Long and Short
Bubbles, Booms and Crashes in the US Stock Market 1792-2024
William Goetzmann, Otto Manninen & James Tyler
NBER Working Paper, February 2026
Abstract:
We examine the historical frequency of stock market booms, crashes, and bubbles in the United States from 1792 to 2024 using aggregate market data and industry-level portfolios. We define a bubble as a large boom followed by a crash that reverses the market's prior gains. Bubbles are extremely rare. We extend the industry-level analysis of Greenwood, Shleifer, and You (2019) through 2024 and replicate their findings out of sample using Cowles Commission industry data from 1871 to 1938. Booms do not reliably predict crashes, but they do predict higher subsequent volatility, increasing the likelihood of both large gains and large losses.
Hidden alpha
Manuel Ammann et al.
Journal of Financial Economics, April 2026
Abstract:
We provide novel evidence suggestive of insider trading through concealed relationships identified using information from over 100,000 Facebook profiles and their 35 million friends. Focusing on connections between fund managers and firm officers, we demonstrate that hidden ties are linked to substantial abnormal returns averaging 135 basis points per month (exceeding 16% alpha annually, t-stat = 3.54) across the universe of mutual funds and public firms. These hidden ties emerge as the most powerful predictor of future stock returns among documented network characteristics, with predictive power increasing over time through the present day. The premium associated with such connections arises not from endogenous selection or familiarity bias; instead, fund managers exhibit specific timing ability in deciding when to hold (or avoid) stocks of firm officers linked through hidden ties. The value of trading information rises with the degree of concealment and is concentrated around earnings and M&A events. The premium is absent in index funds, where strategic stock selection and timing are infeasible. Our findings on the value of hidden ties remain robust across industries, investment styles, time periods, and firm types.
Mimicking Finance
Lauren Cohen, Yiwen Lu & Quoc Nguyen
NBER Working Paper, February 2026
Abstract:
We use frontier advancements in Artificial Intelligence and machine learning to extract and classify the part of key economic agents' behaviors that are predictable from past behaviors. Even the agents themselves might view these as novel (innovative) decisions; however, we show in strong contrast that a large percentage of these actions and behaviors can be predicted — and thus mimicked — in the absence of these individuals. In particular, we show that 71% of mutual fund managers' trade directions can be predicted in the absence of the agent making a single trade. For some managers, this increases to nearly all of their trades in a given quarter. Further, we find that manager behavior is more predictable and replicable for managers who have a longer history of trading and are in less competitive categories. The larger the ownership stake of the manager in the fund, the less predictable their behavior. Lastly, we show strong performance implications: less predictable managers strongly outperform their peers, while the most predictable managers significantly underperform. Even within each manager's portfolio, those stock positions that are more difficult to predict strongly outperform those that are easier to predict. Aggregating across the universe of fund managers each quarter, stocks whose position changes are least predictable additionally significantly outperform stocks whose position changes are most predictable across the universe. Our framework allows researchers to delineate and classify the portion of financial agents' action sets which are predictable from those which are novel responses to stimuli — open to being evaluated for value creation or destruction.
Do Anecdotes Matter? Exploring the Beige Book through Textual Analysis from 1970 to 2025
Shengwu Du et al.
Federal Reserve Working Paper, January 2026
Abstract:
We apply various natural language processing tools to see if the Beige Book is helpful in understanding economic activity. The Beige Book is a gathering of anecdotal compilations of current economic conditions from each Federal Reserve Bank, which is released to the public prior to FOMC meetings. We find that even controlling for lagged GDP growth and other metrics, the Beige Book sentiment provides meaningful explanatory power in nowcasting GDP growth and forecasting recessions, even more so than the yield spread or other news sentiment measures. The results on economic activity even hold in regional panel analysis. The Beige Book offers many more insights on the economy that can be gathered from even simple keyword tabulations. Topic modeling can also inform us about the different factors driving the narrative across particular periods of interest.
AI, Opinion Ecosystems, and Finance
David Hirshleifer et al.
NBER Working Paper, February 2026
Abstract:
Generative AI use for content generation is associated with divergent outcomes on different financial social media platforms: indications of reasoning enhancement on Seeking Alpha, and of belief distortions on WallStreetBets. On Seeking Alpha, adoption is associated with information frictions. AI-assisted postings tilt toward analysis/credibility, and their sentiment positively predicts future returns. Use of AI is associated with more informative retail order flow and lower bid-ask spreads. In contrast, AI adoption on WallStreetBets follows surges in retail buying, and AI-assisted content is associated with emotionality and sentiment contagion. Such content precedes higher trading volume, greater volatility, and more lottery-like return distributions.
Does Peer-Reviewed Research Help Predict Stock Returns?
Andrew Chen, Alejandro Lopez-Lira & Tom Zimmermann
Federal Reserve Working Paper, December 2025
Abstract:
Mining 29,000 accounting ratios for t-statistics > 2.0 leads to cross-sectional return predictability similar to the peer review process. For both, ? 50% of predictability remains after the original sample periods. This finding holds for many categories of research, including research with risk or equilibrium foundations. Only research agnostic about the theoretical explanation for predictability shows signs of outperformance. Our results imply that inferences about post-sample performance depend little on whether the predictor is peer-reviewed or data mined. They also have implications for the importance of empirical vs theoretical evidence, investors' learning from academic research, and the effectiveness of data mining.
Constrained by law: The impact of fiduciary duties on portfolios and prices in US equity markets
Stefano Cassella et al.
Journal of Financial Economics, March 2026
Abstract:
We study the equity market implications of a reform in the fiduciary laws that govern trust investments (prudent man laws), implemented in a staggered fashion across U.S. states from 1985 to 2006. As trusts account for a substantial fraction of institutional equity holdings in our sample period, and since the reform does not pertain to other investors, our empirical setting provides a rare opportunity to study the impact of a regulatory change on institutional investor holdings and relative prices in the U.S. equity market. We show that, before the reform, trusts tilt their portfolios towards prudent stocks. After the law change, trusts undo these tilts, which leads to substantial changes in portfolio performance, investor demand, and stock returns, consistent with a model of inelastic equity markets. More broadly, our paper documents a striking case of investment distortions: while the concept of diversification has been playing a key role in asset pricing theory since the 1950s, fiduciary duties severely constrained trusts' ability to diversify their portfolios for up to half a century later.
Search Spillovers: The SEC's Role in Shaping Information Acquisition
Laura Griffin & Jackie Wegner
University of Colorado Working Paper, November 2025
Abstract:
This study examines whether the SEC's user interface influences which firms' disclosures users acquire. We focus on EDGAR's company search function, which features a predictive dropdown menu that dynamically updates suggestions as users input their search queries. We find that when an S&P 500 firm experiences elevated information acquisition -- measured by EDGAR clicks -- neighboring firms on the dropdown menu also experience a significant increase in information acquisition, compared to control firms in the same industry and of similar size. Consistent with position bias, this spillover effect is stronger for neighboring firms ranked higher on the menu. Finally, we provide suggestive evidence that spillover attention from EDGAR's dropdown menu influences retail investors' trading behavior. These findings highlight the importance of the SEC's user interface design, as seemingly minor design features can have consequences for information acquisition and investor behavior.