Abstract:
Despite being a highly documented phenomenon, the underlying cause of the post-earnings announcement drift is presently unknown. This study tests whether the tendency for investors to hold losses and realize gains--known as the disposition effect--can partially explain the existence of the post-earnings announcement drift anomaly in equity markets. The data for this paper consists of over six thousand companies from 1992 to 2012 for a total of over 140,000 quarterly earnings announcements. This study finds that the post-earnings announcement drift was most severe before 1998. During this period (1992-1998), a trading strategy that purchased (sold) stocks in the highest (lowest) decile of earnings surprises would have earned an annualized abnormal return of 32% (versus 4% for the later period). Furthermore, results show that the post-earnings announcement drift does not occur when the disposition effect is weakest, implying that the disposition effect plays an important role in the drift.