Negativity Bias, Social Media, and Analyst Forecast Performance *

Abstract

This paper examines whether sell-side equity analysts exhibit negativity bias and process social media information asymmetrically. We find that analysts are more responsive to negative firm-specific tweets and use this information to improve their earnings forecasts. Analysts who cover firms with greater sensitivity to negative Twitter information, issue relatively less optimistic and more accurate earnings forecasts, especially for opaque firms. Using an exogenous event that changed the information content of individual tweets and a difference-indifferences design, we establish a causal link between Twitter information and analyst forecast performance. Collectively, these findings suggest that financial markets process negative information more efficiently.

Keywords: social media, Twitter sentiment, earnings forecasts, sell-side analysts, Negativity bias

JEL Classification: G14, G24.