Jump on the Post–Earnings Announcement Drift (corrected)

The authors examined the potential profitability of a strategy that exploits the post–earnings announcement drifts contingent on jump dynamics identified in stock prices around earnings announcements. With long positions in positive-jump stocks and short positions in negative-jump stocks, their hedge portfolio achieved an annualized abnormal return of 15.3% and an annualized Sharpe ratio of 1.52 over the last four decades. Neither conventional risk factors nor common company characteristics explain the abnormal return.The post–earnings announcement drift (PEAD), or earnings momentum, is one of the most robust and persistent anomalies challenging the efficient market paradigm. Previous studies have proposed various trading signals to measure and profit from the surprise in an earnings announcement. In our study, we introduced a new measure of company-level informational shocks based on the realized jump dynamics of stock prices over a three-day window around earnings announcements, and we examined the profitability of an earnings momentum strategy that uses extreme price movements, or jumps, as trading signals.The intuition is that companies’ unobserved “extremely good news” is impounded in unexpectedly large and discrete price hikes, or positive jumps, around earnings announcements, whereas companies’ unobserved “extremely bad news” is reflected in large and abrupt price plunges, or negative jumps, around earnings announcements. Therefore, we formed a hedge portfolio that took long positions in positive-jump companies and short positions in negative-jump companies. Using a particular jump detection method, we found compelling evidence of post–earnings announcement return drifts in the same direction as jumps over the subsequent three months. This strategy yielded a quarterly excess return of 3.63%, equivalent to a 15.3% annualized return, over the sample period of 1971–2009. Our results suggest yet another anomaly associated with PEAD. Market participants seem largely to underreact to, or are simply unaware of, the latent “extremely good (or bad) news” signaled by the direction of jumps around earnings announcements.We conducted an array of tests to show that the jump signal is distinct from two commonly used PEAD trading signals in previous studies: the earnings announcement return and standardized unexpected earnings. We also found evidence that neither common company characteristics (e.g., size, illiquidity, and book-to-market ratio) nor the three Fama–French risk factors, augmented by a momentum factor and a liquidity factor, can explain the abnormal returns.