THE INFORMATION CONTENT OF ANALYSTS' RECOMMENDATIONS REVISITED

Bradley et al. (BCLO, 2014) find evidence that the time stamps reported in I/B/E/S for analysts' recommendations are systematically delayed giving the appearance that recommendations are uninformative. We review the findings of BCLO and extend their analyses along three dimensions. First, we show that time stamp delays are less likely to be associated with all-star analysts, affiliated analysts, and analysts from high reputation banks, but are more likely from independent analysts. Second, we show that recommendations from all-star analysts, analysts working for high reputation banks, and analysts who issued a previous influential recommendation are more likely to be influential. Finally, we examine post-recommendation drift following influential revisions and find post-revision returns of 18(-44) basis points for upgrades (downgrades) over the 2.5 hours following the revision.