Twitter Information, Analyst Behavior, and Market Efficiency

Abstract

Using Bloomberg's daily Twitter Sentiment data for S&P500 firms, we show that Twitter information reduces forecast optimism and improves forecast accuracy of sell-side equity analysts. Negative Twitter information is more influential, and this effect is distinct from the impact of news. Using two exogenous events that changed the information content of individual tweets, we establish a causal relation between Twitter information and analyst behavior. At the aggregate level, Twitter-sensitive firms have smaller earnings surprises and weaker stock market reaction. Collectively, these results suggest that analysts extract useful negative information from Twitter, improving their forecasting performance and overall market efficiency. * We appreciate comments and suggestions from Ran Duchin and Christian Julliard. We are responsible for all remaining errors and omissions. Abstract: Using Bloomberg's daily Twitter Sentiment data for S&P500 firms, we show that Twitter information reduces forecast optimism and improves forecast accuracy of sell-side equity analysts. Negative Twitter information is more influential, and this effect is distinct from the impact of news. Using two exogenous events that changed the information content of individual tweets, we establish a causal relation between Twitter information and analyst behavior. At the aggregate level, Twitter-sensitive firms have smaller earnings surprises and weaker stock market reaction. Collectively, these results suggest that analysts extract useful negative information from Twitter, improving their forecasting performance and overall market efficiency.