Analyst Information Precision and Small Earnings Surprises

Prior research attributes zero and small positive earnings surprises to managers’ incentives for earnings management. In contrast, this study introduces and empirically tests an explanation for zero and small positive earnings surprises based on predictable variation in analyst forecast errors. We argue that analysts trade off competing incentives for forecast optimism and pessimism and they are more (less) likely to issue optimistic (pessimistic) forecasts when their information is less precise. As a result, a consensus forecast contains relatively more optimistic forecasts when analysts face greater earnings forecast uncertainty. Holding constant managers’ reporting incentives, we hypothesize and find that earnings forecast uncertainty is significantly negatively related to the likelihood that a firm reports a small positive instead of a small negative earnings surprise. Similarly, we find that earnings forecast uncertainty is strongly related to zero and small positive earnings surprise incidence. Our results have important implications for studies that categorize firms as "suspect" based on the sign and magnitude of earnings surprises. We highlight the importance of empirically controlling for earnings forecast uncertainty in studies that employ meet/just beat observations to capture constructs related to earnings management.