Information Misweighting and Stock Recommendations

I provide evidence that analysts whose earnings forecast revisions showed signs of greater exaggeration in the past make recommendation changes that lead to lower abnormal returns than their peers. Interpreting stock recommendations as a forecast of future abnormal returns, I show that this evidence is consistent with the hypothesis that analysts who typically exaggerate or overstate the weight of their private information when issuing forecasts also do so when making recommendations. I also show that past earnings forecast provide incremental information about analysts' recommending behavior beyond that contained in past recommendations.