5 Conclusion
I revisit the causes of positive returns around earnings announcements. While the literature provides evidence in support of both attention and information-based explanations, recent work focuses on the latter, raises questions about the former, and does little to reconcile the two. I provide evidence that attention explains a significant proportion of positive returns around earnings announcements using a setting that more cleanly distinguishes the effect of attention from that of new information.
I use earnings notifications, which are short announcements of the date and time of the upcoming earnings announcement and attract investor attention without providing much new information. I find positive returns, abnormal trading volumes, and abnormal EDGAR searches on earnings notification days, consistent with an attention effect. I also find a significant reduction in the level of positive returns around the earnings announcement in the presence of earnings notifications, suggesting that notifications attenuate investor attention to the earnings announcement. My results suggest the attention explanation dominates the information explanations after the earnings announcement but is dominated by the information explanations beforehand. This is consistent with attention increasing the EAP, in part, by mitigating the well-documented pattern of return reversals following the earnings announcement.
