No CrossRef data available.
Published online by Cambridge University Press: 20 February 2023
We analyze a model in which an anomaly is unknown to arbitrageurs until its discovery, and test the model implications on both asset prices and arbitrageurs’ trading activities. Using data on 99 anomalies documented in the existing literature, we find that the discovery of an anomaly reduces the correlation between the returns of its decile-1 and decile-10 portfolios. This discovery effect is stronger if the aggregate wealth of hedge funds is more volatile. Finally, hedge funds increase (reverse) their positions in exploiting anomalies when their aggregate wealth increases (decreases), further suggesting that these discovery effects operate through arbitrage trading.
We thank Thummim Cho (the referee) and Jennifer Conrad (the editor) for their constructive comments. We also thank Nick Barberis, Bruno Biais, Alon Brav, David Brown, Bjorn Eraker, Will Goetzmann, Michael Gofman, Paul Goldsmith-Pinkham, David Hirshleifer, Jon Ingersoll, Wenxi Jiang, Marcin Kacperczyk, Andrew Karolyi, Leonid Kogan, Andrew Lo, Benjamin Loos, Steve Malliaris, David McLean, Alan Moreira, Justin Murfin, Lubos Pastor, Anna Pavlova, Jeff Pontiff, Mark Ready, Jialin Yu, Jianfeng Yu, and seminar participants at Boston University, DePaul University, Georgetown University, HKUST, Johns Hopkins University, PBCSF Tsinghua University, Peking University, Rutgers, SAIF, Temple University, SEM Tsinghua University, University of Florida, University of Toronto, University of Virginia, University of Wisconsin Madison, Yale, the 2019 American Economic Association Meetings (AEA), the 2016 European Finance Association Meetings (EFA), the 2016 European Summer Symposium in Financial Markets, the 2015 China International Conference in Finance (CICF), and the 2015 Northern Finance Association Meetings (NFA) for helpful discussions. Lu is grateful for financial support provided by Bryce Douglas Chair in Corporate Finance. An earlier version of this article was circulated under the title “A Model of Anomaly Discovery.”