Anti-Herding by Hedge Funds, Idiosyncratic Volatility and Expected Returns

Abstract

Utilizing a dataset of 1,899 U.S. hedge funds, we present evidence of anti-herding behavior among hedge fund managers in the U.S. Hedge funds anti-herd primarily based on fundamental information and irrespective of market volatility and credit deterioration conditions although funding illiquidity has a stronger effect on the formation of anti-herding behavior across the majority of hedge fund schemes analyzed. Interestingly, however, we observe a greater deal of heterogeneity across the different hedge fund categories, particularly during crisis periods, with certain hedge fund schemes including Convertible Arbitrage, Equity Market Neutral and Fixed Income Arbitrage experiencing herding driven by the COVID-19 induced market uncertainty. More importantly, we document significant economic implications of anti-herding and show that hedge funds associated with high degree of anti-herding earn significantly higher excess returns over those with low degree of anti-herding, particularly in the intermediate and long horizons up to one year. At the same time, hedge funds that anti-herd experience greater idiosyncratic volatility in subsequent periods, presenting a novel perspective to the relationship between anti-herding, idiosyncratic volatility and expected returns. While the finding of anti-herding in the hedge fund industry is not unexpected as the main attraction of hedge funds is to devise proprietary trading strategies that is based on private information, our findings provide novel insight to the link between idiosyncratic volatility and expected returns in the context of anti-herding in the hedge fund industry.

Keywords: Hedge Funds, Herding Behavior, Fundamental Information, Idiosyncratic Volatility

JEL Classification: G11, G14, G23