy reduced the overall home care workforce by 7.5%, with 70% of this effect driven by a 14.6% reduction in non-US-born workers.
- This labor supply shock led to a 2.9 percentage point (5% relative) decrease in older adults needing assistance receiving any home help, with effects concentrated among Medicaid recipients (10.5% less likely to receive any help, 23.2% less likely to receive formal care), who then substituted towards family care.
Identification Strategy
- The study leverages the quasi-random rollout of the federal Secure Communities policy across US counties between 2008-2013 as a natural experiment.
- It employs difference-in-differences and event study models with time and location fixed effects, utilizing the Callaway and Sant'Anna estimator to account for heterogeneous treatment effects.
Data
The paper uses data from the American Community Survey (ACS) for home care workforce size and sociodemographic characteristics at the CPUMA level, and the Health & Retirement Study (HRS) for older adults' receipt of home care and individual characteristics. County-level Secure Communities activation dates are obtained from Alsan and Yang (2022a, 2022b) and ICE reports.
Patrick Agte, Jitendra Kunar Soni — Development Economics
This paper evaluates a large-scale public healthcare reform in India that added mid-level healthcare workers to village clinics, finding significant reductions in elderly mortality and improvements in public and private healthcare quality.
Finance Application
- The paper's findings on reduced elderly mortality and hospitalizations due to public health interventions are highly relevant for insurance research.
- Health and life insurers could leverage these insights to refine actuarial models, leading to more accurate premium setting for life and health insurance products, especially for elderly populations in developing countries.
- For household finance, reduced health risks and out-of-pocket expenses could alter household savings, consumption, and debt decisions, influencing retirement planning and demand for annuities.
- Furthermore, the impact on longevity risk could inform the pricing of long-duration liabilities in pension funds.
Healthcare PolicyMortality RiskHealth InsuranceHousehold FinanceLongevity RiskEmerging MarketsCausal InferencePublic-Private PartnershipsActuarial ScienceConsumer Behavior
Core finding, identification, data
Core Finding
- The reform, which assigned an additional healthcare worker to village clinics, led to a 10% reduction in all-age mortality rates within two years, primarily driven by a decline in elderly (age 56+) deaths.
- This was achieved by simultaneously improving public sector quality and access, and by inducing private providers to increase their quality due to heightened competition.
- The reform is highly cost-effective, and optimal reallocation of workers could yield even greater mortality reductions.
Identification Strategy
- The study exploits quasi-experimental variation in the staggered rollout of healthcare workers (CHOs) in Rajasthan.
- Due to budget constraints, only two-thirds of eligible vacancies were filled in the first wave.
- Local government officials made ad-hoc assignment decisions based primarily on the CHO's place of residence and clinic locations, leading to quasi-random assignment conditional on clinic location.
- A matched difference-in-differences design with inverse probability weighting is used to compare treated and control subcenters.
Data
The paper uses large-scale administrative health data from India's Pregnancy, Child Tracking and Health Services Management System (PCTS) portal, household census data from the Community Health Integrated Platform (CHIP), original survey data on public and private providers and households, and socioeconomic data from the Socioeconomic High-Resolution Rural-Urban Geographic Platform for India (SHRUG).
Kory Kroft, Isaac Norwich, Matthew J. Notowidigdo, Stephen Tino — Labor Studies
This paper uses administrative data from Canada to show that obtaining permanent residency significantly increases immigrants' job mobility, earnings, and their sorting into higher-wage firms.
Finance Application
- The findings on increased earnings and job mobility post-PR have direct implications for household finance, particularly regarding wealth accumulation and credit risk.
- Immigrants gaining PR might exhibit improved credit scores, higher rates of homeownership, and increased participation in long-term investment vehicles like retirement accounts, as their human capital risk decreases.
- For asset pricing, changes in labor market fluidity and wage growth for a significant demographic could influence the pricing of human capital as an asset, potentially affecting consumption-saving decisions and demand for specific financial products.
- Insurers could also leverage these insights to refine risk models for life, disability, and unemployment insurance products tailored to immigrant populations, reflecting their altered employment stability and income trajectories.
labor economicsimmigrationpermanent residencyjob mobilityearningsfirm sortinghousehold financehuman capitallabor riskmatched employer-employee dataevent studycredit riskwealth accumulationinsurance
Core finding, identification, data
Core Finding
- Permanent residency leads to a sharp, immediate, and persistent increase of approximately 10 percentage points in job switching probability and a 7-8 percent increase in labor earnings after three years.
- A substantial portion (40-50 percent) of the earnings gain is attributed to immigrants sorting into higher-wage firms, with larger effects observed for lower-wage workers.
Identification Strategy
- The study employs an event-study design, exploiting variation in the timing of permanent residency (PR) acquisition among immigrants who initially arrived on temporary visas.
- It compares cohorts with similar temporary visa durations but different PR landing years, assuming common trends conditional on the time spent on a temporary visa.
Data
The research utilizes the Canadian Employer-Employee Dynamics Database (CEEDD), a comprehensive matched employer-employee dataset from Statistics Canada (2004-2019). This dataset links temporary and permanent visa records with individual-level demographic, earnings, industry, and firm financial information.
Ronak Jain, Samuel W. Stemper — Labor Studies
This paper causally estimates the impact of 3G mobile internet expansion on student test scores, technology use, and social well-being among adolescents across 82 countries between 2000 and 2018.
Finance Application
- The finding that 3G internet negatively impacts human capital, especially for disadvantaged groups, has significant implications for household finance and asset pricing.
- Lower human capital translates to reduced future earnings potential, which can exacerbate wealth inequality, impact household savings rates, and influence demand for credit or investment products.
- From an asset pricing perspective, a widespread decline in human capital could signal a long-term drag on aggregate productivity growth, affecting equity risk premiums and the valuation of human-capital-intensive firms, warranting an examination of sector-specific stock performance relative to 3G rollout timing.
Human CapitalTechnology AdoptionEducationInequalityHousehold FinanceAsset PricingProductivityBehavioral FinanceRisk Management
Core finding, identification, data
Core Finding
- The introduction of 3G coverage leads to substantial increases in smartphone ownership and internet usage among adolescents, resulting in significant declines in test scores in math, reading, and science, equivalent to a quarter-year of learning.
- These negative effects are more pronounced for students from disadvantaged backgrounds and are driven by exposure during adolescence, also reducing social connectedness and sense of belonging.
Identification Strategy
- The study employs a difference-in-differences framework, leveraging variation in the timing and geography of 3G coverage at the subnational (country-by-urbanicity) level.
- It includes country-by-urbanicity, country-by-year, and urbanicity-by-year fixed effects to isolate causal impacts, and validates findings with event studies (showing no pre-trends), permutation tests, and alternative DiD estimators.
- Instrumental variables (local lightning strike frequency and prior 2G coverage) are also used as a complementary approach.
Data
The paper uses student test scores from over 2.5 million students in 82 countries from the Programme for International Student Assessment (PISA) between 2000 and 2018. Geospatial data on 3G coverage comes from Collins Bartholomew (2007-2018), aggregated to country-by-urbanicity-by-year cells, combined with global population and urbanicity data from CIESIN and Global Human Settlement Layer.
Jordan D. Rosenthal-Kay — Development Economics
This paper measures urban costs globally using a structural model and geospatial data, finding that high urban costs in developing countries hinder economic development and climate change adaptation, and that road paving is a cost-effective policy.
Finance Application
- This research offers several arbitrage opportunities for finance.
- Real estate investors could use the estimated urban cost elasticities to identify cities with high growth potential (low costs) or hidden risks (high costs) for development and long-term investment.
- Insurance companies can integrate these urban cost metrics, especially their link to climate change adaptation, into their climate risk models to better price property and casualty insurance, or assess long-term liabilities in regions vulnerable to climate shocks.
- Furthermore, the methodology of using granular geospatial data to infer unobservable economic parameters could be adapted by financial institutions to create proprietary indicators for local economic health, housing supply constraints, or infrastructure quality, informing credit risk assessment for municipal bonds or mortgage portfolios.
Urban EconomicsReal EstateClimate RiskGeospatial DataEconomic DevelopmentInfrastructureHousehold FinanceAsset Pricing
Core finding, identification, data
Core Finding
- The paper finds that cities in developing countries exhibit urban cost elasticities 35% higher than rich world cities, driven by higher commuting costs and less elastic floorspace supply, leading to sprawl and steeper skyline slopes.
- Lowering these costs to US levels could raise welfare in developing nations by 66%, with a significant portion (20pp) from general equilibrium effects like labor reallocation.
- Furthermore, high urban costs significantly hinder a nation's ability to adapt to climate change, amplifying welfare losses from rising global temperatures.
Identification Strategy
- The paper employs a structural model of cities and an agricultural sector.
- It identifies commuting cost elasticity from city skyline slopes, floorspace supply elasticity using instrumental variables (model-generated fundamental productivity for wages, controlling for population density and geophysical covariates), and land supply elasticity from time-series variation in urban growth (area and income), instrumenting city GDP with DMSP-OLS nightlights.
Data
The study uses geospatial data from the Global Human Settlement Layer (GHSL) and Urban Centre Database (UCDB) for city boundaries, built volume, and population. It also leverages VIIRS and DMSP-OLS nighttime luminosity for GDP extrapolation, Google Earth Engine, European Space Agency WorldCover data, NASA Digital Elevation Model, and OpenLandMap for geophysical characteristics. Additional data include World Bank 'Doing Business' survey, World Bank ICP, LEHD-LODES, and Zillow ZHVI for validation in the US.
Neil Thakral, Linh T. Tô — Labor Studies
This paper formulates child penalty estimation as an age-period-cohort (APC) identification problem, clarifying the challenges of non-identifiability of linear trends in fixed effects and demonstrating biases from omitting cohort effects.
Finance Application
- The 'child penalty' directly impacts household income and wealth accumulation, influencing savings rates, investment decisions (e.g., human capital vs. financial assets), and demand for financial products like life insurance or college savings plans.
- The identified biases from omitted cohort effects suggest that previous studies on household financial decisions around childbirth might be mis-specified, leading to biased estimates of financial behavior changes.
- Understanding these penalties, especially across cohorts and demographics, is crucial for designing and pricing income protection or parental leave insurance products, and could even have aggregate implications for asset prices through shifts in consumption and investment patterns.
Child penaltyAge-Period-Cohort modelIdentificationFixed effectsLabor economicsHousehold financeEarningsGender inequalityEvent studyOmitted variable bias
Core finding, identification, data
Core Finding
- The fundamental challenge in child penalty estimation is the non-identifiability of linear trends in age, period, and cohort fixed effects.
- Omitting cohort fixed effects, a common practice, leads to substantial and varying biases in estimated child penalties and life-cycle profiles, with the null hypothesis of zero cohort effects being strongly rejected for both women and men.
Identification Strategy
- The paper frames child penalty estimation as an Age-Period-Cohort (APC) identification problem, characterizing the number of restrictions needed for identification, showing that one additional restriction suffices for consecutive cohorts.
- It highlights that individual fixed effects introduce two independent linear trend indeterminacies and demonstrates that period effects (child penalties) are point-identified if a slope restriction on them is correctly specified, even if other restrictions are misspecified.
Data
The study utilizes large U.S. samples from the CPS Basic Monthly files and March supplements (1968–2020), the ACS (2000–2019), the PSID (1968–2019), and the NLSY (1979–2018), combining repeated cross-sections and longitudinal panel data.
Marco Palladino, Antoine Bertheau, Alexander Hijzen, Astrid Kunze, Cesar Barreto, Doğan Gülümser, Marta Lachowska, Anne Sophie Lassen, Salvatore Lattanzio, Benjamin Lochner, Stefano Lombardi, Jordy Meekes, Balazs Murakozy, Oskar Skans — Labor Studies
This paper quantifies the role of gender-specific firm wage premiums in explaining private-sector gender wage gaps across 11 countries, decomposing the gap into sorting (women working at lower-paying firms) and pay-setting (women receiving lower premiums within the same firm) channels.
Finance Application
- The firm-level gender wage gap components (sorting and pay-setting) could inform ESG investing by identifying firms with higher labor-related risks or better gender equality practices, potentially impacting firm valuation and stock returns.
- In household finance, these insights can explain gender differences in savings and investment behavior, as women's sorting into lower-wage, part-time jobs and lower rent-sharing could lead to lower wealth accumulation.
- Insurance companies could use these granular firm-level wage dynamics to refine pricing for long-term care, disability, or retirement products, reflecting gender-specific career paths and income trajectories.
Gender wage gapFirm wage premiumsLabor economicsMatched employer-employee dataSortingPay-settingRent-sharingESG investingHousehold financeLabor riskInternational comparison
Core finding, identification, data
Core Finding
- Firm wage premiums account for a substantial share (15-32%) of gender wage gaps, with sorting being the predominant driver in most countries, increasing with age and linked to part-time work.
- Pay-setting gaps are largest in high-wage firms, where women receive only 89% of the rent-sharing benefits men receive from firm surplus gains.
Identification Strategy
- The study uses a harmonized research design across matched employer-employee datasets, applying the Kitagawa-Oaxaca-Blinder decomposition to gender-specific Abowd, Kramarz and Margolis (AKM) two-way fixed effects models.
- Firm effects are normalized using either firm productivity data (identifying low-surplus firms) or worker exit rates (identifying low-rank firms) to enable cross-gender comparisons.
Data
The paper uses harmonized matched employer-employee administrative data from 10 European countries (Denmark, Finland, France, Germany, Hungary, Italy, the Netherlands, Norway, Portugal, and Sweden) and Washington State, USA, covering mostly the period 2010–2019, including high-quality information on hourly wages and work hours.
Randi Hjalmarsson, Louis-Pierre Lepage, Matthew J. Lindquist, Conrad Miller — Labor Studies
This paper investigates the causal impact of a criminal record on labor market outcomes in Sweden, analyzing how it affects employment, earnings, and job sorting across firms with varying hiring propensities for individuals with criminal records.
Finance Application
- This research offers significant insights for household finance by exploring how criminal records impact access to credit, insurance, and wealth accumulation, potentially leading to persistent financial exclusion.
- For corporate finance and asset pricing, a firm's 'WCR-friendliness' (propensity to hire individuals with criminal records) could be developed as a novel ESG factor, influencing firm valuation, cost of capital, or investor sentiment.
- Managerial characteristics, such as prior exposure to individuals with criminal records, could be explored as a driver of firm-level ESG performance or labor-related risk, which might be priced in equity markets.
criminal recordslabor marketsjob sortingfirm heterogeneitycausal inferenceevent studySwedenhousehold financeESGcorporate governanceinsurancecredit accessmanagerial bias
Core finding, identification, data
Core Finding
- A criminal record causally reduces months employed by 2% and annual earnings by 5%, with larger effects for more serious or subsequent charges.
- This is largely due to sorting away from firms less willing to hire individuals with criminal records and lower monthly wages across all firm types.
- Manager exposure to individuals with criminal records significantly increases a firm's likelihood to hire such workers without detectable productivity loss.
Identification Strategy
- The causal effect is identified using an event study design that compares labor market outcomes for adults charged with an offense for the first time to a matched group of adults suspected of a similar offense but not charged.
- This leverages the fact that only charged individuals acquire a visible criminal record in Sweden, providing a credible counterfactual for the labor market trajectory absent the record.
Data
The study uses comprehensive Swedish register data from 1990-2015, including criminal convictions (Convictions Register, Suspects Register), employer-employee matched data (RAMS, Company Register), demographic information, and a novel dataset of Swedish job advertisements (Platsbanken) to identify firms signaling background checks. O*NET data is used for occupational characteristics.
Atila Abdulkadiroglu, Parag A. Pathak, Christopher R. Walters — Labor Studies
This paper compares the effectiveness of student reallocation policies versus school resource augmentation policies (like closures) in improving student achievement in New York City high schools.
Finance Application
- This research highlights that capacity constraints in high-quality assets (schools) limit the effectiveness of reallocating existing resources (students).
- In finance, this could inform studies on how limited supply of high-performing investment vehicles (e.g., ESG funds, alternative assets, or specific active managers) affects portfolio optimization and investor outcomes.
- The finding that simple metrics (like graduation rates) effectively identify underperforming schools suggests that basic financial ratios or past performance metrics might be sufficient for identifying underperforming stocks or funds for divestment, rather than relying on complex, costly quantitative models.
- This could inform research on the value of simple heuristics in investment decision-making and the impact of supply-side interventions (e.g., creating new high-quality funds) versus demand-side interventions (e.g., improving investor-fund matching platforms).
education economicsresource allocationschool choicematching marketscausal inferenceprogram evaluationcapacity constraintsasset allocationportfolio managementfund selectionperformance measurement
Core finding, identification, data
Core Finding
- The study finds that while sophisticated student reallocation policies can yield modest educational gains, they are limited by the scarcity of highly effective schools.
- Resource augmentation policies, such as closing low-performing schools and reallocating capacity, achieve comparable improvements with less systemic disruption, primarily because school quality differences are predominantly vertical (overall quality) rather than horizontal (student-specific match effects).
Identification Strategy
- The paper employs a quasi-experimental approach by estimating school effects using both OLS value-added models and two-step rank-ordered control function models, which leverage students' rank-ordered preference lists to control for unobserved tastes and selection bias.
- These estimates are then used in counterfactual simulations of various reallocation and resource augmentation policies, with school parameters refined using an empirical Bayes procedure.
Data
The study uses administrative data from the New York City Department of Education for students enrolled in NYC public schools from 2003-04 to 2012-13, focusing on 8th graders and their high school outcomes, including Regents math scores, high school graduation, and college attendance.
Katherine Richard, Lea J. Bart — Public Economics
This paper quantifies the persistent and far-reaching consequences of increased work requirement sanction severity in the U.S. safety net on program participation, labor supply, and household financial resources for vulnerable families.
Finance Application
- This research offers direct insights for household finance by quantifying the impact of exogenous income and safety net shocks on vulnerable families.
- Researchers could model how these severe and persistent financial resource reductions affect household consumption, savings, debt accumulation (e.g., demand for high-interest credit, payday loans), and default rates on consumer loans or mortgages.
- For asset pricing, the localized and aggregate effects of reduced consumer spending and increased economic instability in affected communities could be linked to the performance of local businesses, real estate markets, and the credit risk of financial institutions operating in these areas.
- The decline in Medicaid enrollment also highlights an increased exposure to health-related financial shocks, impacting demand for private insurance or medical debt.
Household FinanceIncome ShocksPovertySocial Safety NetLabor SupplyConsumptionCredit RiskAdministrative DataPolicy ReformDifference-in-DifferencesEconomic InstabilityInsurance Demand
Core finding, identification, data
Core Finding
The study finds that harsher work requirement penalties in the Temporary Assistance for Needy Families (TANF) program lead to persistent declines in TANF, SNAP, and Medicaid enrollment for beneficiaries and their households, a decrease in formal employment due to reduced job entry, and a 73% larger reduction in cumulative financial resources over two years, which is not offset by increased labor market earnings.
Identification Strategy
- The paper leverages a natural experiment from a Michigan policy reform in October 2011 that increased the duration of TANF work sanctions for second and third violations.
- It uses a difference-in-differences model to compare outcomes of individuals sanctioned before vs. after the reform, and a triple difference model with a control group of first-time sanctioned individuals to account for changing economic conditions and anticipation effects.
Data
The paper uses a novel micro-data panel with administrative records from the Michigan Department of Health and Human Services (MDHHS) for all individuals enrolled in Michigan TANF between 2009-2019, combining monthly TANF, SNAP, and Medicaid enrollment records with quarterly Unemployment Insurance earnings records and rich demographics.
Andreas R. Kostøl, Matthew Merkle, Andreas Myhre, Mark R. Whitmeyer — Public Economics
This paper demonstrates how hours constraints influence workers' labor supply responses to annual tax incentives, leading to mid-year work stoppages and affecting participation elasticities.
Finance Application
- The finding that hours constraints significantly affect labor supply and income streams has direct implications for household finance, as it suggests greater income volatility and less flexibility in adjusting to financial shocks.
- This could inform models of household consumption-saving decisions, potentially leading to higher demand for precautionary savings or liquidity.
- In asset pricing, the aggregate effect of widespread hours constraints could influence the elasticity of aggregate labor income to economic conditions, thereby impacting the labor income risk premium and the cross-section of asset returns.
- For insurance, the observed mid-year work stoppages due to tax incentives highlight a specific type of income interruption that could be addressed by tailored income protection products, especially for workers in less flexible employment settings.
labor supplyhours constraintstax incentivesincome elasticityhousehold financeasset pricingincome riskconsumption smoothingdisability insurancelabor economicspublic finance
Core finding, identification, data
Core Finding
- The study finds that hours constraints induce workers to stop working mid-year as their cumulative taxable income crosses into a new tax bracket.
- It estimates a labor force participation elasticity of 0.8 among marginally attached workers, with at least 60% of these workers being hours-constrained.
- This hours mismatch helps reconcile macro and micro evidence on intertemporal participation elasticities, with the intertemporal elasticity of substitution of employment reaching 1.3 when all workers are hours-constrained.
Identification Strategy
- The identification strategy relies on a novel empirical approach that non-parametrically identifies the intertemporal extensive margin labor supply response from monthly employment data.
- It uses annual kinks in the Norwegian tax and transfer system, particularly a large kink in the disability insurance (DI) system.
- The core method involves constructing a participation probability function (PPF) based on year-to-date cumulative income and comparing the observed PPF around the kink to a counterfactual PPF obtained by extrapolating from data to the left of the kink.
Data
The paper utilizes Norwegian monthly employment data, which includes detailed information on earnings, contracted hours, hourly wages, and bonuses. This data is linked to monthly files on DI recipients (providing information on DI benefits and pre-DI earnings) and administrative data from Statistics Norway (educational attainment, annual earnings, and cash transfers from tax registers). The analysis focuses on data from 2015-2017.
Neil Thakral, Linh T. Tô — Gender in the Economy
This paper formulates child penalty estimation as an age-period-cohort (APC) identification problem, demonstrating that omitting cohort fixed effects leads to substantial biases in estimated child penalties and life-cycle profiles.
Finance Application
- The APC identification problem and the biases from omitted cohort effects could be highly relevant in household finance when studying long-term financial decisions, savings, or investment behaviors across generations.
- For instance, analyzing the impact of birth cohorts on retirement savings, intergenerational wealth transfers, or housing market participation, where age, calendar time, and birth cohort effects are intertwined, could reveal significant biases if cohort effects are ignored.
- Similarly, in insurance, understanding how different birth cohorts face varying health or longevity risks over time could lead to mispricing of life or health insurance products if cohort-specific trends are not properly identified and separated from age and period effects.
age-period-cohortchild penaltyidentificationfixed effectsomitted variable biasevent studylabor economicshousehold financeintergenerational wealthinsurancelong-term trends
Core finding, identification, data
Core Finding
- The study finds that the common practice of omitting cohort fixed effects in child penalty estimation introduces significant biases, which vary depending on the data structure.
- When cohort effects are properly accounted for, estimated child penalties for women are moderately reduced, and for men, they are nearly halved.
- The null hypothesis of zero cohort effects is strongly rejected, indicating substantial and systematic heterogeneity across birth cohorts.
Identification Strategy
- The paper frames child penalty estimation as an age-period-cohort (APC) identification problem, highlighting the multicollinearity between age, period (event time), and cohort (age at first birth).
- It shows that identification requires additional linear restrictions beyond standard normalizations, with the number of restrictions depending on the greatest common divisor of cohort gaps.
- The authors empirically test and reject the common practice of omitting cohort fixed effects, demonstrating its biasing effects.
Data
The empirical analysis uses a large U.S. sample combining the CPS Basic Monthly files and March supplements (1968–2020) with the ACS (2000–2019). It also utilizes longitudinal panel data from the Panel Study of Income Dynamics (PSID) (1968–2019) and the National Longitudinal Survey of Youth (NLSY) (1979–2018).
Lena E. Hensvik, Peter Fredriksson, Doğan Gülümser — Gender in the Economy
This paper investigates how outside job opportunities differentially affect men's and women's wages and job mobility, finding that men's wages increase while women's do not, contributing to the gender pay gap, primarily due to differences in wage renegotiation behavior.
Finance Application
- The finding that women are less likely to negotiate wages, even when presented with outside options, could translate to financial negotiations.
- For instance, women might be less likely to negotiate interest rates on loans (mortgages, credit cards), investment fees, or insurance premiums, potentially leading to higher financial costs or lower returns for women.
- This could contribute to a gender wealth gap, and researchers could investigate gender differences in negotiation outcomes for various financial products.
- Furthermore, if women's human capital is less responsive to external opportunities in terms of wage growth, this could imply a different valuation of human capital in financial planning models or for lenders assessing creditworthiness, affecting optimal asset allocation or credit access for women.
Gender DifferencesLabor EconomicsNegotiation BehaviorWage GapJob MobilityOutside OptionsHousehold FinanceBehavioral FinanceHuman CapitalInsurance MarketsSwedenRegister Data
Core finding, identification, data
Core Finding
- The core finding is that improved outside job opportunities (proxied by family networks) lead to higher within-job wage growth for men but not for women in Sweden.
- This differential response is attributed to women's lower propensity to renegotiate wages, rather than differences in the quality of external offers or job mobility, as both genders show similar mobility responses to these opportunities.
Identification Strategy
- The identification strategy leverages quasi-experimental variation in access to information about external job opportunities through family networks (siblings and parents).
- This approach ensures that the quality of external offers is balanced across genders, allowing the authors to isolate the impact of gender-specific responses to these opportunities.
- They use fixed effects (worker-by-establishment, establishment-by-year, occupation-by-time) to control for various confounding factors.
Data
The study uses comprehensive Swedish register data from 2006-2018, including linked employer-employee data, demographic information (LOUISE register), firm and establishment registers (FtgAst), wage and occupation data (WSS), and multi-generational population-wide birth records (Flergenerationsregistret) to establish family connections.
Price V. Fishback, Christopher Vickers, Yiyu Xing, Nicolas L. Ziebarth — Gender in the Economy
This paper investigates the labor market effects of the staggered repeal of gender-specific workweek restrictions (GHRs) in US states during the 1960s and 1970s on both men and women.
Finance Application
- The findings on spousal labor supply complementarities after policy changes could inform household finance models of joint labor supply and its impact on household savings, investment decisions, and retirement planning.
- The observed shifts in labor income (hours and wages) due to regulatory changes provide a natural experiment to quantify labor income risk and evaluate the effectiveness of various insurance products or household self-insurance strategies.
- Furthermore, understanding how gender-specific labor regulations affect human capital retention and labor costs could be relevant for firm valuation and ESG investing, particularly in assessing the 'social' pillar related to gender equality and labor practices.
Labor economicsGender economicsPolicy evaluationDifference-in-differencesHousehold financeLabor supplyWagesHours workedHuman capitalRegulatory riskSpousal effectsIncome riskInsurance marketsFirm valuationESG investingHistorical data
Core finding, identification, data
Core Finding
- The repeal of GHRs led to similar labor market effects for both men and women, including an increase in the average workweek length, fewer people leaving previously affected industries, and a fall in both hourly and annual earnings.
- While effects for women are consistent with an expanded female labor supply, the observed decline in male earnings alongside increased hours suggests that the repeal also expanded male labor supply, likely due to complementarities in spousal leisure choices.
Identification Strategy
- The study employs a difference-in-differences (DiD) framework, exploiting the staggered state-level repeals of GHRs across US states.
- It uses event-study specifications to analyze dynamic effects and controls for individual demographics, state-by-occupation-by-industry fixed effects, year-by-occupation-by-industry fixed effects, and other state-level policies like fair employment and overtime laws, focusing on manufacturing and mercantile sectors.
Data
The paper primarily uses data from the March Current Population Survey (CPS) from 1962 to 1980. It also incorporates historical information on state-level maximum hours standards for women, Fair Employment Practices (FEP) laws, and overtime compensation laws.
Andrea Di Giovan Paolo, Giacomo Marcolin — Gender in the Economy
This paper uses the staggered implementation of the Pregnancy Discrimination Act (PDA) of 1978 to identify and quantify the role of fertility-related concerns in driving gender discrimination in labor markets.
Finance Application
- This research offers insights for the insurance industry, particularly in developing Employment Practices Liability Insurance (EPLI) products.
- The finding that anticipated future costs, even with weak enforcement, drive employer behavior suggests that EPLI premiums could be dynamically priced based on the perceived stringency and enforcement of state-level labor laws, rather than just historical litigation rates.
- For asset pricing, this could inform sector-specific factor models; industries with a high proportion of fertile-age women might exhibit different risk premia or stock price reactions to changes in labor regulations, as these policies directly impact their labor costs and human capital management.
- In household finance, the documented negative impact on women's employment and hiring due to such policies highlights a significant source of income risk that could influence optimal household savings, investment, and human capital accumulation strategies for women in their childbearing years.
Labor EconomicsGender DiscriminationEmployment ProtectionPregnancy Discrimination Act (PDA)Quasi-Experimental DesignDifference-in-DifferencesHiring CostsFertilityHuman CapitalLabor Market FrictionsPublic PolicyESGInsuranceCorporate FinanceHousehold Finance
Core finding, identification, data
Core Finding
- The PDA, by increasing the expected costs of firing pregnant workers, led to a significant decline in employment and hiring of fertile-age women (4.5-8.6 percentage points) but had no significant effect on firing rates of pregnant women or women's wages.
- This suggests weak enforcement of firing protection, prompting employers to reduce hiring to avoid anticipated future costs, and that fertility-related concerns explained a substantial portion (29.3% overall, 78.1% for non-mothers) of the gender employment gap.
Identification Strategy
- The paper employs a quasi-experimental difference-in-differences (DID) design, exploiting the staggered enactment of similar pregnancy discrimination policies by some US states prior to the federal PDA in 1978.
- States that adopted such policies earlier serve as a comparison group (PDA-control) for states only 'treated' by the federal PDA in 1978 (PDA-treated), allowing for a comparison of outcomes for fertile-age women before and after the federal law's passage, relative to the control group's trend.
Data
The paper primarily uses individual-level survey data from the Current Population Survey (CPS) Annual Social and Economic Supplement (ASEC) and the Panel Study of Income Dynamics (PSID) for the period 1973-1984. It also compiles a novel dataset on US states' pregnancy-discrimination policies passed prior to 1978.
James Albrecht, Per-Anders Edin, Raquel Fernández, Jiwon Lee, Peter Thoursie, Susan Vroman — Gender in the Economy
This paper develops and estimates a structural model of household parental leave decisions in Sweden, analyzing the interplay between economic incentives and evolving cultural norms following the 2002 "daddy month" reform.
Finance Application
- The paper's emphasis on social norms as a driver of labor supply and human capital decisions offers a rich avenue for household finance research, particularly in modeling consumption, savings, and investment choices over the life cycle.
- Evolving gender norms and wage penalties for parental leave could significantly alter intra-household financial allocations, risk-taking, and wealth accumulation patterns.
- For insurance, these insights could refine actuarial models for income protection and disability insurance, as the changing career trajectories and income stability of men and women due to parental leave and social norms directly impact risk assessment and pricing.
Household FinanceSocial NormsParental LeaveGender EqualityLabor SupplyWage PenaltiesHuman CapitalConsumptionSavingsInsuranceStructural EstimationRegression Discontinuity
Core finding, identification, data
Core Finding
- Endogenously evolving cultural norms play a major role in parental leave uptake, significantly increasing men's leave and decreasing women's.
- While direct economic incentives from the reform and changes in wage penalties have some effect, cultural changes, amplified by income-related changes, are crucial for explaining shifts in parental leave behavior.
- Counterfactual policies show that only policies directly mandating non-transferable leave have a significant impact on men's share of parental leave, largely through their effect on social norms.
Identification Strategy
- The paper develops a structural choice model of a unitary household, estimated using the method of simulated moments to match 56 key empirical moments (mean/SD of leave, zero leave percentage, share taken by men, wage ratio correlation) for pre- and post-reform periods.
- It also employs a regression discontinuity design around the 2002 "daddy month" reform to examine immediate causal effects, comparing model-derived implications with RD estimates.
Data
The study utilizes comprehensive Swedish administrative data from the Social Insurance Agency (Parental Leave Registers), the Multi-Generation Register, and LISA (Longitudinal Integration Database for Sickness Insurance and Labor Market Studies). These sources provide detailed information on parental leave spells, births, parents' education levels, annual earnings, and other demographic characteristics, supplemented by Wage Structure Statistics for monthly earnings.
Kanal Mangal, Niharika Singh — Gender in the Economy
This paper examines gender dynamics in entry into highly competitive Indian civil service jobs, focusing on participation, performance, and persistence over multiple exam cycles in Tamil Nadu, India.
Finance Application
- The paper's findings on gender-specific persistence costs due to marriage market pressures could inform household finance models of human capital investment, explaining gender gaps in long-term career paths requiring sustained effort, such as becoming a CFA or a specialized medical professional.
- This mechanism could also explain gender differences in long-term savings and investment in risky assets, as women facing earlier marriage pressures might have shorter perceived investment horizons.
- Furthermore, the experimental design could be adapted to study how information about career progression and social norms affects financial planning for retirement or investment in entrepreneurial ventures among different demographic groups.
Household FinanceHuman CapitalGender GapLabor EconomicsSocial NormsPersistenceDecision MakingMarriage MarketCareer ChoiceBehavioral EconomicsField Experiment
Core finding, identification, data
Core Finding
- Despite outperforming men in initial attempts, women are less likely to persist in exam-taking over time and are underrepresented among top-scorers in the Indian civil services.
- This gender gap in persistence and top-scorer representation is primarily driven by marriage market pressures and limited household support, which constrain women's ability to sustain long-term effort and repeated attempts in these highly competitive careers.
Identification Strategy
- The study uses unique administrative data to track a cohort of first-time applicants over multiple exam cycles, providing a pipeline analysis of participation and performance.
- It supplements this with survey data and an information experiment, where respondents were randomly informed about the difficulty of success and cross-randomized with a 'household prime' to assess how marriage market timing and household obligations influence persistence decisions.
Data
The paper utilizes unique administrative data from the Tamil Nadu Public Service Commission (TNPSC) covering all applicants to civil service jobs from 2012-19, including applications, test scores, and placements. It also uses original survey data from active candidates preparing for these exams in 2022.
Iacopo Morchio, Christian Moser — Gender in the Economy
This paper develops an equilibrium search model with endogenous firm pay, amenities, and hiring, using linked employer-employee data from Brazil to analyze the micro sources and macro consequences of the gender pay gap.
Finance Application
- This paper's insights into compensating differentials and gender-specific amenity valuations offer several avenues for finance research.
- In asset pricing, firms with superior amenity profiles (e.g., work-life balance, parental leave policies) might attract and retain higher-quality human capital, potentially leading to more stable earnings, lower operational risk, and thus a 'human capital premium' or better ESG scores reflected in stock valuations.
- For household finance, understanding how amenity preferences influence women's labor supply and wage choices can inform models of household savings, investment decisions, and debt accumulation, especially regarding gender-specific financial planning needs.
- In corporate finance, firms' strategic choices between offering higher wages versus better amenities could be linked to capital allocation decisions, long-term growth strategies, and even M&A activity, as different human capital strategies might appeal to different investor bases.
Gender Pay GapCompensating DifferentialsLabor Market SearchFirm HeterogeneityAmenitiesEmployer-Employee DataEquilibrium ModelPolicy EvaluationHuman CapitalESGHousehold FinanceCorporate FinanceAsset Pricing
Core finding, identification, data
Core Finding
- The study reveals that a significant gender pay gap in Brazil is largely driven by women sorting into lower-paying employers.
- Compensating differentials, where women prioritize non-pecuniary job amenities (e.g., flexibility) over higher pay, explain approximately half of this gap.
- Counterfactual policies aimed at equal treatment, while reducing pay gaps, are found to decrease overall output and worker welfare due to adverse incentive effects on firms' decisions.
Identification Strategy
- The model parameters are identified through a constructive proof using linked employer-employee data.
- This involves: 1) interpreting gender-specific employer fixed effects from an AKM-like wage equation, 2) estimating gender-specific employer ranks based on revealed preferences and firm size distributions, 3) identifying labor market objects (job offer and separation rates) from observed worker flows, 4) recovering firm-level parameters (productivity, gender wedges, amenity valuations) by inverting equilibrium recruiting intensities and firm profits, and 5) pinning down economy-wide elasticities for vacancy and amenity costs using aggregate labor share and pay-profit gradients.
Data
The primary data source is the Brazilian linked employer-employee register Relação Anual de Informações Sociais (RAIS) for the period 2007-2014. This dataset provides detailed information on workers (gender, education, age, tenure, work hours, parental leaves) and establishments (identifiers, earnings, sector, occupation). Firm financial data from Bureau van Dijk's Orbis Historical database is also used to measure firm productivity.
Carolina Arteaga, Gustavo J. Bobonis, Paola Salardi, Dario Toman — Gender in the Economy
This paper studies the impact of specialized domestic violence courts (SDVCs) in Puerto Rico on judicial protection for victims of intimate partner violence (IPV) and offender recidivism, leveraging their staggered implementation.
Finance Application
- This paper's findings on the impact of legal interventions on victim protection and recidivism could be highly relevant for household finance and insurance research.
- In household finance, researchers could investigate how access to SDVCs affects victims' financial stability, credit scores, employment, and housing security, quantifying the economic benefits of such legal support.
- For insurance, the reduction in IPV recidivism could translate to lower claims for health insurance (due to reduced physical/mental harm) and property insurance (due to reduced property damage), allowing for analysis of the financial value of social justice programs for insurers and policyholders.
- The mediation analysis on judge characteristics could inspire studies on the impact of decision-maker biases or training in financial institutions on consumer outcomes or compliance.
Household FinanceInsuranceLegal ReformSocial ImpactGender EconomicsCausal InferenceDifference-in-DifferencesGeographic DiscontinuityJudicial SystemVictim ProtectionRecidivism
Core finding, identification, data
Core Finding
- The introduction of SDVCs significantly increases the probability of judges issuing protection orders by 8 percentage points and reduces victim and offender reappearance rates within one year by 1.7 and 2.4 percentage points, respectively.
- These positive effects are more pronounced for cases involving children and in more remote areas, and are largely driven by the assignment of judges who prioritize victim-centered justice.
Identification Strategy
- The study employs a differences-in-differences (DiD) design, leveraging the staggered opening of SDVCs across judicial regions in Puerto Rico over time, using a fixed effect counterfactual (FECt) estimator.
- Additionally, a geographic discontinuity design (GDD) is used by comparing outcomes for petitioners residing just across judicial region boundaries with and without SDVCs, providing a local average treatment effect.
Data
The paper uses individual-level administrative micro-data from the Automated Protection Order System (APOS) of the Puerto Rico Judicial Branch (2014-2021), covering the universe of civil domestic violence cases. This is complemented by administrative data on judges and survey data from active judges (July-August 2019) on their training, perspectives, and priorities.
Andreas J. Beerli, Andrea Hofer, Ursina Schaede — Gender in the Economy
This paper investigates how an immigration-induced increase in full-time labor supply affects firms' demand for part-time workers and, consequently, maternal labor force participation.
Finance Application
- This research offers insights for asset pricing by suggesting that firms' labor market structures, particularly their 'coordination costs' for flexible work arrangements, could be a priced factor.
- Firms in industries with higher coordination costs might exhibit different sensitivities to labor supply shocks, impacting their profitability and stock returns.
- In household finance, the increased labor market risk for mothers due to reduced part-time job availability could influence household savings, debt decisions, and demand for income protection.
- Insurers could develop new products tailored to mitigate career interruption risks for caregivers in labor markets where firms are less willing to offer part-time roles.
Labor EconomicsPart-time WorkMaternal Labor Force ParticipationImmigrationFirm Labor DemandCoordination CostsChild PenaltyQuasi-ExperimentDifference-in-DifferencesHousehold FinanceAsset PricingInsurance
Core finding, identification, data
Core Finding
- An immigration reform that increased the supply of full-time workers in Swiss border regions led to a 6% drop in mothers' labor force participation.
- This is primarily driven by firms reducing their demand for part-time workers, suggesting that providing part-time jobs incurs coordination costs for firms, which they avoid when full-time labor is readily available.
Identification Strategy
- The study employs a difference-in-differences (DiD) design, exploiting a gradual liberalization of access to the Swiss labor market for cross-border workers (CBW) in the early 2000s.
- This created a localized, exogenous shock to full-time labor supply in municipalities within 15 driving minutes of the border (treated group) compared to those farther away (control group).
Data
The paper uses linked Swiss social security registers and Census data (1994-2010) for individual labor force participation and earnings, the Swiss Business Census (1995-2008) for establishment-level worker composition, the Swiss Job Market Monitor for job vacancy data, and the Swiss Earnings Structure Survey for hourly wages.
Lars Johannessen Kirkebøen, Edwin Leuven, Magne Mogstad, Jack Mountjoy — Economics of Education
This paper investigates the causal impact of college enrollment, particularly in elite programs, on assortative mating patterns, household formation, and economic outcomes in Norway.
Finance Application
- The findings on assortative mating and its impact on household income and wealth accumulation have direct implications for household finance, informing models of household consumption, saving, and investment decisions, especially for high-net-worth households.
- The differential effects on fertility and marriage timing by gender and elite education could refine life-cycle financial planning models and intergenerational wealth transfer studies.
- For insurance, understanding how educational choices affect household income stability and composition could lead to more nuanced risk assessments and product designs for life, disability, or long-term care insurance, potentially allowing for differentiated pricing based on educational background and assortative matching patterns.
Assortative MatingHuman CapitalHousehold FinanceEducationMarriage MarketIncome InequalityFertilityRegression DiscontinuityLabor EconomicsWealth AccumulationInsurance
Core finding, identification, data
Core Finding
- College enrollment itself, rather than pre-selection, is the dominant driver of assortative mating by institution and field of study.
- Elite professional programs significantly propel marginally admitted women into elite household formation, leading to higher own and partner earnings, but also delayed and reduced fertility.
- For men, elite admission has smaller own earnings gains and no partner or fertility effects.
Identification Strategy
- The study employs a fuzzy regression discontinuity design, exploiting admission discontinuities in Norway's centralized college application system.
- Applicants near unpredictable admission cutoffs are effectively randomized into different programs, allowing for the identification of causal effects of enrollment by comparing outcomes of applicants just above versus just below the admission threshold for their preferred program.
Data
The paper uses detailed Norwegian population register data, including specific education types (institution and field), labor market outcomes, marriage/cohabitation records, and application data from the Norwegian Universities and Colleges Admission Service for the years 1998-2004.
Sumit Agarwal, Jeffrey B. Liebman — Economics of Health
This paper investigates the causal effect of income on fertility using a randomized controlled trial of monthly cash transfers to low-income women.
