Get access

Does the Stock Market See a Zero or Small Positive Earnings Surprise as a Red Flag?


  • An earlier version of the paper was presented at 2008 American Accounting Association Annual meeting. We thank the participants for valuable comments. We also thank Merle Erickson (editor), an anonymous referee, Shen Jianfeng, Rick Johnston, Inmoo Lee, Mian Mujtaba, He Wen, Jenny Wenjun Zhang, and other seminar participants at the National University of Singapore for their valuable comments. We gratefully acknowledge the contribution of Thomson Financial for providing earnings per share and earnings forecast data, available through the Institutional Brokers Estimate System (I/B/E/S). These data are provided as part of a broad program to encourage academic research.


This study shows that firms collectively incur a cost for managing earnings and analyst expectations to meet earnings forecasts. We compare the coefficient in the regression of abnormal stock returns on earnings surprise (the earnings response coefficient [ERC]) across ranges of earnings surprises. The ERC for earnings surprises in the range [0, 1¢] is significantly lower than ERCs for earnings surprises in adjacent ranges for firm-quarters in the early and mid 2000s, but not for those in the 1990s. The results are robust to controlling for the sign of estimated discretionary accruals and the trajectory of analyst earnings forecasts. We further find that investors are right to be skeptical about earnings surprises in the range [0, 1¢]. The relation of future earnings surprise with current earnings surprise is more negative for current earnings surprises in that range than for those in any other range. Evidence also suggests analysts react negatively to earnings surprises in that range.

Get access to the full text of this article