The Information Content of CEOs' Personal Social Media: Evidence from Stock Returns and Earnings Surprise

Abstract

We study the personal Twitter accounts of 226 CEOs, and examine whether the content across 127,916 Tweets hold information about future firm performance. The content of these Tweets mostly deals with CEOs' personal activities and opinions, and the analysis further employs machine learning algorithms to identify and exclude Tweets that have relevance for firm operations. The results show that a high proportion of positive words in the CEOs' Tweets predict positive future abnormal returns. This pattern is especially strong before earnings release. Using Tweets posted before earnings announcements, a daily calendar-time spread portfolio that buys stocks with a high proportion of positive words and sells stocks with a low proportion delivers a return of 10.8 to 19.6 basis points per day. Furthermore, the cumulative proportion of positive words can predict quarterly earnings surprise, earnings and cash flows. These results suggest that the content of personal Tweets can elicit the moods of CEOs, and that executives tend to spend time doing enjoyable activities when they have achieved the performance targets.

Keywords: Information Content, Social Media, CEOs' Tweets

JEL Classification: G14, G34