Finance Application
- This research offers valuable insights for household finance, particularly regarding how income shocks influence long-term financial planning and consumption patterns related to family formation.
- Increased fertility due to cash transfers could drive demand for child-related financial products like college savings plans (529s), life insurance, or larger mortgages, impacting asset allocation decisions and the performance of consumer discretionary sectors.
- Furthermore, the intentional nature of fertility responses could inform models of household savings and debt accumulation, especially for low-income segments, and predict shifts in demand for financial advisory services or microfinance products.
household financeincome shocksfertilityrandomized controlled trialconsumptiondemographicssocial safety netsconsumer behaviorhealth economics
Core finding, identification, data
Core Finding
- A randomized monthly cash transfer significantly increased pregnancies and live births among low-income women, particularly multiparous women aged 30-35, suggesting that children are 'normal' goods for this population.
- The effect was driven by new, intentional pregnancies, evidenced by increased prenatal vitamin prescriptions and infertility evaluations, rather than changes in fecundity or abortion rates.
Identification Strategy
- The identification strategy relies on a randomized controlled trial (RCT) where low-income applicants in Chelsea, Massachusetts, were randomly assigned via a lottery to receive monthly cash benefits or be in a control group.
- The probability of winning varied based on socioeconomic criteria, allowing for a causal estimation of the pure income effect on fertility, isolated from other confounding factors.
Data
The paper uses detailed electronic health record data from three major health systems in the Boston area for 2019-2022, linked to lottery participants' application data. This includes comprehensive reproductive and obstetric care information, diagnoses, and visit types, providing granular insights beyond vital records.
Atila Abdulkadiroglu, Parag A. Pathak, Christopher R. Walters — Economics of Education
This paper compares the effectiveness of student reallocation policies (school matching) versus resource augmentation policies (school closures and new school creation) in improving student achievement in urban school systems, using data from New York City high schools.
Finance Application
- The paper's findings on the relative importance of 'vertical' (general quality) versus 'horizontal' (student-specific match) school effects could significantly refine models of housing market dynamics, where school quality is a key amenity.
- If general school quality is paramount, policies like school closures and new school creation would have more predictable and widespread impacts on property values and local tax bases than complex student matching algorithms.
- This also informs household finance by providing quantified impacts on human capital formation (achievement, graduation, college attendance), which can be integrated into models of lifetime earnings, savings behavior, and education investment decisions (e.g., 529 plans, private schooling choices).
- Furthermore, the analysis of school closures and their local economic effects could be used to assess the credit risk of municipal bonds issued by school districts or cities undergoing similar reforms.
Education EconomicsSchool ChoiceHuman CapitalHousehold FinanceUrban EconomicsPolicy EvaluationTreatment EffectsMatching MarketsHousing MarketsMunicipal Finance
Core finding, identification, data
Core Finding
- While sophisticated reallocation policies can generate modest educational gains, these are constrained by limited seats at highly effective schools.
- Resource augmentation policies, such as closing low-performing schools and reallocating capacity, achieve comparable improvements with less systemic disruption.
- This is because school quality differences are predominantly 'vertical' (benefiting all students) rather than 'horizontal' (student-specific match effects), making resource augmentation a more promising means of improving student outcomes.
Identification Strategy
- The paper estimates school effects using two strategies: an OLS Value-Added Model (VAM) assuming selection on observables, and a two-step rank-ordered control function model that leverages student preferences (rank-ordered lists for school choice) to control for unobserved student tastes.
- Counterfactual policy effects are then simulated by substituting empirical Bayes posterior mean predictions of average treatment effects and match effects into a school assignment process (student-proposing deferred acceptance algorithm for reallocation, and various closure rules for resource augmentation).
Data
The study uses administrative data from the New York City Department of Education (NYC DOE) for students enrolled from 2003-04 through 2012-13. This includes middle school demographics, test scores, high school enrollment, Regents test scores, high school graduation status, college attendance, and school application records (rank-ordered lists, priorities).
Barton Willage, James M. Flynn, Bethany I. Lemont — Economics of Health
This paper evaluates Louisiana's "modified-subscription system" for direct-acting antivirals (DAAs) to eliminate Hepatitis C, finding significant increases in diagnoses and treatment, reductions in mortality and liver transplants, and a positive marginal value of public funds.
Finance Application
- The paper's analysis of a subscription model for high-cost pharmaceuticals offers insights for insurance research, examining how such payment structures impact health insurer profitability, risk pooling, and premium setting, especially for chronic diseases.
- The significant reduction in long-term healthcare costs and mortality could inform asset pricing research on the valuation of regional healthcare providers, pharmaceutical companies, or even municipal bonds in states adopting similar public health initiatives.
- Furthermore, the calculation of the Marginal Value of Public Funds could be adapted to evaluate the financial efficiency of various financial regulations or social safety net programs, quantifying their long-term economic benefits and impact on household financial stability.
healthcare financeinsurancepublic financesubscription modelcausal inferencehealth economicssocial welfarecost-benefit analysispharmaceuticalshealth policy
Core finding, identification, data
Core Finding
- Louisiana's Hepatitis C Elimination Plan (LAHCEP), implemented in 2019, dramatically increased HCV diagnoses (e.g., 3,300% in 2020) and DAA prescriptions (e.g., 260% in 2019), leading to an estimated treatment of 67.4% of all patients in the state within four years.
- This intervention resulted in 11-13% reductions in HCV-related mortality and a 27% reduction in liver transplants, with the program more than paying for itself due to avoided medical costs.
Identification Strategy
- The study uses a synthetic control method (SCM) to create a "Synthetic Louisiana" as a counterfactual, along with event-study and difference-in-differences (DiD) models.
- Inference employs weighted least squares and Wild Cluster Bootstrap, with placebo tests and robustness checks to validate the causal effects.
Data
The paper uses data from the Centers for Disease Control and Prevention (HCV diagnoses), Medicaid's State Drug Utilization Data (DAA prescriptions), the National Vital Statistics System (mortality data), and the Scientific Registry of Transplant Recipients (liver transplant data).
Xinming Du, Lei Li, Eric Zou — Economics of Health
This paper reveals a cascading mechanism where international trade-induced deforestation leads to declining health outcomes in distant cities due to increased air pollution.
Finance Application
- This research provides a robust framework for quantifying telecoupled environmental and social risks, which is highly relevant for ESG investing and climate risk assessment.
- Financial institutions could use the 'area-of-effect' model to map physical climate risks (e.g., air pollution from deforestation) to specific geographic portfolios, informing real estate valuations, mortgage lending risks, and corporate bond pricing for firms in agricultural supply chains.
- Insurers could integrate these causal links into actuarial models to better price health and life insurance products, accounting for environmental externalities like transboundary air pollution exposure, potentially leading to new insurance offerings for climate-related health impacts or differentiated premiums based on a household's 'downwind risk' profile.
ESGClimate RiskHealth RiskSupply Chain FinanceInsuranceEnvironmental EconomicsDeforestationAir PollutionMortalityTradeCausal InferenceGeospatial AnalysisHousehold Finance
Core finding, identification, data
Core Finding
- The study finds that agricultural export shocks cause substantial local agricultural expansion and a one-for-one decline in forest cover.
- This deforestation, particularly the loss of natural filtration by forests, leads to increased air pollution and over 700,000 premature deaths in Brazil over two decades, predominantly from cardiovascular and respiratory causes, representing a $0.18 loss in statistical life value per $1 agricultural exports.
Identification Strategy
- The paper employs a two-stage identification strategy.
- First, it uses a shift-share research design, leveraging variation in regions' exposure to global demand shocks due to historical export capacity, to causally link agricultural exports to deforestation.
- Second, it develops an 'area-of-effect' (AoE) model that simulates how wind carries air pollutants across space and time, exploiting quasi-random variation in atmospheric connections to establish a causal link between upstream deforestation and downstream health outcomes.
Data
The paper utilizes Brazil's Comex database for export data, MapBiomas for land use, BDMEP and ERA5 for meteorological data (temperature, precipitation, wind), IEMA for air quality (PM2.5, PM10, O3, NO2, SO2, CO), NASA's FIRMS for forest fire activity, and SIM for mortality microdata.
Cyrus Aghamolla, Jash Jain, Richard T. Thakor — Economics of Health
This paper examines how hospital systems' transition to public equity markets via IPOs impacts their financial performance, operational decisions, and market power, and the implications for healthcare costs and quality.
Finance Application
- This research offers several avenues for finance applications.
- In **asset pricing**, the sustained profitability and market power gains post-IPO could be used to study long-term abnormal returns of healthcare IPOs, or how local market concentration (a non-financial factor) predicts cross-sectional stock returns and valuation multiples for publicly traded healthcare firms.
- For **household finance**, the finding that hospital IPOs lead to higher health insurance premiums (5.9% increase) directly impacts household budgets; researchers could analyze how these rising healthcare costs affect household savings, medical debt accumulation, and consumption decisions, particularly for vulnerable populations.
- In **insurance research**, the paper provides a clear mechanism for how hospital consolidation and pricing power affect health insurers; this could inform studies on insurer profitability, premium setting strategies, and the design of health plans (e.g., narrow networks) in response to evolving hospital market dynamics.
Healthcare financeIPOsPublic equity marketsHospital acquisitionsMarket powerHealth insurance premiumsProfitabilityCorporate financeDifference-in-differencesHealthcare costsPatient outcomesHousehold financeInsurance economicsAsset pricing
Core finding, identification, data
Core Finding
- Hospital systems going public experience dramatic and persistent increases in profitability, net income, and net patient revenues, driven by expanded capacity, equipment, and staffing.
- These systems use raised capital to accelerate strategic acquisitions of nearby hospitals, enhancing their regional market power and allowing them to demand higher reimbursement rates from insurers, thereby increasing prices for hospital services without a significant decline in care quality.
Identification Strategy
- The study employs a staggered difference-in-differences (DID) specification at the hospital-year level (Equation 1), comparing hospitals whose systems undertook an IPO within the past five years to a control group of privately-owned hospitals.
- Robustness is ensured through dynamic treatment effects (Callaway and Sant'Anna, 2021) and a propensity-score matched sample.
- For system-level acquisition analysis, a Poisson specification is used.
Data
The paper uses hospital-level financial and operational data from the HCRIS database (1997-2022), health outcomes and patient satisfaction data from CMS Hospital Compare and HCAHPS, IPO and financing data from Compustat, hospital M&A data from the Health Care Pricing Project (2001-2014), and firm-level health insurance premiums from Form 5500 filings.
Leila Agha, Na'ama Shenhav, Myles Wagner — Economics of Health
This paper examines the social costs of work disruptions caused by female physicians' childbirth on their patients' access to care and health outcomes, particularly for children.
Finance Application
- This research offers significant insights for the insurance industry, particularly health insurers.
- The 'scarring' effect on children's preventative care (vaccinations, lead testing) due to PCP disruptions implies higher future healthcare costs, which could inform health insurance premium setting, risk modeling, and network design.
- For asset pricing and ESG research, the findings suggest that healthcare systems or large medical groups with inadequate parental leave policies for physicians may face increased operational risks (e.g., patient attrition, lower quality of care, potential litigation) and reputational damage, impacting their valuation and cost of capital.
- Investors could integrate these 'social costs' into ESG frameworks to assess human capital management in the healthcare sector.
Health EconomicsLabor EconomicsParental LeaveWork-Life BalanceHealthcare AccessPatient OutcomesInsurance ClaimsESGHuman CapitalSocial CostsCausal InferenceHealth Insurance
Core finding, identification, data
Core Finding
- Female physicians experience an 85% reduction in billed visits with Medicaid patients in the quarter following childbirth, with a rebound thereafter.
- This disruption leads to persistent negative spillovers for child patients, including a 23% lower likelihood of seeing their usual physician for up to two years, a 50% reduction in vaccination claims, and a 42% reduction in lead testing.
- In contrast, adult patients experience only muted and transitory effects, suggesting children are more vulnerable to provider interruptions.
Identification Strategy
- The study employs a triple difference event study design.
- It compares the outcomes of mother and father primary care physicians (PCPs) and their patients before and after childbirth, relative to gender gaps among non-parent PCPs.
- This approach isolates the causal effect of new mothers' work interruptions from other time-varying changes or general trends among female physicians.
Data
The paper uses administrative insurance claims data from California's Medicaid program (Medi-Cal) from 2011-2019, linked with California Vital Statistics Birth Records Data (2007-2017) to identify physician childbirths. It also incorporates data from the California Physician License Survey (2010-2022) for physician work hours and characteristics.
Boaz Abramson, Pablo de Llanos Artero, Lu Han — Real Estate
This paper constructs a new micro-geographic repeat-rent index to estimate the impulse responses of rents to monetary policy shocks, finding that monetary tightening increases both real and nominal rents.
Finance Application
- This paper's findings have significant implications for real estate asset pricing, particularly for REITs and private real estate funds, as it suggests a counter-intuitive positive correlation between interest rate hikes and rental income, affecting cash flow projections and valuations.
- In household finance, the documented shift from homeownership to renting due to monetary tightening informs models of household portfolio choice, housing affordability, and the demand for various mortgage products.
- Furthermore, these dynamics could influence the performance and pricing of mortgage-backed securities (MBS) by impacting mortgage origination volumes and prepayment speeds.
Monetary PolicyRentsHousing MarketReal EstateInflationRepeat-Rent IndexLocal ProjectionsInstrumental VariablesHousehold FinanceAsset PricingREITsMortgage MarketHomeownership
Core finding, identification, data
Core Finding
- Contractionary monetary policy increases both real and nominal rents.
- Specifically, a 25 basis point increase in the 30-year fixed rate mortgage leads to a 1.7% (real) and 1.4% (nominal) increase in rents 12-24 months following the shock.
- This effect is primarily driven by a shift in household demand from the owner-occupied market to the rental market, which is accommodated by real-estate investors.
Identification Strategy
- The paper employs a local projection instrumental variable (LP-IV) framework.
- The instrumented monetary policy indicator is the 30-year fixed rate mortgage, and exogenous shocks are identified using the Bauer and Swanson (2023b) monetary policy surprises series, which are residuals from regressing high-frequency interest rate changes around FOMC meetings on pre-announcement economic and financial variables.
Data
The primary data source is rental listing data compiled by Altos Research (2011-2022) used to construct a new repeat-rent index. Housing transaction data from Corelogic (2000Q1-2024Q2) is used to analyze sales volume and composition, complemented by FRED data for mortgage rates and PCE inflation, and BLS data for county-level unemployment rates.
Cameron M. Ellis, Meghan I. Esson — Economics of Health
This paper investigates how private equity firms in the ambulance industry increase profits through 'cream-skimming' by strategically exploiting regulations and shifting high-cost patients to government-backed providers, leading to negative public health outcomes.
Finance Application
- This research offers crucial insights for asset pricing by highlighting how PE-driven profit maximization in regulated public-private markets can create significant negative externalities that are not typically priced.
- Investors in PE funds or publicly traded healthcare assets could use this to assess the sustainability of reported profits, particularly if they rely on cost-shifting rather than efficiency gains, potentially leading to mispricing.
- For insurance research, the findings directly impact health and life insurance markets; increased fatalities and shifted healthcare costs could lead to higher premiums or claims for health and life insurers, prompting a re-evaluation of risk models and regulatory frameworks for reimbursement in mixed public-private healthcare systems.
Private EquityHealthcareRegulationCream-SkimmingPublic GoodsExternalitiesAsset PricingInsuranceDifference-in-DifferencesOperating Profits
Core finding, identification, data
Core Finding
- Private equity (PE) ownership of ambulance companies increases operating profits by 50% primarily by reducing Advanced Life Support (ALS) runs and increasing Basic Life Support (BLS) runs.
- This is achieved by firing higher-paid paramedics and shifting ALS calls to local fire departments, resulting in an estimated 200 additional traffic fatalities in Arizona and a 7% increase nationally, with no evidence of improved operating efficiency.
Identification Strategy
- The study employs a staggered difference-in-differences framework, specifically the imputed DiD strategy by Borusyak et al. (2024), exploiting the staggered PE acquisitions of the two largest national private ambulance companies.
- Heterogeneous treatment effects are identified by analyzing firms' differential ability to cream-skim, measured by the geographic overlap of their Certificate of Necessity (CON) operating areas with government-operated ambulance services.
Data
The paper uses novel data from Arizona's ambulance industry, including detailed annual cost reports (ARCR) from 2007-2017 for all ambulance companies, geographic operating areas (CONs) from the Arizona Department of Health Services, and private equity ownership status from PitchBook. It also utilizes the Fatality Analysis Reporting System (FARS) for traffic accident fatalities in Arizona (2010-2017) and nationally (2008-2017), and Wayback Machine for national operating areas of Rural/Metro.
Sven Damen, Matthijs Korevaar, Stijn Van Nieuwerburgh — Real Estate
Residential properties with the lowest rent levels consistently provide the highest investment returns to their owners across the United States, Belgium, and The Netherlands, challenging traditional risk-return assumptions.
Finance Application
- This research provides a strong empirical basis for developing specialized real estate investment strategies or funds focused on affordable housing, potentially offering 'impact alpha' for ESG-conscious investors.
- The findings challenge the efficient market hypothesis in a specific asset class, suggesting that market frictions and segmentation can sustain abnormal returns, which could be explored in other illiquid asset markets.
- It also highlights how household financial constraints directly impact asset pricing and market structure, offering avenues for research into the interplay of micro-level constraints and macro-level asset returns.
Affordable HousingReal Estate ReturnsMarket SegmentationLimits to ArbitrageRisk PremiumHousing PolicyProperty InvestmentRental MarketESG InvestingAlpha
Core finding, identification, data
Core Finding
- Housing units in the low-rent tier of the rental market consistently earn significantly higher returns (an 'alpha') compared to properties in the high-rent tier, with annual total returns 1.74% to 4.08% higher.
- This alpha is not explained by systematic or regulatory risk, but rather by market segmentation and limits to arbitrage, including financial constraints of low-income renters and small landlords, and reputational risk/diseconomies of scale for large institutional investors.
Identification Strategy
- The paper uses a cross-country comparative approach (US, Belgium, Netherlands) to establish the robustness of its findings across diverse housing market settings and time periods.
- It employs detailed micro-data on rents, costs, and prices, and leverages state-level variations in regulatory risk (e.g., tenant protection indices constructed with GenAI) and ownership data to identify the mechanisms driving the return differential.
Data
The paper uses detailed micro-data from the United States (Fannie Mae, Freddie Mac, CredIQ, MSCI Real Capital Analytics, FBI UCR), Belgium (Federal Public Service Finance, EPC database, Koeter Vastgoed Adviseurs, EU-SILC), and The Netherlands (Statistics Netherlands Woonbase, Realstats, Koeter). These datasets provide information on rents, property values, costs, tenant characteristics, and investor types.
Gi Heung Kim — Real Estate
This paper empirically examines the equilibrium impacts of a proposed policy called "decoupling" in the US real estate market, which would require buyers and sellers to each pay their respective brokers.
Finance Application
- This structural model of intermediary incentives and market equilibrium could be directly applied to household finance to analyze the impact of financial advisor or mortgage broker compensation structures on consumer outcomes.
- For instance, investigating how commission-based vs. fee-only models for financial advisors affect investment product recommendations and client portfolio performance.
- In insurance, it could model how agent commissions influence policy sales and consumer welfare for complex products like annuities, providing insights for regulatory design.
real estatebroker incentivescommissionsdecouplingequilibrium modelconsumer welfarehousehold financefinancial intermediationmarket efficiencycompetition policy
Core finding, identification, data
Core Finding
- Decoupling reduces total commissions paid by 53%, as sellers no longer need to offer high commissions to attract buyers and brokers compete more intensely for price-sensitive buyers.
- This leads to a significant consumer welfare gain of 4% of the total transaction value, driven by sellers passing through commission savings to house prices, benefiting low-income, price-sensitive buyers disproportionately.
Identification Strategy
- The paper identifies key parameters using sellers' pre-determined loan characteristics (loan-to-value ratios, loan terms) at the time of their purchases as instruments for house prices and commissions.
- This exploits the fact that sellers with greater mortgage burdens are more likely to desire higher house prices and lower commissions, and is robust to controlling for market conditions and unobserved property/broker quality.
Data
The study uses transaction-level data from CoreLogic MLS (including list prices, property characteristics, brokerages, agents, and commission offers), supplemented with CoreLogic Deeds, Loan-Level Market Analytics (LLMA), and public Home Mortgage Disclosure Act (HMDA) data for financial and demographic information on sellers and buyers.
Heidi Artigue, Patrick Bayer, Fernando V. Ferreira, Stephen Ross — Real Estate
This paper examines the long-term impact of retaining versus losing one's home after a mortgage delinquency during the Great Recession on homeownership, consumption, and financial well-being.
Finance Application
- The finding that significant housing wealth accumulation for distressed homeowners does not translate into increased consumption or improved creditworthiness, largely due to restricted access to home equity lending post-crisis, has direct implications for household finance and mortgage markets.
- This suggests that the "wealth effect" from housing is highly conditional on credit availability for credit-constrained households, warranting models that incorporate heterogeneous access to HELOCs and cash-out refinancing.
- For asset pricing, the quasi-random nature of mortgage modifications offers a unique opportunity to causally identify the impact of such interventions on mortgage default rates and prepayment behavior, which could lead to more accurate pricing and risk management strategies for mortgage-backed securities (MBS) and other structured products.
HomeownershipGreat RecessionMortgage DelinquencyMortgage ModificationHousehold WealthConsumptionCreditworthinessNatural ExperimentQuasi-Experimental DesignHousing MarketFinancial CrisisCredit ConstraintsMortgage-Backed Securities
Core finding, identification, data
Core Finding
- Homeowners who received mortgage modifications during the Great Recession were significantly more likely to retain homeownership and accumulated an average of $83,000 more in housing wealth by 2022 compared to those who did not.
- Despite this substantial wealth difference, there was little long-term impact on creditworthiness, consumption, or neighborhood income, suggesting that home loss due to a severe economic shock does not necessarily lead to broader financial ruin.
Identification Strategy
- The study uses a quasi-experimental design by comparing seriously delinquent homeowners who received a mortgage modification between 2010-2013 with those who did not.
- This leverages the "chaotic modification approval process" during the Great Recession, which made the receipt of a modification "close to random" among delinquent borrowers, effectively serving as a natural experiment.
- The authors demonstrate highly similar pre-trends in financial outcomes between these treatment and control groups.
Data
The study utilizes a unique panel dataset of 380,000 individuals from seven major U.S. markets who originated mortgages between 2004-2008. This data combines HMDA and Dataquick Inc. records with individual credit reports collected at three-year intervals from 2004 to 2022, supplemented by Census tract, FHFA house price indices, and IRS income data for neighborhood characteristics.
Zoe B. Cullen, Bobak Pakzad-Hurson, Ricardo Perez-Truglia — Personnel Economics
This paper investigates how information frictions influence salary negotiation behavior and outcomes in the U.S. tech sector using field experiments and a theoretical model.
Finance Application
- This research offers insights for household finance by demonstrating how behavioral biases and information frictions (e.g., underestimating negotiation success, fear of backlash) lead to suboptimal wealth accumulation through forgone salary gains.
- Financial advisors could implement 'negotiation encouragement' interventions for clients, particularly young professionals in high-earning finance roles (e.g., private equity, hedge funds), to significantly boost their long-term human capital value and savings.
- In asset pricing, understanding these frictions in labor markets can inform models of human capital as an asset, where mispricing due to negotiation failures could affect firm valuations through labor costs.
- For insurance, the 'fear of backlash' could be framed as a perceived risk, suggesting a market for 'negotiation insurance' products that mitigate this risk, encouraging individuals to pursue higher compensation.
Labor EconomicsBehavioral EconomicsInformation FrictionsField ExperimentNegotiationCompensationWage GapHuman CapitalHousehold FinanceBehavioral FinanceExperimental EconomicsLabor MarketsRisk AversionWealth Accumulation
Core finding, identification, data
Core Finding
- Extensive-margin uncertainty (fear that employers are not open to negotiation) is a significant barrier, with a light-touch informational intervention increasing negotiation attempts by 7.3 percentage points and compensation gains by 5.7 percentage points.
- In contrast, intensive-margin uncertainty (lack of skills) is not a major barrier, as a heavily discounted coaching service had minimal impact.
- The theoretical model demonstrates that policies encouraging negotiation can enhance welfare and equity.
Identification Strategy
- The identification strategy relies on two randomized controlled field experiments.
- The 'encouragement treatment' involved a light-touch informational intervention providing factual data on negotiation rates and success, targeting extensive-margin uncertainty.
- The 'coaching treatment' offered a substantial discount (80%+) on a negotiation coaching service, targeting intensive-margin uncertainty.
- Causal effects are identified by comparing outcomes between treatment and control groups.
Data
The study utilizes survey and experimental data from over 3,100 active job seekers in the U.S. tech sector, collected in partnership with levels.fyi. This dataset includes detailed information on participants' backgrounds, negotiation attempts, and compensation outcomes.
Konhee Chang — Real Estate
This paper examines the impact of large-scale corporate single-family rental landlords on residential segregation, housing affordability, and household welfare in American suburbs.
Finance Application
- This research offers insights for real estate asset pricing by quantifying the "corporate landlord premium" (9% acquisition premium) driven by local scale economies, suggesting a need to model institutional investor-specific valuation factors.
- For household finance, the findings on financial constraints and welfare impacts of rental supply changes are crucial for understanding household portfolio decisions, particularly for wealth-constrained and minority groups, and for evaluating the long-term effects of housing policies on wealth accumulation and intergenerational mobility.
- The observed incumbent homeowner out-migration due to perceived disamenities could also inform models of neighborhood amenity valuation and its impact on property risk premiums for insurers and investors.
Real Estate FinanceHousehold FinanceAsset PricingUrban EconomicsResidential SegregationRental MarketsHomeownershipFinancial ConstraintsCorporate LandlordsSpatial EquilibriumWelfare AnalysisProperty Valuation
Core finding, identification, data
Core Finding
- Corporate landlord entry increases rental supply in suburbs, reducing residential segregation by enabling financially constrained, disproportionately non-White renters to access desirable neighborhoods.
- However, this reallocates owner-occupied homes to rentals, increasing home prices, which prices out middle-wealth homebuyers and leads to incumbent homeowners moving out due to perceived disamenities, ultimately hurting the median household's welfare.
Identification Strategy
- The study uses property-level event studies, exploiting variation in corporate SFR landlord acquisition timing and geographic clustering to identify causal effects on property outcomes.
- It employs GMM with corporate SFR landlord exposure as an instrument to estimate elasticities within a quantitative spatial equilibrium model, and a repeat-sales design to quantify landlords' willingness to pay for geographic concentration.
Data
The analysis relies on a newly constructed property-level panel dataset integrating housing deeds, assessor records, MLS rental listings, and Data Axle address histories for tenant demographics, alongside commercial mortgage data (TREPP), Survey of Consumer Finances (SCF), and American Community Survey (ACS) data.
Vincent Rollet — Real Estate
This paper analyzes how zoning regulations and the high fixed costs of redevelopment influence urban growth, housing supply, and affordability in New York City using a dynamic spatial equilibrium model.
Finance Application
- The paper's insights into the slow, heterogeneous impact of zoning and redevelopment costs could be crucial for real estate asset pricing, particularly for REITs or private equity funds.
- It highlights that long-term zoning outlook and the 'option value' of redevelopment significantly affect land and property valuations.
- In household finance, the findings on affordability and welfare gains, especially for lower-income households, could inform models of housing wealth accumulation, mortgage risk, and credit access, as local policy directly impacts household balance sheets and migration patterns.
- For insurance, the persistence of city structure due to high redevelopment costs could influence long-term property risk assessment and underwriting in urban areas, especially concerning climate change impacts in flood zones.
Urban EconomicsZoningReal EstateRedevelopmentHousing SupplyAffordabilitySpatial EquilibriumDynamic ModelsFixed CostsHousing PricesRentsAsset PricingHousehold FinanceReal Estate InvestmentUrban PlanningLocal PolicyMigrationProperty Insurance
Core finding, identification, data
Core Finding
- Zoning strongly constrains city growth, but the effects of relaxing regulations materialize slowly (decades) due to large fixed costs of redevelopment, which rise sharply with the size of existing buildings.
- Affordability benefits from zoning reform largely accrue to households outside the rezoned neighborhoods due to migration.
- Removing all zoning could increase floorspace by 58% but only moderately decrease residential rents (-17%) over 40 years.
Identification Strategy
- The study builds the first parcel-level panel dataset of NYC buildings, zoning, and floorspace prices.
- It estimates a dynamic spatial equilibrium model of floorspace supply and demand, validated using quasi-experimental variation from recent zoning reforms (difference-in-differences design for upzoning effects).
- Fixed costs are estimated via a nested fixed-point algorithm (extending Rust, 1987), and demand parameters are calibrated using observed rents, commuting flows, and agglomeration spillovers from large new building constructions.
Data
The paper uses a parcel-level panel dataset (2004-2022) combining property tax records, scraped online listings (StreetEasy), building permits, certificates of occupancy, real estate sales data (NYC Department of Finance), rent estimates (NOPVs), and annual zoning maps. Additional data includes smartphone data (Advan), commuting flows (LEHD-LODES), Google Maps commuting times, historical building permits, and historical sales data.
Nathaniel Baum-Snow, Abdollah Farhoodi, Lu Han — Real Estate
This paper quantifies the welfare costs of rent growth and the prospects for reducing these costs by estimating migration costs and demand systems for neighborhoods using a structural dynamic model and panel data on Texas renters.
Finance Application
- 1. **Household Finance**: The estimated moving costs and neighborhood substitution elasticities are critical for modeling household financial resilience.
- High moving costs imply renters are 'sticky' and more exposed to local rent shocks, impacting their budget constraints and potentially increasing default risk on other financial obligations (e.g., credit cards, auto loans).
- This could inform models of household consumption, savings, and debt accumulation under housing market frictions. 2. **Real Estate Asset Pricing**: Insights into renter mobility and demand elasticities can refine real estate investment strategies.
- Investors could use these to assess risk and return of rental properties in different submarkets, considering how renter stickiness or fluidity affects vacancy rates, rent growth potential, and tenant stability, leading to more nuanced pricing models for residential rental assets. 3. **Mortgage and Insurance Markets**: For mortgage lenders and housing-related insurers, the findings can improve risk assessment.
- If renters in certain areas are less mobile and more vulnerable to rent increases, this could affect landlords' ability to service mortgages or increase claims on landlord insurance policies, especially in low-income, high-rent-growth areas, leading to better underwriting models.
Household FinanceReal EstateRental MarketsMigration CostsHousing AffordabilityDynamic Structural ModelsInstrumental VariablesMachine LearningUrban EconomicsConsumer Mobility
Core finding, identification, data
Core Finding
- The paper quantifies substantial and heterogeneous migration costs for renters, including a fixed cost of about $40,000, which decreases with household income.
- Moving costs are significantly higher between than within commuting zones, and younger/lower-income individuals face lower moving costs.
- Rent growth disproportionately burdens low-rent neighborhoods in large metropolitan areas, exacerbated by these mobility frictions and varying substitution elasticities across neighborhoods.
Identification Strategy
- The paper employs a structural dynamic model of location choice and endogenous migration.
- A key identification strategy involves a novel shift-share instrumental variable for rents, constructed from the interaction of neighborhood-level housing supply elasticity and the overall trend in Texas rental prices, exploiting differential supply responses to common demand shocks.
- The estimation also uses a neural net machine learning algorithm for predicting migration decisions and a control function approach to address endogeneity.
Data
The study uses an individual-level panel dataset of residential locations for the near universe of renters in Texas from 2010-2019, combining migration histories from the Infutor marketing data set, housing tenure information from CoreLogic, Zillow Observed Rent Index (ZORI) data, and American Community Survey (ACS) micro data and aggregates.
Jing Cai, Sai Luo, Shing-Yi Wang — Personnel Economics
This paper uses a field experiment in a Chinese automobile manufacturing firm to compare the effects of financial incentives (signing bonuses) and increased monitoring on worker effort and retention.
Finance Application
- This research provides valuable insights for understanding human capital dynamics and firm valuation.
- In asset pricing, these findings could inform models of firm-specific human capital risk, particularly for industries reliant on worker effort and retention, impacting equity valuation and credit risk.
- For household finance, the differential effects of bonuses on retention and hours worked could be used to model how unexpected income or job security influences household consumption, savings, and debt decisions, especially for financially constrained individuals.
- In insurance, the trade-offs between monitoring and retention could help design better employment practices to mitigate operational risks related to labor turnover, informing underwriting for business interruption or key-person insurance.
labor economicsfield experimentincentivesmonitoringworker effortretentionhuman capitalmoral hazardgift exchangefirm performanceoperational riskasset pricinghousehold financeinsurance
Core finding, identification, data
Core Finding
- The study finds that both financial incentives and monitoring increase worker effort but through different channels and with contrasting retention outcomes.
- Signing bonuses lead to longer working hours and reduced quit rates but do not improve performance quality, consistent with a gift-exchange mechanism.
- In contrast, increased monitoring improves performance as evaluated by managers but significantly raises quit rates, suggesting workers dislike close oversight.
Identification Strategy
- The paper employs a randomized controlled trial design with two main experiments.
- The financial incentives experiment randomly assigned new hires to receive either a standard compensation package or a one-time signing bonus (pre-hire), with a second-stage randomization of a 'surprise bonus' to disentangle selection from moral hazard.
- The monitoring experiment randomly assigned production-line stations to either increased monitoring (additional visits by an independent team) or a control group with standard monitoring.
Data
The study utilizes administrative data from the firm (worker performance, earnings, hours, tenure, demographics), station-level data on monitoring visits, and survey data from baseline applications (worker characteristics, interviewer assessments) and endline follow-up surveys (work satisfaction, well-being, social networks).
Edoardo Di Porto, Christian Dustmann, Chiara Giannetto, Lorenzo Incoronato — Personnel Economics
This paper investigates how a 2014 Italian salary cap on public sector managers affected their career trajectories, the selection of talent, and overall public sector productivity.
Finance Application
- This research offers valuable insights for asset pricing by highlighting how regulatory interventions, such as wage caps, can degrade human capital and productivity in state-owned enterprises (SOEs).
- This could be a significant factor in assessing the credit risk and equity valuation of SOEs, as well as influencing sovereign risk premiums if public sector efficiency broadly impacts national economic performance.
- In household finance, the findings underscore how policy-induced career instability in the public sector could alter individuals' long-term financial planning, savings behavior, and demand for wealth management or insurance products.
- For insurance, the outflow of high-earning public managers could negatively impact the contribution base for public pension funds, increasing their solvency risk and potentially affecting the broader financial stability that insurers rely on.
Public Sector EconomicsLabor EconomicsHuman CapitalState-Owned EnterprisesManagerial IncentivesProductivityWage CapsTalent RetentionSovereign RiskCredit RiskHousehold FinancePension FundsPolicy Risk
Core finding, identification, data
Core Finding
- The salary cap led to a 10 percentage point increase in the likelihood of high-productivity public sector managers transitioning to private sector jobs, resulting in an estimated 2% decline in public management productivity for a mere 0.1% reduction in public employment costs.
- This talent drain also had negative spillover effects, increasing the likelihood of high-productivity co-workers leaving.
Identification Strategy
- The study employs a difference-in-differences strategy, comparing public-in-private managers earning above the €240,000 cap (treatment group) to a matched control group (earning €150,000-€240,000) before and after the 2014 reform.
- Matching is performed one-to-one within age-by-gender cells, and a placebo test using a 2009 reform date validates the empirical design.
Data
The paper utilizes unique matched employer-employee data from the Italian Social Security Institute (INPS), covering both private sector workers and public sector employees (specifically 'public-in-private' workers and broader public sector data from 2014). Managerial quality is proxied by worker fixed effects derived from AKM wage regressions.
Larissa Fuchs, Matthias Heinz, Pia Pinger, Max Thon — Personnel Economics
This paper uses a field experiment and a survey experiment to demonstrate how highlighting job flexibility and career advancement in job advertisements causally affects the size and composition of applicant pools and shapes young professionals' beliefs about the work environment.
Finance Application
- Financial firms, especially in competitive sectors like fintech or quantitative trading, could leverage these findings to optimize talent acquisition strategies for STEM professionals by tailoring job ad messaging.
- Emphasizing 'flexibility' (e.g., remote work, work-life balance) might attract a more diverse applicant pool, while 'career advancement' (e.g., rapid promotion, skill development) could target specific ambitious profiles.
- This signaling mechanism could also extend to how financial institutions communicate with investors or employees; for instance, subtle cues in ESG reports or employee benefit disclosures might shape investor perceptions of human capital management and long-term firm value.
- Furthermore, in household finance, understanding how different demographic segments respond to 'flexibility' versus 'growth' narratives could inform the marketing of investment or insurance products.
human capitaltalent acquisitionjob advertisementsfield experimentRCTsignalingbelief updatinggender differenceswork-life balancecareer developmentSTEM talentlabor economicsorganizational behaviorfintechESG
Core finding, identification, data
Core Finding
- Highlighting job flexibility and career advancement in job advertisements significantly increases applications for entry-level STEM positions (by 44% and 35% respectively), but not for senior-level roles.
- Flexibility attracts both male and female applicants, while career advancement primarily attracts men.
- These treatments also lead to belief updating among potential applicants, with flexibility improving perceived work-life balance and career advancement improving perceived career benefits but lowering perceived work-life balance.
Identification Strategy
- The study employs a randomized controlled trial (RCT) within a large European technology firm.
- Job advertisements for STEM positions are randomly assigned to one of three conditions: control, flexibility highlight, or career advancement highlight.
- Each job ad is posted sequentially under these three conditions for 10-day intervals, allowing for within-job-ad randomization and causal inference using job-ad and time fixed effects.
- A complementary survey experiment among STEM students uses random assignment of these job ad versions to study belief updating.
Data
The field experiment data includes information on 105 job ads and 1,583 applications, capturing application date, applicant gender, place of residence, recruiter ratings (good fit, interview invitation), and anonymized CV data. The survey experiment data comprises responses from 2,136 STEM students who were shown job ads and then asked about their beliefs regarding job characteristics and the working environment.
Manisha Jain, Corina Mommaerts, Jeffrey Weaver — Personnel Economics
This paper evaluates the effectiveness of the Work Opportunity Tax Credit (WOTC), a large federal wage subsidy program for disadvantaged workers, finding precise null effects on hiring, employment, and earnings due to informational frictions at the firm level.
Finance Application
- The findings on the ineffectiveness of wage subsidies due to informational frictions and legal liability concerns have significant implications for corporate finance and ESG investing.
- For corporate finance, this suggests that firms' labor policies, especially those influenced by government incentives, may not translate into expected operational changes or value creation if information flow within the firm is siloed.
- In asset pricing, one could examine whether firms with high WOTC utilization (pure transfers) exhibit different risk-adjusted returns or stock price reactions to labor policy announcements compared to firms that genuinely integrate such policies into hiring.
- For household finance, the null effect on earnings for disadvantaged workers implies that these subsidies do not improve their financial stability, suggesting a need for research into alternative financial support mechanisms or the efficacy of financial literacy programs for this demographic.
wage subsidieslabor economicspublic policycorporate financeESGinformational frictionsemploymentearningsfirm behaviorpolicy evaluationsocial safety netasset pricing implicationshousehold finance implicationsinsurance implications
Core finding, identification, data
Core Finding
- The Work Opportunity Tax Credit (WOTC) has no meaningful impact on hiring, employment, or earnings for eligible disadvantaged workers, despite providing substantial subsidies to firms.
- Instead, the subsidies primarily act as a pure transfer to firms, with most benefits concentrated among a small number of large firms, and 97% of WOTC-subsidized hires being "windfall wastage." The lack of effect is attributed to informational frictions, where hiring managers are often unaware of a candidate's WOTC eligibility due to concerns about legal liability and screening processes.
Identification Strategy
- The paper employs four complementary quasi-experimental approaches: a difference-in-differences (DiD) design exploiting a 2007 WOTC eligibility expansion for SNAP recipients, a regression discontinuity (RD) design leveraging sharp age-based eligibility cutoffs, another DiD approach examining the impact of electronic filing for WOTC applications in 2017, and a firm-level event study analyzing the staggered adoption of WOTC by firms.
- These methods are used to isolate causal effects on labor market outcomes.
Data
The study uses linked administrative data from over 13 million individuals in Wisconsin spanning two decades, including quarterly employment and earnings records, SNAP, TANF, unemployment insurance, and criminal justice records. This is linked to individual-level records of over 800,000 WOTC applications and certifications. Additionally, the authors collected original survey data from 170 WOTC-utilizing firms regarding their hiring processes.
Yi Chen, Fabiano Dal-Ri, Thomas Jungbauer, Daniela Scur — Personnel Economics
This paper develops a theoretical model and provides empirical evidence for employee poaching driven by asymmetric employer learning, showing how managerial compensation reflects information rents and facilitates talent reallocation.
Finance Application
- The concept of 'information rent' for managers, tied to their knowledge of other workers, could be a powerful signal in asset pricing.
- Firms that successfully poach managers with high information rents (e.g., from larger firms with high-ability workers) might experience positive abnormal returns, reflecting the market's underestimation of this human capital acquisition, or vice versa for firms losing such managers.
- In corporate finance, this insight could inform M&A strategies, where the value of a target firm might include a premium for its 'poachable' managers and their embedded human capital networks.
- For risk management and insurance, the vulnerability of a firm to losing information-rich managers could be quantified and potentially hedged, leading to new key-person insurance products that explicitly price this 'information value' risk.
human capitalmanagerial compensationinformation asymmetrylabor marketsfirm valueasset pricingcorporate financerisk managementpoachingtalent reallocationBrazil
Core finding, identification, data
Core Finding
- The paper finds that firms poach managers not only for their track record but also for their personnel-specific information about workers.
- In equilibrium, more productive firms poach managers, whose compensation increases in the quality of their information about workers (information rents).
- This leads to efficient talent reallocation (more able workers to more productive firms) but efficiency is not fully achieved due to information frictions.
- Empirically, poached managers earn higher salaries, which correlate with the demand for and supply of information (firm size, worker ability), and raided workers are of higher ability.
Identification Strategy
- The study uses an event-study design on administrative data from Brazil, defining 'poaching events' as direct job-to-job moves of managers.
- It tracks co-workers of poached managers (raided workers) into the new firm and compares these events to non-managerial poaching and worker referrals.
- Worker ability and firm wage premia are proxied using the Abowd et al. (1999) two-way fixed effect decomposition (AKM model), estimated from pre-poaching data to avoid endogeneity, and tested using event-study and OLS regressions.
Data
The paper utilizes the Relação Anual de Informações Sociais (RAIS) dataset from Brazil (2003-2017), a linked employer-employee administrative dataset covering all formal sector employment contracts, including worker earnings, contract dates, occupation codes, education levels, and establishment identifiers.
