Anomaly time.

We examine the timing of returns around the publication of anomaly trading signals.  Using a database that measures when information is first publicly released, we show that anomaly returns are concentrated in the first month after information release dates, and these returns decay soon thereafter.  We also show that the academic convention of forming portfolios in June underestimates predictability because it uses unnecessarily stale information, which makes some anomalies appear insignificant.  In contrast, we show many anomalies do predict returns if portfolios are formed immediately after information releases.  Finally, we develop guidance on forming portfolios without using stale information.

Gambling preferences, options markets, and volatility.

with Ben Blau and Ryan Whitby

This study examines whether the gambling behavior of investors affects volume and volatility in financial markets. Focusing on the options market, we find that the ratio of call option volume relative to total option volume is greatest for stocks with return distributions that resemble lotteries. Consistent with the theoretical predictions of Stein (1987), we demonstrate that gambling-motivated trading in the options market influences future spot price volatility. These results not only identify a link between lottery preferences in the stock market and the options market, but they also suggest that lottery preferences can lead to destabilized stock prices.

(Not) everybody's working for the weekend: a study of mutual fund manager effort.

We develop a novel measure of effort and revisit the fundamental questions of asset management: how do incentives relate to effort; and how does effort affect performance?  Using unique observations of daily work activity, we define mutual fund manager effort as the ratio of weekend work to weekday work.   We find that investment advisors with stronger competitive incentives exert more effort on weekends.  Focusing on within-advisor variation, we find that more effort follows poor performance, outflows and higher volatility.  Regarding future performance, we show that more effort is associated with higher future returns, especially for mutual funds with strong competitive incentives, higher active share, and lower turnover.  Finally, we demonstrate a causal link between effort and performance using exogenous variation in effort due to weather conditions.

Institutional investors, short-selling constraints, and information acquisition.

with Jesse Davis

This paper highlights an important interaction between short-sale prohibitions and information acquisition.  Investors faced with a short-sale prohibition acquire less information when the likelihood of trading is low.  Unconstrained investors react to their competitors' constraint and acquire more information.  Equilibrium prices are non-linear in fundamentals and price informativeness is asymmetric as a result of the short-sale prohibition.  The novel predictions of the model are tested using unique measures of information acquisition from sophisticated market participants.  When likely bound by their short-selling prohibition, investors decrease their information acquisition activity by up to 16%.

Mutual fund shorts and the benefits of acquiring information.

We study the performance and information acquisition behavior of mutual funds for both their long and short positions.  We show that managers acquire relatively more information about their shorts because the benefit of acquiring information about shorts is larger.  Mutual funds' shorts also generate better returns than their longs.  Moreover, with respect to short positions, performance and information acquisition are inversely related.  Though surprising, we demonstrate that this finding aligns with standard theory as the inverse relation follows from managers acquiring less information about certain high performing short positions, the clear winners.

Failures-to-deliver: when short sale constraints bind.

We study the impact of Failures-to-Deliver (FTD) on price efficiency and future returns.  When a stock experiences high FTDs, this (i) predicts high future FTDs, (ii) reduces price efficiency, such as increasing bid-ask spreads and price-impacts, and (iii) predicts lower future abnormal returns.  We show that among a large set of anomalies, high FTDs result in higher future anomaly returns, driven primarily in the short-leg.  Further, we show the effect of FTDs on anomaly returns is time-variant; high FTDs do not effect anomaly returns immediately, but instead lengthen the period of time in which anomaly returns are earned.  Our results all suggest that FTDs represent a limit to arbitrage driven by illiquidity in the equity lending market.  

Anomaly returns and information acquisition.

Using conventional methods (i.e., annual rebalancing) to study anomaly returns shows no relationship between abnormal returns and information acquisition.  However, using a more focused event-time approach to measure anomaly returns following precise information release dates, we show that abnormal returns are drastically reduced by increased levels of information acquisition from sophisticated investors.  This result is robust across a large subset of anomalies, controlling for confounding factors such as size, and even within recent time-periods.  Our results suggest that anomaly returns are not entirely spurious or the result of data-mining and are not wholly driven by omitted risk-factors.  Instead, a large portion of anomaly returns are the result of mispricing and the slow-diffusion of value-relevant information. 

Predicting anomalies.

Mutual fund flows and information acquisition.

The financial sophistication of CEOs and CFOs.