Detecting GAAP violation: implications for assessing earnings management among firms with extreme financial performance

Abstract

My paper presents a model to detect earnings management among firms experiencing extreme financial performance, and compares the model's performance to that of discretionary accrual models. I found that the model provides timely assessments of the likelihood of manipulation, and that model-based trading strategies earn significant abnormal returns. I also present evidence suggesting that the specification of discretionary accrual models could be enhanced by adding lagged total accruals and a measure of past price performance as explanators. The evidence arises from studying actual instances of earnings management. Its implications are in line with the Guay et al. (1996, p. 104) conjecture that accrual models which take into account managers' incentives, and recognize that discretionary accruals reverse, have a better chance of identifying discretionary accruals. The results have implications for researchers investigating managers' accrual decisions in contexts such as security offerings and financial distress, where extreme performance limits the usefulness of accrual models.