Alejandro Herrera-Caicedo, Jessica Jeffers, Elena Prager — Law and Economics
This paper empirically investigates whether common leadership between firms contributes to collusive agreements, particularly in the labor market.
Finance Application
- This research offers significant insights for asset pricing, household finance, and insurance.
- In asset pricing, common leadership could serve as a novel risk factor or signal for potential anti-competitive behavior, impacting firm valuation and regulatory risk, which could be integrated into ESG screening or M&A due diligence.
- For household finance, the findings highlight how corporate governance structures can directly influence labor market outcomes, affecting household income stability and wealth accumulation, informing models of labor income risk.
- In insurance, D&O liability underwriters could use common leadership metrics to assess the likelihood of antitrust litigation or regulatory fines, adjusting premiums or coverage based on the extent and nature of interlocked leadership.
corporate governanceantitrustcollusionlabor marketsboard interlocksexecutive compensationfirm valuationregulatory riskESGD&O insurancehousehold incomeevent study
Core finding, identification, data
Core Finding
- The probability of a collusive agreement between two firms increases by 12 percentage points after the onset of common leadership, compared to a baseline rate of 1.2 percent.
- This effect is robust to controls for market competition and is most pronounced within one to three years after common leadership begins, suggesting a causal link from common leadership to collusion.
Identification Strategy
- The study employs a difference-in-differences and event study regression framework, using firm-by-year and firm-pair fixed effects to identify the effect of common leadership on collusion.
- This design leverages within-pair variation over time to minimize bias from omitted variables that might simultaneously drive both common leadership and collusive behavior.
Data
The paper utilizes unsealed court evidence from the Silicon Valley no-poaching cases to construct a direct measure of collusive agreements at the firm pair level. Common leadership data is sourced from BoardEx, detailing professional histories, while labor market competition is measured using LinkedIn data on worker flows.
Bocar A. Ba, Patton Chen, Tony Cheng, Martha C. Eies, Justin E. Holz — Economics of Crime
This paper evaluates the impact and fiscal sustainability of Durham, NC's civilian crisis response program (HEART) as an alternative to traditional policing for nonviolent 911 calls.
Finance Application
- This research offers several avenues for finance.
- In insurance, the demonstrated reduction in crime and improved response times could be integrated into property and casualty insurance underwriting models, leading to differentiated premiums for homes and businesses in areas with such programs.
- For household finance, increased public trust and safety could influence household location choices and property values, impacting mortgage demand and housing market dynamics.
- In municipal finance, the fiscal sustainability of these programs could affect municipal bond ratings and yields, as cities with effective, self-sustaining public safety innovations might be perceived as less risky investments.
public safetycrisis responsepolicing alternativescrimepublic trustcost-effectivenessmunicipal financeinsurancehousehold financeproperty values
Core finding, identification, data
Core Finding
- The HEART program reduces crime reports, arrests, and response times, primarily through civilian phone and in-person responses.
- It also fosters public trust, leading to an increase in future 911 calls, and is found to be a fiscally self-sustainable intervention based on a contingent valuation survey.
Identification Strategy
The study employs a difference-in-differences design to causally evaluate the HEART program's impact, comparing outcomes in areas or periods affected by the program to those not affected, while controlling for other confounding factors.
Data
The paper utilizes administrative data from Durham, North Carolina's HEART program, including 911 call records, crime reports, arrest data, and response times. It also incorporates an original contingent valuation survey to assess the program's perceived value and fiscal sustainability.
Aviv Caspi, Charlie Rafkin — Law and Economics
This paper uses a randomized controlled trial to evaluate the impact of free legal assistance on eviction outcomes for tenants, identifying the crucial role of emergency rental assistance and directly eliciting tenant demand for legal services.
Finance Application
- This research offers significant insights for household finance and insurance.
- In household finance, the high elicited WTA for legal aid, even beyond its direct monetary impact, suggests that households value non-pecuniary benefits like stress reduction and navigation of complex systems, which could inform the design and marketing of financial advisory services or debt counseling.
- The finding that legal aid's effectiveness is highly complementary to direct financial assistance (ERAP) highlights the importance of integrated financial support, suggesting that emergency savings or liquidity products could be more impactful when paired with legal or advisory services.
- For insurance, the strong demand for legal aid could indicate a market for 'eviction insurance' or legal expense insurance products, where the elicited WTA could directly inform pricing models.
- Furthermore, understanding how legal aid mitigates eviction risk and its impact on tenant financial stability (e.g., credit scores) could help lenders and insurers better assess and price credit risk for vulnerable populations.
Household FinanceEvictionLegal AidWillingness to PayRandomized Controlled TrialFinancial DistressCredit RiskInsuranceHousing MarketsPolicy EvaluationBehavioral Economics
Core finding, identification, data
Core Finding
- Providing an attorney reduces tenant eviction judgment rates by 23 percentage points (37%), but this effect shrinks by 75% and becomes statistically insignificant once concurrent emergency rental assistance (ERAP) expires.
- Tenants' elicited willingness to pay (WTP) for an attorney is double the attorney's price and eight times their implied income impact, and this high WTP is robust to tests for inattention, misperceptions, or budget constraints.
Identification Strategy
- The study employs a randomized controlled trial (RCT) assigning tenants to receive an offer of full attorney representation or control.
- A key identification strategy leverages a sharp policy variation: the expiration of the Emergency Rental and Utilities Assistance Program (ERAP) midway through the RCT, allowing for a pre-/post-ERAP comparison of attorney effectiveness.
- Tenant demand is elicited using real-stakes, incentivized multiple price lists for Willingness to Accept (WTA) cash versus a lawyer, with embedded tests for behavioral biases and inattention.
Data
The paper utilizes administrative court data on eviction judgments, nonsuits, writs, and judgment amounts. It also collects data from endline surveys on informal outcomes (e.g., moves, out-of-pocket payments, bargaining) and baseline surveys for direct elicitation of tenant demand, beliefs about lawyer effectiveness, and trust games. Administrative payment records from the Memphis/Shelby County ERAP are also used.
John MacDonald, Wilson Hernández Breña, Anthony Braga, Charles Branas — Economics of Crime
This paper evaluates the impact of street cleaning interventions on gun violence and street crimes in Philadelphia using a block randomized controlled trial.
Finance Application
- This research offers insights for asset pricing and insurance by highlighting the disconnect between superficial environmental improvements and actual crime reduction.
- For asset pricing, it suggests that investments in urban revitalization projects focused solely on aesthetics (like litter removal) might not translate into the expected uplift in property values or local business performance if underlying crime risks persist.
- For insurance, it implies that relying on visual cues of neighborhood quality (e.g., cleanliness) without considering actual crime data could lead to mispricing of property or business interruption insurance premiums, as the perceived risk might not align with the true risk of loss due to crime.
Urban EconomicsCrimeRandomized Controlled TrialProperty ValuesInsuranceMunicipal FinanceESGEnvironmental QualitySocial Impact
Core finding, identification, data
Core Finding
- Street cleaning interventions significantly reduced visible litter by 16% but had statistically insignificant effects on gun violence and street crimes.
- This suggests that basic environmental cleanups alone may be insufficient to reduce violence, emphasizing the need for more comprehensive place-based crime prevention strategies.
Identification Strategy
- The study employs a block randomized controlled trial (RCT) design.
- Street segments were randomly assigned to biweekly cleaning, monthly cleaning, or a no-treatment control group, allowing for causal inference regarding the impact of street cleaning on litter and crime outcomes.
Data
The paper uses gun crime and street crime incident counts reported by the Philadelphia Police Department (PPD) and monthly litter index scores collected by trained observers using a 4-point Likert scale.
Gianmarco Daniele, Francesca Calamunci, Giovanni Mastrobuoni, Daniele Terlizzese — Economics of Crime
This paper examines whether incarcerated women serving sentences in women-only prisons in Italy exhibit lower recidivism rates compared to those in mixed-gender prisons.
Finance Application
- The paper's findings on institutional design impacting 'recidivism' could be applied to financial misconduct or default rates.
- For instance, one could investigate if specialized financial institutions or advisory firms catering specifically to women (e.g., women-led wealth management, microfinance for female entrepreneurs) lead to lower rates of financial fraud, loan defaults, or higher savings rates among their clients, using a similar distance-based instrumental variable for access.
- The Policy Relevant Treatment Effects (PRTE) methodology could also be used to optimize the placement of financial literacy programs or credit facilities in underserved communities to maximize positive financial outcomes for specific demographic groups.
recidivismprisonsgenderinstrumental variablespolicy evaluationtreatment effectsinstitutional designsocial economicspublic policy
Core finding, identification, data
Core Finding
- The study finds that women-only prisons significantly reduce three-year recidivism by 8 to 16 percentage points.
- This effect is particularly pronounced when the control group consists of women's sections within men's prisons that house a smaller number of women, suggesting that a critical mass of women in a facility is a key driver of the positive outcomes.
Identification Strategy
- The identification strategy uses an instrumental variable approach, leveraging a quasi-random institutional assignment rule based on the difference in distance between an inmate's residence and the closest mixed-gender prison versus the closest women-only prison.
- This instrument exploits the tension between the Prison Administration's preference for women-only prisons and the goal of minimizing distance to the inmate's residence, allowing for causal inference by isolating exogenous variation in prison assignment.
Data
The paper utilizes individual inmate data from the Department of Prison Administration of the Italian Ministry of Justice for female prisoners incarcerated between 2012 and 2022, covering 10,222 incarcerations. This is supplemented by annual surveys on prison conditions (2017-2019) from the Italian NGO Antigone.
Vincent Rollet — Urban Economics
This paper analyzes how zoning regulations and the existing building stock influence urban evolution and redevelopment in New York City using a novel parcel-level panel dataset and a dynamic spatial equilibrium model.
Finance Application
- The insights on slow-moving supply elasticities and heterogeneous redevelopment costs could inform real estate asset pricing models, particularly for REITs or private equity funds specializing in urban development, by predicting returns under various zoning policy scenarios.
- For household finance, the findings on affordability and welfare gains for different income groups could be integrated into models of housing wealth accumulation and household location choices, especially when considering regulatory changes.
- Insurers could use the analysis of fixed costs and slow redevelopment to better assess long-term property risks and the pace of adaptation to climate change in urban environments.
Urban EconomicsReal EstateZoningRedevelopmentHousing SupplySpatial EquilibriumDynamic ModelsRegulationAffordabilityNew York CityProperty ValuesConstruction CostsAsset PricingHousehold Finance
Core finding, identification, data
Core Finding
- Zoning strongly constrains city growth, but its effects materialize slowly over decades due to large fixed costs of redevelopment, which rise sharply with existing building size.
- Relaxing zoning leads to heterogeneous increases in floorspace supply and affordability benefits, primarily in high-price, low-density areas, with significant affordability gains accruing to households outside rezoned neighborhoods due to migration.
- A frictionless model would significantly overstate the impact of zoning relaxation.
Identification Strategy
- The paper leverages quasi-experimental variation from recent zoning reforms (upzoning) in a difference-in-differences design at the parcel level to identify the effects of relaxing regulatory constraints.
- It then integrates these empirical findings into a dynamic spatial equilibrium model, estimated using methods from empirical industrial organization, to extrapolate long-term and general equilibrium effects.
Data
The study constructs a unique parcel-level panel dataset for NYC (2004-2022) combining cadastral maps, property tax records, scraped online listings (StreetEasy), building permits, certificates of occupancy, real estate transaction data, rent data (NOPVs), zoning maps, smartphone data (Advan), commuting flows (LEHD-LODES), and historical real estate data.
Lei Ma — Urban Economics
This paper develops and estimates an equilibrium model of segmented housing markets to quantify the causes and distributional consequences of the shift towards larger homes in new construction and evaluate housing policies aimed at improving affordability.
Finance Application
- The paper's detailed equilibrium model of segmented housing markets, incorporating heterogeneous demand and supply constraints, could be adapted to analyze the pricing and risk of real estate-backed financial products, such as REITs or MBS, by understanding how demand and supply shocks propagate across different housing segments.
- In household finance, the granular modeling of household preferences and housing choices could inform studies on how housing wealth accumulates across demographic groups, how housing affordability policies impact household balance sheets and consumption, and how these effects vary with market segmentation.
- For insurance, the quantification of zoning regulatory costs and their impact on property values could help refine property insurance pricing and underwriting models, especially in areas with stringent regulations that affect reconstruction costs and market values.
Housing marketsReal estateZoning regulationsHousing supplyHousing demandHeterogeneous preferencesHousehold financeAsset pricingUrban economicsPolicy evaluationSubsidiesMarket segmentationMicrodataStructural model
Core finding, identification, data
Core Finding
- The shift towards larger homes in new construction is driven by both demand from high-income households and, more significantly, by zoning density restrictions that limit smaller home construction.
- Relaxing these restrictions could substantially increase the supply of small homes and lower prices.
- While demand-side subsidies for young, low-income households provide targeted welfare gains to recipients, they can hurt non-recipients due to rising prices and minimal supply response, whereas supply-side subsidies for small home construction lead to modest, untargeted welfare gains by crowding out larger home construction.
Identification Strategy
- The demand model uses a revealed preference approach, identifying household preferences from observed housing choices across demographic groups, and addresses price endogeneity with an instrumental variable based on the availability of similar housing nearby (BLP/BFM methods).
- The supply model estimates zoning regulatory costs using a residual approach, inferring them from the gap between housing prices and observed production costs, and exploits variation in development choices across parcels and house types.
Data
The paper uses property-level characteristics from CoreLogic, individual-level demographics and residential addresses from L2 voter registration records, parcel-specific development costs from City of Atlanta building permits, zoning codes from CoreLogic and manually collected development standards from MuniCode.com, and neighborhood-level land value estimates from Davis, Larson, Oliner, and Shui (2021).
Nathaniel Baum-Snow, Abdollah Farhoodi, Lu Han — Urban Economics
This paper quantifies the welfare costs of rent growth for renters by estimating migration costs and housing demand using a dynamic structural model and individual-level panel data on Texas renters.
Finance Application
- The paper's quantification of substantial, heterogeneous moving costs and the welfare burden of rent growth for renters offers several finance applications.
- In household finance, these costs can be integrated into models of housing tenure choice, consumption smoothing, and financial resilience, explaining why households might endure high housing costs rather than move, impacting their savings and debt accumulation.
- For asset pricing, understanding how renter mobility and demand elasticities vary by region and household type can improve the valuation of residential REITs and real estate-backed securities by providing better forecasts of rental income stability and local market risk.
- In insurance, the identified moving costs and welfare losses could inform the design and pricing of new products like 'relocation insurance' or enhance the value proposition of existing renters' insurance, particularly for vulnerable populations facing housing affordability shocks.
Housing affordabilityRentersMigrationMoving costsDynamic structural modelNeural networksInstrumental variablesWelfare costsHousehold financeReal estateUrban economicsTexas
Core finding, identification, data
Core Finding
- The paper finds that moving costs are substantial (fixed costs around $40,000, higher for cross-CBSA moves) and vary significantly by household type (lower for younger, higher-income individuals).
- These costs, combined with limited availability of affordable substitutes, lead to significant welfare burdens from rent growth, particularly for lower-income renters in large metropolitan areas.
Identification Strategy
- The paper uses a novel shift-share instrumental variable to identify the causal effects of rental prices on migration flows and welfare.
- The instrument interacts neighborhood-level housing supply elasticity with the overall trend in Texas rental prices, assuming it affects rents only through supply-side responses and is uncorrelated with unobserved amenity changes.
- A structural dynamic model of location choice, estimated using a neural net and PPML regressions, is employed to quantify migration costs and demand.
Data
The paper uses an individual-level panel dataset of residential locations for nearly all renters in Texas from 2010-2019, created by combining Infutor marketing data (migration histories) with CoreLogic housing tenure information. It also incorporates Zillow Observed Rent Index (ZORI) data and American Community Survey (ACS) data for neighborhood characteristics.
Robert Collinson, Deniz Dutz, John Eric Humphries, Nicholas S. Mader, Daniel I. Tannenbaum, Winnie van Dijk — Urban Economics
This paper provides the first comprehensive causal analysis of how eviction impacts children's home environment, school engagement, educational achievement, and high school completion in two major U.S. cities.
Finance Application
- The causal links between eviction and long-term educational and housing stability outcomes for children have significant implications for household finance, insurance, and real estate.
- In household finance, these findings could inform models of mortgage default risk and the design of targeted financial assistance programs to prevent evictions, potentially improving loan performance.
- For insurance, new products could emerge, such as 'education interruption insurance' or 'housing stability insurance,' while property insurers could adjust risk assessments for landlords based on eviction prevention efforts.
- In real estate, ESG investors could use these insights to evaluate the social impact and long-term value of rental properties, favoring those with lower eviction rates and robust tenant support.
evictionhousing instabilitychildren's outcomeseducationhomelessnesscausal inferenceinstrumental variableshousehold financesocial impactreal estateESG investinginsurance
Core finding, identification, data
Core Finding
- Eviction significantly increases children's residential mobility, homelessness, and likelihood of doubling up with grandparents or other adults.
- It also disrupts school engagement, leading to increased absences, school changes, and reduced high school course credits and graduation rates, with more disruptive effects observed for older children and boys.
Identification Strategy
- The study employs an instrumental variables (IV) approach that leverages the random assignment of eviction court cases to judges who vary systematically in their leniency.
- This allows for the estimation of the causal impact of an eviction order on children's outcomes by comparing compliers, defined as children whose case outcome would have changed had their case been assigned to a different judge.
Data
The study links near-universal eviction court records from Cook County, IL (2000-2016) and New York, NY (2007-2017) to administrative public school records, homelessness services data, and restricted Census data (2000 and 2010 Decennial Censuses for Chicago).
Shoshana Vasserman, Cody Cook, Aboudy Kreidieh, Hunt Allcott, Neha Arora, Freek van Sambeek, Andrew Tomkins, Eray Turkel — Urban Economics
This paper evaluates the short-run impacts of congestion pricing in New York City on traffic speeds, air quality, commerce, and welfare, finding significant speed improvements both within and outside the priced zone but limited effects on other outcomes.
Finance Application
- This research offers insights for several finance areas.
- In real estate, the observed speed improvements and potential changes in accessibility could influence commercial property valuations and REIT performance, especially for assets within or near the CBD.
- For insurance, reduced congestion might lead to fewer traffic accidents or less severe ones, impacting auto insurance claims and pricing models.
- Furthermore, the welfare gains from improved mobility could be factored into the valuation of infrastructure projects or public-private partnerships focused on urban transportation, affecting investors in municipal bonds or infrastructure funds.
- Finally, changes in commuting costs and time could influence household budgets and consumption, potentially impacting credit risk or savings behavior.
Congestion PricingUrban EconomicsTrafficSynthetic ControlReal EstateInsuranceInfrastructure FinancePolicy EvaluationTransportationWelfare Economics
Core finding, identification, data
Core Finding
- Congestion pricing in NYC led to an approximate 10% increase in average speeds within the Central Business District (CBD), with the most significant improvements occurring during peak afternoon hours.
- Importantly, these speed gains extended beyond the CBD to the broader metropolitan area, and the majority of welfare gains accrued from "unpriced" trips that benefited from reduced congestion.
- The policy had little-to-no discernible short-run effects on air quality, visits to shops and restaurants, or foot traffic.
Identification Strategy
- The study employs a synthetic control method (Xu, 2017) to estimate the causal effects of congestion pricing.
- New York City's CBD is treated, while other major US cities (e.g., Philadelphia, Chicago, Boston, Atlanta, Baltimore) serve as controls to construct a synthetic counterfactual for NYC's traffic and non-traffic outcomes.
- This approach allows for the estimation of the Average Treatment Effect on the Treated (ATT) by comparing NYC's post-treatment outcomes to its synthetic counterpart.
Data
The paper utilizes aggregated and anonymized Google Maps trip data (Sept. 2024-June 2025) for segment-level (traversal speeds, fuel consumption) and origin-destination level (travel times, speeds) outcomes. Non-traffic data includes ambient air quality from PurpleAir sensors, transaction data from MBHS3 (credit/debit cards) for restaurants and retail, and foot traffic data from GPS devices (Advan Neighborhood Patterns).
Enrico Moretti, Harrison Wheeler — Urban Economics
This paper quantifies the economic costs and distributional impacts of traffic noise in the U.S. using quasi-experimental variation from noise barrier construction and estimates the potential benefits of electric vehicle adoption.
Finance Application
- This research offers crucial insights for real estate asset pricing by quantifying how noise externalities impact property values.
- REITs and other real estate investors could integrate noise exposure data into their valuation models to identify undervalued or overvalued assets, especially in urban areas.
- For household finance, mortgage lenders could adjust underwriting criteria or offer differentiated products based on noise levels, impacting household wealth accumulation and location decisions.
- Insurers could explore noise pollution's link to property damage (e.g., from vibrations) or health outcomes, potentially influencing property or health insurance premiums and product design.
Real EstateHousing PricesEnvironmental EconomicsExternalitiesTraffic NoiseElectric VehiclesProperty ValuationHousehold WealthESG InvestingUrban EconomicsMortgage MarketsInsurance
Core finding, identification, data
Core Finding
- The study finds that reduced traffic noise exposure leads to significant increases in house prices, with a 6.8% increase for properties within 100 meters of a noise barrier.
- The aggregate economic cost of traffic noise in the U.S. is estimated at $110 billion, disproportionately affecting lower-income and minority households.
- A broad shift to electric vehicles could yield noise reduction benefits of $77.3 billion nationwide, concentrated among low-income families in dense urban areas.
Identification Strategy
- The identification strategy relies on quasi-experimental variation from the construction of noise barriers.
- A difference-in-differences model compares changes in prices for properties near constructed barriers (treatment group) to those further away (control group).
- A triple-difference model further refines this by comparing constructed barriers to proposed-but-not-built barriers, allowing for richer controls for time-varying, barrier-specific, and distance-bin-specific heterogeneity.
Data
The paper uses transaction-level housing price data and assessor data from CoreLogic, location-specific traffic noise estimates from the U.S. Department of Transportation National Transportation Noise Map, and sound barrier inventory data from the Florida Department of Transportation (FDOT). It also incorporates census tract-level demographic data from the American Community Survey (ACS), wind data from NCEI, Google Maps Platform for air quality, and MRLC Consortium data for land cover.
Marcos Agurto, Sudipta Sarangi, Danila Serra — Children and Families
This paper evaluates the effectiveness of social media-based role model interventions and teacher information campaigns in promoting STEM education and college enrollment among high school girls in Peru.
Finance Application
- This research demonstrates how targeted social media interventions and influencer campaigns can significantly alter educational and career preferences, which could be directly applied to household finance.
- For instance, an RCT could test if Instagram reels featuring successful female financial advisors or entrepreneurs can increase financial literacy, savings rates, or investment intentions among young women, potentially reducing gender gaps in financial market participation.
- Similarly, insurance companies could use such interventions to promote awareness and adoption of specific insurance products (e.g., health or life insurance) among demographics with low penetration, by showcasing relatable role models benefiting from these products, thereby influencing risk management behaviors.
randomized controlled trialsocial mediarole modelsbehavioral economicseducationgender gaphousehold financefinancial literacyhuman capitalinfluencer marketing
Core finding, identification, data
Core Finding
- The interventions had a mixed impact: a short-term negative effect on self-reported engineering preferences for lower-performing girls, but a significant positive effect on engineering preferences (measured by scholarship applications) for top-performing girls, especially when combined with teacher information.
- Overall, the Instagram role model content led to a large increase in college enrollment for all girls, regardless of major, likely due to a positive shift in perceptions of university life.
Identification Strategy
- The study employs a randomized controlled trial (RCT) design across 73 high schools in Peru, divided into two treatment groups and one control group.
- Interventions included Instagram reels featuring female engineering role models and WhatsApp messages for teachers discussing gender stereotypes in STEM, allowing for causal inference on the impact of these social media-based interventions.
Data
The paper uses baseline and follow-up survey data from 787 high school students (547 girls) and administrative college enrollment data from Peru.
Sean E. McCulloch, Matthew P. Schaelling, Matthew Turner, Toru Kitagawa — Urban Economics
This paper investigates the causal impact of sewer access on urban population density, income, and literacy in developing world cities, leveraging a quasi-experimental design based on natural drainage basin divides.
Finance Application
- This research offers insights for real estate and mortgage markets by highlighting how basic infrastructure like sewer access directly influences population density, and thus, potentially property values and collateral.
- Lenders could integrate sewer access metrics into their underwriting models for developing market real estate.
- For ESG investing, the paper underscores the social impact of infrastructure, allowing investors to better assess the 'S' in ESG for public or private infrastructure projects.
- Furthermore, the findings could inform municipal bond analysis, as improved urban density and economic activity (even without demographic shifts) can impact local government revenues and creditworthiness.
Real EstateUrban EconomicsInfrastructureDeveloping MarketsPopulation DensityProperty ValuationESG InvestingMunicipal FinanceSpatial Econometrics
Core finding, identification, data
Core Finding
- Increased sewer access has a substantial effect on population density, with a 1% increase in sewer connections leading to a 6% increase in tract population density.
- However, it has only minor effects on tract mean income and literacy, suggesting that sewer networks are critical for urban economic geography and primarily benefit existing residents or demographically similar migrants, rather than attracting more affluent newcomers.
Identification Strategy
- The identification strategy exploits the fact that moving sewage uphill is more costly than downhill, creating differential costs of sewer access for neighborhoods on opposite sides of natural drainage basin divides.
- The authors use an indicator for being 'outside' the central basin (requiring uphill sewage movement) and its interactions with horizontal and vertical displacement from the basin divide as instrumental variables to estimate the causal effect of sewer access.
Data
The study utilizes census tract-level data from Brazil, Colombia, South Africa, Jordan, and Tanzania, combined with UN DESA World Urbanization Prospects data for city centers and Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) digital elevation maps to delineate drainage basins and elevation profiles.
Milena Almagro, Aradhya Sood — Urban Economics
This paper quantifies the relative contributions of household preferences versus choice set restrictions (such as discrimination) to residential segregation in 1940 Minneapolis using a novel two-sided housing matching model and historical data.
Finance Application
- The findings on how preferences and restrictions drive segregation have significant implications for real estate asset pricing and household finance.
- In asset pricing, understanding these drivers can inform models of local property value dynamics, risk premiums for real estate investment trusts (REITs) in historically segregated areas, and the pricing of mortgage-backed securities.
- For household finance, the restricted choice sets for certain demographic groups directly impact their ability to accumulate wealth through homeownership, potentially leading to persistent disparities in household balance sheets and access to credit, which could be modeled to understand intergenerational wealth gaps.
housing marketssegregationdiscriminationracial covenantschoice setspreferencesinstrumental variablesspatial economicshousehold financereal estateurban economicshistorical datacausal inferenceasset pricing
Core finding, identification, data
Core Finding
- The study finds that three-quarters of the observed residential segregation in 1940 Minneapolis was driven by household preferences for neighborhood demographic composition, while choice set restrictions (including de jure racial covenants) accounted for the remaining one-quarter.
- Specifically, racial covenants reduced the probability of non-White and White Eastern/Southern European immigrant households living in covenanted areas by 12.6 and 13.9 percentage points, respectively, and minority households were 8 percentage points less likely to be admitted to an average development.
Identification Strategy
- The paper employs an instrumental variable (IV) approach to identify the causal effects of racial covenants.
- For reduced-form estimates, it instruments racial covenants with an 'urbanization frontier index' (the share of built-up area from the previous period in neighboring developments closer to the city center), relying on the random availability of agricultural land for development.
- For the structural model, it uses racial covenants as an excluded shifter for admission rules and instruments housing prices with steep land slopes, and minority share with a shift-share migration instrument based on 1924 immigration quotas.
Data
The paper constructs a novel dataset by linking 1940 individual-count US census data, historical street grids, Hennepin County assessor data, and racial covenant data from the Mapping Prejudice Project for the Minneapolis metro area. It also incorporates geographic data on wetlands, elevation, industry zoning, and streetcar networks.
Sofia Amaral, Selim Gulesci, Maria Micaela Sviatschi, Alejandra Ramos, Sarita P. Ore-Quispe, Aixa Garcia-Ramos — Children and Families
This paper evaluates a randomized controlled trial in Mozambique showing that an intervention addressing gender-based violence (GBV) in schools reduces violence by teachers/staff and increases girls' school enrollment, especially when girls are trained to be proactive.
Finance Application
- The core insight that victim proactivity is crucial for an intervention's positive impact could be applied to household finance or insurance.
- For instance, in insurance, studying how proactive reporting of minor claims (e.g., property damage) by policyholders, incentivized by specific training or communication, affects long-term claim frequency or severity could optimize policy design.
- In household finance, examining how financial literacy programs, combined with fostering proactive behaviors like seeking financial advice or reporting fraud, impact household financial resilience and investment outcomes, especially in contexts with information asymmetry, could yield significant results.
- This could also extend to corporate governance, where empowering employees to proactively report misconduct might be key to reducing corporate fraud.
Randomized Controlled TrialGender-Based ViolenceEducationHuman CapitalBehavioral EconomicsInformation AsymmetryMoral HazardPrincipal-Agent ProblemProactive BehaviorReporting MechanismsSocial Norms
Core finding, identification, data
Core Finding
- The intervention, consisting of Gender Focal Point (GFP) training for school personnel and student training, reduced violence perpetrated by teachers/school staff against girls by 67% across all treated schools.
- Crucially, girls' school enrollment increased by 10% only in schools where girls received specific training, suggesting that victim proactivity (e.g., reporting) is essential for educational attainment improvements from GBV reduction.
Identification Strategy
- The study uses a clustered-randomized controlled trial involving 326 primary schools in Mozambique.
- Schools were randomized into a control group and three treatment groups: GFP training plus student training for girls only (T1), GFP training plus student training for boys only (T2), and GFP training plus student training for both genders (T3).
- This cross-randomization allows for identifying the causal effects of GFP training and the specific targeting of student training on GBV and educational outcomes.
Data
The paper uses baseline and endline survey data from adolescents (grades 6 or 7), GFP surveys, teacher surveys, and administrative data on school records (enrollment). These data capture violence experiences, attitudes, social desirability, and GFP/teacher activities and knowledge.
Cache Ellsworth, Ian Fillmore, Adrian Haws, Joseph Price — Children and Families
This paper leverages a natural experiment of unexpected oil discoveries on Oklahoma homesteads to estimate the long-run causal effects of parental wealth shocks on children's education, labor supply, migration, and marriage outcomes.
Finance Application
- This research offers valuable insights for household finance by demonstrating how unexpected wealth shocks can profoundly shape intergenerational human capital and marriage patterns, which are crucial determinants of long-term financial well-being and wealth accumulation.
- Researchers could explore how these shocks influence household savings rates, investment choices (e.g., real estate vs. financial assets), risk-taking behavior, and demand for insurance products across generations.
- The localized nature of the wealth shock could also inform studies on how regional commodity booms affect local asset prices and financial market participation.
Wealth ShocksIntergenerational MobilityHuman CapitalEducationLabor SupplyMarriageHousehold FinanceNatural ExperimentHistorical DataOil DiscoveriesAsset AccumulationRisk AversionReal Estate
Core finding, identification, data
Core Finding
- Parental oil wealth shocks significantly increased children's education (1.1 additional years, 17 percentage points more likely to graduate high school).
- Sons were more likely to be in the labor force and earned 14% higher occupational income, while daughters delayed marriage and married more educated spouses.
- These effects are attributed to reduced demand for child labor and increased access to schooling due to parental migration to urban areas.
Identification Strategy
- The study leverages a natural experiment where homesteaders claimed land without knowing about hidden oil fields.
- It compares children of neighboring homesteaders within six-by-six mile survey townships, where some land had oil and others did not.
- This highly local variation separates individual wealth shocks from community-wide resource boom effects, and homesteaders' characteristics are balanced across treatment status, supporting a quasi-random assignment.
Data
The paper constructs an intergenerational panel dataset by linking land and oil records (Oklahoma Geological Survey, BLM Federal Land Patents) to U.S. census records (1900, 1910, 1930, 1940 U.S. Census) using the Census Tree Project.
Eric V. Edmonds, Priya Mukherjee, Nikhilesh Prakash, Nishith Prakash, Shwetlena Sabarwal — Children and Families
This paper evaluates the impact of a talk therapy intervention on the mental health and human capital outcomes of adolescents at risk of school dropout in a low-resource setting in Nepal using a large randomized controlled trial.
Finance Application
- This research highlights how mental health interventions can significantly improve individual well-being and emotional regulation, which are crucial for sound decision-making in household finance.
- For instance, improved emotional regulation could lead to more rational savings and investment decisions, reduced impulsive spending, or better management of financial stress, impacting wealth accumulation and debt.
- Insurers could leverage such interventions to reduce health-related risks, potentially lowering claims for mental health services or improving overall longevity, which is relevant for life and health insurance product design and pricing.
- Furthermore, the finding that mental health benefits don't automatically translate to educational attainment in low-resource settings suggests that financial literacy or vocational training programs might need to be integrated with mental health support to maximize human capital development and future earning potential, influencing microfinance or impact investing strategies.
household financebehavioral economicshuman capitalmental healthinsurancerisk managementadolescentslow-resource settingsrandomized controlled trialfinancial decision-making
Core finding, identification, data
Core Finding
- Individual talk therapy significantly reduced psychological distress, improved emotional regulation, and enhanced life perspectives among adolescents, with high participation rates even without prior diagnosis.
- However, these mental health benefits did not translate into improved school attendance or cognitive outcomes, suggesting that mental health interventions alone may be insufficient to address structural barriers to educational attainment in such contexts.
Identification Strategy
- The study employs a large-scale randomized controlled trial (RCT) in Nepal, assigning adolescents at risk of school dropout to one of four groups: therapy-only, education nudge-only, therapy-plus-nudge, or a pure control group.
- The impact of therapy uptake is estimated using Two-Stage Least Squares (2SLS), with random assignment serving as the instrumental variable to address potential endogeneity in therapy participation.
Data
The study uses administrative records and endline survey data from 1,635 adolescents across 40 government schools in seven municipalities in Nepal. Baseline data on dropout risk, gender, and age were collected from school records, and mental health and educational outcomes were measured via surveys and administrative records.
Ronja Helénsdotter — Children and Families
This paper examines peer effects in substance abuse, self-harm, and educational attainment among youths in Swedish residential treatment facilities using a novel instrumental variable approach.
Finance Application
- The identified 'contagion' of risky behaviors through peer effects could be highly valuable for insurance and household finance.
- Insurers could leverage peer group characteristics (e.g., in shared living, rehabilitation, or even workplace settings) to refine risk assessments and pricing for health, life, or disability insurance, particularly for individuals with prior risk factors.
- In household finance, the methodology could be adapted to study how peer exposure to risky financial behaviors (e.g., speculative trading, excessive debt, or even fraud) influences individual financial decisions, especially in contexts where individuals are exogenously grouped, such as new employee cohorts or residential communities.
Peer EffectsSocial InfluenceSubstance AbuseSelf-HarmInstrumental VariablesCausal InferenceHealth EconomicsYouth OutcomesInsurance RiskHousehold FinanceBehavioral Finance
Core finding, identification, data
Core Finding
- The study reveals strong reinforcing peer effects in substance abuse and self-harm: a 1-standard-deviation increase in exposure to peers with a history of substance abuse (self-harm) significantly increases the risk of hospitalization for these issues post-discharge by 17.9 (12.1) percentage points, respectively.
- These effects are primarily driven by social influence among same-sex peers and negatively impact educational attainment.
Identification Strategy
- The paper employs an instrumental variable (IV) design, exploiting the quasi-random variation in peer group composition within facility-by-year cells.
- The share of peers with a particular history on a youth's first day at the facility serves as an instrument for the total peer exposure during the entire placement spell, controlling for facility-by-year fixed effects.
Data
The research utilizes novel Swedish register data covering over 16,000 youths admitted to state-owned residential treatment facilities between 2000 and 2020. This includes administrative records from the National Board of Institutional Care, linked with national registers on hospitalizations, deaths, adult addiction treatment, out-of-home placements, legal proceedings, and detailed demographic/socioeconomic characteristics.
C. Kirabo Jackson, Julia A. Turner, Jacob Bastian — Children and Families
This paper investigates the short-run labor market and aggregate earnings impacts of Universal Pre-Kindergarten (UPK) programs across several U.S. states and cities, finding significant increases in labor force participation, employment, and earnings, particularly for women.
Finance Application
- The significant increase in labor force participation and earnings, especially for women, could impact household financial decisions, such as savings rates, debt accumulation (e.g., mortgage applications, consumer credit), and demand for financial planning services.
- For asset pricing, the boost in local economic activity and disposable income could translate into higher local real estate values and a 'UPK premium' in housing markets, affecting property-backed securities or local REITs.
- The potential for UPK programs to be self-sustaining through tax revenues could also influence the creditworthiness and pricing of municipal bonds issued by implementing localities.
Universal Pre-Kindergartenlabor supplyemploymentearningshousehold incomeeconomic stimulusdifference-in-differencesevent studyhousehold financereal estatemunicipal bondswomen in workforcechildcare
Core finding, identification, data
Core Finding
- Universal Pre-Kindergarten (UPK) programs, implemented across nine states and cities from 1995-2020, significantly increased Pre-K enrollment and led to a 1% rise in labor force participation, a 1.2% increase in employment, and a 1.7% growth in hours worked, resulting in higher aggregate earnings.
- These effects were strongest for mothers but extended to other women, and each dollar spent on UPK generated $8-$20 in additional aggregate earnings, potentially covering program costs through increased tax revenues.
Identification Strategy
- The study employs a stacked difference-in-differences approach within an event-study framework, leveraging the staggered adoption of UPK programs across different states and cities.
- It compares changes in outcomes in UPK-implementing areas before and after implementation to changes in areas without UPK, controlling for pre-trends, regional shocks, and other policy changes.
Data
The paper utilizes data from the American Community Survey (ACS), Current Population Survey (CPS) (ASEC and October Supplement), National Institute for Early Education Research (NIEER) for Pre-K enrollment and spending, and the American Time Use Survey (ATUS) for time-use data. It also incorporates state-level policy controls from the University of Kentucky Center for Poverty Research.
Jocelyn S. Wikle, Riley Wilson — Children and Families
This paper examines how earlier access to public kindergarten impacts mothers' labor supply decisions in the short and long term, and the factors influencing these effects.
Finance Application
- The findings have direct implications for household financial planning and wealth accumulation, especially for mothers.
- Increased, even if temporary, maternal earnings could influence household savings rates, debt management, and investment in children's human capital.
- For insurance, temporary employment boosts could increase demand for employment-linked products like life, disability, and health insurance.
- Insurers could analyze local kindergarten policies (e.g., full-day mandates) to better assess household income stability and tailor product offerings or pricing, considering the persistent non-childcare 'child penalty' that affects long-term female wealth accumulation and retirement planning.
Labor SupplyMaternal EmploymentChildcareKindergartenGender InequalityHousehold FinanceHuman CapitalRegression DiscontinuityPolicy EvaluationEarningsWealth GapInsurance Demand
Core finding, identification, data
Core Finding
- Earlier kindergarten access increases maternal employment by 0.23 percentage points, with effects persisting for two years, particularly for highly educated mothers and in regions with full-day kindergarten.
- However, these gains are transitory and do not lead to persistent changes in career trajectories or fully ameliorate the 'child penalty,' as other household management duties and gender norms continue to constrain mothers' labor force attachment.
Identification Strategy
- The study employs a Regression Discontinuity (RD) design, exploiting children's kindergarten eligibility cutoffs.
- It compares mothers whose children are born just before the eligibility cutoff (gaining earlier access) to those born just after (missing earlier access), assuming that birth timing relative to the cutoff is quasi-random and demographics are smooth at the threshold.
Data
The paper links several administrative and survey datasets, including the Census Household Composition Key (CHCK), SSA Numident, Master Address File (MAF), Linked Employer-Household Dynamics (LEHD) job-level files, American Community Survey (ACS), monthly Current Population Survey (CPS), and American Time Use Survey (ATUS, for descriptive evidence).
Ankita Patnaik, Michael Levere, Isabel Musse, Gina Livermore — Children and Families
This paper evaluates the long-term impact of a randomized controlled trial (PROMISE) providing intensive services to disadvantaged youth with disabilities on their economic and health outcomes.
Finance Application
- The findings on the importance of early employment for long-term economic well-being can inform household finance research on human capital development and financial resilience among vulnerable populations, particularly regarding lifetime income and wealth accumulation.
- For insurance, the observed reduction in healthcare expenditures could refine actuarial models for health insurance, while the impact on SSI/SSDI participation can inform disability insurance pricing and policy design.
- The cost-benefit analysis also offers insights into designing cost-effective financial literacy or vocational training programs.
Household FinanceHuman CapitalDisabilitySocial Safety NetEmploymentHealthcare CostsRCTInsuranceProgram Evaluation
Core finding, identification, data
Core Finding
- The PROMISE intervention improved youth's employment outcomes and reduced healthcare expenditures, with early paid employment experiences identified as the most critical mediator for these long-term improvements.
- However, the program was costly, resulting in a negative net benefit per family primarily due to high administrative costs.
Identification Strategy
- The study employs a randomized controlled trial (RCT) involving over 12,000 Supplemental Security Income (SSI) recipients with disabilities, randomly assigning them to a treatment or control group.
- This design allows for a causal intent-to-treat estimation of the intervention's impact, further explored through mediation analysis to pinpoint specific mechanisms.
Data
The paper primarily utilizes administrative data from the Social Security Administration (SSA) for SSI/SSDI payments and youth characteristics, Centers for Medicare & Medicaid Services (CMS) for healthcare expenditures, and the Internal Revenue Service (IRS) for annual earnings. These are supplemented by two waves of surveys conducted 18 months and 5 years after youth enrollment.
Juanita Bloomfield, Ana I. Balsa, Alejandro Cid, Philip Oreopoulos, Alejandrina Cristia — Children and Families
This paper evaluates a multi-component, over-the-phone intervention combining teleoperator calls, messages, a chatbot, and an AI tool to improve early childhood outcomes and family well-being in Uruguay.
Finance Application
- This paper's findings are highly relevant for household finance and insurance.
- Financial institutions could adapt similar multi-component digital interventions to improve financial literacy, encourage savings, or facilitate the uptake of beneficial financial products (e.g., micro-insurance, small loans) among vulnerable populations, especially by leveraging the demonstrated success in increasing access to government cash transfers.
- Insurers could integrate such programs to reduce parental stress and improve early childhood outcomes, potentially leading to healthier families and lower future claims costs, thereby informing actuarial models and product design for underserved segments.
