Uncertainty Resolution Before Earnings Announcements

Abstract

We document that 72% of the earnings announcement premium is realized before, rather than after, earnings releases. It is possible that large pre-announcement returns are caused by uncertainty resolution before the announcement. To test the uncertainty resolution hypothesis, we examine the pattern of pre-announcement returns in a cross-section of stocks. There is compelling empirical evidence to support the hypothesis: pre-announcement returns are significantly higher for firms with high uncertainty, and when the aggregate market uncertainty is high. Finally, we present two distinct channels for uncertainty resolution: information acquisition by investors and information supply by analysts.