Relative Timeliness of Good vs. Bad News Inferred from Stock Returns

Abstract

There are mixed views about whether firm managers voluntarily disclose good news in a more timely fashion than they do bad news. Our study investigates this issue by inferring managers' strategic disclosure behavior from firms' stock returns in the earnings announcement vs. non-announcement windows in the recent time period (1996-2005). We find that large firms' non-announcement excess returns are significantly lower than their announcement returns and vice versa for a typical small firm. Because excess return captures news arrived during the measurement window, our results suggest that for large firms bad news is relatively more timely than good news and for small firms good news is relatively more timely than bad news. If private information search is as likely to uncover good news as it is to uncover bad news, our evidence suggests that large firms preempt bad news, whereas small firms preempt good news. Furthermore, we find the relative timeliness of large firms' bad news only in the recent period, not in an early time period, suggesting that preemptive bad-news disclosure of large firms is a recent phenomenon.