Household FinanceBehavioral EconomicsDigital InterventionsRandomized Controlled TrialVulnerable PopulationsFinancial InclusionParental StressEarly Childhood DevelopmentSocial BenefitsFintechInsurance
Core finding, identification, data
Core Finding
- The intervention significantly increased weekly parental engagement in stimulating activities, parental knowledge of language stimulation, and reduced parental stress (effects around 0.20 standard deviations).
- It also improved parents' linguistic interactions with children (higher average word rate and pitch range by 0.37 standard deviations) and increased access to social benefits, including cash transfers and employment support programs (0.30 standard deviations).
Identification Strategy
- The study employs a randomized controlled trial (RCT) design, assigning 1360 families to either a treatment group (receiving the full intervention) or a control group (restricted chatbot access).
- Randomization was stratified by the child's age and mother's education, allowing for causal inference of the intervention's impact.
Data
The paper uses data from two follow-up telephone surveys (at 4 and 8 months post-intervention), WhatsApp audio recordings of parent-child conversations (analyzed by AI for language metrics), and baseline socioeconomic data. Outcomes include access to government transfers, parental investment and knowledge, parental well-being and stress, and child anthropometry.
Livia Chitu, Sofia Gori, Refet S. Gürkaynak — Capital Markets and the Economy
This paper compares market-based and bank-based external finance premia in the US and euro area, revealing that market finance is highly integrated across regions, while bank finance remains country-dependent.
Finance Application
- This research provides a clear framework for distinguishing the pricing and risk characteristics of market-based versus bank-based corporate debt.
- In asset pricing, this could lead to models that explicitly incorporate the funding source (bond vs. loan) as a state variable influencing corporate credit spreads and their sensitivity to monetary policy shocks.
- Researchers could explore whether the 'excess bond premium' is systematically lower or less volatile for firms with diversified market access, or if the 'bank-sovereign doom loop' observed in bank finance has spillover effects on corporate bond pricing for firms heavily reliant on domestic banks.
- This could also inform the design of optimal capital structures for firms operating in monetary unions with varying degrees of financial market integration.
corporate bondsbank loansmonetary policy transmissionexternal finance premiumeuro areaUnited Statesfinancial integrationasset pricingcorporate financeheterogeneityevent study
Core finding, identification, data
Core Finding
- The paper finds that corporate bond spreads (market finance premia) in both the euro area and the US show little dependence on the issuer's state or country of origin, and monetary policy transmission to these spreads is homogeneous.
- Conversely, bank loan spreads, even for the same bond-issuing firms, are strongly determined at the country level, highlighting a fundamental difference in how market versus bank finance is priced and transmits monetary policy.
Identification Strategy
- The identification strategy relies on an event study methodology, regressing changes in corporate bond spreads around FOMC/ECB announcements on 'pure monetary policy surprises' (Jarociński and Karadi, 2020).
- It further examines the interaction of these surprises with a dummy for firms located in 'lower-rated' states or countries to identify differential transmission effects, controlling for various fixed effects and firm characteristics.
Data
The study uses a unique micro-level dataset combining bond-level data (ICE BofAML, Bloomberg, Moody's CreditEdge), firm-level balance sheet information (LSEG Datastream, Orbis), and bank loan data (AnaCredit for the euro area), along with investor composition data (ECB SHSS), covering 2006-2023 at daily frequency.
Alexander Copestake, Divya Kirti, Maria Soledad Martinez Peria, Yao Zeng — Monetary Economics
This paper examines how payments interoperability increases digital payment adoption and usage, particularly in fragmented markets, using India's Unified Payments Interface (UPI) as a natural experiment.
Finance Application
- This research offers direct insights for household finance by demonstrating how interoperability-driven digital payment adoption can expand credit access, particularly for underserved segments like entrepreneurs and hawkers, by generating verifiable transaction data.
- In asset pricing, the framework could be applied to analyze the valuation of fintech firms or digital asset platforms, where interoperability (or lack thereof) influences network effects and market power.
- For insurance, increased digital payment data could enable the development and pricing of novel micro-insurance products for individuals and small businesses, whose financial behavior becomes more transparent.
PaymentsFintechInteroperabilityNetwork EffectsFinancial InclusionDigital PaymentsCredit MarketsIndiaQuasi-ExperimentCausal InferenceHousehold FinanceMarket Structure
Core finding, identification, data
Core Finding
- Interoperability significantly increases digital payment usage, especially in regions with higher initial fragmentation, by expanding effective network size without requiring consolidation on a single platform.
- It led to over a 50% increase in total digital payment usage nationally and boosted lending in more fragmented districts.
Identification Strategy
- The study uses a natural experiment from India where a major digital wallet platform integrated with UPI.
- It employs a heterogeneous adoption design, comparing districts with varying ex-ante fragmentation levels.
- Robustness is checked using matching on observables and an instrumental variable approach, where distance to pre-determined 'hub' cities (where the incumbent platform launched early) instruments for ex-ante fragmentation.
Data
The paper utilizes novel data covering the universe of payments on India's UPI, transaction data from a major Indian fintech firm, NPCI data on ATM cash withdrawals as a proxy for cash usage, and household-level borrowing data from the Consumer Pyramids Household Survey (CPHS).
Tobias Adrian, Christopher Erceg, Marcin Kolasa, Jesper Lindé, Pawel Zabczyk — Monetary Economics
This paper uses a DSGE model to analyze the macroeconomic and fiscal consequences of Quantitative Easing (QE) in deep versus shallow liquidity traps, considering the role of initial financial conditions, QE size, and commitment aspects.
Finance Application
- The paper's detailed analysis of central bank balance sheet dynamics, duration risk, and the interaction with fiscal positions offers rich insights for finance.
- For **asset pricing**, the findings on term premium compression and central bank losses due to unexpected interest rate hikes could inform models of bond market liquidity, the pricing of duration risk in fixed income, and the equity risk premium, especially during periods of unconventional monetary policy.
- In **household finance**, the differential impact of QE on tax revenues and consumption, compared to fiscal stimulus, could be used to model household savings and investment decisions, particularly how perceived future tax burdens or inflation expectations influence long-term financial planning.
- For **insurance**, the explicit modeling of central bank losses from duration risk when interest rates rise unexpectedly provides a direct framework to assess and manage interest rate risk for life insurers and pension funds, which hold large portfolios of long-duration assets and are highly sensitive to yield curve shifts and central bank policy changes.
Quantitative EasingMonetary PolicyFiscal PolicyLiquidity TrapCentral Bank Balance SheetGovernment DebtTerm PremiumInflationDSGE ModelDuration RiskFinancial StabilityAsset PricingHousehold FinanceInsurance
Core finding, identification, data
Core Finding
- QE provides substantial macroeconomic and fiscal benefits in deep liquidity traps, significantly boosting output and inflation while reducing public debt.
- However, in shallow liquidity traps, QE's benefits are smaller, and it carries a greater risk of overheating the economy and generating sizable central bank losses, especially with low initial term premiums, large QE size, or strong commitment to low policy rates.
- Despite potential central bank losses, the consolidated fiscal position generally improves, though these gains are less visible than the losses.
Identification Strategy
- The paper employs a DSGE model featuring bond market segmentation to allow QE to affect term premiums, behavioral discounting to mitigate the forward guidance puzzle, and a nonlinear Phillips Curve to capture asymmetric inflation responses and overheating risks.
- Shocks, including discount factor shocks for liquidity traps and cost-push/consumption demand shocks for recovery scenarios, are calibrated to match US data volatility from 1960-2019, enabling the analysis of QE's differential impacts under various economic conditions and policy commitments.
Data
The study calibrates its DSGE model using empirical evidence from the U.S. and Euro area for monetary and fiscal policy transmission. Stochastic shocks are calibrated to match US data for the 1960-2019 period, specifically targeting the unconditional standard deviations and correlations of output growth, core PCE inflation, nominal wage growth, and hours worked per capita.
Cosmin L. Ilut, Ralph Luetticke, Martin Schneider — Impulse and Propagation Mechanisms
This paper develops and estimates a two-asset HANK model where agents respond to both idiosyncratic and aggregate uncertainty, with the latter modeled as ambiguity to allow for tractable estimation.
Finance Application
This framework, particularly the interaction of ambiguity aversion and financial frictions in a heterogeneous agent setting, could be applied to: 1) Asset Pricing: Investigate how ambiguity aversion affects the pricing of specific illiquid assets beyond aggregate capital, such as private equity, real estate, or structured products, especially during periods of heightened macroeconomic uncertainty. 2) Household Finance: Analyze how different household segments (e.g., by wealth, age, or labor income risk) adjust their allocation to various asset classes (e.g., stocks, bonds, housing) and their demand for insurance products in response to perceived ambiguity about future economic growth or inflation. 3) Risk Management: Model how financial institutions, facing ambiguity about future liabilities or investment returns, adjust their hedging strategies, capital requirements, or pricing of long-term contracts.
HANKAggregate UncertaintyAmbiguity AversionAsset PricingCapital PremiumHousehold FinancePortfolio ChoiceIncomplete MarketsBusiness CyclesBayesian EstimationMacro-Finance
Core finding, identification, data
Core Finding
- Aggregate uncertainty shocks, modeled as ambiguity, are the primary driver of business cycle fluctuations and asset premia in an estimated HANK model.
- Specifically, 3.2% of the 5.5% average capital premium is attributed to aggregate uncertainty, with heterogeneity in household portfolios being crucial for generating this sizable equilibrium compensation.
Identification Strategy
- The model's tractability stems from modeling aggregate uncertainty as ambiguity, which allows first-order effects on steady state and linear dynamics, enabling standard Bayesian estimation.
- The relative importance of aggregate uncertainty shocks versus idiosyncratic risk shocks is identified by their distinct impact on investment dynamics, with only aggregate uncertainty generating a recession with a strong protracted investment slump.
Data
The paper uses quarterly US data from 1985Q1 to 2019Q4, including growth rates of consumption, investment, hours, inflation, the nominal interest rate, and the capital premium. Data sources include the St. Louis FED - FRED database, World Inequality Database (2023), Gomme et al. (2011) for capital premium, and Wu and Xia (2016) for shadow federal funds rates.
Tobias Adrian, Christopher Erceg, Marcin Kolasa, Jesper Lindé, Pawel Zabczyk — International Finance & Macroeconomics
This paper uses a DSGE model to analyze the macroeconomic and fiscal impacts of quantitative easing (QE) under varying economic conditions, particularly distinguishing between deep and shallow liquidity traps and considering uncertainty.
Finance Application
- The paper's insights on central bank losses and term premium dynamics under QE are highly relevant for financial markets.
- Asset pricing models could incorporate the state-dependent effects of QE on bond risk premiums and the yield curve, especially in scenarios of low initial term premiums or large QE programs.
- For household finance, understanding how QE impacts the consolidated fiscal position and tax revenues could inform models of household saving and investment decisions, particularly concerning long-term fiscal sustainability and inflation hedging.
- Insurance companies, with their long-duration liabilities, could use the findings on central bank losses and rapid interest rate hikes in shallow traps to model interest rate risk and optimize asset-liability management strategies.
Quantitative EasingMonetary PolicyFiscal PolicyLiquidity TrapCentral Bank Balance SheetTerm PremiumInflation DynamicsDSGE ModelGovernment DebtFinancial StabilityInterest Rate Risk
Core finding, identification, data
Core Finding
- Quantitative Easing (QE) offers substantial macroeconomic benefits and improves the consolidated fiscal position in deep liquidity traps, even with unexpected faster recovery.
- However, in shallow liquidity traps, QE's benefits are smaller, and it carries a greater risk of overheating and significant central bank losses, especially if initial term premiums are low or QE size is large.
- The paper also finds QE to be fiscally more efficient than conventional fiscal stimulus for a given output boost, despite potential visible central bank losses.
Identification Strategy
- The study employs a DSGE model with segmented bond markets to allow QE to affect term premiums, incorporating behavioral discounting to address the forward guidance puzzle and a nonlinear Phillips Curve to capture state-dependent inflation dynamics and overheating risks.
- It uses stochastic simulations, calibrating shocks to match historical macroeconomic volatility and co-movements in the US (1960-2019), to assess risks under uncertainty.
Data
The model's parameters are calibrated using US data from 1960-2019 to match unconditional standard deviations of output growth, PCE inflation, nominal wage growth, and hours worked, as well as their correlations with output growth. Steady-state ratios are also targeted based on US and Euro area empirical evidence.
Andrea Carriero, Davide Pettenuzzo, Shubhranshu Shekhar — Forecasting & Empirical Methods
This paper comparatively analyzes the accuracy of Large Language Models (LLMs) against traditional macro time series forecasting approaches for macroeconomic variables.
Finance Application
- The methodology of using LLMs for time series forecasting could be directly applied to predicting financial market variables such as stock returns, volatility, bond yields, or commodity prices.
- For instance, an LLM could be trained on a vast corpus of financial news, earnings call transcripts, and market data to forecast future equity risk premia or credit spreads.
- The finding that LLMs perform better in periods of structural breaks (like post-COVID) suggests their utility in forecasting market regimes or identifying turning points in asset prices, which is crucial for dynamic asset allocation strategies or risk management in insurance portfolios, especially for anticipating large claims or market dislocations.
Large Language ModelsTime Series ForecastingMacroeconomicsBayesian VARFactor ModelsForecasting AccuracyFinancial MarketsAsset PricingRisk ManagementMachine LearningDeep Learning
Core finding, identification, data
Core Finding
- The paper finds that while LLMs (specifically Moirai and TimesFM) can be competitive against simple AR benchmarks, their performance is broadly comparable to, or slightly inferior to, established econometric models like Bayesian VARs and Factor Models.
- LLMs show less reliability at times, being prone to occasional unreasonable forecasts, but perform relatively better in the post-Covid-19 era, suggesting an advantage in handling structural breaks.
Identification Strategy
- The study conducts a rigorous comparative evaluation of LLMs against traditional macro-prediction methods (Bayesian VARs, Factor Models, BART, and AR(1) benchmark) using a pseudo out-of-sample forecasting exercise.
- It assesses performance based on Root Mean Squared Forecast Errors (RMSFEs) across various forecast horizons for a comprehensive set of macroeconomic variables from the FRED-MD database, evaluating both zero-shot and fine-tuned LLM performance.
Data
The paper primarily uses the FRED-MD database, a repository of over one hundred monthly macroeconomic variables from the Federal Reserve Economic Data (FRED) system, spanning January 1959 to December 2023. The LLMs themselves are pre-trained on various large time series datasets, some of which overlap with FRED-MD.
Olivier De Jonghe, Daniel Lewis — Forecasting & Empirical Methods
This paper proposes a new model to non-parametrically identify relationship-specific supply and demand shocks in bipartite networks using price and quantity innovations, and applies it to European credit market data.
Finance Application
- This methodology has high potential for application in other finance areas involving bipartite networks with observable prices and quantities.
- In **Asset Pricing**, it could be used to disentangle firm-specific supply shocks (e.g., bond issuance decisions) and investor-specific demand shocks (e.g., liquidity needs, sentiment) in corporate bond markets, explaining cross-sectional variation in bond yields and trading volumes.
- In **Household Finance**, applying it to mortgage markets (households-lenders) could identify heterogeneous demand shocks (e.g., household income changes) and supply shocks (e.g., lender capital constraints) for mortgage contracts, shedding light on housing market dynamics.
- In **Insurance Research**, the model could analyze commercial property & casualty markets (firms-insurers) to identify firm-specific demand shocks (e.g., risk profile changes) and insurer-specific supply shocks (e.g., underwriting capacity), explaining variations in premiums and coverage.
Credit MarketsSupply and DemandHeterogeneous ShocksBipartite NetworksIdentificationEconometricsCorporate FinanceAsset PricingHousehold FinanceInsurance Research
Core finding, identification, data
Core Finding
- The model successfully identifies heterogeneous supply and demand shocks at the firm-bank relationship level, revealing significant within-firm/bank heterogeneity not explained by conventional fixed effects.
- This unexplained heterogeneity correlates strongly with economically meaningful relationship-level characteristics and macroeconomic policy measures, and the elasticities vary considerably across countries and over time, reflecting diverse market dynamics during different economic periods.
Identification Strategy
- The identification strategy non-parametrically identifies the matrix of elasticities (A) linking price and quantity innovations to supply and demand shocks.
- This is achieved by exploiting two covariance matrices: one measuring price and quantity innovations across firms (holding bank fixed), and another across banks (holding firm fixed).
- This approach is inspired by heteroskedasticity-based identification in SVARs but adapted for bipartite networks.
Data
The primary dataset used is the AnaCredit dataset, a harmonized euro area loan-level credit registry from the ECB, covering 11 largest euro area countries from 2019-2023. It includes term loans, credit lines, and revolving credit. Additional data includes Monetary Policy and Central Bank Information shocks, and a Macroprudential Policy index.
Rohan Kekre, Moritz Lenel — International Asset Pricing
This paper develops a general equilibrium model to disentangle the sources of exchange rate fluctuations, finding that persistent relative demand shocks are the primary driver of dollar/G10 exchange rate variance.
Finance Application
- The paper's structural decomposition of exchange rate movements into demand, supply, and intermediation shocks offers a valuable framework for analyzing other asset markets.
- For instance, researchers could apply this to disentangle drivers of equity risk premia or corporate bond spreads, identifying whether movements are due to persistent shifts in investor preferences (demand), changes in fundamental economic productivity (supply), or financial intermediary capacity (intermediation).
- The insight that persistent demand shocks lead to high asset price volatility but low predictability (low R-squared) could be generalized to understand the behavior of long-horizon returns in fixed income or commodity markets, emphasizing the role of long-lived shifts in underlying economic fundamentals.
exchange ratesnatural ratesprice of riskgeneral equilibriumdemand shocksintermediation shockssupply shocksinterest rate differentialsuncovered interest parityasset pricinginternational financemacro-financeyield curveexcess bond premium
Core finding, identification, data
Core Finding
- Persistent shocks to relative demand, reflected in persistent interest rate differentials, account for roughly 75% of the variance in the dollar/G10 exchange rate, leading to a strong dollar when U.S. yields and consumption are high.
- While demand shocks explain long-run movements and comovements, shocks to currency intermediation are crucial for generating high-frequency deviations from uncovered interest parity and explaining dollar appreciation during crises, even when U.S. interest rates are low.
Identification Strategy
- The authors calibrate a two-country general equilibrium model by disciplining demand and supply shocks to match the stochastic properties of 10-year yields and output per capita in the U.S. and G10 countries.
- Currency intermediation shocks are identified by assuming the excess bond premium is a direct measure of arbitrageur risk aversion, up to scale, and their volatility is set to match the observed volatility of real exchange rate changes.
- The model is then inverted to recover the sequence of these shocks that rationalize the exact historical paths of 10-year yields, output per capita, and the excess bond premium.
Data
The study uses end-of-quarter nominal yields (Libor rates and interest rate swaps, and government bond yields) by currency and tenor from Bloomberg, nominal exchange rates, consumer price index, national accounts, and population from the IMF International Financial Statistics for the U.S. and G10 countries (Australia, Canada, Denmark, Euro Area, Japan, New Zealand, Norway, Sweden, Switzerland, U.K.) over the 1991-2020 period. It also uses the excess bond premium of Gilchrist and Zakrajsek (2012) and other proxies for risk (VIX, global factor in risky asset prices, Treasury basis).
Martin Aragoneses, Sagar Saxena — Entrepreneurship
This paper examines the magnitude, allocation, and impact of government-backed venture capital (VC) investment, particularly by the European Investment Fund (EIF), on firm dynamics and aggregate productivity in Europe.
Finance Application
- This research offers insights for asset pricing by demonstrating how a large, policy-driven institutional investor (EIF) can influence private market dynamics and firm valuations.
- It suggests exploring how government as an LP affects the risk-return profiles of VC funds, potentially creating 'policy-induced' alpha or altering liquidity premiums in private equity.
- For household finance, understanding how such industrial policies affect entrepreneurial success and wealth accumulation, especially for young vs. mature firms, could inform models of household portfolio choice and intergenerational wealth transfer.
- The findings on financial market imperfections could also be relevant for pricing credit risk in private debt markets, particularly for firms in sectors or regions targeted by such policies.
Venture CapitalIndustrial PolicyGovernment InvestmentFirm DynamicsFinancial FrictionsAggregate ProductivityPrivate EquityEntrepreneurshipEuropean Investment FundPolicy EvaluationMicro-Macro LinkAsset PricingHousehold Finance
Core finding, identification, data
Core Finding
- Empirically, government-backed VC in Europe is substantial, provides larger funding per firm, and leads to stronger firm growth when focused on young firms, despite a policy bias towards older 'scale-ups'.
- Theoretically, a structural model shows that the optimal allocation of VC funds (early vs. later stage) to maximize aggregate productivity depends critically on the degree of financial market imperfections, suggesting that in markets with mild frictions, reallocating funds to early-stages is more cost-effective than simply increasing the budget.
Identification Strategy
- The empirical analysis uses a stacked (staggered) difference-in-differences specification to estimate dynamic firm-level effects.
- It compares EIF-backed firms to matched peers, employing Propensity Score Matching based on firm age, size, sector, and financial ratios to ensure parallel trends and control for selection bias.
Data
The paper digitizes annual reports from the European Investment Fund (EIF) for fund-level data and capital commitments. It links this with micro-level deal data from PitchBook on private equity and venture capital transactions, and firm-level financial and characteristic data from Orbis, supplemented by Eurostat and CompNet for macro characteristics.
Satyajit Chatterjee, Burcu Eyigungor — Inequality and Macroeconomics
This paper develops a heterogeneous-agent macro model with a competitive credit card industry to explain observed patterns in credit card contract terms, usage, and default rates, and to analyze policy implications like interest rate caps.
Finance Application
- The structural model of credit card supply and demand, incorporating competitive search and heterogeneous agents, could be used to stress-test credit card Asset-Backed Securities (ABS) portfolios under various macroeconomic scenarios or regulatory changes (e.g., different rate cap levels).
- The findings on the drivers of MPCs and their heterogeneity could inform models of household portfolio rebalancing and consumption responses to wealth shocks, impacting predictions for equity market participation or demand for liquid assets.
- The detailed modeling of default behavior and its relationship to utilization rates could also be applied to underwriting models for other unsecured consumer loans or for designing more effective financial literacy programs.
credit cardsconsumer credithousehold debtdefault riskinterest ratescredit limitsmarginal propensity to consume (MPC)heterogeneous agentscompetitive searchrate capswelfare economicsstructural estimation
Core finding, identification, data
Core Finding
- The model demonstrates that heterogeneity in discount factors and default costs is crucial to explain observed patterns in credit card terms, usage, and default rates, particularly the puzzlingly large gap between interest rate spreads and default frequencies.
- It also finds that a 10% interest rate cap, as proposed in legislation, would reduce credit limits for risky borrowers and be welfare-reducing for them, while also implying high marginal propensities to consume (MPCs) due to household impatience.
Identification Strategy
- The model is estimated by matching observed patterns in credit-limit-to-income ratios and default frequencies across credit score quintiles from administrative data.
- The key innovation is extending a canonical heterogeneous-agent macro model with a competitive credit card industry and *ex-ante heterogeneity* in discount factors and default costs, which are then calibrated to fit these empirical moments.
Data
The paper primarily uses administrative data from the Federal Reserve System's Y-14M reports, specifically a 1-in-200 anonymized random sample of new, revolving, general-purpose, unsecured credit card accounts originated between June 2014 and May 2015. It also references Storesletten, Telmer, and Yaron (2004) for earnings process calibration and the 2013 Survey of Consumer Finances for liquid assets distribution comparison.
Justin Katz, Paul S. Willen — CRIW Pre-Conference, Summer 2025
This paper examines how the distribution and scarcity of buildable land, particularly large parcels, influence residential investment, construction productivity, and house price growth in New England.
Finance Application
- This research offers valuable insights for asset pricing, particularly for Real Estate Investment Trusts (REITs) and real estate development funds, by highlighting how land fragmentation and scarcity of large parcels can constrain supply and drive up prices, affecting investment returns and development strategies.
- In household finance, the findings directly impact housing affordability and household wealth accumulation, explaining regional disparities in housing equity and influencing mortgage demand and default risk.
- For insurance, understanding the dynamics of housing supply and price growth due to land constraints can inform property valuation, replacement cost estimation, and long-term risk modeling for homeowners insurance, especially in densely populated or growing areas.
housing supplyland usereal estatehouse pricesconstruction productivityurban economicsparcel sizeresidential investmenthousehold wealthREITsmortgage marketshousing affordability
Core finding, identification, data
Core Finding
- The paper establishes five empirical facts: most buildable parcels are small and large ones are scarce, large parcels have become scarcer over time and in more populous markets, and markets with fewer large parcels experience higher house price growth but lower residential development relative to price growth.
- Counterfactual simulations suggest that recombining small buildable parcels into larger ones, while holding total land fixed, would increase housing supply, improve construction productivity, and reduce house price growth.
Identification Strategy
- The study uses a parcel-level panel dataset to track land development and aggregates data to the zip code level.
- It estimates a log-linear model for development odds, controlling for various demographic characteristics and using year, year-by-state, and county fixed effects.
- The authors acknowledge the challenge of identifying development due to re-parceling and the absence of a direct instrument, relying on robust specifications.
Data
The primary data source is a parcel-level panel of public tax assessor records from the Warren Group, covering Massachusetts, Connecticut, and Rhode Island from 2007 to 2021. This is supplemented by CoStar listings for buildable land for sale and public data from FHFA (house price indices) and the American Community Survey (demographics).
Palaash Bhargava, Sandra E. Black, Jeffrey T. Denning, Robert W. Fairlie, Oded Gurantz — Household Finance
This paper examines how college attendance and its associated costs impact the financial health and debt portfolios of both parents and students, using comprehensive administrative data from California.
Finance Application
- This research provides rich empirical regularities and a robust identification strategy directly applicable to household finance.
- The findings on parental debt reallocation and credit score changes due to college costs can inform models of household portfolio choice, intergenerational wealth transfers, and consumption smoothing under educational expenditure shocks.
- Specifically, the observed substitution between HELOCs and educational loans based on homeownership could refine models of housing wealth utilization and its impact on retirement savings.
- For asset pricing, understanding how these parental financial adjustments affect aggregate household balance sheets could influence predictions of consumer spending and demand for various financial assets, especially those sensitive to household liquidity and credit risk.
Household FinanceIntergenerational TransfersDebtCredit ScoresDelinquencyHigher Education FinanceParent PLUS LoansHome Equity LoansConsumption SmoothingCausal InferenceRegression Discontinuity
Core finding, identification, data
Core Finding
- The study finds that parents primarily reallocate debt rather than increasing total indebtedness when their child attends college.
- Higher-income parents shift from other debt types to educational loans, while lower-income parents take on more educational loans, experience reduced delinquency on non-educational debt, and see improved credit scores.
- Furthermore, grant aid (reducing college price) causally decreases parental educational loans and home equity loans (HELOCs), and lowers parental debt delinquency, with effects varying by homeownership status.
Identification Strategy
The paper employs two main identification strategies: an event-study framework to analyze changes in parental finances around the child's first FAFSA submission, using a Callaway Sant'anna (2021) estimator for staggered treatment; and a regression discontinuity (RD) design that exploits sharp GPA-based cutoffs for Cal Grant eligibility to identify the causal effect of college price changes on parental debt and financial health.
Data
The research utilizes administrative data from the universe of Free Application for Federal Student Aid (FAFSA) submissions by California residents (2006-2015), linked at the individual level to detailed credit and debt records from a large credit bureau via the University of California Consumer Credit Panel (2004-2023).
Titan Alon, Sena Coskun, Jane Olmstead-Rumsey — Gender in the Economy
This paper documents a rising gender divergence in sectors of work in the U.S., primarily driven by married women's changing preferences for certain sector characteristics, which paradoxically contributes to a decline in the gender earnings gap.
Finance Application
- This research highlights how changing preferences for job amenities (e.g., flexibility, childcare compatibility) among married women influence sector choice.
- This could be applied to understanding household financial decision-making, such as savings rates, investment choices, and insurance demand, by linking these decisions to the characteristics of the sectors household members work in.
- For example, households with members in 'flexible' sectors might have different risk tolerances or need for liquidity.
- The finding that 'more gender equal countries are more segregated' suggests a complex interaction between social norms, labor markets, and potentially corporate behavior, which could inform ESG investing strategies, particularly the 'S' component.
- Insurers could also tailor products based on sector-specific risk profiles and gender-specific preferences, especially concerning income protection or disability insurance.
gender economicslabor economicssegregationpreferencesdiscriminationearnings
Core finding, identification, data
Core Finding
- Gender segregation by sector has increased significantly in the U.S. since 1980, a trend almost entirely driven by married women.
- This rise is largely explained (59%) by changing preferences of married women for sectors offering amenities like part-time work and compatibility with childcare.
- While increasing segregation, these changing preferences also contribute to a reduction (23%) in the gender earnings gap, as women increasingly choose higher-paying sectors that align with their preferences.
Identification Strategy
- The paper uses a structural model based on Hsieh et al. (2019) to estimate sector-specific discrimination and preferences for different cohort-groups (men, single women, married women).
- The model assumes men face no discrimination and innate human capital is similar across groups.
- Counterfactuals are then performed by fixing preferences, discrimination, or technology to 1980s levels to decompose their contributions to changes in segregation and earnings gaps, leveraging the idea that earnings gaps reflect selection effects and relative sectoral shares, given discrimination, reveal relative preferences.
Data
The study primarily uses data from the Current Population Survey (CPS) for employment, hours, earnings, gender, marital status, and age, focusing on 5 cohorts aged 25-34 from 1975-2019 across 11 sectors. It also uses EU Labour Force Survey data for cross-country comparisons.
David Burgherr — Economics of Social Security
This paper investigates how mandatory pension contributions affect total savings and portfolio composition, finding significant crowding-out of private financial assets, especially for less liquidity-constrained individuals.
Finance Application
- This research directly informs household finance by showing how mandatory savings policies impact household portfolio allocation and liquidity management.
- It suggests that policies forcing illiquid savings might increase financial fragility for liquidity-constrained households, potentially increasing their demand for short-term credit or emergency funds.
- For asset pricing, understanding the shift from liquid financial assets to illiquid pension accounts could influence demand for different asset classes, especially if such mandates are widespread or change over time.
- Insurers could also use these insights to better model demand for liquidity-providing products or assess risk for households with varying levels of illiquid retirement savings.
Household FinanceSavings BehaviorPension PlansCrowding-OutPortfolio AllocationLiquidity ConstraintsRegression DiscontinuityAdministrative DataWealth ManagementRetirement Planning
Core finding, identification, data
Core Finding
- Mandatory pension plans have limited effects on total savings, with an estimated crowd-out rate of 73%.
- Workers largely offset mandatory contributions by reducing private non-retirement savings, primarily in liquid financial assets (bank accounts, stocks, bonds, and other investments) by CHF 260, while other asset types (business, property, other wealth, debt) are largely unaffected.
- Liquidity-constrained workers are less able to reduce private savings, leading to an increase in their total savings by CHF 200-350.
Identification Strategy
- The study employs a sharp regression discontinuity (RD) design, leveraging the exogenous variation in mandate coverage created by an earnings threshold in the Swiss occupational pension plan.
- Workers just above this threshold are mandated to contribute, while those just below are not, allowing for causal inference by comparing otherwise similar individuals.
Data
The paper uses detailed administrative income, wealth, and savings tax microdata from the canton of Bern, Switzerland, spanning 2005-2017. This data provides comprehensive information on various asset types (financial, business, property, other wealth, and debt) and allows for household-level analysis.
Greg Buchak, Gregor Matvos, Tomasz Piskorski, Amit Seru — Real Estate
This paper uses a calibrated dynamic structural search model to analyze how reduced real estate agent commissions, stemming from the NAR settlement, impact home prices, housing turnover, and consumer welfare.
Finance Application
- This paper's structural model and its findings on transaction costs and asset valuation have significant implications for asset pricing and household finance.
- The 'valuation effect' mechanism, where lower future transaction costs increase an asset's durable value and thus its current price, could be applied to other illiquid assets like private equity, venture capital, or even collectibles, to understand how changes in trading frictions (e.g., improved secondary markets, lower brokerage fees) affect their fundamental valuation and equilibrium prices.
- In household finance, the model could be extended to analyze how transaction costs in other major household assets (e.g., cars, investment properties) influence household portfolio allocation, mobility decisions, and wealth accumulation, especially for credit-constrained households.
- The finding that benefits accrue disproportionately to existing owners could inform policy discussions on wealth inequality and access to asset markets.
Household FinanceReal EstateTransaction CostsAsset ValuationStructural ModelSearch FrictionsConsumer WelfareHousing AffordabilityMarket EquilibriumAgent Commissions
Core finding, identification, data
Core Finding
- Contrary to common belief, reducing real estate agent commissions generally leads to higher house prices.
- This occurs because lower future transaction costs increase the perceived value of housing as a durable asset, boosting demand and allowing sellers to command higher prices.
- While overall consumer welfare increases, these benefits primarily accrue to existing homeowners, with financially constrained prospective buyers potentially being crowded out by rising prices.
Identification Strategy
- The paper employs a calibrated dynamic structural search model of the housing market, building on Buchak et al. (2022).
- It identifies the effects of commission changes through counterfactual simulations, comparing equilibrium outcomes (prices, welfare, turnover) under various fee structures (e.g., seller pays all fees, decoupled fees) against a baseline 6% commission.
- The model is calibrated to match key empirical moments like house prices, time on market, and seller-to-buyer ratios.
Data
The study calibrates its model using data from various sources, including HUD, public Zillow data, and NAR surveys, to match observed house prices, time on market, average buyer viewings, and the seller-to-buyer ratio.
You Suk Kim, Feng Liu, David Zhang — Real Estate
This paper examines how changes in government-sponsored enterprise (GSE) guarantee fees (g-fees) affect the volume and distribution of home purchases across different borrower income and credit score groups.
Finance Application
- This research directly informs household finance by quantifying how government interventions in mortgage markets can exacerbate or mitigate wealth inequality through homeownership.
- For asset pricing, the finding that g-fee changes alter the *composition* of home buyers (allocative effect) suggests that the risk profile of underlying mortgage pools (and thus MBS tranches) will change, impacting MBS pricing and risk premia.
- In insurance, the interaction between GSE g-fees and private mortgage insurance (PMI) could be explored: if g-fees subsidize certain borrower segments, it might crowd out PMI demand or shift risk burdens, warranting research into optimal public-private risk sharing.
Household FinanceReal Estate FinanceMortgage MarketsGovernment-Sponsored Enterprises (GSEs)SecuritizationMortgage-Backed Securities (MBS)Policy EvaluationDifference-in-DifferencesCredit RiskIncome InequalityHousing SearchFinancial Intermediation
Core finding, identification, data
Core Finding
- GSE g-fee pricing has significant distributional effects on home purchases, which may be regressive in the income dimension.
- A 100 basis point decline in g-fees (leading to a 27-32 bps decline in fee-adjusted interest rate) increases relative home purchase originations by 13.9%.
- These effects are largely driven by an allocative mechanism, where changes in financing costs alter which borrowers successfully purchase homes, rather than solely increasing overall buyer entry.
Identification Strategy
- The study employs a stacked difference-in-differences (DID) design, exploiting discontinuous changes in g-fees (Loan-Level Price Adjustments, LLPAs) across specific credit score cutoffs (e.g., 640, 680, 720, 760, 780) following a major policy adjustment in January 2023.
- It compares home purchase volumes for borrowers just above and below these cutoffs, pre- and post-policy change, while controlling for various fixed effects and unobserved heterogeneity.
Data
The paper uses confidential HMDA data from 2022-2023, which includes loan origination dates, credit scores, property values, and loan amounts. It also utilizes the National Mortgage Database (NMDB), a 1-in-20 sample of first-lien mortgages with information on non-mortgage debts from Experian.
Basil Halperin, J. Zachary Mazlish — Impulse and Propagation Mechanisms
This paper documents four robust facts about macroeconomic expectation biases across different forecast horizons and countries, showing that while long-term expectations overreact more, it is short-term expectations that are most strongly associated with economic fluctuations.
Finance Application
- The finding that short-term expectations, despite exhibiting less overreaction, are the primary drivers of economic and stock market fluctuations has significant implications for asset pricing.
- It suggests that models focusing solely on long-term extrapolative biases might miss key short-term market dynamics.
- Researchers could develop new asset pricing factors based on short-term expectation revisions across different asset classes or explore how these biases affect the pricing of short-dated derivatives.
- In household finance, understanding how households form short-term expectations about inflation or income could explain their immediate consumption and saving responses, or their timing of market entry/exit.
- For insurance, the horizon-dependent biases could lead to mispricing of policies with different durations, as insurers' short-term risk assessments might be underreactive while long-term ones are overreactive.
Macroeconomic expectationsBehavioral financeForecast biasOverreactionUnderreactionForecast horizonAsset pricingStock marketGDP growthInflationInvestmentConsumptionSticky informationCostly recall
Core finding, identification, data
Core Finding
- The paper establishes four key facts: (1) expectations under-react at short horizons (one year or less), (2) over-react at long horizons (two years or more), (3) are "too extreme" at all horizons, and (4) overreaction and over-extremity increase with forecast horizon.
- These patterns hold across 89 countries and multiple macroeconomic variables.
- Crucially, despite stronger overreaction in long-term expectations, short-term expectations are most predictive of fluctuations in GDP, investment, and the stock market.
Identification Strategy
- The paper primarily documents empirical facts using panel regressions of forecast errors on forecast revisions and lagged forecasts, pooled across countries and macroeconomic variables.
- It employs z-scoring of forecasts (with an expanding window) and country-variable fixed effects to control for heterogeneity and prevent volatility from dominating results.
- An out-of-sample forecasting exercise demonstrates the predictive power of these identified biases.
Data
The study uses survey data on macroeconomic expectations from Consensus Economics, covering 89 countries from 1989 onwards, with forecast horizons from 0 to 10 years, for GDP growth, inflation, consumption growth, and investment growth. Realized outcomes are sourced from the World Bank's World Development Indicators (WDI), and stock market data from WRDS "Daily World Indices."
Sterre Kuipers, Hyunju Lee, Radek Paluszynski — International Trade and Macroeconomics
This paper develops a quantitative model to identify the joint evolution of time-varying trade and financial frictions by matching observed gross trade and capital flows, and analyzes their effects on the US trade deficit.
Finance Application
- The paper's methodology of backing out time-varying financial frictions (capital controls) from observed gross capital flows could be directly applied in asset pricing to understand how implicit market access costs affect cross-border asset valuations and capital allocation.
- For instance, researchers could use this approach to identify frictions impacting specific international bond or equity markets, explaining deviations from covered interest parity or equity home bias.
- In household finance, the identified persistence of financial friction shocks could inform models of optimal international portfolio diversification and long-term wealth management for individuals facing varying degrees of capital mobility.
GlobalizationInternational TradeFinancial IntegrationCapital FlowsTrade DeficitFinancial FrictionsTrade CostsQuantitative ModelRisk SharingInternational Macroeconomics
Core finding, identification, data
Core Finding
- The reduction in trade costs is identified as a primary driver and prerequisite for both trade and financial globalization, increasing the trade deficit.
- Financial integration, while also crucial for gross capital flows, can either widen or reduce the trade deficit and its shocks have more persistent effects compared to the short-lived impacts of trade cost shocks.
Identification Strategy
- The paper identifies time-varying trade costs, financial frictions, and foreign output shocks by matching three observed data series: gross trade flows to GDP, net exports to GDP, and gross financial positions to GDP.
- This is achieved using a Maximum Likelihood estimation combined with an iterative procedure to infer the paths of these 'wedges' in a quantitative international trade and financial integration model, building on the Chari, Kehoe and McGrattan (2007) methodology.
Data
The study uses US data against the Rest of the World for the period 1973-2024. Key data series, including trade and capital flows over GDP, are sourced from the International Transactions (ITA) data by the Bureau of Economic Analysis (BEA).
Zhengyang Jiang, Arvind Krishnamurthy, Hanno Lustig — International Asset Pricing
This paper demonstrates that standard frictionless models fail to explain key exchange rate puzzles, but introducing small 'wedges' in bond Euler equations, interpreted as home bias, convenience yields, or financial repression, can resolve these discrepancies.
Finance Application
- The concept of 'Euler equation wedges' could be broadly applied in asset pricing to diagnose and resolve puzzles in other markets where frictionless arbitrage is assumed but empirical deviations persist, such as in commodity futures or credit default swaps.
- In household finance, the 'home bias' interpretation of wedges could be used to quantify the implicit costs or convenience yields that drive household portfolio choices, explaining why individuals might under-diversify or favor local investments.
- For insurance research, the 'financial repression' wedge offers a framework to analyze how regulatory mandates (e.g., requiring insurers to hold government bonds) distort investment decisions and asset valuations within the insurance sector, impacting capital requirements and risk management.
Exchange RatesAsset PricingInternational FinanceEuler EquationsMarket FrictionsHome BiasConvenience YieldsFinancial RepressionPredictabilityRisk PremiaMacro-Finance
Core finding, identification, data
Core Finding
- Standard international real business cycle models, assuming frictionless trading of home and foreign risk-free bonds, cannot simultaneously reconcile the Backus-Smith puzzle (counter-cyclical exchange rates), the Fama puzzle (deviations from UIP), and the Meese-Rogoff puzzle (lack of exchange rate predictability).
- The paper analytically shows that 'wedges' in the bond Euler equations, representing small frictions (e.g., 40 basis points), are necessary to align models with these empirical facts, without requiring extreme market segmentation.
Identification Strategy
- The paper's identification strategy is theoretical, deriving necessary conditions (bounds) for exchange rate cyclicality, predictability, and Fama regression coefficients under frictionless Euler equations.
- It then demonstrates how the introduction of 'wedges' into these equations, representing various market frictions, relaxes these bounds, allowing the model to match observed empirical regularities that are otherwise impossible to explain.
Data
The paper relies on stylized facts and empirical regularities from the exchange rate literature, such as the negative Fama b-coefficient, low R-squared values for exchange rate predictability, and the Backus-Smith puzzle. It uses empirically plausible values for calibration (e.g., R^2 = 5%, b = -1, std(∆st+1) = 10%) and references existing asset pricing models (Bansal and Yaron, Campbell and Cochrane) for distributions of SDF volatility.
Ararat Gocmen, Clara Martinez-Toledano, Vrinda Mittal — Household Finance
This paper examines how the growth of private capital markets, driven by high-net-worth individuals' early-stage investments, contributes to rising economic inequality in the U.S., leveraging federal tax reforms as a quasi-exogenous shock.
Finance Application
- This research offers several insights for finance.
- The documented excess returns of HNWIs in early-stage private markets, especially the highly skewed distribution with a few outsized gains, is crucial for understanding the private equity premium and portfolio allocation for ultra-high-net-worth individuals.
