Do News Releases Help Investors Disaggregate Earnings? An Examination of Information Lost Through Earnings Aggregation

Abstract

This paper examines how news items released throughout the previous quarter help investors predict and understand earnings. It focuses on how earnings aggregates transactions and how this aggregation may lead to information loss. Through the disclosure of these news items that reveal these underlying transactions, the market may be able to predict future earnings surprises more accurately. Using a novel database that allows identification of these news-items for firms, I find that these news-items decrease the reaction to earnings surprise in the earnings announcement window (-1, +1) and decrease post-earnings announcement drift (+2, +60). These results suggest that transactional-level news may help investors understand earnings through disaggregation. Moreover, this paper compares the effect of these news releases to that of Form 8-Ks and 10- Qs/10-Ks. While Form 8-Ks similarly reduce investors’ reaction to earnings in the short-term window, they are ineffective in reducing the mispricing over the longer window.