This entry pertains to the NBER Asset Pricing and New Developments in Long-Term Asset Management (LTAM) Spring 2025 Meeting. I summarize the presentations that I listened to, followed by some general reflections.

Introduction

It was my first time back in Chicago after my graduation, and the city was in spring bloom. It was great to be back at the Gleacher Center, at which I spent countless days during my PhD.

Asset Pricing (AP) Session

• Maarten Meeuwis (MIT Sloan) presented “Time-Varying Risk Premia and Heterogeneous Labor Market Dynamics.” The paper argues that rising risk premia causes lower labor earnings, especially for low-paid workers, mainly through higher rates of job loss. They show this using administrative earnings data empirically and build a search-and-matching model where time-varying risk premia depress both hiring and job retention, particularly for workers with weaker bargaining positions. The core idea, which traces back to Hall (2017), is that employment is an investment decision for firms. So when discount rates (risk premia) rise, firms cut back on hiring and are quicker to fire, especially lower-paid workers whose employment is closer to their outside option.
• Kilian Huber (Chicago Booth) presented our paper on the program “Climate Capitalists.” David Thesmar (MIT Sloan) gratefully presented a discussion that help us interpret the perceived cost of capital using an approach in his recent paper.
• Francesca Bastianello (Chicago Booth) presented “Mental Models and Financial Forecasts” showing that financial analysts form forecasts using “sparse mental models”: they pay selective attention to a few firm-specific variables and choose valuation methods (like DCF or P/E multiples) that fit the information they focus on. Discussant Stefano Giglio (Yale) provided a useful roadmap of additional questions that the paper’s dataset could explore, including how analysts’ attention shifts across macro shocks. A broader question is whether we should care about analyst beliefs per se, or treat them as a window into the belief formation processes of broader investor populations.
• Moto Yogo (Princeton) presented “Upgrading Credit Pricing and Risk Assessment through Embeddings.” The paper proposes replacing traditional credit ratings with machine-learned “firm embeddings” estimated from bond holdings of mutual funds and insurance companies. These embeddings capture richer information than credit ratings and the distance to default, explaining variation in credit spreads, changes in spreads, and spread volatility more accurately. In the session, there was an active debate about what credit ratings are supposed to achieve. Some pointed out that ratings are meant to be through-the-cycle measures of credit risk, not real-time market valuations. This led to the broader question of whether fitting spreads better actually aligns with the regulatory and informational role of ratings.
• Alessandro Melone (OSU) presented “The Unintended Consequences of Rebalancing” which shows that mechanical rebalancing by large institutional investors creates predictable, non-fundamental price pressure across markets. The authors build empirical proxies for rebalancing intensity (Threshold and Calendar signals) and show they predict short-term stock and bond returns, with impacts reverting in about two weeks. Valentin Haddad (UCLA) compared the findings to that of Etula, Rinne, Suominen, and Vaittinen (2020), which connects flows today to future liquidity needs. I also agreed with the discussant that a more comprehensive assessment of costs and benefits could be nice.
• Wenhao Li (USC) presented “Granular Treasury Demand with Arbitrageurs” that presents a model of the Treasury market with both inelastic investors and arbitrageurs together. Adi Sunderam (HBS) gave a discussion that touched upon many broader questions surrounding demand system asset pricing, such as the interpretation of elasticity as a structural parameter, the construction of the instrument, and the paper’s relation to the literature.

New Developments in Long Term Asset Management (LTAM) Session

• Stefan Nagel (Chicago Booth) presented “Leaning against Inflation Experiences” in which he argues that a big chunk of the long-run swings in real interest rates—like the drop from the 1980s to today—can be explained by monetary policy “leaning against” inflation expectations that are based on people’s own lived experiences rather than rational forecasts. Since people update their inflation beliefs very slowly, central banks have to keep real rates high (or low) for a long time to shift expectations, which in turn moves real rates persistently. This experience-based learning dynamic helps explain real rate trends without needing to invoke changes in the natural rate of interest (r*). As is the case with Stefan’s other papers, this piece challenges the established consensus and raises big questions: if people are anchored mostly by past inflation experiences, does central bank communication even work?
• Marius Guenzel (Wharton) presented “What Drives Very Long-Run Cash Flow Growth Expectations?” in which the authors build a huge new dataset on analysts’ very long-run firm growth expectations (terminal growth rates) and finds that they behave very differently from short-term forecasts. The punchline seems to be that long-run growth expectations are much more subjective and idiosyncratic than we might expect.
• Ishita Sen (HBS) presented “The Market for Sharing Interest Rate Risk: Quantities and Asset Prices” where the authors study how interest rate risk is shared across sectors (banks, pension funds, insurers) in the UK swap market, using near-universal trade data. It shows that pension funds and insurers consistently buy duration (receive fixed), while banks and corporations sell it (pay fixed). But because each sector sticks to different maturities, dealers are left holding maturity-specific risks. The idea that maturity segmentation creates persistent risk for intermediaries feels important, especially for thinking about financial stability. But this begs the question: if preferred habitats are so sticky, can regulation ever really “fix” imbalances, or do you just shove risks from one pocket of the system to another?

Takeaways

1. Compared to a few years ago, there are many more papers grounded in demand system asset pricing, big data applications, and machine learning techniques. Embeddings, portfolio choice models, and asset flows are now front and center.
2. There’s a still emphasis on big empirical exercises, but I think there’s more ambition now to reshape foundational narratives: how inflation expectations are formed, how creditworthiness is assessed, how market structure influences asset prices.