- Asset pricing models could incorporate this heterogeneity in private asset returns and investor access.
- Household finance could explore how tax incentives like QSBS influence portfolio choices, wealth accumulation, and intergenerational wealth transfer strategies among the wealthy, potentially leading to new models of optimal portfolio choice under specific tax regimes and access to private markets.
private capitalprivate equityventure capitalhigh-net-worth individualswealth inequalityincome inequalitytax incentivesQSBSasset returnsportfolio choicehousehold financeasset pricingentrepreneurial finance
Core finding, identification, data
Core Finding
- The expansion of the Qualified Small Business Stock (QSBS) tax exclusion significantly increased HNWIs' early-stage investments, leading to a 3.5 percentage point higher probability for eligible companies to stay private and a 7.2% increase in the income gap between HNWIs and other earners.
- These excess returns on private investments, compared to public markets, explain 2% and 14% of the growth in the top 0.5% share of post-tax income and wealth, respectively, suggesting a feedback loop where inequality fuels further private market participation.
Identification Strategy
- The study employs a difference-in-differences design, exploiting the quasi-exogenous shock of federal tax reforms (QSBS exclusion expansion in 2009-2010, made permanent in 2015) that incentivized HNWIs to invest in early-stage companies.
- They use both company-level variation (QSBS-eligible vs. ineligible firms) and state-level variation (based on the ex-ante number of resident HNWIs) to identify the causal effects on private market participation and inequality.
Data
The paper primarily uses Pitchbook data for private capital market activity (financing, investors, company valuations), complemented by IRS Statistics of Income for state-level income inequality, the Survey of Consumer Finances (SCF) for wealth distribution, Forbes 400 rich lists, and the GEOWEALTH-US database for the number of HNWIs per state.
Goetz von Peter, Bryan Hardy, Patrick McGuire, Torsten Ehlers, Sonya Zhu — International Finance and Macroeconomics Data Session
This series of articles from the BIS Quarterly Review aims to enhance the understanding and use of BIS international financial statistics by providing compact explanations of key concepts and corresponding data in international finance.
Finance Application
- This comprehensive view of international finance could be applied in asset pricing to better model global risk factors and their impact on cross-border equity and bond returns, especially differentiating between residence and nationality of issuers.
- In household finance, it could inform studies on the vulnerability of households with foreign currency-denominated debt to exchange rate shocks.
- For insurance, it offers a framework to assess systemic country risk for political risk insurance and trade credit insurance, by revealing the true consolidated exposures of multinational entities.
International FinanceBIS StatisticsResidence vs NationalityCurrency DimensionBank ExposuresCountry RiskGlobal CreditFinancial StabilityMultinational BanksBond MarketsDerivatives Markets
Core finding, identification, data
Core Finding
- The series collectively demonstrates that traditional residence-based statistics often obscure the true global interconnectedness and risk exposures in international finance.
- By adopting a nationality perspective and incorporating currency and geographical dimensions from BIS data, a more accurate picture emerges of multinational banks' and non-banks' consolidated balance sheets, foreign currency debt vulnerabilities, and the drivers of cross-border financial positions.
Identification Strategy
- The methodological approach involves a systematic comparison of financial statistics compiled under different conceptual frameworks: the standard residence-based view versus a nationality-based view that groups balance sheets by headquarters.
- This is complemented by granular data on currency denomination and geographical booking locations from BIS international banking, debt securities, and derivatives statistics to reveal hidden exposures and interdependencies.
Data
The paper primarily utilizes various Bank for International Settlements (BIS) statistics, including those on international banking (locational and consolidated), debt securities, and derivatives markets, often comparing residence-based and nationality-based views.
Young Soo Jang, Sharjil Haque, Jessie Jiaxu Wang — Corporate Finance
This paper investigates how banks' financing of nonbank lenders (Business Development Companies, BDCs) influences monetary policy transmission to firms, revealing a quantity-price tradeoff during monetary tightening.
Finance Application
- This research highlights how monetary policy impacts corporate credit risk and borrowing costs through indirect channels, which could be integrated into asset pricing models to better predict corporate bond spreads, private credit fund returns, and private equity valuations, especially during tightening cycles.
- The mechanism of non-bank lenders amplifying borrowing costs could also be relevant for household finance, investigating if similar 'price channel' amplification occurs for household debt (e.g., mortgages, consumer loans) from fintech lenders.
- Insurance companies, as major institutional investors in private credit, could use these insights to refine their asset allocation strategies and risk management for private credit portfolios.
Monetary Policy TransmissionNonbank LendingPrivate CreditBusiness Development Companies (BDCs)Credit ChainsCorporate FinanceCredit RiskInterest RatesFinancial DistressAsset PricingHousehold FinanceInstitutional Investors
Core finding, identification, data
Core Finding
- During monetary tightening, banks reallocate credit to BDCs, which then lend to firms.
- This intermediation mitigates credit contraction (quantity channel) by sustaining lending volume but amplifies monetary transmission by elevating borrowing costs for firms and increasing their financial distress risk (price channel).
- Bank-reliant BDCs respond more strongly to tightening, passing higher funding costs to borrowers, who exhibit greater growth but weaker interest coverage ratios.
Identification Strategy
- The paper employs a difference-in-differences approach using a dummy for the 2022 monetary tightening cycle and changes in the effective Federal Funds rate, robust to monetary policy shocks (Jarociński and Karadi (2020), Bauer and Swanson (2023)).
- It uses granular bank-internal credit ratings and bank×credit rating×year-quarter fixed effects to control for risk and lender heterogeneity.
- For BDC lending to firms, it uses a Khwaja and Mian (2008)-style identification with borrower×year-quarter×loan type fixed effects on overlapping borrowers.
Data
The study uses the Federal Reserve's supervisory Y-14 loan-level data for bank loan information and Refinitiv's BDC Collateral dataset for BDC financials and investments.
Francesca Bastianello, Paul Décaire, Marius Guenzel — Corporate Finance
This paper uses large language models (LLMs) to extract and analyze the detailed mental models of financial professionals from equity analyst reports, linking their reasoning to quantitative forecasts and asset pricing patterns.
Finance Application
- The LLM-based methodology for extracting structured mental models from unstructured text could be applied to various financial contexts.
- For instance, it could be used to analyze central bank communications to infer policymakers' attention and beliefs about inflation or financial stability, impacting fixed income markets.
- In household finance, it could extract mental models from financial advice forums to understand retail investor biases and their impact on portfolio choices.
- For asset pricing, the framework could be extended to corporate earnings call transcripts to measure management's mental models regarding future growth and risk, potentially predicting stock returns or M&A activity.
Mental ModelsFinancial ForecastsLarge Language ModelsLLMsTextual AnalysisAttentionBelief FormationAsset PricingBehavioral FinanceAnalyst ReportsForecast BiasReturn Predictability
Core finding, identification, data
Core Finding
- Analysts exhibit sparse and rigid mental representations, with their attention allocation tightly linked to chosen valuation methods.
- Differences in attention, more than valuation methods, drive changes in valuations and disagreement.
- Analysts overreact to firm-specific topics and underreact to macro-related ones, which translates into predictable asset price patterns where overreacted topics predict lower realized returns and underreacted topics predict higher returns.
Identification Strategy
- The paper's methodological innovation involves a multi-step LLM prompting strategy and diagnostic tools akin to bootstrapping to reliably extract detailed lines of reasoning (topic, valuation channel, time horizon, sentiment) and valuation methods from unstructured text in analyst reports.
- This approach addresses limitations of single-step prompting and ensures comprehensive and stable output.
Data
The study uses a near-universe of 2.1 million equity analyst reports from 43 brokerage houses over a 26-year period (2000-2025) from Refinitiv Eikon, with a subset of over 300,000 reports used for detailed reasoning extraction, yielding 11.8 million lines of reasoning.
Joseph G. Haubrich — Capital Markets and the Economy
This paper applies stochastic inventory theory to determine the optimal buffer size for Federal Reserve reserves, balancing interest rate volatility against balance sheet costs.
Finance Application
- The methodology of applying stochastic inventory theory and a 'revealed preference' calibration to buffer management could be highly valuable in corporate finance.
- For instance, it could be used to model optimal corporate cash holdings, balancing the cost of illiquidity (e.g., costly external financing, missed investment opportunities) against the opportunity cost of holding cash.
- In asset pricing, this framework could inform models of liquidity risk premiums, where firms' optimal cash buffers influence their resilience to market shocks and thus their valuation.
- For insurance, it could optimize capital reserves, trading off regulatory capital costs against the risk of insolvency from unexpected claims.
Monetary PolicyCentral BankingBalance SheetReservesInventory TheoryStochastic ModelsFederal Funds RateQuantitative Tightening (QT)Liquidity Management
Core finding, identification, data
Core Finding
- The optimal buffer of reserves for the Federal Reserve, calculated using a revealed preference approach and stochastic inventory theory, is estimated to be relatively small (ranging from $8.1 billion to $356.1 billion) compared to the total level of ample reserves.
- This suggests the FOMC implicitly values the cost of reserves falling below 'ample' as significantly higher (approximately 3.3 times) than the cost of maintaining a large balance sheet.
Identification Strategy
- The paper employs a 'revealed preference approach' to calibrate the optimal buffer.
- It equates the critical fractile (the probability of the fed funds rate rising above its target or the IORB) to the historical frequency of such events.
- This observed frequency is then used in the first-order condition of a stochastic inventory model, which is inverted using the estimated distribution of supply shocks (changes in reserves or Treasury General Account) to determine the optimal buffer size.
Data
The study utilizes the Effective Fed Funds Rate (EFFR), Interest on Excess Reserves (IOER)/Interest on Reserve Balances (IORB), and Federal Funds Target Range data from the Federal Reserve. It also uses confidential daily data from the Federal Reserve FR-34 report and publicly available weekly balance sheet data for reserve changes, as well as daily changes in the Treasury General Account (TGA).
Dominic Cucic, Rajkamal Iyer, Sotirios Kokas, Jose-Luis Peydro, Stefano Pica — Corporate Finance
This paper examines how increased deposit insurance coverage during a crisis distorts deposit and credit allocation, disproportionately benefiting weaker banks and leading to riskier lending.
Finance Application
- The findings on how government guarantees distort risk pricing and capital allocation could be applied to asset pricing by examining how implicit government backstops for 'too-big-to-fail' corporations (beyond banks) affect their bond yields and equity valuations, leading to misallocation of capital in the broader economy.
- In household finance, the observed shift in depositor behavior could inform research on how retail investors' perceived safety of various investment vehicles (e.g., money market funds, short-term bond funds) influences their portfolio choices during crises, especially if some are seen as implicitly guaranteed.
- For insurance research, the mechanism of disrupted market discipline could be explored in government-subsidized insurance markets (e.g., flood or crop insurance), analyzing how changes in coverage affect insured parties' risk-taking behavior and the pricing of private insurance alternatives.
Deposit InsuranceFinancial CrisisBank RunsMarket DisciplineCredit AllocationLoan-to-Deposit RatioBank VulnerabilityMoral HazardFinancial StabilityHousehold BehaviorRisk Pricing
Core finding, identification, data
Core Finding
- Individual depositors impose market discipline on weaker banks through withdrawals of uninsured deposits, but expanded insurance coverage disrupts this mechanism.
- This disproportionately benefits weaker banks, enabling them to expand lending to lower-quality firms that subsequently default without appropriate risk pricing, highlighting a trade-off between short-term financial stability and the costs of increasing deposit insurance coverage.
Identification Strategy
- The study exploits a natural experiment from Denmark's deposit insurance reform during the Global Financial Crisis (GFC) in October 2008, which eliminated the DKK 300,000 coverage limit and introduced unlimited deposit insurance.
- It compares deposit and credit allocation in 2008:Q3 (limited insurance) versus 2008:Q4 (unlimited insurance), using a threshold design around the DKK 300,000 limit and individual-bank fixed effects to control for depositor sorting.
- For credit, it uses firm-bank level fixed effects to isolate supply-side effects.
Data
The paper uses comprehensive administrative datasets from Statistics Denmark, including a deposit register covering over 6 million individuals (2004-2010), a credit register for non-mortgage loans to 100,000 non-financial firms, and administrative data on bank and firm balance sheets, income statements, and individual depositors' wealth and income from tax records. Quarterly supervisory data is also used.
Mary Amiti, Anil K Kashyap, Anna Kovner, David Weinstein — Capital Markets and the Economy
This paper quantifies the search costs for firms to switch banks using confidential supervisory data, finding significant interest rate dispersion among similar loans not explained by risk.
Finance Application
- The methodology for estimating search costs from price dispersion could be applied to other financial markets.
- In asset pricing, this could inform models of corporate bond pricing, where firms with higher switching costs might face higher and more volatile funding costs, impacting their bond yields and equity valuations.
- In household finance, the framework could quantify search costs for mortgages or auto loans, explaining price dispersion and its impact on household wealth and default risk.
- For insurance, it could reveal why consumers pay different premiums for similar coverage, influencing insurer pricing strategies and market efficiency.
BankingCorporate FinanceSearch CostsLoan PricingCredit SupplyMarket FrictionsInformation AsymmetryMacrofinance
Core finding, identification, data
Core Finding
- The paper documents significant variation in interest rates for observably similar commercial and industrial loans, which traditional risk factors cannot explain.
- A search cost model reveals that these costs are highest for smaller and riskier borrowers, leading to substantial price dispersion.
- A counterfactual exercise estimates that a bank failure forcing borrowers to switch lenders would incur a deadweight loss of at least 11% of the bank's loans.
Identification Strategy
- The authors develop a search cost model, adapting methodologies from industrial organization, to estimate borrower switching costs from observed interest rate dispersion.
- They group loans into 'bins' based on multiple observable characteristics (risk rating, loan size, maturity, purpose, interest rate regime) to control for risk, and then attribute the remaining within-bin interest rate variation to search costs, estimated via a maximum likelihood approach.
Data
The study uses a unique matched U.S. bank-firm dataset from the Federal Reserve's Dodd-Frank Act mandated stress testing (FR Y-14), specifically the H.1 schedule, which includes comprehensive data on commercial and industrial loans from the largest U.S. banks.
Ginha Kim, James Traina — Capital Markets and the Economy
This paper develops a new method to measure firm-level markups by integrating cost share analysis with implied cost of capital techniques, revealing stable average markups but increasing dispersion driven by financing advantages.
Finance Application
- This framework directly links financial market conditions (cost of capital) to real economic outcomes (markups/market power).
- Asset pricing researchers could use the negative relationship between markups and ICC to construct new factors or signals for stock returns, testing whether firms with persistently lower financing costs (and thus higher markups) exhibit different risk-adjusted returns or valuation multiples.
- Furthermore, the increasing dispersion in markups driven by financing advantages could inform models of firm heterogeneity and its implications for aggregate market returns or cross-sectional anomalies.
MarkupsCost of CapitalProduction FunctionFirm HeterogeneityAsset PricingMarket PowerImplied Cost of CapitalSuperstar FirmsCorporate Finance
Core finding, identification, data
Core Finding
- The study finds that average markups for US public firms remained stable and competitive from 1970 to 2022, but markup dispersion significantly increased, with top-percentile markups rising.
- A strong negative relationship exists between markups and the implied cost of equity capital, suggesting that firms with financing advantages can sustain higher markups, supporting the "superstar firm" hypothesis.
Identification Strategy
- The paper's methodological innovation combines the cost share approach to production function estimation (from industrial organization/macro) with firm-specific implied cost of equity capital (ICC) estimates (from finance).
- ICC is recovered by applying the present value relation to market values and expected future profits, derived from an earnings forecast model (Gebhardt et al., 2001; Hou et al., 2012), thereby resolving identification problems in production function estimation by accurately measuring capital costs.
Data
The paper uses a panel of US public firms from 1970 to 2022 from Compustat North America, including accounting data for earnings, sales, operating expenses, interest expense, depreciation, invested capital, and market values.
Kinda Cheryl Hachem, Martin Kuncl — Capital Markets and the Economy
This paper develops a theoretical model of financial regulation incorporating shadow banking, bailouts, and imperfect information, demonstrating that state-contingent regulation is more susceptible to circumvention than non-contingent regulation.
Finance Application
This paper's insights on regulatory circumvention, particularly for state-contingent instruments, could be applied to: 1. **Asset Pricing**: Investigate how the *perceived* state-contingency of financial instruments (e.g., catastrophe bonds, contingent capital, or certain derivatives) is affected by shadow activities, and how this divergence from *actual* state-contingency impacts their pricing and risk premia. 2. **Household Finance**: Explore how regulations designed to protect households in specific adverse states (e.g., mortgage forbearance, student loan relief, or insurance policy triggers) might be circumvented by financial institutions through complex product structures or hidden fees, leading to suboptimal household outcomes and systemic risk. 3. **Insurance**: Analyze how the effectiveness of state-contingent insurance contracts (e.g., business interruption insurance, cyber insurance) is undermined by insurers' shadow activities that reduce their actual payout obligations during stress events, potentially leading to underestimation of systemic risk in the insurance sector.
financial regulationshadow bankingstate-contingent regulationbailoutsinformation asymmetryregulatory circumventioncontingent convertible bondsliquidity riskpecuniary externalitiesbank capital
Core finding, identification, data
Core Finding
- The paper finds that the threat of shadow activities disproportionately constrains state-contingent regulation compared to non-contingent regulation.
- Under imperfect information and lack of commitment, shadow activities are triggered with positive probability, leading to larger bailouts, particularly when state-contingent regulation is circumvented, as its costs directly lower asset sale prices in stress states.
- Empirical evidence supports that banks use credit lines to circumvent state-contingent CoCo regulation.
Identification Strategy
- The theoretical model identifies the differential impact of shadow activities on two types of regulation: non-contingent (ex-ante liquidity ratio) and state-contingent (ex-post haircuts).
- It explicitly models two distinct shadow activities with different cost structures and impacts on bank balance sheets.
- The empirical section uses a regression framework to identify the association between CoCo issuance (a state-contingent regulatory instrument) and a proxy for shadow activities (fee/commission income, specifically credit lines), controlling for bank-specific factors and capital ratios, and further examines pricing effects to infer the nature of the shadow activity.
Data
The empirical analysis uses data on Contingent Convertible Bonds (CoCos) issued by European banks from Bloomberg and Thomson Reuters Eikon. Financial statement information, including capital ratios and income components, is sourced from Bloomberg and S&P Capital IQ.
T. Niklas Kroner — Monetary Economics
This paper demonstrates that investor attention significantly influences financial market reactions to macroeconomic news, particularly CPI releases, leading to stronger responses and short-term overreaction.
Finance Application
- The methodology for quantifying investor attention from news coverage and search data could be adapted to study attention around corporate events (e.g., earnings announcements, M&A deals) and its impact on stock prices, trading volume, and post-event drift.
- The finding that high attention leads to overreaction suggests that attention-driven mispricing could exist in other asset classes, creating opportunities for attention-based trading strategies.
- Furthermore, the framework could be used in household finance to analyze how retail investor attention to specific investment products or financial scams influences their behavior and market outcomes.
Investor attentionMacroeconomic newsCPIInflationMarket reactionsEvent studyHigh-frequency dataOverreactionMonetary policyAsset pricingBehavioral finance
Core finding, identification, data
Core Finding
- The market reaction to Consumer Price Index (CPI) releases sharply increased during the 2021-2023 inflation surge, with bond yields and inflation expectations responding significantly more strongly than to other macro announcements.
- This heightened sensitivity is primarily driven by increased investor attention to CPI news, which also leads to market overreaction in the short term.
Identification Strategy
- The study employs a high-frequency event study design, comparing market reactions to 16 major macroeconomic news announcements across low- and high-inflation periods.
- It constructs pre-announcement investor attention measures (CPI-IA) from Bloomberg Terminal news coverage and other sources, and uses a simple information acquisition model to distinguish attention-based explanations from alternative mechanisms like changes in monetary policy response or signal informativeness.
Data
The paper uses intraday financial data from LSEG Tick History (interest rates, inflation swap rates, S&P 500, VIX, commodities, corporate bonds, FX, Bitcoin) from 2009-2023. Macroeconomic news surprises are constructed from Bloomberg's U.S. Economic Calendar. Investor attention measures are derived from Bloomberg Terminal news coverage, Dow Jones Newswires, Google Searches, and Mainstream Media. Various uncertainty measures are also used as controls.
Jonathon Hazell, Stephan Johan Hobler — Impulse and Propagation Mechanisms
This paper uses a high-frequency narrative approach, combining investment bank reports and asset prices, to measure the causal effect of the 2021 US fiscal deficits on inflation expectations.
Finance Application
- This paper's high-frequency narrative event study methodology is directly transferable to asset pricing research.
- It could be used to quantify the impact of specific regulatory announcements (e.g., SEC rulings, antitrust decisions) on affected company stock prices or industry-specific ETFs.
- The approach of combining qualitative narrative data (investment bank reports) with high-frequency asset price movements offers a powerful way to isolate and measure the causal effects of various policy or idiosyncratic shocks on market expectations and asset valuations, extending beyond fiscal policy to areas like climate policy, trade policy, or technological shifts.
Macro-financeFiscal PolicyInflationEvent StudyHigh-Frequency DataAsset PricesInflation SwapsGovernment BondsPolitical Risk
Core finding, identification, data
Core Finding
- The Georgia Senate election runoffs, by signaling future fiscal deficits (a $450 billion shock or 2.1% of GDP), led to an increase in expected price levels of 0.22-0.38% over 2021-22.
- This implies that the 2021 deficits contributed significantly (20-61%, baseline 34%) to the post-Pandemic inflation, a response consistent with standard HANK and FTPL models, especially given loose monetary policy.
Identification Strategy
- The identification strategy is a high-frequency narrative event study.
- The Georgia Senate election runoffs in early 2021 are identified as the event releasing news about future deficits.
- The size of this deficit shock is quantified using new narrative data from investment bank reports, and the market's response is measured using high-frequency changes in inflation forecasts derived from inflation swaps and other asset prices around the event window.
Data
The paper uses three main datasets: hand-collected narrative data from 20 investment banks and macroeconomic research groups (for shock sizing), high-frequency asset prices (inflation swaps, TIPS, dividend futures, government bonds) at daily and intra-daily levels (for market response), and election probabilities from online betting exchanges like PredictIt and BetFair (for probability of Democrat victory).
Ina Simonovska, Gonzalo E. Basante Pereira — International Finance & Macroeconomics
This paper develops a theoretical and empirical framework to quantify financial constraints for young firms by examining their capital structure evolution across countries with varying contract enforcement levels.
Finance Application
- This framework could be applied to study how changes in legal and institutional environments (e.g., bankruptcy reforms, contract law improvements) impact the pricing and issuance of corporate bonds and private credit, especially for small and medium-sized enterprises.
- Researchers could analyze how these institutional factors affect the risk premia on debt for young firms, influencing their cost of capital and, consequently, their equity valuations.
- The methodology could also be adapted to household finance to understand how local legal enforcement affects household-level entrepreneurship, access to credit for family businesses, and the capital structure decisions of closely held firms, thereby impacting household wealth dynamics.
Corporate FinanceCapital StructureYoung FirmsFinancial ConstraintsContract EnforcementEconomic DevelopmentLeverage DynamicsFirm GrowthInternational Finance
Core finding, identification, data
Core Finding
- Stronger contract enforcement significantly increases long-term leverage for young firms more so than for mature ones.
- The paper documents a novel pattern where short-term leverage rises while long-term leverage falls during a firm's early life cycle, a compositional shift masked in total leverage measures.
- Weaker contract enforcement prolongs this early life cycle duration, meaning firms in less developed economies take longer to become unconstrained.
Identification Strategy
- The identification strategy relies on a novel decomposition that separates institutional distortions from fundamental productivity differences, leveraging the insight that initial leverage ratios depend on enforcement quality but are invariant to entrepreneurial talent.
- Empirically, an age-period-cohort (APC) framework is used to isolate age effects from confounding factors, comparing leverage dynamics across countries grouped by World Bank recovery rates, which measure contract enforcement.
Data
The paper utilizes firm-level balance sheet data from the ORBIS database (compiled by Bureau van Dijk) and cross-country measures of contract enforcement from the World Bank's Doing Business Database, specifically recovery rates in insolvency proceedings.
María José Arteaga Garavito, Riccardo Colacito, Mariano Max Croce, Biao Yang — International Finance & Macroeconomics
This paper develops novel high-frequency climate attention indices from newspaper tweets across developed and emerging economies to study the impact of climate news shocks on international capital flows, currencies, and asset prices.
Finance Application
- The high-frequency, country-specific climate attention indices could be used to construct novel climate risk factors for global bond markets or commodity markets, testing their explanatory power for cross-sectional returns.
- The firm-level emission data combined with country-level climate sensitivity offers a rich dataset to study how climate news affects firms' cost of capital, ESG ratings, or green investment decisions in different regulatory regimes.
- For insurance, these indices could inform the pricing of catastrophe bonds or climate-related insurance products, especially by differentiating between physical and transition risk news.
Climate FinanceAsset PricingInternational FinanceExchange RatesCapital FlowsText AnalysisSocial MediaESGClimate RiskNews ShocksEmpirical
Core finding, identification, data
Core Finding
- Countries experiencing more severe climate news shocks tend to see capital inflows and currency appreciation.
- Brown stocks in highly exposed countries suffer large and persistent negative returns after global climate news shocks.
- International current accounts respond to these shocks, with highly sensitive countries hedging through international financial markets and receiving resources (declining net exports) when adverse news materializes.
Identification Strategy
- The study constructs high-frequency climate attention indices by applying textual analysis (cosine similarity, BERTopic, multilingual sentiment classifiers) to over 23 million tweets from major national newspapers across 25 countries (2014-2022).
- It identifies 'top days' of climate news (top 15% of global index innovations) and examines the subsequent behavior of currencies and stocks, distinguishing between physical and transition risk news.
Data
The paper uses over 23 million tweets from national newspapers in 25 countries (2014-2022), bilateral exchange rate data from Bloomberg, firm-level emission intensity data from Trucost (16,774 firms), quarterly international trade data, and external measures like the Notre Dame Global Adaptation Initiative vulnerability scores and World Cities Database for latitudes.
Charles Engel, Steve Pak Yeung Wu — International Finance & Macroeconomics
This paper demonstrates that standard empirical exchange rate models, augmented with global risk and liquidity measures, now fit U.S. dollar exchange rates well in the 21st century, attributing this improvement to increased central bank credibility and better monetary policy.
Finance Application
- The paper's methodology of identifying monetary policy regime shifts (e.g., via Markov-switching Taylor rules) and correlating them with model fit could be applied to other asset markets.
- For instance, researchers could investigate if "disconnect puzzles" in equity or credit markets are resolved when accounting for shifts in regulatory credibility or market liquidity regimes.
- The identified importance of global risk and liquidity measures for currency dynamics also suggests their potential as leading indicators for cross-asset correlations and tail risks in multi-asset portfolios, informing dynamic hedging or tactical asset allocation strategies.
Exchange ratesMonetary policyCentral bank credibilityFinancial crisesAsset pricing (currencies)Macro-financeRegime shiftsEconometricsGlobal riskLiquidity
Core finding, identification, data
Core Finding
Standard empirical exchange rate models, augmented with global risk and liquidity variables, now explain U.S. dollar exchange rate movements very well in the 21st century, resolving the "exchange-rate disconnect puzzle." This improved fit is attributed to a regime shift towards more credible monetary policy, which reduced the scope for self-fulfilling expectations that previously obscured the link between fundamentals and exchange rates.
Identification Strategy
- The paper employs a multi-pronged approach to identify the impact of monetary policy credibility.
- It uses rolling regressions and Meese-Rogoff/Clark-West out-of-sample tests to show a time-varying fit of exchange rate models.
- Crucially, it estimates Taylor rules with Markov-switching models to identify regime shifts in monetary policy credibility (specifically, when the Taylor principle is satisfied) and correlates these shifts with the improved model fit.
- Further evidence comes from event studies around inflation targeting adoption dates and analysis of exchange rate reactions to inflation news.
Data
The paper uses monthly exchange rate data from the Federal Reserve, CPI from IMF IFS, and government bond interest rates from GFD. It constructs global risk measures from various FRED series and liquidity measures from CIP deviations or Bianchi et al. (2021). For monetary policy analysis, it uses quarterly data on policy rates, inflation, GDP growth, and output gaps, alongside a central bank independence index from Romelli (2024) and Bloomberg inflation forecasts.
Tim de Silva, Eugene Larsen-Hallock, Adam Rej, David Thesmar — Forecasting & Empirical Methods
This paper shows that non-linearities in forecast revisions and errors, fat tails in underlying processes, and mean-reversion in extreme realizations are consistent with a model where forecasters ignore fat tails in their data-generating processes.
Finance Application
- The insight that forecasters ignore fat tails can be applied to explain other asset pricing anomalies beyond momentum, such as value or accruals, where misperceptions of extreme events in firm fundamentals could lead to systematic mispricing.
- In household finance, this mechanism could explain suboptimal savings or investment decisions if households misestimate the persistence of fat-tailed income or asset return shocks.
- For insurance, it suggests that risk models assuming Gaussian distributions might systematically underestimate extreme losses, leading to underpricing of tail risks in various insurance products.
Behavioral FinanceExpectations FormationFat TailsNon-LinearityForecast ErrorsForecast RevisionsAsset PricingMomentumKalman FilterBounded Rationality
Core finding, identification, data
Core Finding
- The paper documents three empirical facts: (i) a strongly non-linear relationship between forecast revisions and future forecast errors (underreaction in the bulk, overreaction in the tails), (ii) fat tails in the distributions of underlying firm-level variables, and (iii) mean-reversion in extreme realizations.
- It proposes a model where forecasters, assuming a Gaussian DGP, incorrectly apply a linear Kalman filter, which quantitatively explains these non-linearities.
Identification Strategy
- The identification relies on documenting robust non-linear relationships in a large panel of firm-level forecasts and then building a parsimonious theoretical model where forecasters ignore the fat-tailed nature of the underlying data-generating process (DGP).
- The model's ability to replicate the empirical facts is tested quantitatively, and further supported by an online forecasting experiment with a controlled fat-tailed DGP, and by showing similar non-linearities in macroeconomic forecasts and stock return momentum.
Data
The study uses a large panel of analyst forecasts for 22 firm-level variables from LSEG IBES (2000-2023), data from an online forecasting experiment, macroeconomic forecasts from the Survey of Professional Forecasters (1968-2025), and CRSP monthly stock returns (1927-2023).
Mikhail Chernov, Magnus Dahlquist, Lars A. Lochstoer — International Asset Pricing
This paper re-evaluates currency risk premiums by incorporating floating-regime emerging markets and disentangling priced from unpriced risks using a mean-variance efficient portfolio.
Finance Application
- The methodology of constructing an UMVE portfolio and disentangling priced from unpriced risks could be applied to other asset classes like equities, commodities, or fixed income to identify true risk factors versus unpriced common factors, leading to more robust factor models and improved portfolio construction.
- Portfolio managers could leverage this approach to hedge out unpriced risks in their existing strategies, potentially boosting Sharpe ratios and enhancing risk-adjusted returns across multi-asset portfolios.
- Furthermore, the concept of "unpriced risk" could be explored in household finance to understand how individuals might unknowingly take on risks that do not offer a premium, informing better financial advice and portfolio structuring.
currency marketsrisk premiumsfactor modelsmean-variance efficiencyemerging marketsunpriced riskSharpe ratiotransaction costsasset pricingportfolio optimizationconsumption risk
Core finding, identification, data
Core Finding
- The paper demonstrates that a small set of floating-regime emerging market currencies substantially expands the investment frontier.
- It finds that prominent currency trading strategies contain significant "unpriced risks" (e.g., related to geographical factors) that, when hedged out using a real-time unconditional mean-variance efficient (UMVE) portfolio, substantially increase Sharpe ratios (e.g., carry strategy SR increases from 0.71 to 1.29).
- The truly "priced risk" is related to long-run U.S. consumption exposure.
Identification Strategy
- The methodological innovation involves constructing a real-time, out-of-sample unconditional mean-variance efficient (UMVE) portfolio for G10 and floating-regime emerging market currencies.
- This UMVE factor is then used to conditionally disentangle priced from unpriced risks in various currency trading strategies, allowing for the hedging of unpriced risks and testing the true drivers of risk premiums, while also incorporating transaction costs into the UMVE construction and testing.
Data
The paper uses daily spot and one-month forward exchange rates for 75 currencies (narrowed to 59) from the WM Refinitiv database (Refinitiv Eikon) for the period 1996-2023, backfilled to 1985. It also incorporates CPI data from OECD and IMF, and classifies currencies based on Ilzetzki, Reinhart, and Rogoff (2019) IRR scores.
Erik Loualiche, Alexandre R. Pecora, Fabricius Somogyi, Colin Ward — International Asset Pricing
US monetary policy transmits internationally by influencing investment funds and global banks to reallocate capital towards currencies with higher factor-based risk exposures, impacting firm-level outcomes.
Finance Application
- The methodology of linking macro shocks to micro-level financial decisions via factor exposures could be applied to other asset classes.
- For instance, one could examine how monetary policy affects equity sector flows or bond fund allocations based on their exposure to equity or credit risk factors.
- The granular CLS data on institutional currency flows could inspire similar data collection or proxy development for other markets to study market microstructure and liquidity provision by different player types in response to macro news.
Monetary PolicyExchange RatesFactor ModelsCurrency RiskInternational FinanceBankingCorporate FinanceCapital FlowsRisk-Taking ChannelMarket Microstructure
Core finding, identification, data
Core Finding
- An unexpected easing of US monetary policy induces investment funds to sell safe and buy risky currencies, and global US banks to tilt foreign loan origination towards currencies with greater systematic currency risk.
- These effects persist for several months, leading to increased leverage and real investment for firms operating with high-risk currencies, demonstrating that currency factor exposures are a key lens for understanding international monetary policy transmission.
Identification Strategy
- The paper identifies US monetary policy shocks using high-frequency changes in Federal Funds futures prices around FOMC announcements (Kuttner, 2001; Bernanke and Kuttner, 2005).
- It then uses interaction terms between these shocks and currency-specific risk measures (carry and dollar betas) to capture differential responses across currencies.
Data
The study utilizes hourly exchange rate data from Bloomberg, detailed currency flow data by institution type from Continuous Linked Settlement (CLS) Group, global corporate loan origination data from Thomson Reuters DealScan, and corporate balance sheet data from Compustat North America and Global.
Peter Karadi, Marek Jarociński — Workshop on Methods and Applications for Dynamic Equilibrium Models
This paper identifies and disentangles three distinct components of high-frequency surprises around Fed announcements—monetary policy (MP), central bank information (CBI), and Fed response to news (FRN) shocks—and estimates their macroeconomic and financial impacts.
Finance Application
- The disentangled shocks could be used to refine event-study methodologies for various asset classes (e.g., corporate bonds, commodities, real estate, currencies), informing factor models or risk premia estimation by distinguishing between different types of Fed communication.
- Understanding how CBI shocks (reflecting Fed's economic outlook) versus FRN shocks (market misperceptions of Fed's rule) impact long-term interest rates could help model household mortgage refinancing decisions or investment in long-duration assets.
- The high-frequency identification and use of heteroskedasticity across FOMC/non-FOMC events could inspire new ways to identify different types of information shocks in other markets, helping to understand price discovery and market efficiency around scheduled announcements.
Monetary policyCentral bank communicationHigh-frequency dataEvent studiesAsset pricingInterest ratesStock pricesVARHeteroskedasticityFinancial markets
Core finding, identification, data
Core Finding
- The paper confirms the robust presence of central bank information (CBI) shocks, which cause a temporary boom in activity and prices.
- Monetary policy shocks, purified of CBI and FRN influences, generate impulse responses in line with theoretical predictions (downturn in activity, reduction in price level, deterioration of financial conditions).
- FRN shocks play a role in daily data but have marginal impact at the monthly level in the baseline VAR.
Identification Strategy
- Identification is achieved by leveraging (i) the high-frequency co-movement of interest rate and stock price surprises, (ii) the predictability of surprises based on public news, and (iii) heteroskedasticity between FOMC and non-FOMC announcements.
- Specifically, MP shocks are associated with negative co-movement between unexplained interest rate surprises and stock prices, FRN shocks with negative co-movement between predictable interest rate surprises and stock prices, and CBI shocks with positive co-movement between interest rate and stock price surprises.
Data
The paper uses high-frequency financial market data from TickData, Datascope Tick History, and Pi Trading, including Eurodollar/SOFR futures and S&P500 futures. It extends previous datasets by including surprises around non-FOMC events like Fed chair speeches and minutes releases, covering 1988:01-2024:09.
Vadim Elenev, Lu Liu — Macro, Money and Financial Frictions
This paper develops a quantitative macro-finance model to analyze how different mortgage structures (fixed-rate vs. adjustable-rate) affect financial stability and risk sharing between households and banks, especially in response to interest rate shocks.
Finance Application
- This model could be extended to study the optimal design of other long-term debt contracts, such as corporate bonds or commercial real estate loans, considering the interaction of borrower default risk, lender funding structures (e.g., sticky deposits vs. market-based funding), and interest rate exposure.
- Researchers could analyze the impact of different macroprudential tools (e.g., capital buffers, LTV/DTI limits) on the stability of specific credit markets, like leveraged loans or private credit, under varying interest rate environments.
- The framework also offers a lens to evaluate how central bank policies, such as quantitative tightening, differentially affect bank and non-bank financial intermediaries with diverse asset-liability structures.
mortgage marketsfinancial stabilityrisk sharingadjustable-rate mortgagesfixed-rate mortgagesinterest rate riskcredit riskhousehold financebankingmacro-financemonetary policymacroprudential regulationdefault riskintermediary asset pricing
Core finding, identification, data
Core Finding
- The paper finds a U-shaped relationship between mortgage fixation length and financial stability risks, measured by intermediary net worth volatility.
- Intermediate fixation lengths (3-5 years) minimize banking sector volatility and optimize aggregate risk sharing, balancing the trade-offs between household exposure to rising payments (ARMs) and bank exposure to interest rate risk (FRMs).
- ARMs provide net worth hedging by concentrating defaults when intermediary net worth is high, while FRMs benefit from sticky deposit rates.
Identification Strategy
- The identification strategy relies on a calibrated quantitative macro-finance model of the U.S. economy.
- The authors conduct counterfactual simulations by varying mortgage fixation lengths (from pure ARMs to FRMs) and introducing different correlations between aggregate income and interest rate shocks to assess their impact on financial stability and risk sharing.
- Cross-country empirical evidence is presented as illustrative support for the model's predictions.
Data
The paper uses U.S. macroeconomic data (Treasury rates, unemployment, GDP), financial data (Call Reports for deposit rates, FRED for delinquency and charge-off rates), household survey data (SCF for LTV, income, housing shares), and international data (BIS for credit shares, Bloomberg for bank equity indices, various sources for house prices and government debt maturities).
Kristian Blickle, Xu Lu, Jian Li, Yiming Ma — Macro, Money and Financial Frictions
This paper empirically documents and theoretically models how deposit flightiness fluctuates over time, showing that large deposit inflows (e.g., from QE) attract more rate-sensitive depositors, increasing financial instability risks from subsequent monetary policy tightening.
Finance Application
- This research could inform asset pricing models by incorporating time-varying bank funding costs and fragility into the valuation of bank equity and credit spreads, especially for banks with high exposure to flighty deposits.
- In household finance, the depositor-level data and insights into heterogeneous convenience values could be used to model household portfolio allocation decisions between bank deposits and alternative short-term investments (e.g., money market funds) under different monetary policy regimes.
- For insurance, understanding the dynamics of deposit flightiness could help assess the systemic risk contributions of financial institutions that rely heavily on such funding, influencing capital requirements or stress testing scenarios for insurers with significant bank counterparty exposures.
BankingFinancial StabilityMonetary PolicyQuantitative EasingDeposit FlowsInterest Rate SensitivityBank RunsHeterogeneityHousehold FinanceAsset Pricing
Core finding, identification, data
Core Finding
- The flightiness of bank deposits, measured by their sensitivity to interest rates, has fluctuated significantly over time, reaching historical highs after the Covid-19 crisis.
- This elevated flightiness is linked to central bank reserve expansions (Quantitative Easing) and low interest rates, as these conditions attract investors with lower convenience value for deposits, making the marginal depositor more rate-sensitive and increasing the risk of bank runs during policy rate hikes.
Identification Strategy
- The paper identifies deposit flow sensitivity by instrumenting bank-level deposit rates using supply-side factors from the industrial organization literature, specifically per unit asset fixed costs and salary expenses.
- This approach isolates the causal effect of deposit rate changes on flows, assuming these instruments affect deposit rates through supply-side costs rather than depositor demand.
Data
The study uses Call Report data for bank-level characteristics, regulatory data (FR2052) on deposits by counterparty and account type, and novel transaction-level data from over 1,400 U.S. depository institutions (2015-2022) to track fund movements between banks and between banks and investment options. It also uses aggregate data like the Fed funds rate and outstanding reserves from FRED.
Chuck Fang, Kairong Xiao — Macro, Money and Financial Frictions
This paper analyzes how granular portfolio holdings of major U.S. bond investors and bond issuance dynamics transmit and amplify conventional and unconventional monetary policy to long-term yields.
Finance Application
- The random-coefficient demand system framework could be applied to other asset classes, such as equities or real estate, to understand how heterogeneous investor preferences and substitution patterns drive asset prices and market liquidity.
- The insights into how retail inflows to mutual funds and banks amplify monetary policy could inform household finance research on the macroeconomic impact of household investment decisions.
- For insurance research, the detailed modeling of liability-driven investment (LDI) strategies and MBS convexity could be used to analyze systemic risk from insurer behavior under various regulatory and interest rate environments, or to optimize asset-liability management for pension funds.
Monetary PolicyBond MarketsPortfolio HoldingsDemand SystemsMarket ClearingInvestor BehaviorQuantitative EasingTerm StructureCredit SpreadsDurationInsurance CompaniesMutual FundsBanksBond Issuance
Core finding, identification, data
Core Finding
- The paper finds that mutual funds and banks act as a "helping hand" during expansionary monetary policy, increasing bond purchases due to retail inflows and liability hedging, which amplifies the Fed's impact on long-term yields.
- This elevated demand is primarily absorbed by new bond issuances, not dealers, and the "helping hand" effect is three times more influential than risk-free rates in driving corporate bond issuance.
- The dynamic supply response of bond issuance also explains the observed long-run reversal of monetary policy effects.
Identification Strategy
The identification strategy for bond yields relies on two sets of instruments: 1) bond characteristics (rating, duration, coupon) and those of peer bonds, assuming their supply is exogenous to investors, and 2) flow-induced trading by other investors, where idiosyncratic flows (residualized against common factors) to other investors create plausibly exogenous buying/selling pressure on bonds held by the investor of interest.
