Do Investors Under‐React to Information in Analysts' Earnings Forecasts?

  In this paper, we highlight important consequences of the choice of deflator for conclusions regarding apparent patterns in stock returns. In particular, we focus on the use of share price as a deflator. Previous research has documented that there are patterns in stock returns related to the magnitude of share price, and consequently this relation has the potential to confound inferences about information variables that are scaled by share price. We show analytically the potentially misleading effects of using share price as a deflator in abnormal returns tests and identify several studies in the literature that have deflated candidate anomaly variables by price. We then select one previous finding—evidence of abnormal returns to a strategy based on share‐price‐scaled analysts' forecasts—as an exemplar of the problem. Taken as a whole, our findings indicate that after controlling for the systematic negative relation between share price and subsequent abnormal return there is no apparent mis‐weighting by investors of the information in analysts' earnings forecasts or abnormal returns to an investment strategy based on analysts' forecasts. These findings suggest that caution is warranted in the selection of deflators in anomaly studies and in the interpretation of findings of abnormal returns to share‐price‐scaled information variables.