Swarm intelligence? Stock opinions of the crowd and stock returns

We find that crowds’ analyses of stocks, disclosed on a social investment platform, provide explanatory power for stock returns. Exploiting a novel dataset that contains more than 14.9 million individual stock assessments for 10,452 stocks over the period from August 1, 2007, to July 15, 2015, our study shows that social networks may add valuable information for explaining future and abnormal stock returns. We find that a portfolio based on social media opinions yields a monthly excess return of 3.3%. We provide a theoretical rationale for our findings based on the argument that the platform is subject to fewer institutional restrictions and is designed more efficiently for prediction than financial markets.