Data
The paper uses a comprehensive dataset of granular portfolio holdings from 2003 to 2022, covering mutual funds (Morningstar, CUSIP-level), insurance companies (NAIC filings, CUSIP-level), banks (FFIEC Call Reports, maturity bucket level), primary dealers (New York Fed, maturity bucket level), and the Federal Reserve (SOMA and SMCCF, CUSIP-level). This data spans Treasury notes and bonds, agency mortgage-backed securities (MBS), and corporate bonds, representing nearly $40 trillion in holdings.
Thummim Cho, Christopher Polk, Robert Rogers — Asset Pricing
This paper introduces "discounted alphas," a novel valuation framework that values equity based on the present value of future abnormal returns, circumventing the need for stock-level cost-of-equity estimates.
Finance Application
- The "discounted alphas" methodology could be applied to value complex, illiquid assets like private equity investments or real estate, where traditional DCF is challenging due to uncertain cash flows and discount rates; instead, one could model future abnormal returns based on observable deal characteristics.
- In household finance, the framework could analyze how retail investors' trading behavior interacts with identified mispricing pockets, potentially revealing if they exacerbate or exploit these inefficiencies.
- For insurance, it could be adapted to value complex insurance liabilities or portfolios of policies by modeling expected "alphas" from policy characteristics, offering a robust alternative to traditional actuarial valuation methods.
Equity ValuationDiscounted Cash Flow (DCF)Market EfficiencyAsset PricingMisvaluationAlphaPrivate EquityInstitutional InvestorsStock CharacteristicsCost of Equity
Core finding, identification, data
Core Finding
- The discounted alphas framework identifies economically important variation in fundamental value not captured by traditional DCF methods.
- It reveals that discretionary buy-and-hold funds and private equity funds successfully exploit CAPM misvaluation, yet overall, firm equity values are found to be "almost efficient" by Black's (1986) definition.
Identification Strategy
- The core methodological innovation is the "discounted alphas" identity, an exact mathematical identity that expresses fundamental value as the current price plus the present value of all future buy-and-hold alphas.
- This approach uses risk-adjusted alphas (derived from stock characteristics) to correct market prices, avoiding the imprecise estimation of stock-specific discount rates and building value from scratch.
Data
The paper uses monthly stock price data from CRSP, annual accounting data from CRSP/Compustat Merged (CCM), pre-Compustat book equity data from Davis, Fama, and French (2000), and factor data from Kenneth French's data library, covering 1953m6-2023m12.
Francesco D’Acunto, Michael Weber, Xiao Yin — Behavioral Macro
This paper demonstrates how extrapolative subjective income expectations, driven by unexpected income shocks, influence household consumption, debt accumulation, and default decisions, and can generate aggregate credit cycles.
Finance Application
- This research provides a micro-foundation for sentiment-driven phenomena in asset pricing, suggesting that periods of widespread extrapolative optimism about income could lead to overvaluation in consumer-sensitive asset classes, followed by corrections.
- For household finance, lenders could integrate survey-based expectation data into credit risk models to better predict default probabilities for unsecured loans and mortgages, especially during economic expansions.
- Insurance providers could design income protection products that account for consumers' tendency to over-extrapolate income shocks, offering behavioral nudges or dynamic coverage to mitigate excessive debt accumulation.
Income ExpectationsHousehold DebtConsumptionBehavioral FinanceBehavioral MacroeconomicsCredit CyclesDefaultSurveysMicro-to-MacroKalman FilteringDiagnostic Expectations
Core finding, identification, data
Core Finding
- Consumers form extrapolative income expectations, systematically overreacting to unexpected income innovations, which leads to increased current spending and debt accumulation.
- This behavioral bias results in higher default rates when actual income falls short of optimistic expectations.
- A structural model incorporating these diagnostic expectations successfully reproduces aggregate household debt cycles and default patterns observed in the data, including those reminiscent of the 2008-2009 Global Financial Crisis.
Identification Strategy
- The study leverages a unique panel dataset that combines individual-level transaction data, credit registry information, and customized survey data on income expectations.
- The identification strategy exploits within-individual variation over time, controlling for individual fixed effects and various demographics, to analyze the impact of unexpected income shocks on subsequent expectation errors, consumption, debt, and default.
- Objective income shocks are estimated using a Kalman filter approach to isolate subjective biases.
Data
The paper uses matched transaction-level data from a large Chinese commercial bank, credit-registry data from the People's Bank of China, and repeated income expectations elicited via customized surveys from the bank's customers.
Yashar Barardehi, Vincent Bogousslavsky, Dmitriy Muravyev — Big Data and High-Performance Computing for Financial Economics
This paper re-evaluates retail trading costs, particularly in options, by accounting for the widespread use of limit orders and introducing a novel adjustment to effective spread calculations.
Finance Application
- This research directly contributes to market microstructure and household finance.
- The refined measurement of retail trading costs could be applied to institutional trading data to provide more accurate estimates of execution quality across different market participants.
- The findings on persistent execution skill among retail traders could inform the design of financial literacy programs or automated trading tools to help less sophisticated investors minimize costs.
- Furthermore, the strategic use of limit orders by retail investors challenges traditional assumptions in asset pricing models about retail liquidity provision.
Retail TradingOptionsTrading CostsLimit OrdersMarket OrdersExecution QualityMarket MicrostructureForm 606Trader BehaviorLiquidity
Core finding, identification, data
Core Finding
- Retail investors extensively use non-marketable limit orders (41% of option volume), which significantly reduces their actual trading costs.
- True effective spreads for options average 1.07%, substantially lower than the conventional 2.59% due to limit order usage and an adjustment for aggressive limit orders setting the midquote.
- Execution performance is persistent, with more sophisticated traders achieving lower costs.
Identification Strategy
- The paper identifies true trading costs by using verified trade direction and a novel adjustment to the effective spread: when an aggressive limit order sets the NBBO, the midquote is calculated using the second-best bid/ask price to avoid understating costs.
- Order types (market vs. limit) are inferred by comparing trade prices to pre-trade quote midpoints.
Data
The study uses a trader-level dataset from Bogousslavsky and Muravyev (2024) covering 451,299 option and 607,922 stock retail trades from 2020-2022, merged with public market data (OPRA for options, TAQ for stocks). It also incorporates Form 606 reports from retail brokers for aggregate limit order usage.
Xuelin Li, Sijie Wang, Jiajie Xu, Xiang Zheng — Entrepreneurship
This paper examines how venture capital (VC) activism, particularly by smaller and more concentrated VCs, influences strategic experimentation and project prioritization in life sciences startups, often leading to slower progress and reduced exit likelihood.
Finance Application
- The paper's findings on how VC activism can lead to suboptimal project prioritization and increased failure risk could be applied to asset pricing by examining whether public markets systematically misprice early-stage biotech firms based on their VC backing profile, potentially leading to an 'innovation discount' for firms with highly active VCs.
- In household finance, the conflicts of interest between VCs and founders regarding project selection and control could inform studies on entrepreneurial decision-making under external pressure and its impact on founder wealth and career trajectories.
- For insurance, the identified increase in failure risk for certain projects due to VC prioritization could be used to develop more refined risk models for R&D or intellectual property insurance products, differentiating premiums based on the VC's concentration and investment horizon.
Venture CapitalStartupInnovationExperimentationLife SciencesESGAgency CostsPortfolio ManagementProject PrioritizationEntrepreneurship
Core finding, identification, data
Core Finding
- Active involvement by smaller, more concentrated VCs drives early pipeline prioritization, advancing a few projects while holding back most of the R&D portfolio.
- While this can lead to 'blockbuster' IPOs for successful projects, it reduces overall exit likelihood by sacrificing diversification benefits and increases failure risk for high-risk, long-horizon technologies due to conflicts of interest arising from VCs' investment horizons and concerns about portfolio cannibalization.
Identification Strategy
- To address endogeneity, the authors use two main identification strategies: first, they exploit limited partners' (LPs) staggered adoption of ESG objectives as an instrument for shifts in VCs' portfolio focus, forcing VCs to concentrate investments.
- Second, they leverage the introduction of direct flights between VC and startup headquarters as an exogenous shock for on-site engagement and monitoring intensity.
Data
The study uses granular project-level development data from the Cortellis Drug Discovery Intelligence Platform, VC investment data from Pitchbook (and VentureXpert for robustness), IPO and M&A dates from SDC Platinum and Jay Ritter's dataset, and T-100 Domestic Segment Database for airline route data.
Christopher M. Hair, Sabrina T. Howell, Mark J. Johnson, Siena Matsumoto — Entrepreneurship
This paper shows that incorporating cash flow data into credit underwriting significantly improves loan approval and interest rates for younger entrepreneurs, who are often disadvantaged by traditional FICO-based models.
Finance Application
- The 'Tail Analysis for Comparative Outcomes' (TACO) method could be directly applied in household finance to evaluate the distributional impact of new credit scoring models (e.g., using utility payments, rent history) on different demographic groups (e.g., young adults, minorities) for mortgages or consumer loans.
- In asset pricing, the improved risk assessment for small businesses using cash flow data could inform the pricing of private credit or small business loan-backed securities, potentially revealing new factors for firm-level risk premiums.
- For insurance, the enhanced ability to assess risk for younger businesses could lead to more accurate underwriting and pricing of small business insurance products, such as business interruption or general liability policies.
AgeEntrepreneurshipCredit AccessFintechCash FlowFICOLendingSmall BusinessMachine LearningAlternative DataUnderwritingTail Analysis for Comparative Outcomes (TACO)Financial Constraints
Core finding, identification, data
Core Finding
- Younger entrepreneurs (under 40) experience a 2.4 percentage point higher chance of loan approval and a 1.8 percentage point lower APR from cash flow-intensive lenders, particularly those with low FICO scores.
- This benefit stems from cash flow data's ability to identify creditworthy 'diamonds in the rough' among younger applicants, rather than increased risk-taking by lenders, as cash flow variables are more predictive for this demographic.
Identification Strategy
- The paper uses two quasi-experimental designs: first, exploiting within-applicant assignment to multiple lenders (some cash flow-intensive, some not) with application fixed effects to control for applicant characteristics; second, leveraging true random assignment of applications to loan officers at one lender, where officers vary in their reliance on cash flow data.
- It also introduces 'Tail Analysis for Comparative Outcomes' (TACO) to nonparametrically assess subgroup benefits from model changes.
Data
The study utilizes application, origination, and loan performance data from three U.S. fintech companies (two lenders and one platform) serving small businesses. This includes FICO scores, bank statement-based cash flow variables (revenues, withdrawals, balances, distress indicators), and borrower characteristics like owner age, firm age, and industry. Business survival data is also collected for a subset of applicants.
George-Marios Angeletos, Chen Lian, Christian K. Wolf — Micro Data and Macro Models
This paper compares the relationship between fiscal deficits and inflation in Heterogeneous Agent New Keynesian (HANK) models versus the Fiscal Theory of the Price Level (FTPL), finding that HANK reproduces FTPL's predictions more robustly but with quantitatively different outcomes.
Finance Application
- The paper's findings have direct implications for asset pricing and household finance.
- The prediction of a front-loaded and short-lived inflation response from fiscal deficits in HANK models, compared to FTPL, could inform the optimal duration and structure of inflation-hedging instruments (e.g., TIPS) for institutional and retail investors.
- Furthermore, the dampening effect of tax-base feedback and long-term debt on inflation could be incorporated into sovereign bond pricing models to better understand inflation risk premia across countries with varying fiscal structures.
- For household finance, the non-Ricardian nature of households and their differential responses to fiscal transfers could influence demand for inflation-indexed annuities or long-term care insurance, as perceived wealth effects alter saving incentives.
MacroeconomicsFiscal PolicyMonetary PolicyInflationHANKFTPLSovereign DebtHousehold FinanceAsset PricingQuantitative ModelsRicardian EquivalenceInflation Hedging
Core finding, identification, data
Core Finding
- The paper's core finding is that HANK models, despite different underlying mechanisms, reproduce the qualitative predictions of FTPL regarding deficits and inflation, but are more robust to policy assumptions.
- Quantitatively, an empirically-disciplined HANK model predicts that cumulative inflation from unfunded fiscal deficits is about half of what simple FTPL arithmetic suggests, primarily due to the endogenous expansion of the tax base and front-loading of inflation with long-term debt.
Identification Strategy
- The paper employs a quantitative, model-based approach.
- It simulates the effects of exogenous, surprise fiscal deficit shocks, specifically modeling them as one-off lump-sum transfers to households (e.g., CARES and ARP Act stimulus checks), within a calibrated HANK framework.
- The model's parameters are disciplined using empirical evidence on household behavior, NKPC characteristics, and government debt structure.
Data
The paper uses various empirical estimates for model calibration, including intertemporal marginal propensities to consume (iMPCs), household wealth holdings, fiscal transfer incidence, the slope and backward-lookingness of the New Keynesian Phillips Curve (NKPC), average labor tax rates, and the U.S. government debt-to-GDP ratio and average debt maturity. Specific fiscal stimulus amounts from the CARES and ARP Acts are used for a quantitative application.
Bianca He, Lauren I. Mostrom, Amir Sufi — Macroeconomics and Productivity
This paper introduces novel measures of customer capital investment using financial statements, salary data, and textual analysis of SEC filings, demonstrating its significant contribution to firm value and intangible capital.
Finance Application
- This research offers a more granular understanding of intangible assets crucial for asset pricing and corporate finance.
- In asset pricing, customer capital could be integrated as a novel intangible factor, explaining cross-sectional stock returns or risk premia, especially for firms in high-growth or platform industries.
- For corporate finance, the detailed measurement and valuation of customer capital can refine M&A due diligence, capital budgeting for marketing initiatives, and improve credit risk assessment by highlighting the stability derived from strong customer relationships, potentially impacting bond yields or corporate insurance premiums.
Intangible CapitalCustomer CapitalSales and MarketingFirm ValuationAsset PricingCorporate FinanceTextual AnalysisMachine LearningM&ATobin's Q
Core finding, identification, data
Core Finding
- The paper establishes that customer capital, measured through novel methods including sales/marketing expenses and textual analysis of SEC filings, is a quantitatively large and growing component of intangible capital.
- It demonstrates that this investment significantly explains cross-industry variation in firm valuation (Tobin's Q) and acquisition prices, particularly for customer-related intangible assets, unlike residual SG&A.
Identification Strategy
- The study identifies the value of customer capital by exploiting cross-industry variation in sales and marketing intensity and business model characteristics (e.g., platform, online sales, high-tech).
- It uses a modified capitalization methodology for publicly traded firms and purchase price allocation data from M&A transactions to link these investments to intangible asset values, providing both market-based and transaction-based evidence.
Data
The paper utilizes Capital IQ for sales and marketing expense data, Revelio Labs for sales/marketing and engineer salaries, annual 10-K SEC filings (processed with Google Gemini 1.5 Flash for textual analysis), Compustat for baseline firm data, Business Valuation Resources' (BVR) DealStats database for Purchase Price Allocations in M&A, I/B/E/S for analyst forecasts, and Aswath Damodaran for asset betas.
Tolga Caskurlu, Gerard Hoberg, Gordon M. Phillips — Innovation
This paper introduces a novel text-based measure of technology sectoral disruptions (TSDs) from patent data, finding that these unexpected shocks create wealth, primarily for small firms, with market and insider reactions occurring with significant lags.
Finance Application
- The concept of unexpected, wealth-creating, and long-lasting TSDs could be integrated into dynamic asset pricing models to explore a 'disruption risk premium' or 'disruption alpha' for early identifiers.
- For household finance, the delayed and differential market reaction to TSDs could inform studies on individual investor behavior, examining whether retail investors systematically under- or over-react to these shocks, impacting their long-term wealth.
- In insurance, the unexpected nature of TSDs presents a challenge for traditional risk modeling; this could lead to the development of new insurance products or hedging strategies for firms exposed to technological disruption risk, especially those that are slow to adapt.
PatentsInnovationTechnologySectoral DisruptionAsset PricingCorporate FinanceR&DInsider TradingFirm ValuationSmall FirmsLarge FirmsText AnalysisMachine LearningCreative DestructionMarket EfficiencyJEL: O31JEL: O34JEL: D43JEL: F13
Core finding, identification, data
Core Finding
- Technology Sectoral Disruptions (TSDs) are largely unexpected by market participants, corporate insiders, and analysts, with price discovery and behavioral responses occurring months to years after the TSDs become measurable.
- Small firms significantly benefit from TSDs, exhibiting increased R&D, asset growth, equity issuance, and valuations, consistent with theories of creative destruction and smaller firm innovation.
- In contrast, large firms generally reduce R&D and capital expenditures, experiencing declining valuations and sales growth.
Identification Strategy
- The paper identifies TSDs as plausibly exogenous shocks by demonstrating their unexpected nature.
- This is established by showing that stock returns, insider trading activity, and analyst forecasts only react to TSDs with significant lags (months to years) after they are publicly measurable, indicating a lack of prior awareness.
- The TSD measure itself is constructed using a dynamic text-based spatial model of patents, specifically Google Cloud's 64-dimensional patent embeddings, to quantify the correlated movement of technology bundles across multiple industries over time (a text-based analog to covariance).
Data
The study utilizes patent data from the Google Cloud Public Database (1890-2022), linked to public firms via the Kogan et al. (2017) database. Financial data comes from CRSP for monthly stock returns and Compustat (1951-2020) for firm financials. Insider trading data is from the Thomson Financial Insider Filing database (1986-2022), and analyst forecast data is from the I/B/E/S database (1985-2020).
Tolga Caskurlu, Gerard Hoberg, Gordon M. Phillips — Digital Economics and Artificial Intelligence
This paper constructs a novel measure of technology sectoral disruptions (TSDs) using patent text data and examines their impact on firm behavior and stock returns, finding that TSDs are largely unexpected and disproportionately benefit smaller firms.
Finance Application
- The text-based methodology for identifying 'Technology Sectoral Disruptions' (TSDs) could be adapted to identify other types of unexpected, sector-wide shocks from non-traditional data sources (e.g., news articles, social media, regulatory filings) and examine their impact on asset prices, firm financing, and household wealth.
- For instance, one could identify 'Regulatory Sectoral Disruptions' from legislative text and analyze their effect on bond markets or specific insurance lines.
- In household finance, the unexpected nature of TSDs could be used to study how households adjust their human capital investments or portfolio allocations when their primary industry faces a sudden, unanticipated technological shift.
TechnologyInnovationSectoral DisruptionsPatentsAsset PricingCorporate FinanceFirm BehaviorInsider TradingAnalyst ForecastsSmall FirmsLarge FirmsText AnalysisMachine Learning
Core finding, identification, data
Core Finding
- The core finding is that Technology Sectoral Disruptions (TSDs) are largely unexpected by economic agents (investors, insiders, analysts) and generate significant, long-lasting wealth, primarily for small firms.
- Small firms respond by increasing equity issuance, R&D, and asset growth, while large firms tend to reduce R&D and capital expenditures, experiencing declining valuations.
Identification Strategy
- The paper's identification strategy relies on the unexpected nature of TSDs, which are measured using a dynamic text-based spatial model of patents.
- The authors show that stock returns, insider trading, and analyst forecasts only react to TSDs with significant lags (months to over a year) after they become measurable, suggesting that TSDs act as plausibly exogenous shocks to corporate decision-makers.
Data
The paper uses patent data from Google Cloud Public Database (1890-2022), linked to public firms via Kogan et al. (2017) (KPSS) database. It also uses CRSP for stock returns, Compustat for firm financials (1951-2020), Thomson Financial Insider Filing database (1986-2022) for insider trading, and I/B/E/S for analyst forecast data (1985-2020).
Marco Cosconati, Yi Xin, Fan Wu, Yizhou Jin — Industrial Organization
This paper develops and estimates a novel empirical model of imperfect competition in selection markets, where firms have heterogeneous information about consumers, varying cost structures, and differentiated products, using data from the Italian auto insurance market.
Finance Application
- The framework for analyzing imperfect competition under heterogeneous firm-level information is highly relevant for household finance and other financial markets.
- It could be applied to study competition among mortgage lenders, credit card providers, or FinTech platforms, where firms possess varying capabilities in assessing borrower risk (e.g., using proprietary AI/ML models).
- This could inform policy on data sharing regulations (like open banking) by analyzing their impact on market concentration, pricing for different risk segments, and consumer welfare.
- The methodology could also be adapted to analyze information asymmetry among institutional investors in less transparent asset markets, such as private credit or corporate bonds, and the role of public information disclosures in enhancing market efficiency.
Information AsymmetryImperfect CompetitionAuto InsuranceRisk RatingConsumer SurplusData SharingRegulationHousehold FinanceCredit MarketsFinTechEconometricsSelection Markets
Core finding, identification, data
Core Finding
- The study finds substantial differences in risk rating precision across insurers, with less accurate algorithms often correlating with more efficient cost structures.
- A simulated centralized risk bureau, which equalizes information access, significantly reduces prices (21.6%), boosts consumer surplus (15.7%), and lowers average costs (12 euros/contract) by increasing competition and improving insurer-insuree matching, primarily benefiting low-risk consumers.
Identification Strategy
- The paper's methodological innovation involves a novel fixed-point approach to jointly estimate demand parameters, price distributions, and consumer sorting probabilities from transaction prices.
- It identifies firm-specific signal distributions and pricing coefficients by leveraging the monotonic relationship between offered prices and firms' private signals (similar to auction models) and matching observed price-risk correlations and variances, then recovers cost parameters from firms' profit maximization first-order conditions.
Data
The paper uses a unique market-level dataset from the Italian auto insurance industry, covering 124,428 liability insurance contracts for new customers in Rome from 2013-2021. This dataset includes individual-level demographic information, vehicle characteristics, contract details, transaction prices, and claim records, with consumers tracked across insurers.
Sumit Agarwal, Andrea F. Presbitero, Andre F. Silva, Carlo Wix — Household Finance
This paper examines the redistributive effects of credit card reward programs, finding that financially sophisticated consumers profit at the expense of naive consumers, regardless of income.
Finance Application
- This research offers direct insights for household finance, particularly on how complex financial product design can exploit behavioral biases and lead to wealth transfers.
- The methodology could be adapted to study the redistributive impact of complex investment products (e.g., structured products, annuities) or insurance policies (e.g., whole life insurance with complex riders) where benefits are salient but costs are shrouded.
- The finding on suboptimal balance-matching heuristics could be extended to how households manage multiple debt obligations (mortgages, student loans, personal loans) with varying terms, potentially leading to inefficient repayment strategies.
household financecredit cardsfinancial sophisticationbehavioral economicswealth redistributionconsumer debtrewards programsidentification strategyFICO scores
Core finding, identification, data
Core Finding
- Credit card reward programs lead to a significant annual redistribution of $15 billion from less financially sophisticated (low-FICO, less educated, poorer, high-minority areas) to more sophisticated (high-FICO, more educated, richer, low-minority areas) consumers.
- Sophisticated individuals benefit from rewards and lower interest payments, while naive individuals incur higher unpaid balances and suboptimal repayment behavior, leading to net losses.
Identification Strategy
- The study employs a quasi-experimental design by exploiting bank-initiated credit limit increases.
- It compares spending, repayment, and unpaid balance responses of consumers who received limit increases on reward cards versus non-reward cards, controlling for extensive card- and cardholder-level fixed effects (Bank × ZIP code × Income percentile × FICO percentile) to isolate causal effects.
Data
The paper uses comprehensive monthly supervisory credit card data from the Federal Reserve Board's Y-14M reports, covering over 200 million U.S. credit card accounts from 19 large U.S. banks. This dataset includes detailed account-level information on rewards, interest/fee charges, purchase volumes, repayment amounts, outstanding balances, FICO scores, credit limits, and cardholder demographics.
Alexander Braun, Lauren Cohen, Jiahua Xu DR — Household Finance
This paper quantifies the substantial economic value lost by U.S. households who prematurely terminate life insurance policies instead of selling them in the secondary market, identifies market frictions, and proposes policy solutions.
Finance Application
- This research highlights a significant, unaddressed market inefficiency in household finance.
- Asset pricing researchers could analyze the pricing of life settlement-backed securities, considering the impact of proposed market reforms (e.g., centralized platforms, GSEs) on liquidity and risk premia.
- Household finance studies could investigate the behavioral biases leading to premature policy termination, perhaps designing interventions or nudges to encourage secondary market sales, especially for vulnerable populations.
- Insurance companies could explore new product designs that incorporate mortality-contingent surrender values, potentially reducing adverse selection and improving policyholder welfare.
Core finding, identification, data
Core Finding
- U.S. households forgo an estimated $192 billion annually by lapsing or surrendering life insurance policies instead of selling them in the secondary market.
- The median household loss is $51,500, disproportionately affecting elderly and low-income populations.
- This value is largely captured by intermediaries due to high transaction costs, leaving investors with negative expected profits.
Identification Strategy
- The paper uses an actuarial valuation model to estimate the economic value (EEV) of life insurance policies, incorporating mortality and lapse rates from the Society of Actuaries.
- It then develops a parsimonious theoretical model that formalizes how search and transaction frictions (costs 'c' and intermediary power 'alpha') create an upper bound on the life expectancy of tradable policies, explaining why only policies with short life expectancies and high face values are transacted.
Data
The paper uses mortality data (VBT-01, VBT-15) and lapse data (2009-13 U.S. Individual Life Persistency Report) from the Society of Actuaries (SOA). It also uses policy distribution data from SOA's Individual Life Insurance Mortality Experience Report, total market size from S&P Capital IQ Pro, U.S. Treasury yield curve data, and privately collected U.S. life insurance policy data for cash account interest rates.
Juan Martin Morelli, Matias Moretti, Venky Venkateswaran — Household Finance
This paper develops a spatial banking model calibrated to U.S. microdata to quantify how geographic diversification and competition influence deposit rates, risk premia, and markups, and their implications for lending.
Finance Application
- The paper's framework for quantifying geographic funding risk and market power could be adapted to other financial sectors.
- In asset pricing, one could analyze how geographic concentration of underlying assets (e.g., real estate, local businesses) affects the risk premia and pricing of securitized products like REITs or municipal bonds.
- For household finance, the model could explore how the geographic diversification of a household's lenders impacts their borrowing costs (e.g., mortgage rates) and access to credit, especially in underserved areas.
- In insurance, the spatial model could be used to study how geographic diversification of an insurer's underwriting portfolio affects policy pricing and capital requirements, particularly for risks with localized exposure like natural disasters.
BankingDeposit MarketsGeographic RiskDiversificationMarket PowerInterest RatesLendingSpatial EconomicsStructural ModelInstrumental VariablesBartik InstrumentMonetary Policy Shocks
Core finding, identification, data
Core Finding
- The paper finds that geographic diversification significantly reduces funding risk premia in deposit markets, especially benefiting smaller and poorer counties, while markups changed only modestly.
- Banking consolidation, particularly by larger regional or national banks acquiring local ones, has lowered risk premia but can reduce local lending by reallocating credit to more profitable markets.
Identification Strategy
- The identification strategy primarily relies on Bartik-type instrumental variables.
- For elasticities of substitution (η and θ), the authors use changes in bank teller wages, assuming these affect bank costs but not preferences.
- For the curvature parameter (χ), they exploit variation in the risk term orthogonal to cost shifters.
- For the cash-to-deposits elasticity, they use monetary policy shocks as an instrument.
Data
The paper uses annual data on deposits from the FDIC's Summary of Deposits (SOD), bank-level financial variables from Call Reports, branch-level deposit rates from RateWatch, county-level economic activity data from the Bureau of Economic Analysis (BEA), market interest rates from FRED (HQM Corporate Bond Spot Rate), wage data from the Bureau of Labor Statistics, and household cash/deposit data from the Financial Accounts of the United States (Z.1).
Peleg R. Samuels — Urban Economics
This paper examines how home-biased household real estate investors, often called 'mom-and-pop' landlords, influence local rental housing prices and returns.
Finance Application
- This research could inform asset pricing models for other illiquid, locally-traded assets, such as private businesses or local small-cap stocks, by incorporating the role of home-biased local capital supply.
- In household finance, it highlights how demographic and cultural factors, alongside wealth, shape significant asset allocation decisions in real estate, suggesting avenues to explore these influences on other household portfolio choices.
- For institutional real estate investors (e.g., REITs), the findings suggest that local market dynamics, driven by 'mom-and-pop' landlords, can create specific pricing environments that affect their investment strategies and returns.
Real EstateHousehold FinanceAsset PricingLocal MarketsHome BiasRental HousingWealth EffectsIdentificationShift-Share InstrumentMom-and-Pop Investors
Core finding, identification, data
Core Finding
- The paper finds that increases in local wealth and investor participation rates significantly lower rent-price ratios and future returns.
- Specifically, a one-log-point increase in the local investor share reduces the rent-price ratio by 7 percentage points, and positive wealth shocks lead to subsequent price increases and return declines.
Identification Strategy
The study employs two main identification strategies: (1) For cross-sectional analysis at the neighborhood level, it uses two shift-share instruments: 'Ancestry-induced Real Estate Proclivity' (AREP) based on culturally driven real estate investment tendencies, and average household equity wealth from IRS data to instrument for mean investor wealth. (2) For panel data at the metro-year level, it uses a shift-share instrument for equity wealth shocks, constructed by interacting metro-specific equity-wealth exposure with national equity-market surprises.
Data
The paper constructs a novel dataset by linking the 2013 DataQuick dataset (property-level details) with owner mailing addresses, the Survey of Consumer Finances (SCF), and the Rental Housing Finance Survey (RHFS). It also uses BLS rent growth, FHFA home price appreciation, IRS capital-income data, and Census/ACS micro-samples for ancestry information.
Tarek Alexander Hassan, Stephan Hollander, Aakash Kalyani, Laurence van Lent, Markus Schwedeler, Ahmed Tahoun — International Finance and Macroeconomics Data Session
This paper demonstrates how computational linguistics techniques applied to unstructured corporate text can provide real-time, nuanced insights into firm-level exposure, risk, and sentiment regarding economic shocks, often eluding traditional data sources.
Finance Application
- In asset pricing, these text-based risk and sentiment measures could be used to construct novel, granular risk factors (e.g., 'AI sentiment factor' or 'supply chain risk factor') that explain cross-sectional stock returns or predict sector-specific volatility.
- For insurance research, the distinction between 'bad news' and 'uncertainty' could refine pricing models for corporate insurance products, allowing for more precise underwriting of emerging risks like climate change or cyber threats.
- In household finance, aggregated sentiment from job postings about new technologies could signal future labor market shifts, influencing household investment in education or regional housing market dynamics.
textual analysisnatural language processingearnings callscorporate financeasset pricingrisk measurementsentiment analysiseconomic shocksinnovationpatentsjob postingsfirm behaviormarket efficiencyfinancial accounting
Core finding, identification, data
Core Finding
- Textual analysis of corporate communications (earnings calls, patents, job postings) effectively measures firm-level exposure, risk, and sentiment related to various economic shocks (e.g., geopolitical events, pandemics, technological advancements).
- These measures offer disaggregated, cardinal comparisons of risks and sentiments, enabling a deeper understanding of market and firm responses and the transmission of shocks across firms, sectors, and countries.
Identification Strategy
- The methodological innovation involves applying computational linguistics (keyword-based approaches, training libraries, large language models) to systematically quantify economic concepts like 'topic exposure,' 'topic risk,' and 'topic sentiment' from unstructured text.
- This allows for disaggregation of these measures by firm, sector, country, and specific topic, facilitating the analysis of shock transmission and the distinction between 'bad news' and 'uncertainty' effects.
Data
The primary data sources include 379,227 earnings conference call transcripts from 12,805 firms across 89 countries, patent documents, and online job postings.
Viral V. Acharya, Manasa Gopal, Sascha Steffen — Corporate Finance
This paper examines how corporate reliance on nonbank financing, particularly through Collateralized Loan Obligations (CLOs), increases firm fragility and negatively impacts their access to bank-provided liquidity during market stress.
Finance Application
- This research offers several avenues for asset pricing and risk management.
- First, the 'nonbank dependence' and 'rollover risk' measures could be developed into novel firm-level risk factors to explain cross-sectional variation in corporate bond yields, equity returns, or credit default swap (CDS) spreads, particularly during periods of credit market stress.
- Second, institutional investors, including insurance companies, that hold CLOs or other nonbank debt could use these insights to refine their portfolio construction and risk models, better understanding the indirect systemic exposures and liquidity risks embedded in their holdings.
- Finally, the findings highlight a potential 'fragility premium' in the pricing of bank credit for nonbank-reliant firms, which could be explored in models of bank loan pricing and its implications for financial stability.
shadow bankingnonbank financingcredit linesrollover riskCLOsliquidity riskcorporate financebank lendingoil price shocksystemic riskasset pricingcredit marketsfirm fragility
Core finding, identification, data
Core Finding
- Firms with greater reliance on nonbank financing face reduced availability and higher costs for bank credit lines.
- During the 2014-16 oil price collapse, non-oil-sector firms with maturing nonbank loans (exposed to CLOs affected by the oil shock) experienced significant contractions in nonbank funding, tighter and more expensive bank credit lines, and weaker financial and real performance, despite drawing down existing credit lines.
Identification Strategy
- The study leverages the 2014-16 oil price collapse as a plausibly exogenous shock to the rollover risk of nonbank financing for non-oil-sector firms.
- This shock impacted CLOs heavily exposed to oil and gas firms, leading them to offload loans, including those of 'innocent bystander' non-oil firms.
- This creates exogenous variation in nonbank credit supply for firms whose fundamentals were not directly affected by the oil shock, allowing for causal inference on how banks adjust liquidity provision based on nonbank reliance and rollover risk.
Data
The paper utilizes loan origination and secondary market data from Refinitiv DealScan and LSTA, borrower financial information from Compustat and CapitalIQ, stock returns from CRSP, and CLO tranche and holdings data from Creditflux.
Mara Faccio, Jin Xu — Corporate Finance
This paper examines how limitations on interest deductibility, introduced through interest ceiling rules globally, causally affect the capital structure choices of private firms.
Finance Application
- The causal link between tax policy and private firm leverage has direct implications for private equity valuation and deal structuring, as changes in interest deductibility affect the optimal capital structure and returns for portfolio companies.
- In household finance, these findings are critical for understanding the wealth accumulation and tax planning strategies of entrepreneurs, as their personal financial decisions are often intertwined with their private businesses' capital structure.
- For corporate finance, the methodology could be adapted to study how similar tax changes affect the investment and growth strategies of private firms, which are often less financially constrained than public firms.
Corporate FinanceTaxationCapital StructurePrivate FirmsLeverageDifference-in-DifferencesInterest DeductibilityTax PolicyInternational Finance
Core finding, identification, data
Core Finding
- The study finds that private firms significantly reduce their leverage when faced with new interest ceiling rules that limit interest deductibility.
- This effect is robust across various tests, including firms near limitation thresholds, matched samples, and EU countries where rules were mandated, and is absent in falsification tests using pseudo-reform years.
- More broadly, private firms tend to decrease leverage in response to tax rate cuts and increase it with corporate tax rate hikes.
Identification Strategy
- The primary identification strategy is a difference-in-differences (DiD) framework, exploiting the staggered introduction of interest ceiling rules across 25 countries between 2005 and 2021.
- The treatment group consists of firms whose interest payments exceed the new deductibility cap, while the control group's payments remain below it.
- The analysis includes country-year, industry-year, and firm-consolidation type fixed effects, and employs falsification tests with randomly generated pseudo-reform years.
Data
The paper uses firm-level accounting data from the Orbis database (version "orbis5") for 215,158 unique private firms across 93 countries from 1997-2022. Corporate income tax data and interest limitation rules are hand-collected from Ernst & Young's "Worldwide Corporate Tax Guide[s]", PricewaterhouseCoopers, OECD, and the Tax Foundation. Country-level macroeconomic controls are from the IMF and World Bank.
Gary B. Gorton, Ye Li, Guillermo Ordoñez — Corporate Finance
This paper proposes a theory where complex credit intermediation chains emerge to maximize financing capacity by diluting lenders' incentives to acquire costly information about pledgeable assets through probabilistic asset ownership.
Finance Application
- The model could be applied to asset pricing to analyze the valuation of complex securitized products (e.g., CLOs, CDOs) where multiple layers of intermediation exist, explaining how information dilution affects tranche pricing and liquidity.
- In household finance, it could shed light on how the structure of mortgage-backed securities influences information incentives among originators, servicers, and investors, impacting mortgage credit availability.
- For insurance, the framework could be extended to reinsurance chains, examining how information costs and ownership dilution affect the pricing and systemic risk of complex reinsurance contracts.
Financial IntermediationCredit ChainsDebt vs EquityInformation AsymmetryCostly InformationAsset OwnershipSecuritizationRepo MarketsCapital StructureLiquiditySystemic Risk
Core finding, identification, data
Core Finding
- The optimal financial architecture involves systematically sequencing multiple intermediaries with heterogeneous information costs and asset correlations.
- Debt generates greater financing capacity than equity because its probabilistic asset ownership weakens incentives for information acquisition.
- Intermediation further dilutes this probabilistic ownership, allowing for greater credit capacity, with the optimal chain equalizing information acquisition incentives across lenders.
Identification Strategy
- The paper's methodological innovation is a theoretical model that introduces 'probabilistic asset ownership' as the key differentiator between debt and equity.
- This concept, combined with costly information acquisition, endogenously generates a pecking order theory and rationalizes the emergence and optimal structure of complex intermediation chains as a mechanism to dilute information acquisition incentives.
Data
This is a theoretical paper and does not use empirical data.
Falk Bräuning, Victoria Ivashina — Corporate Finance
This paper empirically demonstrates that corporate revolving credit line drawdowns are highly sensitive to interest rates, acting as a significant counterforce to deposit outflows during periods of banking sector stress.
Finance Application
- This research offers valuable insights for asset pricing by suggesting that the interest rate sensitivity of corporate credit line drawdowns could influence bank equity and bond valuations, especially during monetary policy shifts.
- In household finance, the mechanism could be extended to analyze how households' utilization of credit lines (e.g., HELOCs) responds to interest rate changes, impacting household liquidity and consumption.
- For financial stability, the findings highlight a crucial, often overlooked, counterbalancing force to deposit runs, which can inform stress testing models and regulatory frameworks for both traditional banks and non-bank lenders.
bank liquidityrevolving credit linesinterest ratesbank runsfinancial crisismonetary policyfinancial stabilitycorporate financeempirical methods
Core finding, identification, data
Core Finding
- The core finding is that a 1 percentage point increase in interest rates leads to an 8.2 percentage point reduction in precautionary drawdowns on corporate revolving credit lines.
- This interest rate sensitivity makes revolving line runs less likely in high-interest-rate environments, providing a stabilizing force for bank liquidity that contrasts with the dynamics of deposit runs.
Identification Strategy
- The identification strategy employs a Regression Kink Design (RKD) that leverages interest rate floors embedded in variable-rate credit line contracts.
- This design exploits the non-linear relationship between LIBOR and the effective interest rate when LIBOR falls below the contractual floor, allowing for the causal estimation of interest rate elasticity on credit line utilization, particularly during the high-uncertainty COVID-19 period.
Data
The study primarily utilizes detailed facility-level supervisory FR Y-14Q data from the Federal Reserve, focusing on Commercial and Industrial (C&I) loans from large US bank holding companies, supplemented by public FR Y-9C bank balance sheet data.
Rüdiger Fahlenbrach, Minsu Ko, René M. Stulz — Corporate Finance
This paper examines how political administrations and regulatory environments influence the payout policies (dividends and repurchases) of large banks, finding significant differences compared to smaller banks and industrial firms.
Finance Application
- This research highlights the importance of political risk in financial markets, particularly for heavily regulated sectors like banking.
- In asset pricing, the findings could inspire models that incorporate a 'political risk premium' for bank stocks, especially those of large, systemically important financial institutions, differentiating by political regime or regulatory uncertainty.
- For household finance, understanding how political cycles affect bank payouts could indirectly inform research on credit supply and financial stability, impacting household access to credit.
- In insurance, this framework could be adapted to analyze how political and regulatory shifts affect the payout policies and capital management of large insurance companies, which are also subject to significant government oversight and often have bank-like characteristics.
BanksPayout policyDividendsRepurchasesRegulationPoliticsPolitical uncertaintyStock returnsEvent studyRegression discontinuityDifference-in-differencesFinancial stability
Core finding, identification, data
Core Finding
- Large banks exhibit significantly lower net payout rates under Democratic presidents and during periods of high financial regulation uncertainty, with this effect primarily driven by the greater flexibility and political sensitivity of share repurchases compared to dividends.
- Surprise presidential elections lead to differential stock price reactions for large banks, indicating that political shifts in regulatory stance impact their expected future payouts and valuation.
Identification Strategy
- The study employs a difference-in-differences approach, comparing large banks (more affected by politics/regulation) to smaller banks and industrial firms (controls) across different political administrations.
- It also uses a regression discontinuity design (RDD) by exploiting the sharp regulatory threshold of $50 billion in assets for CCAR banks to identify causal effects of regulation on payouts.
- Furthermore, event studies around 'surprise' presidential elections (Trump 2016, Biden 2020, Trump 2024) are used as unexpected shocks to the regulatory environment.
Data
The paper uses bank-level data from CRSP-FRB link tables, Compustat, Federal Reserve Bank of Chicago, and SNL Financial from 1977-2023. It also includes industrial firm data from merged CRSP-Compustat. Political and regulatory data come from OptionMetrics, SDC Platinum, Baker, Bloom, and Davis (2016) for uncertainty indices, and publicly available Federal Reserve CCAR data.
Jiang Wang — Capital Markets and the Economy
This paper develops a general equilibrium framework to analyze the impact of government trading in financial markets on market outcomes and investor welfare, especially under market incompleteness and asymmetric information.
Finance Application
- This framework could be applied to analyze the welfare implications of central bank asset purchase programs (e.g., quantitative easing) in specific markets, considering how the transparency of their operations affects risk premia and investor welfare.
- It could also model the impact of large institutional investors with non-pecuniary objectives (like sovereign wealth funds or ESG-mandated funds) on market quality and the welfare of other market participants.
- Furthermore, the insights on how government trading affects informational efficiency and liquidity could inform optimal market design in contexts with significant public sector involvement.
Government InterventionFinancial MarketsAsset PricingMarket EfficiencyInvestor WelfareRisk SharingInformation AsymmetryGeneral EquilibriumNoise TradingCentral Bank Policy
Core finding, identification, data
Core Finding
- The paper theoretically demonstrates that government intervention in financial markets, particularly through its own trading, can improve investor welfare by reducing risk premia and facilitating risk sharing, especially when the government acts as a "noise trader" (trading on unobservable noise).
- However, interventions based on public information are ineffective due to offsetting investor trades, and common market performance metrics (like price stability or informational efficiency) can be misleading indicators of welfare.
