A rapidly growing literature claims to refute the efficient-market hypothesis by producing large estimates of long-term abnormal returns following major corporate events. The preferred methodology in this literature is to calculate average multiyear buy-and-hold abnormal returns and conduct inferences via a bootstrapping procedure.
This article reexamines the reliability of recent long-term stock price performance estimates using three large well-explored samples of major corporate events. We find that the popular approach of measuring long-term abnormal performance is not an adequate methodology because it assumes independence of multiyear event-firm abnormal returns. We show that event-firm abnormal returns are positively cross-correlated when overlapping in calendar time.
As such, assuming independence is problematic for any long-term abnormal performance methodology. Moreover, this is likely to be a problem for most event samples, not just the mergers, seasoned equity offerings and share repurchases examined in this article, since major corporate actions are not random. As a result, we strongly advocate a methodology that accounts for the dependence of event-firm abnormal returns, such as the calendar-time portfolio approach.
The primary implication of our results is that most of the evidence against market efficiency contained in recent studies measuring significant long-term abnormal returns following major corporate events is largely irrelevant because these studies assume independence. Our estimates of long-term abnormal performance that account for the positive cross-correlations of event-firm abnormal returns produce very little evidence of long-term abnormal performance.
Merton Miller Prize 2000