When Two Anomalies Meet: The Post–Earnings Announcement Drift and the Value–Glamour Anomaly

This study of the post–earnings announcement drift and the value–glamour anomaly finds that value stocks have greater information uncertainty, exhibit more-muted initial market reactions to earnings surprises, and have better (more positive or less negative) post–earnings announcement drifts than do glamour stocks. A trading strategy based on these findings can generate an average annual abnormal return of 16.6–18.8 percent before transaction costs.For decades, the post–earnings announcement drift and the value–glamour anomaly have fascinated both academics and practitioners in the field of finance. The goal of our study was to link these two anomalies directly by studying the different initial market reactions to earnings announcements and the drift patterns of various value and glamour portfolios and to design a new trading strategy based on the signs of the earnings surprise (+/–/0) and the instant stock price reaction to the earnings news (+/–).We built our study on prior research that explored behavioral and rational explanations for the existence of these two anomalies. In particular, we investigated how information uncertainty may lead to different initial and subsequent reactions to earnings news. We developed several testable predictions. We first predicted that value stocks have greater information uncertainty than glamour stocks. We then predicted that, owing to greater information uncertainty, value stocks have more-muted initial reactions to earnings surprises than do glamour stocks. Finally, we predicted that when earnings news is good, value stocks have better (more positive) post-announcement drifts than glamour stocks; but when earnings news is bad, whether value stocks have better (less negative) drifts than glamour stocks owing to the two opposite effects—the risk premium effect and the delayed reaction effect—is unknown ex ante.To test our predictions, we used data for 1984–2008 from Capital IQ Compustat, CRSP, and I/B/E/S. We sorted the companies into five value–glamour quintiles for each quarter contingent on the sign of the earnings surprise (defined by the difference between the actual earnings and the analyst consensus forecast) and the sign of the instant stock price reaction to the earnings news (measured by the abnormal return over a three-day window around the quarterly earnings announcement—the earnings announcement abnormal return, or EAAR).We made a number of new findings. First, we observed that value stocks have higher information uncertainty than glamour stocks. Second, we found evidence that value stocks have more-muted initial reactions to earnings surprises than do glamour stocks. When three-day EAARs are positive, value stocks have lower (less positive) EAARs than glamour stocks. When EAARs are negative, value stocks have higher (less negative) EAARs than glamour stocks. Third, consistent with our understanding of the delayed reaction effect and the risk premium effect, when earnings news is good, value stocks have better (more positive) drifts than glamour stocks. Fourth, our empirical results suggest, ex post, that when earnings news is bad, value stocks still have better (less negative) drifts than glamour stocks. For value stocks, the risk premium effect induced by high information uncertainty signals dominates the delayed reaction effect arising from the uncertainty. Finally, when EAARs and earnings surprises move in opposite directions, the drift patterns are mixed and smaller in magnitude for both value and glamour stocks.A trading strategy of taking a long position in value stocks when both earnings surprises and EAARs are positive and a short position in glamour stocks when both are negative can generate annual returns (before transaction costs) of 16.6–18.8 percent. This anomaly is mainly a long-side phenomenon; preventing investors from short selling glamour stocks will not prevent them from earning a value premium.