Identification Strategy
- The paper constructs a parsimonious general equilibrium model featuring heterogeneous investors with CARA utility and private information about income risk, alongside a government that trades to maximize investor welfare.
- The impact of government intervention is identified by comparing market outcomes and investor welfare under different government information sets (public, partial, full) and trading strategies (e.g., "noise trading" vs. trading on public signals), allowing for a welfare-based assessment of policy.
Data
The paper is purely theoretical and does not use real-world data; numerical examples are used for illustration of the model's implications.
Olivier Darmouni, Yuqi Zhang — Capital Markets and the Economy
This paper studies capital allocation in the coal power sector, focusing on ownership dynamics and the role of "green finance" and government objectives in the energy transition.
Finance Application
- This research provides a framework to analyze how non-pecuniary factors (like climate externalities or social benefits) and financial frictions influence asset valuations and trading decisions in carbon-intensive industries.
- Asset pricing researchers could use this model to study the pricing of "brown" assets, incorporating investor heterogeneity in ESG mandates and the impact of state ownership on risk premia.
- For instance, one could investigate if firms with broader ESG mandates (higher θ) face different costs of capital or if state-owned entities (with high 's' values) exhibit lower discount rates for assets with social benefits, potentially leading to mispricing relative to purely profit-driven valuations.
- The model's insights on financial frictions affecting asset prices and incentivizing closures (rather than sales) could also inform studies on the effectiveness of "green finance" policies on corporate investment and divestment strategies in other polluting sectors.
Energy transitionclimate financecapital allocationstate ownershipESGexternalitiesfinancial frictionsasset pricingcorporate finance
Core finding, identification, data
Core Finding
- Public equity ownership of European coal plants has declined significantly due to plant scaling down (utilization) rather than sales (reallocation).
- Conversely, state ownership has risen, with national states being slowest to scale down and foreign/local states selling to private firms, suggesting states prioritize social factors like jobs and energy security, which can lead to nationalization and slow the energy transition, especially during energy shortages.
Identification Strategy
- The paper introduces a formal decomposition of ownership changes that distinguishes between capital reallocation (asset sales) and capital utilization (scaling down plants).
- This decomposition, applied to asset-level data, allows them to isolate "leakage" (asset sales) and infer investor mandates and valuations of externalities and social factors by observing differential behavior across investor groups (public equity, private, state) in terms of keeping, selling, or retiring assets.
- They calibrate a model using nine moments derived from this decomposition.
Data
The paper uses a novel dataset combining asset-level data on EU coal power plants (2015-2022) from Beyond Fossil Fuels (BFF) with firm-level shareholder data from S&P Capital IQ and manual collection for unlisted firms. This includes plant location, generation capacity, CO2 emissions (from EU ETS), ownership shares, and shareholder types.
Zhiguo He, Wenxi Jiang, Wei Xiong — Capital Markets and the Economy
This paper investigates whether the high price informativeness of Chinese A-share markets is driven by earnings management rather than genuine fundamentals, proposing a 'manipulate-to-cater' mechanism.
Finance Application
- This research could inform international asset pricing models by highlighting how earnings quality and investor sophistication in emerging markets (like China) affect the pricing of accounting-based factors.
- For household finance, the findings on retail investors' failure to recognize earnings inflation could be used to design financial literacy programs or behavioral interventions to improve retail investor outcomes.
- In insurance, the prevalence of earnings management and its impact on firm fundamentals could influence the underwriting and pricing of Directors & Officers (D&O) liability insurance or credit risk insurance, as managed earnings misrepresent true corporate risk.
earnings managementprice informativenessChina A-share marketNon-Recurring Gains and LossesNRGLsdelisting rulesretail investorsmarket efficiencystock returnspayoutscash flowsnatural experimentasset pricinghousehold financeinsurance
Core finding, identification, data
Core Finding
- Chinese A-share stock prices predict future earnings as well as U.S. stocks, but this is partly due to firms manipulating reported earnings (especially through Non-Recurring Gains and Losses, NRGLs) to align with optimistic market valuations, leading to earnings reversals and weak predictability for payouts and cash flows.
- Investors fail to fully recognize this earnings inflation, resulting in lower subsequent returns.
Identification Strategy
- The study leverages the 2020 delisting rule reform in China as a natural experiment.
- This reform excluded NRGLs from earnings calculations for regulatory delisting decisions, effectively increasing the cost of earnings management via NRGLs.
- This allows the authors to observe changes in NRGL usage and its correlation with market valuation and future earnings/payouts.
Data
The paper uses financial information and stock returns for Chinese A-share firms (1995-2022) from the China Stock Market and Accounting Research (CSMAR) database, including Non-Recurring Gains and Losses (NRGLs). For comparison, it uses U.S. S&P 500 firms (1960-2021) data from Compustat. Additional data sources include iFinD and WIND for expected shell probability (ESP) calculations.
Samuel Antill, Zhaoli Jiang, Neng Wang — Corporate Finance
This paper models how distressed firms use 'creditor-on-creditor violence' (Liability Management Transactions or LMTs) to strip collateral from existing lenders and pledge it to new lenders, mitigating debt overhang and improving ex-ante firm value.
Finance Application
- The paper's insights on debt overhang and renegotiation frictions under asymmetric information could be applied to asset pricing by refining models for distressed debt valuation and credit risk, particularly for syndicated loans where LMTs are prevalent.
- In household finance, the mechanism of 'collateral stripping' to facilitate liquidity could inform studies on mortgage or student loan renegotiations, where information asymmetry about borrower capacity or willingness to pay impedes efficient restructuring.
- For insurance, understanding how LMTs alter corporate default probabilities and recovery rates could impact the pricing of credit default swaps or other credit-related insurance products.
Corporate FinanceDebt RestructuringSecured DebtAsymmetric InformationDebt OverhangLiquidityBankruptcyMechanism DesignCredit Risk
Core finding, identification, data
Core Finding
- The paper finds that LMTs mitigate a debt overhang problem in distressed firms.
- By allowing equity holders to strip collateral from existing debt holders and pledge it to new liquidity providers, LMTs enable equity to internalize more of the benefits from successful liquidity injections, preventing costly bankruptcies and increasing ex-ante firm value.
- This explains why new debt contracts intentionally allow for nonexclusive LMTs.
Identification Strategy
- This is a theoretical paper that develops a formal model of debt renegotiation in distressed firms.
- The methodological innovation lies in introducing asymmetric information about investors' required returns and modeling LMTs as a mechanism that allows equity holders to transfer value from existing debt holders to new liquidity providers, thereby mitigating debt overhang.
Data
The paper is primarily theoretical but motivates its assumptions and findings using empirical observations. It references a self-collected list of 86 LMTs from sources like Credit Sights and S&P Global, and cites other empirical studies on loan contracts and bankruptcy filings.
Kyle F. Herkenhoff, Josh Lerner, Gordon M. Phillips, Francisca Rebelo, Benjamin Sampson — Corporate Finance
This paper measures the real effects of private equity buyouts on worker outcomes, finding that post-buyout employment and wage dynamics are consistent with professional investors increasing productivity and monitoring companies, rather than monopsony or breach of implicit contracts.
Finance Application
- The finding that PE value creation stems from efficient operational reallocation rather than market power exploitation could refine asset pricing models for private equity, suggesting that operational improvements are a key driver of PE returns and firm valuation.
- In household finance, the worker-level earnings losses and reallocation patterns could inform models of labor income risk and optimal household portfolio choice, especially for employees in industries prone to PE activity.
- For insurance, these insights could help price unemployment or income protection products for workers in PE-targeted sectors, by better assessing the stability of their labor income post-buyout.
Private EquityLabor MarketsMonopsonyImplicit ContractsWorker OutcomesProductivityReallocationBuyoutsEmpirical FinanceMicrodata
Core finding, identification, data
Core Finding
- Private equity buyouts lead to significant earnings and employment losses for workers, but these are primarily driven by efficient reallocation of workers from less productive to more productive plants.
- The study finds no evidence that PE-backed firms exploit local labor market monopsony power or breach implicit contracts with targeted high-wage workers; instead, high-wage workers are reallocated to more productive plants, suggesting a focus on efficiency gains.
Identification Strategy
- The paper employs several identification strategies: Coarsened Exact Matching (CEM) for baseline effects, comparing treated workers to a matched control group.
- For monopsony, it uses a within-firm-across-market identification, comparing workers in the same firm across markets with varying labor market concentration.
- For efficient reallocation, it uses a triple-difference strategy comparing treatment and control workers at productive versus unproductive plants.
Data
The paper links CapitalIQ data on private equity buyouts (1993-2013) to the Longitudinal Employer-Household Dynamics (LEHD) database, covering 2.5 million workers. It also uses the 2000 Decennial Census for occupation data and the Census of Manufactures for plant-level total factor productivity (TFP).
Xiao Cen, Winston Wei Dou, Leonid Kogan, Wei Wu — Capital Markets and the Economy
This paper investigates the determinants of compensation and career trajectories for US active equity mutual fund managers, emphasizing the roles of Assets Under Management (AUM), performance, and fund flows.
Finance Application
- This research offers several avenues for finance.
- In asset pricing, understanding how manager compensation is tied to AUM and flows, even from 'luck-driven' sources, could explain manager herding behavior or preferences for popular assets, potentially influencing stock prices beyond fundamentals.
- For household finance, the significant income risk faced by fund managers, particularly during job transitions driven by flows, could inform models of career risk and financial planning for high-earning professionals.
- The finding that self-disclosed compensation structures are misleading also has implications for investor protection and regulatory design in mutual funds.
fund managerscompensationmutual fundsAUMfund flowscareer outcomescausal inferenceinstrumental variableslabor economics
Core finding, identification, data
Core Finding
- US active equity mutual fund managers' compensation is primarily driven by AUM, with return performance directly influencing bonuses beyond its AUM impact.
- Fund flows significantly affect both compensation and career outcomes, where large outflows increase job turnover and compensation decline, and systematic flows impact base pay while idiosyncratic flows influence bonuses.
- Importantly, the study provides causal evidence that AUM and revenue, including 'luck-driven' variations, contractually determine manager compensation, often contradicting self-disclosed compensation structures.
Identification Strategy
- The study employs an instrumental variable (IV) approach to establish a causal link between AUM/revenue and manager compensation.
- The IV, inspired by Koijen and Yogo (2019), captures non-skill-related fluctuations in fund AUM and revenue by aggregating exogenous shifts in demand from other institutions for specific stocks, which are beyond the fund manager's control.
- This allows the authors to isolate the impact of 'luck-driven' AUM variations on compensation.
Data
The study leverages a unique dataset combining the US Census Bureau's Longitudinal Employer-Household Dynamics (LEHD) program for individual compensation and employment histories, with CRSP and Morningstar mutual fund data, and information from LexisNexis, CoreLogic, and SEC filings (SAIs).
Matthew S. Jaremski, Steven Sprick Schuster — Development of the American Economy
This paper investigates how the availability of limited federal deposit insurance through the U.S. Postal Savings System during the Great Depression impacted the stability and closure rates of nearby commercial banks.
Finance Application
- This research offers valuable insights for modern asset pricing and household finance, particularly concerning the impact of new 'safe' government-backed assets like Central Bank Digital Currencies (CBDCs).
- Researchers could model how the introduction of a CBDC might affect the funding costs and liquidity of commercial banks, influencing their equity valuations and the pricing of their debt during periods of market stress.
- In household finance, the 'flight to quality' observed could inform studies on how retail investors reallocate funds between traditional bank accounts, brokerage platforms, and new fintech offerings (with varying insurance levels) during economic uncertainty, impacting household portfolio choices and demand for financial stability.
deposit insurancebank runsfinancial crisesGreat Depressionpostal savingsliquidity riskbank stabilityregulatory competitionflight to qualityhistorical finance
Core finding, identification, data
Core Finding
- Commercial banks located near post offices offering federally-insured postal savings were significantly more likely to close during the Great Depression than banks in areas without such an option.
- This instability was driven by uninsured depositors withdrawing funds from commercial banks and moving them to the safer postal savings system, particularly affecting commercial banks with lower liquidity.
Identification Strategy
- The study employs an instrumental variable approach, leveraging a structural change during World War I when the Postmaster General ordered many 2nd and 3rd class post offices to stop taking deposits.
- The decision to continue accepting deposits was largely determined by whether the post office had at least one depositor (versus zero) in 1913-1915, which serves as an exogenous instrument for the availability of postal savings in 1928, as this initial, small difference in usage was unlikely to influence future commercial bank behavior.
Data
The paper utilizes balance sheet data for over 16,000 commercial banks in 1928, bank closure information from 1929-1935, town-level postal savings data, and county-level demographic and economic characteristics from the U.S. Census Bureau and tax records.
Gustavo S. Cortes, Cameron LaPoint — Development of the American Economy
This paper constructs a new century-long, geographically disaggregated dataset of U.S. building permits to demonstrate that local housing market activity, specifically building permit growth volatility, is a strong predictor of future financial market volatility.
Finance Application
- This research provides a novel, geographically disaggregated factor (BPG volatility) that could be integrated into multifactor asset pricing models to explain cross-sectional stock and bond returns, especially for firms with significant real estate exposure.
- Financial institutions and insurers could develop regional hedging strategies based on BPG volatility to mitigate exposure to localized housing market downturns.
- Furthermore, the BPG volatility could serve as an early warning indicator for macroprudential policy, informing regulators about impending regional financial risks linked to overbuilding.
Housing MarketReal EstateBuilding PermitsFinancial CyclesAsset PricingStock VolatilityBond VolatilityMacro-FinancePredictabilityBig DataMachine LearningRegional EconomicsOption ValueFinancial Crises
Core finding, identification, data
Core Finding
- The central finding is that local building permit growth (BPG) volatility robustly predicts future stock and corporate bond market volatility across a century of data and multiple recession episodes.
- This predictability persists even after controlling for macroeconomic factors and is particularly strong in housing supply-elastic regions, suggesting that building permits serve as a quasi-public signal reflecting investors' beliefs about local economic fundamentals.
Identification Strategy
- The identification strategy leverages a novel, hand-collected, and digitized century-long dataset of monthly building permits across U.S. states and MSAs.
- The authors use GARCH models to extract BPG volatility and then test its predictive power on financial market volatility.
- A theoretical framework models building permits as a call option, where their volatility acts as a noisy rational expectations signal about future dividends, providing a microfoundation for the empirical patterns.
- Principal Component Analysis (PCA) is also used to isolate a 'subprime' factor in BPG volatility that anticipates key events of the Global Financial Crisis.
Data
The paper utilizes a new hand-collected and digitized database spanning 1919-2019 of monthly building permit quantities and valuations for all U.S. states and the 60 largest MSAs, combining data from Dun & Bradstreet's Review and the U.S. Census Bureau's Building Permit Survey. It also incorporates CRSP, Finaeon, SDC Refinitiv, S&P Case-Shiller, Zillow, CoreLogic microdata, Dun & Bradstreet's DUNS Marketing Identifier, and various macroeconomic controls.
Mai C. Dao, Pierre-Olivier Gourinchas, Oleg Itskhoki — Monetary Economics
This paper develops a unifying empirical model to explain the joint dynamics of covered and uncovered currency premia, interest rates, and spot/forward exchange rates, using a partial equilibrium model of currency intermediation.
Finance Application
- The insight that dealer inventory (futures positions) drives significant exchange rate movements and predictable returns could be applied to other illiquid or intermediated asset markets, such as corporate bonds or less liquid derivatives, to identify demand shocks and predict returns.
- The distinction between global (CIP) and currency-specific (UIP) drivers of premia could inform macro hedging strategies or the design of FX intervention policies.
- Furthermore, the finding of predictable currency returns following dealer-driven depreciations could be exploited in quantitative trading strategies.
exchange ratescurrency premiacovered interest parityuncovered interest parityfinancial intermediationdealer banksfutures marketsasset pricingpredictabilitymacro-finance
Core finding, identification, data
Core Finding
- The paper finds that currency market patterns align with a model where intermediary banks, constrained by value-at-risk, supply hedged/unhedged currency.
- In the cross-section, local-currency savings determine interest rates and premia.
- In the time-series, covered premia are driven by global financial conditions, while uncovered premia respond to currency-specific demand shocks, largely explained by dealer banks' net futures positions, which also account for most spot exchange rate variation and lead to predictable currency returns.
Identification Strategy
- The core identification strategy uses changes in net currency futures positions of dealer banks as a proxy for currency-specific demand shocks.
- This reduced-form approach captures shifts in currency demand that drive equilibrium currency premia and exchange rate dynamics, without assuming exogeneity of these shifts or identifying primitive macroeconomic/financial shocks.
Data
The paper uses monthly data on 3-month covered and uncovered interest parity (CIP/UIP) premia for G7+ and emerging market currencies, constructed from money market rates, spot/forward exchange rates, and survey-based exchange rate forecasts. Crucially, it employs weekly net FX futures positions of dealer banks from the CFTC's Traders in Financial Futures (TFF) report, aggregated monthly. Global financial conditions are proxied by VIX, global financial cycle factors, and the broad dollar index.
Greg Buchak, Gregor Matvos, Tomasz Piskorski,, Amit Seru — Monetary Economics
This paper analyzes the secular decline of bank balance sheet lending since the 1970s, identifying its key drivers and implications for monetary policy and financial regulation.
Finance Application
- The model's decomposition of credit market shifts could be used to analyze the determinants of long-term trends and cross-sectional variation in credit spreads across different asset classes (e.g., traditional bank loans, corporate bonds, MBS).
- The insights into saver demand shifts, particularly due to policy changes like 401k/IRA reforms, offer a robust framework to study how household portfolio choices between deposits and securities respond to incentives and influence credit supply.
- Furthermore, the finding that increased capital requirements have modest effects on total lending but large effects on bank balance sheets could inform the design of macroprudential policies by modeling their impact on the entire credit ecosystem, including shadow banking.
Monetary PolicyBank RegulationFinancial IntermediationShadow BankingSecuritizationQuantitative EasingCapital RequirementsLiquidityCredit MarketsHousehold SavingsBorrower Demand
Core finding, identification, data
Core Finding
- The paper finds a secular decline in bank balance sheet lending (from 55% in the 1970s to 33% in 2023), driven primarily by borrower demand shifting to informationally insensitive credit and saver demand moving away from deposits.
- It demonstrates that while higher capital requirements significantly shrink bank balance sheets, their impact on total lending is modest as securities markets absorb the shift, underscoring diminishing regulatory leverage over credit supply.
Identification Strategy
The paper employs several identification strategies: a regression discontinuity design around GSE conforming loan limits to identify borrower demand shifts towards securitized lending; a difference-in-differences approach using the 2001 ERISA reforms to identify saver shifts from deposits to securities; and event studies around post-GFC regulations and Quantitative Easing policies to assess their impact on bank balance sheets and lending.
Data
The study utilizes comprehensive data from the Financial Accounts of the United States, HMDA loan-level mortgage origination data, FDIC Quarterly Banking Profile, Preqin for private debt funds, and bank Call Reports, alongside various market rates and economic indicators.
Stefan Avdjiev, Leonardo Gambacorta, Linda S. Goldberg, Stefano Schiaffi — International Finance & Macroeconomics
This paper analyzes the heterogeneity, evolution, and drivers of the risk sensitivity of global liquidity flows, distinguishing between cross-border bank loans and international debt securities, and across different borrower types and countries.
Finance Application
- The findings on how balance sheet constraints of specific types of financial intermediaries (banks vs.
- NBFIs) influence the risk sensitivity of international debt securities and cross-border loans could be used to model time-varying credit risk premiums in global bond and loan markets.
- Researchers could investigate how the capital and leverage ratios of large institutional investors (e.g., mutual funds, pension funds, insurance companies) affect their demand for specific international assets (e.g., emerging market sovereign bonds, corporate debt) and thus their pricing and liquidity.
- This framework could also be applied to understand how regulatory changes impacting intermediary balance sheets translate into shifts in asset demand and market stability.
global liquidityrisk sensitivitycross-border loansinternational debt securitiesfinancial intermediariesbanksnon-bank financial institutionsbalance sheet constraintscapitalizationleveragerisk migrationVIXemerging marketsadvanced economiesasset pricingcredit marketsinstitutional investors
Core finding, identification, data
Core Finding
- The risk sensitivity of global liquidity flows to global risk (proxied by VIX) varies significantly over time and across different financial instruments and borrower groups.
- Post-GFC, the risk sensitivity of cross-border bank lending declined, while international bond issuance by EME borrowers remained sensitive.
- These shifts are primarily driven by the balance sheet constraints of financial intermediaries (banks and NBFIs) and the migration of riskier borrowers from bank loans to international debt markets, with higher bank capitalization and lower NBFI leverage associated with lower risk sensitivities.
Identification Strategy
- The paper employs a baseline regression model to estimate the relationship between global liquidity components and global/local factors, including the VIX.
- It then extends this by introducing interaction terms with financial health metrics (bank capitalization, NBFI leverage) and a risk migration proxy (IDS share) to analyze their impact on risk sensitivities.
- The study also identifies a structural break in 2009:Q1 and uses recursive estimations to examine the time-varying nature of these sensitivities.
Data
The paper uses quarterly data from Q1 2000 to Q1 2024 from the BIS Locational Banking Statistics (LBS), International Debt Securities Statistics (IDSS), and Consolidated Banking Statistics (CBS). It incorporates bank balance sheet data from Fitch, NBFI vulnerability metrics from the Financial Stability Board (FSB), and portfolio investment data from the IMF Coordinated Portfolio Investment Survey (CPIS). Global risk is proxied by the VIX index, with alternative measures (BEX RA, VSTOXX, MOVE) used for robustness.
Mai C. Dao, Pierre-Olivier Gourinchas, Oleg Itskhoki — International Finance & Macroeconomics
This paper offers a unifying empirical model to explain covered and uncovered currency premia, interest rates, and spot/forward exchange rates across currencies and over time, emphasizing the role of intermediary banks' balance sheet constraints and currency demand shocks.
Finance Application
- The insight that dealer banks' net FX futures positions are a strong predictor of currency returns and exchange rate movements could be applied to developing systematic trading strategies in FX markets, particularly for currencies with high liquidity and active futures markets.
- The model's emphasis on frictional intermediation and balance sheet constraints could be extended to analyze the impact of financial regulations on efficiency and risk pricing in other derivatives markets, such as commodity or credit default swap markets.
- This framework could also inform institutional investors' hedging strategies by distinguishing between idiosyncratic currency risks and systemic financial risks.
Exchange RatesCurrency PremiaCovered Interest Parity (CIP)Uncovered Interest Parity (UIP)Frictional IntermediationDealer BanksFX FuturesAsset PricingMacro-FinanceInternational FinanceBalance Sheet ConstraintsRisk PremiaPredictabilityGlobal Financial Cycle
Core finding, identification, data
Core Finding
- The paper finds that a partial equilibrium model, where intermediary banks face value-at-risk balance-sheet constraints, explains rich empirical patterns in currency premia and exchange rates.
- Specifically, currency-specific demand shocks, captured by dealer banks' net FX futures positions, drive uncovered currency premia and account for most of the variation in spot exchange rates, leading to predictable appreciations after sharp depreciations.
Identification Strategy
- The identification strategy relies on a simple partial equilibrium model of the currency market, where global intermediary banks absorb residual net demand for currency subject to balance-sheet constraints.
- The key innovation is using changes in net currency futures positions of dealer banks as a proxy for currency-specific demand shocks, which are then linked to UIP and CIP premia and exchange rate dynamics.
Data
The paper uses monthly 3-month ahead exchange rate forecasts from Consensus Economics, 3-month money market/interbank deposit rates, and spot/forward exchange rates for G7+ and emerging market currencies. Crucially, it employs net FX futures positions of dealer banks from the CFTC's Traders in Financial Futures (TFF) weekly report, scaled by total open interest, and various global financial condition proxies like VIX and external dollar asset gaps.
Gianluca Benigno, Alessandro Rebucci, Aliaksandr Zaretski — International Finance & Macroeconomics
This paper revisits the Bianchi and Mendoza (2018) framework to analyze the multiplicity of constrained-efficient equilibria and the optimal design of time-consistent macroprudential policies, decomposing them into ex-ante and ex-post components.
Finance Application
- The concept of multiplicity of equilibria and the impact of policy components on asset prices could be applied to asset pricing models to understand 'crisis premia' or volatility spikes during periods of policy uncertainty.
- In household finance, the decomposition of debt taxes into ex-ante and ex-post components could inform the design of dynamic mortgage regulations, analyzing how different policy mixes affect household leverage, default risk, and housing market stability.
- The 'risk sharing component' of the optimal tax could also be explored in insurance research to model how macroprudential policies act as a form of social insurance, influencing the demand for and pricing of private insurance products against financial shocks.
Macroprudential policyFinancial crisesDebt taxationConstrained efficiencyTime consistencyPecuniary externalitiesAsset pricingHousehold debtWelfare analysisEquilibrium multiplicity
Core finding, identification, data
Core Finding
- The paper demonstrates a multiplicity of constrained-efficient equilibria in the workhorse model, with the unconstrained equilibrium being welfare-dominant.
- It quantitatively shows that the optimal time-consistent tax on debt, comprising both ex-ante (macroprudential) and ex-post (crisis-resolution) components, is welfare-improving, and these components are complements rather than substitutes, yielding much lower or even negative welfare gains if implemented alone.
Identification Strategy
- The methodological innovation involves extending the analysis of an established macro-finance framework (Bianchi and Mendoza, 2018) to formally prove the existence of multiple constrained-efficient equilibria.
- It then rigorously decomposes the optimal time-consistent debt tax into its ex-ante and ex-post components, analyzing their individual and joint welfare implications through a calibrated computational model.
Data
The paper is theoretical and computational, relying on a calibrated model economy. It performs quantitative analysis by simulating the decentralized competitive equilibrium (DE) for 101,000 periods, rather than using real-world datasets.
Thomas Mertens, Tony Zhang — Forecasting & Empirical Methods
This paper develops a New Keynesian model using risk-neutral expectations derived from financial market prices to extract the real-time state of the economy, derive benchmark interest rates, and forecast inflation.
Finance Application
- The real-time, high-frequency estimates of natural rates, optimal policy rates, and inflation risk premiums could be directly integrated into quantitative trading strategies for fixed income and interest rate derivatives, such as identifying mispricing in Treasury bonds or OIS contracts relative to the model's benchmarks.
- The decomposition of inflation into demand and supply shocks could inform dynamic asset allocation strategies, allowing investors to tilt portfolios towards sectors or factors that perform well under specific inflation drivers.
- Furthermore, the model's ability to extract perceived central bank reaction functions from market data could be used to refine models of central bank communication and its impact on asset prices, particularly for forward guidance strategies.
New Keynesian modelfinancial marketsrisk-neutral expectationsinflationinterest ratesmonetary policyyield curveinflation risk premiumterm premiumnatural rateinflation forecastingOISinflation swapsreal-time estimation
Core finding, identification, data
Core Finding
- The model accurately extracts the state of the economy from daily financial market data, showing that markets predicted the post-COVID inflation surge by mid-2021 and that inflation risk premiums turned positive thereafter.
- Its inflation forecasts are as accurate as leading alternatives, and it provides real-time estimates of natural and optimal monetary policy rates, along with a decomposition of inflation into demand and markup shocks.
Identification Strategy
- The paper's identification strategy involves matching risk-neutral expectations from a log-linearized New Keynesian model directly with observable financial market prices, specifically inflation swap rates and Overnight Indexed Swap (OIS) rates across various maturities.
- It estimates shock magnitudes, their persistence (short-run and long-run components), and associated risk premiums by minimizing squared errors between model-implied and observed term structures, allowing these parameters to vary daily.
Data
The paper uses daily end-of-day Overnight Indexed Swap (OIS) rates on the federal funds rate and inflation swap rates from Bloomberg, covering maturities from 1 to 30 years. These rates are fitted with Nelson-Siegel curves. It also incorporates observed monthly CPI inflation rates. The sample period for the financial data ranges from January 1, 2014, to November 8, 2024.
Rahul Mazumder, Seth Pruitt, Landon Ross — Forecasting & Empirical Methods
This paper predicts future aggregate risk exposures (betas) using both topic models (what is discussed) and context models (how it is discussed) extracted from 10-K risk disclosures, finding both are significant and complementary predictors.
Finance Application
- This methodology could be applied in household finance to analyze text from personal financial statements or loan applications, using 'what' is disclosed (e.g., income, assets) and 'how' it's described (e.g., cautious vs. confident language) to predict default risk or savings behavior.
- In insurance, it could be used on policy documents or claims narratives to identify subtle linguistic cues that predict claims frequency or severity, or to price emerging risks not captured by traditional actuarial models.
- The 'what vs. how' distinction could also be applied to earnings call transcripts to predict M&A success or innovation outcomes, focusing on the nuanced language used by management.
text analysistopic modelscontext modelstext embeddinggroup lassorisk disclosures10-K filingsasset pricingbeta predictionunstructured datamachine learningcorporate finance
Core finding, identification, data
Core Finding
- The study finds that both topic and context models from 10-K risk disclosures provide significant and distinct predictive information for future aggregate risk exposures, even after controlling for structured firm characteristics.
- Context models, which capture the 'how' of disclosure, are particularly strong predictors for market beta and investment risk, and the information is economically valuable, yielding a 28 basis points per annum certainty equivalent for a Bayesian investor.
Identification Strategy
- The paper's identification strategy relies on two distinct text analysis methodologies: Latent Dirichlet Allocation (LDA) for topic modeling to capture 'what' is discussed, and text embedding with a group lasso regularization for context modeling to capture 'how' it is discussed.
- The group lasso allows for sparse and interpretable selection of relevant context vectors.
- Robustness against overfitting is ensured through a strict out-of-sample analysis, training models on 2006-2014 data and testing on 2015-2022 data.
Data
The paper uses Item 1A risk disclosures from 10-K filings (2006-2022), linked to CRSP/Compustat data. It also incorporates Fama-French portfolio factors (Mkt-RF, SMB, HML, CMA, RMW, Momentum) and macroeconomic factors from FRED (exchange rate, credit spread, term spread).
Paul Goldsmith-Pinkham, Tianshu Lyu — Forecasting & Empirical Methods
This paper re-examines financial event studies through a causal inference lens, demonstrating how factor model misspecification biases abnormal return estimates, especially in long-horizon analyses, and proposes synthetic control methods as a robust solution.
Finance Application
- This paper's methodological insights are directly applicable within asset pricing to improve the robustness of event studies.
- For instance, researchers could re-evaluate the long-run performance of IPOs, M&A, or seasoned equity offerings using synthetic control methods to isolate the true causal impact from factor-driven noise.
- In household finance, it could be used to causally assess the impact of specific regulatory changes on the stock performance of financial institutions or the wealth of specific investor groups.
- For insurance, it offers a way to more accurately measure the market's reaction to new product launches, regulatory shifts, or major disaster events for specific insurers by constructing robust counterfactuals.
Causal InferenceEvent StudiesFactor ModelsSynthetic ControlAbnormal ReturnsAsset PricingEconometricsTreatment EffectsMarket EfficiencyPolitical ConnectionsIndex Inclusion
Core finding, identification, data
Core Finding
- The paper demonstrates that traditional abnormal return estimators in financial event studies are prone to inconsistency due to factor model misspecification, especially in long-horizon analyses where bias accumulates.
- It shows that synthetic control methods provide a more robust approach by directly modeling counterfactual security paths, mitigating these biases without requiring correct specification of the underlying factor structure.
Identification Strategy
- The paper introduces a potential outcomes framework for financial event studies.
- It identifies the causal effect by constructing counterfactual returns using synthetic control methods, which create a weighted combination of control units to match the pre-event characteristics and returns of the treated unit.
- This approach is presented as superior to traditional factor models, particularly when factor structures are misspecified or event timing is non-random.
Data
The paper uses daily returns from Datastream for political connections analysis (revisiting Acemoglu et al., 2016) and Siblis Research data for S&P 500 index inclusion events, matched to CRSP for firm returns. It also uses daily Fama-French returns for simulations.
Victor Degorce, Olivier Accominotti, Jason Cen, David Chambers — International Asset Pricing
This paper provides long-run empirical evidence on the covered interest parity (CIP) condition based on a century of data, showing that deviations are the norm rather than the exception and linking them to financial intermediaries' constraints.
Finance Application
- The paper's long-run data and findings on intermediary constraints could be applied to study the historical evolution of market efficiency and arbitrage limits in other asset classes, such as fixed income or commodities.
- Researchers could investigate how similar regulatory shifts or financial intermediary health metrics (e.g., bank capital ratios, balance sheet capacity) have historically impacted the pricing of derivatives or the effectiveness of arbitrage in bond markets.
- The methodology for constructing no-arbitrage bands and assessing their economic significance could also be adapted to analyze other cross-market arbitrage opportunities, providing a century-long perspective on their persistence and drivers.
Covered Interest ParityCIPFX SwapsArbitrageFinancial IntermediariesBanking RegulationTransaction CostsHistorical DataInternational FinanceAsset PricingForeign Exchange
Core finding, identification, data
Core Finding
- Economically significant deviations from Covered Interest Parity (CIP) have been the norm rather than the exception over the past century, persisting even after accounting for capital controls and transaction costs.
- The only period when CIP deviations were economically insignificant was from the mid-1990s to 2006, immediately preceding the Global Financial Crisis (GFC).
- These long-run patterns are linked to changes in financial intermediaries' constraints, particularly those associated with banking regulation.
Identification Strategy
- The paper employs a panel regression framework to analyze the economic significance of CIP deviations, specifically focusing on 'quarter-end effects' as a proxy for intermediary constraints.
- It leverages historical regulatory changes (e.g., Basel I implementation, post-GFC regulations) as quasi-natural experiments, testing whether CIP deviations are more pronounced during periods and at times (like quarter-ends) when regulatory constraints on banks' balance sheets are stricter.
- This difference-in-differences approach helps isolate the impact of regulatory-induced frictions.
Data
The paper constructs a comprehensive daily dataset of bid and ask quotes for spot and FX swap rates, as well as money market interest rates, covering 19 advanced-economy currencies from 1921 to 2025. Data are sourced from newspaper archives, other historical records, and modern electronic databases, with a systematic cleaning and cross-checking process. Offshore (eurocurrency) interest rates are used where possible to account for capital controls and credit risk comparability.
Mai C. Dao, Pierre-Olivier Gourinchas, Oleg Itskhoki — Macro, Money and Financial Frictions
This paper develops a unifying empirical model to explain covered and uncovered currency premia, interest rates, and exchange rates, emphasizing the role of intermediary banks' balance sheet constraints and dealer net futures positions in both cross-sectional and time-series dynamics.
Finance Application
- The paper's framework, linking intermediary balance sheet constraints and order flow (dealer positions) to asset prices, could be extended to other illiquid or intermediated asset markets like corporate bonds, private credit, or derivatives for other asset classes.
- Specifically, the use of dealer net positions to identify demand shocks and predict subsequent asset returns could be applied to analyze predictability in less liquid fixed income or derivatives markets, informing strategies for institutional investors or market makers.
currency premiaexchange ratescovered interest parityuncovered interest paritydealer balance sheetsfrictional intermediationfutures positionsasset pricingmacro-financepredictable returnsmarket microstructure
Core finding, identification, data
Core Finding
- In the cross-section, a country's local-currency funding gap determines its interest rates and currency premia (e.g., excess savings lead to low rates and negative premia).
- In the time-series, uncovered currency premia are volatile and currency-specific, driven by dealer net futures positions, while covered premia are stable and linked to aggregate financial conditions.
- Currency-specific demand shocks cause sharp spot exchange rate depreciations followed by small, persistent, predictable appreciations.
Identification Strategy
- The paper identifies currency-specific demand shocks using changes in dealer banks' net FX futures positions from the CFTC's Traders in Financial Futures (TFF) report.
- This variable serves as a reduced-form proxy for overall shifts in currency demand that intermediaries must clear, allowing the authors to trace its impact on equilibrium currency premia and exchange rates via a partial equilibrium model of frictional intermediation.
Data
The study uses 3-month money market/interbank deposit rates, spot and forward exchange rates, and 3-month ahead consensus survey exchange rate forecasts for G7+ and emerging market currencies. Crucially, it employs weekly net futures positions of FX dealer banks from the CFTC's TFF report, alongside global financial condition proxies like the VIX and the broad dollar index.
Sinem Hacioglu Hoke, Daniel A. Ostry, Hélène Rey, Adrien Rousset Planat, Vania Stavrakeva, Jenny Tang — International Asset Pricing
This paper provides a high-frequency, granular analysis of the London FX derivatives market, detailing its structure, firm-level hedging/speculative motives, and how different participants transmit aggregate shocks to exchange rates.
Finance Application
- The granular firm-level data and the distinction between hedging and speculative flows could enhance asset pricing models for currencies, particularly in explaining anomalies and the impact of institutional flows on exchange rates.
- The insights into dealer bank 'toll-taking' behavior could inform market microstructure studies on liquidity provision in OTC derivatives.
- Furthermore, the analysis of insurers' FX hedging strategies and their sensitivity to costs offers direct implications for risk management and capital requirements in the insurance sector.
FX DerivativesExchange RatesHedgingSpeculationInstitutional InvestorsMarket MicrostructureMonetary PolicyCredit RiskNon-Bank Financial InstitutionsDealer Banks
Core finding, identification, data
Core Finding
- Pension funds, investment funds, insurers, and non-financial corporations primarily use FX derivatives for hedging, while hedge funds engage in speculation based on carry, momentum, and macro news.
- Dealer banks act as 'toll-takers' by taking offsetting exposures.
- Hedge funds play a key role in transmitting monetary policy shocks, and investment funds contribute to dollar appreciation during credit risk events.
Identification Strategy
- The study employs a two-stage least squares (2SLS) approach, instrumenting sector-level derivatives positions with aggregate shocks.
- Specifically, monetary policy surprises (from Fed, ECB, BoE, BoJ) are used to identify hedge fund behavior, and a US credit spread macro news index is used to identify investment fund behavior, examining their causal impact on exchange rates.
Data
The paper uses 100 million transaction-level FX derivatives data from the UK segment of the European Market Infrastructure Regulation (EMIR) Trade Repository (DTCC and UnaVista) for over 16,000 firms in the UK FX market from January 2015 to December 2020, constructing daily firm-level net FX derivatives exposures.
Valentin Haddad, Zhiguo He, Paul Huebner, Péter Kondor, Erik Loualiche — Macro, Money and Financial Frictions
This paper provides a framework for using causal inference methods to identify portfolio demand and its equilibrium impact in asset markets, under specific assumptions about substitution patterns.
Finance Application
- This framework provides a rigorous foundation for applying causal inference to various asset pricing questions, such as quantifying price impact of central bank asset purchases, understanding demand for sustainable assets, or analyzing the effects of index inclusions.
- It clarifies how to interpret micro-level elasticities versus aggregate substitution patterns, guiding empirical design in studies of market efficiency, liquidity, and investor behavior across different asset classes like equities, corporate bonds, and treasuries.
Causal InferenceAsset PricingDemand ElasticityPrice ImpactInstrumental VariablesDifference-in-DifferencePortfolio ChoiceFinancial MarketsCorporate BondsMonetary Policy
Core finding, identification, data
Core Finding
- The paper establishes that under an assumption of homogeneous substitution conditional on observables, cross-sectional causal inference identifies the *relative* demand elasticity (or multiplier) between assets with similar characteristics.
- To identify aggregate elasticities or substitution patterns across different characteristics, exogenous time-series variation is required, allowing for a decomposition into micro, meso, and macro effects.
Identification Strategy
- The identification strategy relies on instrumental variable (IV) or difference-in-difference (DiD) regressions, using exogenous shocks (e.g., Fed bond purchases or mutual fund flows) as instruments.
- The core methodological innovation is the 'homogeneous substitution conditional on observables' assumption, which allows for consistent estimation of *relative* elasticities by addressing omitted variable bias from cross-asset spillovers.
Data
The empirical examples primarily use data on U.S. investment-grade corporate bonds from the WRDS Bond Returns database (2010-2022), mutual fund holdings and flows from the CRSP Survivor-Bias-Free US Mutual Fund Database, and macroeconomic/factor data from FRED and Kenneth French libraries.
Elisabeth Kempf, Mancy Luo, Margarita Tsoutsoura — International Economics and Geopolitics
This paper investigates how the political ideology of U.S. firm CEOs, relative to foreign governments, influences cross-border trade relationships and firm-level economic outcomes.
Finance Application
- This research highlights a novel 'political alignment risk' that could be priced in asset markets.
- Investors might demand higher risk premia for firms whose CEOs exhibit greater ideological misalignment with key foreign trading partners, especially those with high international trade exposure, leading to cross-sectional differences in stock returns or firm valuations.
- Furthermore, the findings could inform the development of new financial instruments, such as derivatives, to hedge against ideologically-driven supply chain disruptions, or influence risk assessment and pricing in trade credit and political risk insurance markets.
CEO IdeologyPolitical EconomyGlobal TradeFirm NetworksSupply ChainsPolitical RiskAsset PricingCorporate GovernanceInternational Finance
Core finding, identification, data
Core Finding
- U.S. firms are significantly more likely to terminate trade relationships with countries whose government's ideology becomes more distant from that of their CEO after a foreign election.
- This effect is stronger for shorter trade relationships and for more politically engaged CEOs, leading to economic costs such as increased cost-of-goods-sold-to-revenue ratios for import-heavy firms and reduced revenue.
Identification Strategy
- The study employs a quasi-natural experiment using close foreign elections as an exogenous source of variation in ideological distance.
- A difference-in-differences design compares changes in trade patterns between Democratic- and Republican-led U.S. firms trading with the same country around the same foreign election, controlling for election x time, firm x election, and product x time fixed effects to isolate the causal impact of CEO ideology.
Data
The paper combines granular transaction-level trade data from S&P Global's Panjiva database, U.S. CEOs' political affiliations from voter registration records, party ideology scores and election data from the Manifesto Project Database (MPD), and firm-level financial data from Compustat. It also uses the Notable Names Database for CEO prominence and the Global Party Survey Database for alternative ideology measures.
Valentin Haddad, Zhiguo He, Paul Huebner, Péter Kondor, Erik Loualiche — Asset Pricing
This paper provides a comprehensive framework for applying causal inference methods to asset pricing, clarifying how to identify demand elasticities and price impacts while accounting for asset substitution.
Finance Application
- This framework can be applied to rigorously estimate demand elasticities and price impacts across various asset classes (e.g., equities, fixed income, real estate) for different investor types (e.g., sovereign wealth funds, insurance companies, retail investors).
- It provides a structured approach to quantify the impact of policy changes (e.g., central bank interventions, ESG regulations) or market events on asset prices, explicitly accounting for how investors substitute between assets based on their characteristics like duration, credit quality, or sustainability scores.
