Downside Risk and Long-Horizon Stock Return Reversals

Abstract

We argue that long-horizon return reversals [Debondt and Thaler (1985)] reflect a premium for downside risk. Consistent with this, we find that downside betas of past losers are significantly greater than downside betas of past winners, and the inclusion of downside beta in Fama-Macbeth regressions subsumes the reversal effect. We note that downside risk offers a theoretical justification for the “distress risk” explanation for long-horizon return reversals of Fama and French (1996). Consistent with this view, we find that downside beta is more highly correlated with firm characteristics associated with distress (dividend reductions and delisting) and explains long-horizon return reversals better than SMB/HML and other proxies for distress risk in the literature.

Keywords: downside risk, stock reversals, downside beta, contrarian effect, market efficiency

JEL Classification: C13, G12