| Using a sample of 24,340 quarterly earnings announcements over 2004–2008, we find evidence that institutions discover the post-earnings announcement drift (PEAD). Institutions can earn a sixty-day mean abnormal return of 4.355% from their arbitrage trades. But, we don't find enough evidence that institutions take arbitrage use of PEAD. Because that, transient institutions'quarterly ownership changes are positively related to the contemporaneous quarter's earnings surprise and the earnings surprises of the previous three quarters, implying that transient institutions initiate and unwind their arbitrage trades over three quarters.We find that direct and indirect trading costs significantly reduce transient aggressiveness in exploiting PEAD, But don't change institutions'conduct.Consistent with prior research, we find that transient institutions are strong return momentum traders. Transient institutions'momentum-driven trades in a quarter are more than two times as large as PEAD-driven trades, suggesting that the institutions'dominant trading strategy is buy and hold for a short time.Though the institutions do not use the PEAD taking arbitrage, Transient institutions' trading behavior documented in this study do accelerate the markets'reaction to PEAD or ease the PEAD phenomena. |
PDF Full Text Request