Divergent Opinions on Social Media

This study analyzes the informational value of tweets in which opinions diverge from the consensus. I identify them using each firm's most positive and negative intraday social media sentiments. I find that these divergent opinions—specifically, negative ones—predict stock returns without subsequent reversals. Additionally, they contain incremental information on firm fundamentals identified by subsequent revisions to analysts' target prices and earnings forecasts. Finally, I find that return predictability is attributed to the fundamental information contained in the divergent opinions. My analysis sheds light on the role of divergent opinions on social media.