| Summary: | In this paper we carry out cointegration analyses, in order to study whether the relationship between analysts’ recommendations and their projected capital gains (or losses), is consistent with the hypothesis that sell recommendations are costlier than buy recommendations. We find that recommendations that plainly urge the investor to take action (buy, sell) are consistent with their estimated losses. We also find that recommendations react mildly to higher projected losses, and strongly to higher projected capital gains, which is consistent with systematic optimism. Additionally, we could establish that higher projected losses are positively related to dispersion in recommendations. In summary, we got evidence consistent with Womack’s (1996) hypothesis that the cost of issuing a sell recommendation is higher than the cost of a buy recommendation.
|
|---|
