Anomalies, Management Expectations, and Stock Returns

Abstract

We investigate the extent to which managers incorporate public information in their earnings expectations and its implications for capital market efficiency. We find that management earnings forecasts are biased upward (downward) for stocks that are overvalued (undervalued) based on anomaly signals. On average, these biases are more severe than those in consensus analyst forecasts. Anomaly returns are higher (lower) when managers issue forecasts that update the market in the direction that is consistent (inconsistent) with the anomaly information. This return difference is about two percent in the month after anomaly portfolio formation and persists for over 12 months. Consistent with managers' biases exacerbating anomaly-related mispricing, the return difference is more salient during high sentiment periods and when there are firm news events.