Earnings Quality and Future Returns: The Relation between Accruals and the Probability of Earnings Manipulation
Messod Daniel Beneish
Indiana University Bloomington - Department of Accounting
D. Craig Nichols
Cornell University - Samuel Curtis Johnson Graduate School of Management
May 17, 2005
The paper examines the relation between the probability of manipulation, accruals, and future returns. We show that firms that have a high likelihood of earnings manipulation (as measured by the Beneish (1999)'s M-Score) experience lower future earnings, but that investors expect these firms to have higher future earnings. Indeed, we find that investors overestimate next-period return on assets by 490 to 690 basis points (this is significant as the median ROA in the sample 4.6%). We also show that the probability of manipulation is a correlated omitted variable for the earnings forecasting models used in prior research on accrual mispricing and that including the probability of manipulation greatly attenuates the mispricing of accrual persistence. Finally, we show that the probability of earnings manipulation predicts economically significant abnormal returns of approximately 15% per year after controlling for accruals and various controls for risk factors, including a factor compensating for earnings quality differences (Easley and O'Hara (2004), Francis et al. (2005)). We interpret our results that the predictive ability of accruals for returns is greatly diminished in the presence of the M-Score as indicating that accrual mispricing arises because investors are misled by managers' opportunistic management of earnings.
Number of Pages in PDF File: 60
Keywords: Earnings manipulation, accrual mispricing, future returns
JEL Classification: M41, M43, G12, G14working papers series
Date posted: May 19, 2005
© 2013 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo3 in 0.593 seconds