- This can help in designing more effective policy interventions and understanding market microstructure.
Causal InferenceAsset PricingDemand ElasticityPrice ImpactInstrumental VariablesDifference-in-DifferencePortfolio ChoiceSubstitution PatternsCorporate BondsFactor ModelsEconometrics
Core finding, identification, data
Core Finding
- The core theoretical finding is that under assumptions of homogeneous substitution conditional on observables and constant relative elasticity, cross-sectional instrumental variable regressions identify only the relative elasticity (the difference between own-price and cross-price elasticity).
- To identify aggregate elasticities and substitution patterns along specific characteristics, joint estimation using multiple sources of exogenous time-series variation is required.
Identification Strategy
- The identification strategy relies on two key assumptions: (1) homogeneous substitution conditional on observables, meaning assets with the same observables share the same cross-price elasticity, and (2) constant relative elasticity, where the difference between own-price and cross-price elasticity is constant.
- The framework uses standard instrumental variable or difference-in-difference regressions, with instruments providing exogenous variation in prices or demand shocks, such as Fed asset purchases or mutual fund flows.
Data
The paper uses data on investment-grade corporate bonds from the WRDS Bond Returns database (2010-2022) and mutual fund bond holdings and flows from the CRSP Survivor-Bias-Free US Mutual Fund Database. It also references FRED and Kenneth French data library for factor data.
Stefan Nagel — Asset Pricing
This paper argues that the strong out-of-sample performance of Random Fourier Features (RFF) based return prediction in short training windows is an illusion, as it mechanically reduces to a volatility-timed momentum strategy rather than learning complex predictive relationships.
Finance Application
- This paper's insights are critical for evaluating other machine learning models in asset pricing that claim to find complex predictive relationships with short training windows.
- Researchers could apply this framework to test if other ML strategies are implicitly reducing to known, simpler risk premia (e.g., value, size, momentum) due to mechanical properties of the data and model structure.
- The methodology of decomposing complex ML model outputs into interpretable components could also be used to understand the drivers of ML-based factor strategies, distinguishing genuinely novel signals from re-packaged known risk premia.
- This emphasizes the need for sufficiently long training periods or explicit regularization in ML model design for finance.
asset pricingmachine learningreturn predictionRandom Fourier Featureskernel regressionmomentumvolatility timingmodel interpretabilitysample sizeoverfittingfactor models
Core finding, identification, data
Core Finding
- The paper demonstrates that in ridgeless regression with Random Fourier Features (RFF) and short training windows (e.g., 12 months), the high-complexity model does not learn complex predictive relationships.
- Instead, it mechanically approximates a kernel ridgeless regression, which effectively becomes a volatility-timed momentum strategy.
- This strategy performs well historically due to coincidental success of volatility-timed momentum, not due to genuine learning from the data.
Identification Strategy
- The paper's methodological innovation is to analytically show that RFF-based ridgeless regression, when the number of features (P) vastly exceeds the training observations (T), closely approximates a kernel ridgeless regression with Gaussian kernels.
- This re-interpretation reveals that the strategy's weights are mechanically driven by predictor similarity (leading to momentum) and predictor volatility (leading to volatility timing), rather than learned predictive signals.
- This is empirically supported by artificial data experiments where the RFF strategy still produces momentum-like forecasts even when the true data exhibits reversals.
Data
The paper uses the CRSP value-weighted index excess returns and predictor variables from Welch and Goyal (2008), augmented with lagged index returns. It also constructs artificial return data with MA(2) components to simulate reversals.
Joseph Engelberg, Asaf Manela, William Mullins, Luka Vulicevic — Asset Pricing
This paper proposes and validates 'entity neutering,' a method using Large Language Models (LLMs) to anonymize text and mitigate look-ahead bias in financial predictions, while preserving semantic content.
Finance Application
- This methodology is directly applicable to any finance research leveraging LLMs for textual analysis on historical financial data, such as earnings call transcripts, analyst reports, or regulatory filings (e.g., 8-Ks, 10-Ks, 10-Qs).
- Researchers can use entity neutering to extract sentiment or other textual features for predicting stock returns, bond yields, or corporate events, ensuring that their findings are not contaminated by look-ahead bias from the LLM's training data.
- It provides a robust framework to quantify and bound the impact of such bias, enhancing the credibility of LLM-based financial predictions.
Large Language ModelsLook-ahead BiasTextual AnalysisSentiment AnalysisFinancial NewsMD&AReturn PredictabilityEconometricsMethodologyArtificial Intelligence
Core finding, identification, data
Core Finding
- Entity neutering effectively masks identifying information in financial texts, with LLMs unable to recognize the firm or date in 90% of neutered articles.
- The sentiment extracted from raw and neutered text is highly correlated (90% agreement), and both show similar return predictability.
- The difference in predictability between raw and neutered text provides an upper bound on look-ahead bias, found to be at most 19-28% for news articles and 32-71% for MD&A texts, suggesting LLMs can be both the problem and the solution for look-ahead bias.
Identification Strategy
- The methodological innovation is 'entity neutering,' where an LLM (GPT-40 mini) is prompted to remove all identifying information (names, dates, products, industries, contextual clues) from a text, using its best judgment to ensure no expert could identify the subject.
- The effectiveness of this neutering is then rigorously tested by feeding the anonymized text to a *separate instance* of an LLM (or other LLMs like Llama 3, Gemma 3) to see if it can still identify the firm or date, thus validating the anonymization process.
Data
The study uses 1,056,053 financial news articles from the Dow Jones Archive (2000-2009, with a robustness sample extending to 2023) and 283,660 Management Discussion and Analysis (MD&A) sections from firm 10-Q and 10-K filings (2002-2023).
Chuck Fang, Kairong Xiao — Asset Pricing
This paper analyzes the transmission channels of conventional and unconventional monetary policy to long-term bond yields using granular portfolio holdings data of major U.S. bond investors.
Finance Application
- The random-coefficient demand system and decomposition framework could be applied to other asset classes (e.g., equities, private credit) to understand how diverse institutional investors (e.g., pension funds, hedge funds) respond to macroeconomic shocks, and how their heterogeneous preferences and constraints shape equilibrium prices.
- The insights on retail inflows could inform household finance research on direct bond holdings or other retail investment products, modeling how household preferences interact with policy.
- Furthermore, the finding that new issuance, rather than dealer activity, clears the market could be generalized to other markets to study liquidity provision and market structure.
Monetary PolicyBond MarketsPortfolio HoldingsInstitutional InvestorsDemand SystemsQuantitative EasingTerm StructureCredit SpreadsMarket ClearingInvestor Heterogeneity
Core finding, identification, data
Core Finding
- When the Federal Reserve conducts expansionary monetary policy, mutual funds, insurance companies, and banks increase bond purchases, acting as a "helping hand" that amplifies the Fed's impact on long-term yields.
- This elevated demand is primarily absorbed by new bond issuances, not dealers, and the "helping hand" effect is three times more influential than changes in risk-free rates in driving corporate bond issuance.
- The differing speeds of demand and supply responses explain the short-run excess sensitivity and long-run reversal of yields.
Identification Strategy
- The paper develops a demand-based asset pricing framework with random coefficients to model investor portfolio choices.
- Identification relies on two sets of instruments: exogenous bond characteristics and flow-induced trading by other investors, where idiosyncratic flows to an investor are used as plausibly exogenous shocks to yields for that investor.
Data
The paper uses granular portfolio holdings data from 2003Q1 to 2022Q4 for mutual funds (Morningstar), insurance companies (NAIC filings), banks (FFIEC Call Reports), primary dealers (New York Fed), and the Federal Reserve (SOMA, SMCCF), covering Treasury notes and bonds, agency mortgage-backed securities (MBS), and corporate bonds.
Benjamin Knox, Juan M. Londono, Mehrdad Samadi, Annette Vissing-Jorgensen — Asset Pricing
This paper develops a methodology using daily S&P 500 option expirations to identify "equity premium events" characterized by significantly elevated ex-ante equity premium relative to the daily equity term structure.
Finance Application
- This real-time methodology for pricing event risk could be directly applied by quantitative hedge funds or asset managers to develop short-term trading strategies around scheduled macroeconomic announcements or political events, such as options trading strategies that exploit mispricing or hedging strategies to manage event-specific volatility.
- Furthermore, the decomposition framework could inform the design of structured products tailored to specific types of event risk (e.g., inflation risk, monetary policy risk).
- Insurers could also use this to price event-contingent contracts or assess systemic risk exposures related to major economic announcements.
Equity PremiumOptionsMacroeconomic AnnouncementsFOMCCPINFPEvent RiskQuantile RegressionRisk Premia DecompositionMarket EfficiencyRealized Returns
Core finding, identification, data
Core Finding
- Macroeconomic and political events, especially FOMC, nonfarm payrolls, and CPI releases, are associated with significantly elevated ex-ante equity premia, which increased substantially between June 2022 and June 2023.
- However, these option-implied premia are quantitatively smaller than estimates derived from realized excess returns in prior literature, suggesting that realized returns may partly reflect unexpected good news.
- The variation in these event premia is driven by changes in news variance and the sensitivities of the stock market and stochastic discount factor to the news.
Identification Strategy
- The study leverages the rich forward term structure of S&P 500 option prices to derive ex-ante equity premium estimates for daily forward periods.
- It identifies 'equity premium events' by calculating residuals from a quantile (median) regression of these forward premia on term structure variables (term, term squared, and a first-expiration-of-week dummy).
- This data-driven approach isolates abnormal premia without being influenced by outliers, and an asset pricing framework further decomposes these premia into news variance and market/SDF sensitivities.
Data
The paper uses end-of-day S&P 500 option prices from Optionmetrics (October 2016-December 2023), covering daily expirations. It also incorporates 124 U.S. macroeconomic variables from the Bloomberg Economic Calendar and hand-collected historical data for FOMC, NFP, and CPI releases dating back to 1928, along with SPY ETF trades from TAQ for realized variance.
Milena Wittwer, Andreas Uthemann — Asset Pricing
This paper examines how financial intermediaries (dealers) specialize across Canadian bond, stock, and derivatives markets and the impact of this specialization on their trading performance.
Finance Application
- While already a finance paper, its methodology for quantifying multi-market specialization using the Theil index and LEIs could be applied to study institutional investor behavior (e.g., pension funds, endowments) across traditional and alternative asset classes, linking specialization to portfolio performance, risk management, and capital allocation decisions.
- This could also inform regulatory discussions on systemic risk from highly specialized, interconnected financial groups, by mapping their cross-market exposures and potential contagion channels during stress events.
Market MicrostructureFinancial IntermediariesSpecializationAsset PricingMarket SegmentationBroker-DealersBondsStocksDerivativesTheil IndexInstrumental VariablesLegal Entity Identifiers (LEIs)
Core finding, identification, data
Core Finding
- Financial intermediaries exhibit significant specialization across asset classes, which is more pronounced than within-market product specialization.
- This specialization is driven by a combination of firm organization, market frictions, and client demand, and more specialized dealers consistently obtain better prices, indicating a trading advantage.
Identification Strategy
To address endogeneity concerns, the authors employ two strategies: 1) they relate lagged specialization scores (from the previous year) to current trade margins, assuming past specialization is less likely to be influenced by current pricing; and 2) for stock market trades, they use daily client orders as an instrumental variable for dealer specialization in own-account trades, positing that client orders are plausibly exogenous shocks to a dealer's specialization.
Data
The paper utilizes a unique trade-level dataset from 2019 to 2022, covering Canada's fixed-income market (Debt Securities Transaction Reporting System, MTRS2.0) and all exchanges owned by the TMX Group (Toronto Stock Exchange, TSX Venture Exchange, TSX Alpha Exchange, and Montreal Exchange for derivatives). It links trading activity across these markets using Legal Entity Identifiers (LEIs).
Xugan Chen, Allen Hu, Song Ma — Big Data and High-Performance Computing for Financial Economics
This paper investigates how banks strategically develop brand images through advertising and how these efforts influence franchise value and monetary policy transmission.
Finance Application
- This research could be applied to household finance by examining how advertising influences household decisions regarding specific financial products like investment accounts, mortgages, or insurance policies, especially for different demographic segments.
- In asset pricing, the 'brand image' or 'advertising capital' built by financial firms could be explored as a factor influencing their valuation, risk, or expected returns.
- For insurance, the methodology could analyze how advertising shapes consumer perceptions of risk and trust, impacting policy demand and pricing strategies across different insurance lines.
Financial AdvertisingBankingBrand ImageMonetary PolicyFranchise ValueArtificial IntelligenceVideo EmbeddingsBorder Discontinuity DesignMarketingDeposit QuantitiesSpreads
Core finding, identification, data
Core Finding
- Banks consistently build images around pricing, service quality, or life aspirations.
- Advertising intensity and style significantly affect deposit quantities and spreads, particularly during monetary policy transmission, indicating that image building strengthens market positions and boosts franchise value.
Identification Strategy
The study employs a border discontinuity design to identify the causal effects of advertising intensity and style on deposit quantities and spreads.
Data
The paper analyzes over a decade of TV commercial videos using transformers and video embeddings, and utilizes Nielsen data.
Thomas Ernst, Bryan Routledge, Chester S. Spatt, Ariel Zetlin-Jones — Big Data and High-Performance Computing for Financial Economics
This paper analyzes proprietary data from a large brokerage firm to quantify the price improvement retail customers receive from off-exchange (dark) cryptocurrency trades compared to exchange-based trading.
Finance Application
- This research provides a direct parallel to dark pools in traditional equity markets, offering insights into how market fragmentation and intermediation affect price discovery and liquidity in a less regulated asset class.
- In asset pricing, this could inform models of crypto asset valuation by incorporating the impact of off-exchange liquidity on price efficiency and volatility.
- For household finance, it highlights the value proposition of financial intermediaries in nascent markets, prompting research into retail investor decision-making regarding direct exchange access versus brokerage services, and the potential for regulatory arbitrage.
- It also opens avenues to study the competitive landscape between crypto brokers and exchanges, and how best execution rules (if implemented) might reshape the crypto market structure.
cryptocurrencymarket microstructuredark poolsbest executionretail tradingbrokerageliquidityprice improvementmarket efficiencyfinancial intermediation
Core finding, identification, data
Core Finding
- The study reveals $550 billion in dark crypto trades, facilitated by a brokerage firm, frequently offer price improvement over hypothetical 'NBBO' prices, saving customers between $38 and $74 million annually.
- This benefit stems from the broker's ability to access diverse off-exchange liquidity, a mechanism not widely understood or regulated in the cryptocurrency market.
Identification Strategy
- The identification strategy involves comparing actual execution prices from a brokerage firm's proprietary off-exchange trades against a hypothetical 'National Best Bid or Offer' (NBBO) constructed from major U.S. and E.U. cryptocurrency exchanges, as well as fee-adjusted quotes from individual and second-best exchanges.
- Regression analysis is used to link price improvement and price impact to market conditions like order imbalance and wholesaler concentration.
Data
The paper utilizes proprietary trading records of $550 billion in cryptocurrency trades placed through a major brokerage firm from January 2020 to December 2023, covering 19 USD-denominated symbols. It also incorporates quote data from major U.S. and E.U. cryptocurrency exchanges (Coinbase, Kraken, Gemini, Bitstamp, Binance-US, FTX-US) to construct benchmark prices.
Daniel Barth, Phillip J. Monin, Emil Siriwardane, Adi Sunderam — Asset Pricing
This paper investigates agency problems in asset management by examining how investors' inability to observe the true risks taken by hedge fund managers leads to capital misallocation.
Finance Application
- This paper's insights on hidden risk and information asymmetry could be applied to understanding pricing anomalies in less transparent asset classes like private equity or venture capital, where true risk exposures might be systematically underestimated by investors.
- In household finance, it could inform research on how retail investors perceive and react to risk in mutual funds or robo-advisors, especially for complex strategies.
- For insurance research, the framework could be used to study how insurers (as large asset managers) manage and disclose the risks of their own investment portfolios, particularly for illiquid or alternative assets, and how this affects solvency or product pricing.
Hedge FundsAgency ProblemsAsymmetric InformationRisk MeasurementCAPM BetaInvestor FlowsForm PFMarket ShocksPrivate InformationAsset Management
Core finding, identification, data
Core Finding
- Hedge fund managers possess valuable private information about fund risk-taking that investors cannot infer from past returns.
- Investor flows respond to this information only when it is beneficial to managers, indicating that contracting and reputation do not fully resolve the asymmetric information problem.
- This leads to a significant misallocation of capital, with 10-20% of flows potentially misdirected.
Identification Strategy
- The authors leverage a unique regulatory dataset (SEC Form PF) where hedge fund managers report their perceived risk exposures directly to the SEC, but this information is not shared with investors.
- They compare these manager-perceived betas with investor-inferable betas from historical returns.
- The 2020Q1 COVID-19 market crash serves as a natural experiment to observe how investors update their perceptions of fund risk when hidden risks are revealed by extreme market movements.
Data
The primary data source is confidential regulatory data from the U.S. Securities and Exchange Commission (SEC) Form PF, covering most U.S. hedge fund assets. Additional data includes the CRSP Value-Weighted index for market returns and Form ADV for manager ownership information.
Manav Chaudhary, Julie (Zhiyu) Fu, Haonan Zhou — Asset Pricing
This paper develops an empirically tractable equilibrium model of the U.S. Treasury market to quantify investor sensitivities to yields and macroeconomic factors, decompose yield movements by investor, and analyze investor responses during economic episodes.
Finance Application
- The demand-system approach and optimal GIV estimator could be applied in asset pricing to analyze price formation and liquidity in other concentrated markets like private credit or municipal bonds, identifying key investor types and their macro sensitivities.
- In household finance, the methodology could model granular household portfolio rebalancing across diverse asset classes (e.g., stocks, real estate) in response to economic shocks, providing insights into their 'flight-to-quality' behavior beyond Treasuries.
- For insurance research, the framework could quantify how specific regulatory changes or interest rate environments impact life insurers' demand for long-duration assets, informing their asset-liability management and systemic risk analysis.
Treasury marketyield movementsinvestor demandprice elasticitymacro multiplierGranular Instrumental Variables (GIV)flight-to-safetyFederal Reserveforeign investorsfinancial crisisquantitative easingasset pricingmarket liquidity
Core finding, identification, data
Core Finding
- The U.S.
- Treasury market is relatively inelastic with a macro multiplier of 1.0, but exhibits significant heterogeneity in price elasticities across investor sectors.
- A structural shift occurred post-2008, with foreign investors' influence declining and the Federal Reserve emerging as a state-contingent liquidity provider.
- Contrary to conventional wisdom, foreign investors do not exhibit flight-to-safety behavior for Treasuries; instead, domestic investors drive countercyclical price movements.
Identification Strategy
- The paper employs an optimal Generalized Method of Moments (GMM) estimator, an extension of Granular Instrumental Variables (GIV), to address simultaneity bias.
- It identifies idiosyncratic sector-specific demand shifters as instruments, assuming their orthogonality across sectors.
- The estimator optimally weights moment conditions, prioritizing shocks from larger sectors with more volatile holdings, to enhance instrument relevance and efficiency.
Data
The study utilizes quarterly sector-level Treasury holdings and transactions from the Financial Accounts (Z.1), complemented by country-level Treasury International Capital (TIC) data for foreign investors and Call Reports for individual commercial banks. Treasury bond yields are measured using the CRSP US Treasury Database.
Balint Szoke, Jonathan Payne, Clemens Lehner, Jack Shurtleff — Asset Pricing
This paper estimates the historical funding cost advantage of the U.S. government from 1860-2024 by constructing new corporate and Treasury bond datasets and adjusting for tax treatments and embedded options.
Finance Application
- The detailed methodology for adjusting bond prices for complex tax effects and embedded options (like flower bonds or callability) could be adapted to price other fixed-income securities with similar features, such as municipal bonds with specific tax treatments or convertible bonds with embedded options.
- The extensive, corrected historical dataset provides a rich laboratory for studying long-term trends in credit risk, liquidity premia, and government financing costs, which could inform asset pricing models for rare disasters or long-run risk.
- The finding that debt supply primarily impacts short-term spreads could refine models of government debt and asset pricing, suggesting that long-term bond risk premia are driven more by traditional risk factors than by supply-demand imbalances.
Corporate BondsTreasury BondsYield CurvesFunding AdvantageHistorical DataTax EffectsEmbedded OptionsKernel Ridge RegressionDebt-to-GDPAsset PricingFixed Income
Core finding, identification, data
Core Finding
- The U.S. government's funding advantage emerged with the 1862-65 National Banking Acts, staying high at ~1.5% until 1920, then dropping.
- Existing measures have exaggerated this advantage, especially during the 1970s-80s, due to mismeasurement of flower bonds and inflation risk premia.
- Quantity-driven spreads are primarily a short-term phenomenon, with debt-to-GDP increases forecasting spread declines for maturities less than 1 year, but not for long-term bonds.
Identification Strategy
- The authors construct zero-coupon yield curves for highest-grade corporate and government bonds using a new micro-level dataset spanning 1860-2024.
- They deploy a Kernel Ridge estimator and make crucial adjustments for bond heterogeneity, including differential tax treatments (e.g., tax exemptions, capital gains tax advantage) and embedded options (flower bonds, call options, exchange privilege) using a generalized pricing formula to ensure like-for-like comparisons.
Data
The paper uses a new historical dataset of US corporate bonds (1840-2024) compiled from historical newspapers, business magazines, and existing databases (Lehman Brothers, Merrill Lynch). It also utilizes a comprehensive US Treasury debt dataset (1790-2024) combining Hall et al. (2018) and CRSP Treasury Database, along with Moody's credit ratings.
Oliver Hellum, Theis I. Jensen, Bryan T. Kelly, Lasse H. Pedersen — Big Data and High-Performance Computing for Financial Economics
This paper proposes and analyzes the "Common Task Framework" (CTF), a collaborative and competitive research process, demonstrating its effectiveness in enhancing innovation, reducing research costs, and improving comparability, with a specific application to financial economics.
Finance Application
- Beyond asset pricing, the CTF could be applied to other finance sub-fields.
- For instance, in credit risk, a CTF could identify the best models for predicting corporate defaults using shared financial statements and market data, with a success metric based on out-of-sample default prediction accuracy.
- In insurance, a CTF could optimize premium pricing by evaluating models that predict claims based on shared policyholder and demographic data, using metrics like RMSE.
- In household finance, a CTF could be used to identify models forecasting household financial distress or savings behavior, fostering innovation in consumer credit risk and financial planning tools.
innovationcommon task frameworkasset pricingresearch methodologycompetitioncollaborationmachine learningSharpe ratiopricing kernelreproducibilityincentives
Core finding, identification, data
Core Finding
- The Common Task Framework (CTF), defined by shared data, a predefined success metric, and a leaderboard, incentivizes researcher effort, increases innovation, and curbs misrepresentation of findings.
- An economic model shows these benefits arise from reduced research costs and improved comparability.
- The paper applies CTF to asset pricing, systematically evaluating existing models and proposing a competition for new ones, finding that newer machine-learning methods generally outperform classic factor models and that ensemble methods further improve performance.
Identification Strategy
- The methodological innovation is the Common Task Framework (CTF) itself, which structures research as a collaborative and competitive process.
- The paper develops an economic model to formally analyze how CTF's elements (shared data, common success metric, leaderboard) impact researcher incentives, innovation, and comparability.
- It then applies this framework by replicating and comparing 18 asset pricing models on a common dataset with a unified metric, effectively creating a 'leaderboard' for existing finance research.
Data
The paper primarily uses a comprehensive set of global equity characteristics and returns from Jensen et al. (2023), available via WRDS. This dataset includes 153 'original JKP characteristics' and an expanded set of 402 characteristics. The data covers non-micro US stocks from 1952/01 to 2023/12.
Simona Abis, Bo Bian, Huan Tang — Big Data and High-Performance Computing for Financial Economics
This paper examines how Apple's App Tracking Transparency (ATT) policy, by restricting user-level data sharing, impacted financial information flows and market efficiency.
Finance Application
- While this paper is already in finance, its insights into regulatory-induced data degradation could be applied to other finance subfields.
- For asset pricing, the mechanism suggests that regulatory risk to data access could be priced, potentially creating new factors related to a firm's data vulnerability or reliance on alternative data.
- In household finance, similar data restrictions could impact consumer credit scoring models or personalized financial advice, leading to higher borrowing costs or reduced access to services for individuals.
- For insurance, changes in data availability (e.g., from wearables or smart home devices due to privacy rules) could significantly alter risk assessment and pricing for health, life, or property insurance, potentially creating new challenges for actuarial science.
privacy regulationsdata sharingfinancial marketsapp tracking transparencyATT policyalternative datamarket efficiencyforecast errorsstock pickingmutual fundssell-side analystsprice informativenessregulatory riskmobile dataasset pricinginformation asymmetry
Core finding, identification, data
Core Finding
- The study finds that ATT significantly reduced the predictive power of mobile-generated signals for firm performance and investor behavior.
- This led to a decline in mutual funds' stock-picking ability for mobile-data-exposed firms, increased analyst forecast errors for those relying on mobile data, and ultimately weakened price efficiency, particularly for firms most exposed to mobile-data-informed analysis.
Identification Strategy
- The paper leverages Apple's App Tracking Transparency (ATT) policy, introduced in 2021, as a natural experiment.
- ATT sharply restricted the sharing of user-level data across mobile applications on iOS, effectively severing cross-app tracking and decreasing the signal-to-noise ratio of mobile-generated alternative data.
- This policy shock allows for causal inference on the impact of privacy regulations on financial information.
Data
The paper uses granular mobile app usage data from Apptopia, firm financials from Compustat, mutual fund portfolio holdings from CRSP Survivorship-Bias-Free Mutual Fund dataset, and sell-side analyst forecasts and reports from I/B/E/S and LSEG.
Jason Allen, Milena Wittwer — Big Data and High-Performance Computing for Financial Economics
This paper documents the prevalence and pricing implications of multi-asset trade bundling by institutional investors in the Canadian fixed income market, highlighting its role in dealer-client relationships and the impact of electronic trading platforms.
Finance Application
- The insights into trade bundling and its cost implications could be applied to understanding liquidity provision and market efficiency in other OTC markets, such as derivatives or private credit, where relationship banking is crucial.
- Researchers could investigate if similar bundling strategies exist in equity block trading or M&A advisory, and how they affect deal pricing and counterparty selection.
- The methodology for quantifying relationship strength and transaction costs could also be adapted to study the value of long-term relationships in insurance underwriting or household financial advisory services.
OTC marketsfixed incometrade bundlingtransaction costsdealer-client relationshipsmarket microstructureelectronic tradingliquidityfinancial innovationasset pricing
Core finding, identification, data
Core Finding
- The paper finds that institutional investors commonly bundle trades, often across asset classes, with their most-favored dealers.
- These bundled trades, particularly 'switches' (simultaneous buy/sell of different assets), tend to save transaction costs, especially when intermediated by a relationship dealer.
- While electronic platforms facilitate bundling, switches on these platforms are not necessarily cheaper than bilateral trades, suggesting a convenience premium paid by investors.
Identification Strategy
- The study identifies trade bundling by observing multiple transactions between the same dealer and client within a five-second window involving different assets.
- It quantifies transaction costs by comparing actual trade prices to daily average benchmark prices for single assets and uses regression analysis with bond-day and dealer-client fixed effects to isolate the impact of bundling on costs, controlling for various factors.
Data
The paper utilizes trade-level data from the Market Trade Reporting System 2.0 (MTRS2.0) provided by the Canadian Investment Regulatory Organization (CIRO), covering the near-universe of the Canadian fixed income market from 2016 to 2023. This dataset includes unique Legal Entity Identifiers (LEIs) for institutional clients and dealers, as well as detailed trade information (time, price, quantity, asset characteristics).
Sean S. Cao, Zhongling Qin, Tao Shu — Big Data and High-Performance Computing for Financial Economics
This paper investigates how lightly regulated "Other Events" (OE) disclosures in 8-K filings lead to sentiment distortion, stock mispricing, and disproportionately harm retail investors.
Finance Application
- The paper's insights into sentiment distortion in lightly regulated disclosures, especially for intangible information, could be applied in asset pricing to examine the valuation of firms heavily reliant on intangible assets (e.g., AI, biotech) and whether their stock prices exhibit greater sentiment-driven volatility or long-term reversals.
- In household finance, the identified vulnerability of retail investors could inform research on how retail traders react to complex, less-regulated disclosures in emerging asset classes like cryptocurrencies or NFTs, potentially leading to new models of retail investor behavior under information asymmetry.
- For insurance, one could investigate if similar disclosure gaps exist in voluntary ESG reports or climate risk disclosures by insurers, and how such potentially distorted sentiment impacts policyholder trust, capital allocation decisions, or the pricing of insurance-linked securities.
disclosure regulationsentiment analysis8-K filingsretail investorsstock mispricingcorporate eventstextual analysisLLMintangible assetsregulatory scrutinyinsider tradingoption grantsseasoned equity offeringsmergers and acquisitions
Core finding, identification, data
Core Finding
- Lightly regulated "Other Events" (OE) disclosures in 8-K filings exhibit sentiment distortion, leading to initial stock mispricing that subsequently reverses, and negatively predicting firms' future financial performance.
- This distortion is most pronounced for hard-to-verify intangible information and is strategically timed to benefit managers and firms around corporate events like insider sales, option grants, seasoned equity offerings, and stock mergers, with unsophisticated retail investors being particularly vulnerable.
Identification Strategy
- The paper employs a quasi-natural experimental design by comparing lightly regulated "Other Events" (Item 8.01) disclosures with more strictly regulated non-OE disclosures within the same 8-K filings.
- Regulatory scrutiny is measured using SEC comment letters.
- Methodologically, it uses topic modeling (Latent Dirichlet Allocation) and ChatGPT for content analysis and interpretation, along with nearest-neighbor matching to control for event type differences.
Data
The study utilizes a comprehensive dataset of 8-K filings from SEC EDGAR (2004-2020), financial data from Compustat and CRSP, insider trading data from Thomson Financial Insider Filing, CEO option grant data from ISS Incentive Lab, M&A data from Thomson Reuters SDC Platinum, and retail order flow data from TAQ.
Songrun He, Linying Lv, Asaf Manela, Jimmy Wu — Big Data and High-Performance Computing for Financial Economics
This paper introduces ChronoBERT and ChronoGPT, large language models trained exclusively on chronologically consistent data to avoid lookahead bias, demonstrating competitive performance in NLP tasks and stock return prediction.
Finance Application
- This methodology of chronologically consistent LLMs could be extended to analyze long-term trends in corporate disclosures (e.g., 10-K/Q filings) without the risk of future information contaminating historical analyses of firm characteristics or risk factors.
- In household finance, it could be used to study the evolution of consumer sentiment or financial literacy from historical media, ensuring that models don't "know" about future crises when interpreting past narratives.
- For insurance, these models could analyze historical news or regulatory changes to predict future risk events or policy changes, providing a robust framework for backtesting underwriting strategies.
Large Language ModelsChronological ConsistencyLookahead BiasTraining LeakageAsset PricingFinancial NewsStock ReturnsNatural Language ProcessingBacktesting
Core finding, identification, data
Core Finding
- The authors find that chronologically consistent LLMs (ChronoBERT and ChronoGPT) achieve strong performance on NLP benchmarks and generate Sharpe ratios comparable to much larger, non-chronologically constrained models like Llama 3.1 in predicting next-day stock returns from financial news.
- A key insight is that lookahead bias in this asset pricing application is modest, and the earliest models perform nearly as well as later, more knowledgeable models, suggesting a "temporal alignment" where models calibrated to their specific historical period perform best.
Identification Strategy
- The methodological innovation is the creation of "chronologically consistent" LLMs (ChronoBERT and ChronoGPT).
- This is achieved by training models exclusively on text data available up to a specific point in time, with incremental training for subsequent years, and rigorously filtering data to prevent any future information leakage.
- The consistency is validated by testing the models' inability to predict events or presidents occurring after their knowledge cutoff date.
Data
The models are pretrained on a corpus of 7 billion tokens of chronologically organized English text (historical web content, archived news, scientific publications) up to 1999, with incremental training on 65 billion tokens up to 2024. For the asset pricing application, they use the Dow Jones Newswire dataset (news headlines, full articles with microsecond timestamps) from 2007-2023, merged with CRSP stock returns.
Aleksandar Andonov, Andy Li, Paul Smeets — Entrepreneurship
This paper empirically investigates whether Development Financial Institutions (DFIs) achieve their stated objectives of fostering economic development, entrepreneurship, innovation, and sustainability through venture capital investments, differentiating between developed and developing economies.
Finance Application
- The findings on DFI risk aversion and preference for later-stage, more experienced GPs/funds, despite their mandates for early-stage and ecosystem building, could inform research on LP behavior and its impact on GP strategy in private markets.
- This could extend to examining how the *governance structure* of different public LPs (e.g., subnational vs. multilateral DFIs, or even public pension funds) correlates with the risk-return profile of the funds they back, and whether this leads to different market outcomes (e.g., crowding out private capital, impact on innovation).
- The paper's detailed framework for mapping non-financial objectives to measurable impact indicators could also be adapted to evaluate the *real-world impact* of private impact funds more broadly, beyond just financial returns, by developing new metrics for 'impact alpha' or 'impact beta' in private markets.
Development financeVenture capitalImpact investingPrivate equityLimited partnersGeneral partnersMarket failuresESGInnovationEntrepreneurshipDeveloped economiesDeveloping economiesDifference-in-difference
Core finding, identification, data
Core Finding
- DFIs show mixed results in achieving their impact objectives.
- In developing economies, they are more likely to target industries with positive externalities, support underrepresented fund managers, and improve return transparency.
- However, DFIs are less likely to invest in early-stage funds or deals and do not significantly improve firm growth, innovation, or sustainability outcomes compared to conventional VCs.
- In developed economies, their impact is even more muted, and they exhibit higher risk aversion.
Identification Strategy
- The paper employs a stacked difference-in-difference (DiD) method.
- For each DFI-backed firm, it identifies a matched control firm backed by conventional investors, based on exact matches for investment year, deal stage, industry, country, and nearest neighbor matching on lagged total assets.
- This allows for dynamic comparisons of outcomes (e.g., profitability, employment, patenting, sustainability) while addressing selection bias and treatment heterogeneity.
Data
The study constructs a comprehensive database of 344 DFIs with VC investments across 57 economies, using Preqin for fund and deal data. Firm-level financial information and patenting activity are sourced from Orbis Historical Database, exit outcomes from SDC Platinum, environmental disclosure from Trucost, and corporate scandal data from RepRisk.
Harold L. Cole, Daniel Neuhann, Guillermo Ordoñez — Macroeconomics Within and Across Borders
This paper analyzes optimal primary sovereign bond auction protocols under asymmetric information about default risk and demand, proposing a novel 'partially discriminating' mechanism and providing empirical evidence from Mexico.
Finance Application
- The theoretical framework on auction design, asymmetric information, and price discovery can be extended to other asset classes beyond sovereign debt.
- For instance, the 'partially discriminating' auction mechanism could be adapted for corporate bond issuance or even for designing new market mechanisms in decentralized finance (DeFi) to optimize liquidity and price efficiency under information asymmetry.
- The empirical methodology for quantifying information content (elasticity of secondary prices, marginal R^2) could be applied to study price discovery in various primary and secondary markets for corporate securities, IPOs, or even real estate auctions.
Sovereign DebtAuctionsMarket MicrostructureAsymmetric InformationPrice DiscoveryPublic FinanceGovernment BondsUniform Price AuctionDiscriminatory Price AuctionPartial DiscriminationWinner's CurseVolatility
Core finding, identification, data
Core Finding
- Governments face a trade-off in sovereign bond auctions between extracting inframarginal surplus (discriminatory pricing) and encouraging broader participation by mitigating the winner's curse (uniform pricing).
- A novel 'partially discriminating' protocol is proposed to balance this by combining features of both.
- Empirically, discriminatory auctions are found to be more informative and lead to lower price volatility than uniform auctions, particularly when information asymmetry is endogenous.
Identification Strategy
- The paper leverages a natural experiment: Mexico's 2017 policy change from discriminatory to uniform price auctions for its Cetes (Treasury Bills).
- This allows for a direct comparison of information revelation and price volatility under different auction protocols, measured by the elasticity of secondary market prices to primary price surprises and the marginal R-squared of auction prices in explaining secondary market variance.
Data
The study uses Mexican Federal Treasury Bills (Cetes) and Bondes D auction data from May 2003 to June 2024, alongside secondary market prices reported by the Bank of Mexico's PIP and Valmer systems.
Jonathan Payne, Balint Szoke — Macro Public Finance
This paper develops a structural model and uses historical data to show that the US government's funding advantage, often attributed to treasuries' safe-asset properties, is fragile, policy-dependent, and influenced by financial regulation and fiscal policy.
Finance Application
- This framework could be applied to analyze the "liquidity premium" in other asset classes, such as highly-rated corporate bonds or certain derivatives, by modeling how institutional demand (e.g., from pension funds, insurance companies) driven by specific regulations or internal risk management policies creates a funding advantage for issuers.
- Researchers could also explore how changes in corporate governance or capital structure (analogous to fiscal policy) might erode this "advantage" by altering the asset's hedging properties for institutional investors.
Convenience YieldsGovernment DebtFinancial RepressionSafe AssetsFinancial RegulationFiscal PolicyAsset PricingBankingMacro-FinanceYield CurveBond-Stock BetaLiquidity Premium
Core finding, identification, data
Core Finding
The paper demonstrates that government funding advantage arises from the financial sector's ability to use treasuries for risk hedging, amplified by financial regulation that creates "captive demand." However, fiscal policies that destabilize treasury prices (e.g., through devaluation) erode this advantage, leading to a "trilemma" where a government cannot simultaneously achieve high funding advantage, financial sector stability, and fiscal policies that destabilize treasury prices.
Identification Strategy
- The paper uses a stochastic neoclassical growth model with a two-period structure (morning/afternoon markets) to microfound financial sector frictions and regulatory constraints.
- This model endogenizes the connections between financial regulation, fiscal policy, and government debt's role as a hedging asset.
- Empirical evidence is drawn from historical US yield curve estimates and bond-stock betas, analyzing how these interact with major regulatory and fiscal events.
Data
The paper utilizes new historical yield curve estimates for high-grade corporate bonds and US treasuries from 1860-2024 (Lehner et al., 2025), along with market value of government debt-to-GDP, bond-stock betas, stock market volatility, and dummies for regulatory eras. It also references Eurozone credit default swap (CDS) spreads for a comparative analysis.
Itamar Drechsler, Hyeyoon Jung, Weiyu Peng, Dominik Supera, Guanyu Zhou — Household Finance
This paper analyzes the economics of credit card banking using granular account-level data to explain why credit card interest rates are so high, decomposing revenues, costs, and risk premia.
Finance Application
- The methodology of using FICO-sorted portfolios to estimate systematic default risk betas and then pricing that risk via Fama-MacBeth could be applied to other granular credit markets, such as auto loans, student loans, or even small business loans, to understand their risk premia.
- The finding that marketing expenses create pricing power suggests that researchers could explore how advertising intensity in different financial product markets (e.g., mortgages, insurance, brokerage services) impacts pricing spreads and firm profitability, potentially revealing unexploited market power or competitive inefficiencies.
credit cardshousehold financeconsumer creditdefault riskrisk premiummarket poweroperating expensesFama-MacBethbankingROA
Core finding, identification, data
Core Finding
- Credit card interest rates are high due to a combination of substantial operating expenses (especially marketing, which generates pricing power) and a significant default risk premium.
- While default rates are high, they only partially explain the rates.
- After accounting for all costs and a 5.3% default risk premium, credit card lending still yields a substantial "alpha" of 1.17% to 1.44% compared to the overall banking sector.
Identification Strategy
- The paper uses a two-stage Fama and MacBeth (1973) approach to estimate the default risk premium.
- In the first stage, they estimate the beta of each FICO portfolio to systematic default risk by regressing the monthly change in its charge-off rate on the change in the aggregate credit card portfolio's charge-off rate.
- In the second stage, they regress the portfolios' ROAs on these estimated betas to find the compensation for default-risk exposure.
Data
The paper primarily uses a comprehensive supervisory dataset of 330 million monthly individual credit card accounts from the Federal Reserve's Y-14M reports (2015-2023), covering 90% of the US market. This is augmented with Call Reports for bank-level data and Mergent Fixed Income Securities Database (FISD) for corporate bond data.
Julia Fonseca, Lu Liu, Pierre Mabille — Real Estate
This paper studies the general equilibrium effects of mortgage lock-in on housing markets and evaluates a policy designed to alleviate these effects using a spatial housing ladder model.
Finance Application
- This paper directly informs household finance by quantifying the impact of mortgage lock-in on household mobility, housing consumption, and wealth accumulation across different housing ladder segments.
- For asset pricing, the heterogeneous price effects of lock-in across housing types and geographic areas could be used to model differential risk and return profiles for geographically segmented real estate investment trusts (REITs) or mortgage-backed securities (MBS) portfolios.
- The model's ability to track mortgage rates and loan balances could also be used to stress-test MBS portfolios under various interest rate shock scenarios, assessing prepayment and default risk changes due to lock-in.
mortgage lock-inhousing marketshouse priceshousehold mobilityfixed-rate mortgagesspatial equilibriummonetary policyconsumer creditpolicy evaluationhousehold finance
Core finding, identification, data
Core Finding
- Mortgage lock-in, caused by rising interest rates, increases house prices, particularly in expensive areas, because locked-in borrowers (disproportionately downsizers) would otherwise demand less housing.
- A temporary $10,000 tax credit to starter-home sellers modestly unlocks mobility but paradoxically increases trade-up home prices due to spillover demand.
Identification Strategy
- The paper provides causal evidence by exploiting individual-level variation in the timing of mortgage originations and aggregate variation in mortgage rates.
- It instruments household-specific mortgage rate deltas with aggregate mortgage rate deltas determined by current and origination-month rates to identify the effect of lock-in on mobility and house prices.
Data
The study uses a state-of-the-art consumer credit panel (Gies Consumer and Small Business Credit Panel - GCCP), Zillow house price index, CoreLogic Property Deeds data, Home Mortgage Disclosure Act (HMDA) for loan characteristics, American Community Survey (ACS), Panel Study of Income Dynamics (PSID), Freddie Mac Primary Mortgage Market Survey (PMMS), and FRED data.
Savannah G. Noray — Gender in the Economy
The provided image is a message inviting the reader to access a paper via a link, not the paper itself.
Finance Application
Cannot assess potential applications without the paper's content.
N/Apaper not provided
Core finding, identification, data
Core Finding
Not available as the paper content is not provided.
Identification Strategy
Not available as the paper content is not provided.
Data
Not available as the paper content is not provided.