with Gregory Brown and Christian Lundblad (Review of Financial Studies 2022)
Abstract: Hedge fund positions are an important component of crowded trades. These vehicles are particularly active, take highly concentrated positions, and utilize leverage and short sales. Using a database of hedge fund holdings, we measure the degree of security-level crowdedness. The difference between the average returns on portfolios sorted by high versus low crowdedness portfolios is sizable and their variation is distinct from other traditional risk factors. Further, hedge fund exposures to crowdedness are often significant and they help to explain downside "tail risk", as funds with higher exposures experience relatively larger drawdowns during periods of industry distress.
with Riccardo Colacito, Mariano M. Croce and Steven Ho (American Economic Review 2018)
Abstract: We study the response of international investment flows to short- and long-run growth news. Among developed G7 countries, positive long-run news for domestic productivity induces a net outflow of investments, in contrast to the effects of short-run growth shocks. We document that a standard Backus, Keho, and Kydland (1994) (BKK) model fails to reproduce this novel empirical evidence. We augment this model with Epstein and Zin (1989) preferences (EZ-BKK) and characterize the resulting recursive risk-sharing scheme. The response of international capital flows in the EZ-BKK model is consistent with the data.
with Tilan Tang and Zhiyan Wang
Abstract: This study examines the impact of anti-SLAPP (anti-Strategic Lawsuits Against Public Participation) laws on corporate social responsibility (CSR). Using the staggered adoption of these laws across U.S. states from 1995 to 2019 as a quasi-natural experiment, we find that anti-SLAPP laws significantly improve firms' CSR performance. Consistent with their role in protecting free speech, these laws are associated with increased negative media coverage and heightened activist pressure on CSR issues. The effects are stronger for firms with high ESG reputation risk and those headquartered in states with constituency statutes. Improvements span multiple CSR dimensions, particularly those most visible to public scrutiny. Overall, our findings suggest that legal protections for speech enhance external monitoring and incentivize greater corporate social responsibility.
with Anthony Diercks and Mehrdad Samadi
Abstract: We build and evaluate a language-server protocol for Dynare using a benchmark of seventeen error types across five canonical DSGE models. Providing structured model context improves repair pass rates by as much as fifty-two percentage points for mid-tier language models after five turns, demonstrating that domain-aware tooling can make AI-assisted economic modeling more reliable, reproducible, and accessible.
with Bill Marcum, Deon Strickland, and Tilan Tang
Abstract: Using Zillow data and instrumental-variable methods, we decompose ZIP-code housing returns and examine how persistent local housing risk changes households’ systematic exposure and portfolio volatility across macroeconomic regimes.
with Kyre Lahtinen and Deon Strickland
Abstract: We analyze more than 530,000 daily observations across 119 state markets and five commodities. Controlling for futures-market information, we use satellite vegetation measures, weather, drought, and network structure to identify how regional information innovations propagate through geographic and transportation links.
Abstract: In a New-Keynesian model subject to the zero lower bound (ZLB), constrained monetary policy endogenously results in time-varying equity risk premia and equity-bond market correlations. Liquidity traps at the ZLB are characterized by negatively skewed and increasingly uncertain consumption growth, labor growth, and inflation. Investors with recursive preferences price the liquidity traps, resulting in rising equity risk premiums. Real bond yields and equity returns become negatively correlated at the ZLB, while positive in normal times. The model provides a general equilibrium foundation for 1) the time-varying comovement amongst macroeconomic quantities and asset prices observed during the the Great Recession and 2) why real bonds ceased to provide investors with insurance at the ZLB, precisely when they valued it most.
with Gregory Brown, Juha Joenväärä and Christian Lundblad
Abstract: Using a novel data set containing fund returns and characteristics for almost all activist hedge funds, we study the risk-reward characteristics of activist hedge funds. Our preliminary analysis shows a large return gap between activist hedge funds and activist campaigns. Currently we seek to answer: Do activist hedge funds create value and/or risk for long-term investors? Who reaps the rewards from activist campaigns: shareholders, fund managers, and/or fund investors? Are fund risk-adjusted returns related to short-term and/or long-term activist objectives and tactics?
with Mike Bancks
Abstract: We develop a privacy-preserving federated learning framework for patient-outcome prediction across heterogeneous healthcare sites. The project uses synthetic and open-access clinical data with a masked discrete diffusion architecture to model incomplete patient records without centralizing raw data. We benchmark federated performance against local and pooled baselines and study how site heterogeneity, informative missingness, and governance constraints affect accuracy, privacy, and deployability.
with Mehrdad Samadi
Abstract: We construct a corpus of nearly 29,000 articles from twenty-two finance journals published since 2000. Language-model perplexity distinguishes top-journal articles from other articles with 76.6 percent accuracy. Observable article characteristics explain only about one quarter of that predictive performance, while more than 70 percent of the gap is associated with textual patterns.
with Tilan Tang
Abstract: We examine when validation of language-based executive measures supports generalization to previously unrated CEOs, using open-language CEO personality measure as the setting. Replication and simulation evidence shows that performance can appear strong when test pages come from executives represented during training, while the signal weakens when evaluation CEOs are excluded from fitting. The project develops a stricter validation standard centered on new people, verified text ownership, and invariance to irrelevant comparison cases.
with Anthony Diercks
Abstract: We characterize the welfare implications of the Federal Reserve's new monetary policy framework. We first show that the dynamics of the new framework can be model with a Taylor rule that features super-inertial price level targeting. Unconditionally, this closes the deflationary bias. Then we show the new policy stabilizes prices at the ZLB. The price stabilization reduces consumption volatility and is welfare enhancing. In addition, the new policy reduces the equity risk premium, which results in higher asset valuations and exacerbates inequality.
with Roberto Camassa, Bong Jae Chung, Richard McLaughlin, and Ashwin Vaidya (Advances in Mathematical Fluid Mechanics 2010)
Abstract: We study the orientational behavior of a hinged cylinder suspended in a water tunnel in the presence of an incompressible flow with Reynolds number (Re), based on particle dimensions, ranging between 100 and 6000 and non-dimensional inertia of the body (I*) in the range 0.1–0.6. The cylinder displays four unique features, which include: steady orientation, random oscillations, periodic oscillations and autorotation. We illustrate these features displayed by the cylinder using a phase diagram which captures the observed phenomena as a function of Re and I*. We identify critical Re and I* to distinguish the different behaviors of the cylinders. We used the hydrogen bubble flow visualization technique to show vortex shedding structure in the cylinder’s wake which results in these oscillations.