Publication | Open Access
Detecting Earnings Management
5.7K
Citations
0
References
1994
Year
Earnings ManagementAccountingAccounting PracticeTest StatisticsFinancial PerformanceExtreme Financial PerformanceBusinessFinancial StatementFinancial AccountingAccounting ProblemFinanceQuantitative Management
Models for detecting earnings management typically rely on discretionary accruals, yet no systematic comparison of their relative performance existed before this study. The study compares the relative performance of competing discretionary‑accrual models by evaluating the specification and power of commonly used test statistics. The authors assess test‑statistic performance by measuring type‑I error rates on random firm‑year samples and on samples of firms with extreme financial performance, including SEC‑targeted firms and artificially manipulated firms.
This paper evaluates alternative models for detecting earnings management. The paper restricts itself to models that assume the construct being managed is discretionary accruals, since such models are commonly used in the extant accounting literature. Existing models range from simple models in which discretionary accruals are measured as total accruals, to more sophisticated models that separate total accruals into a discretionary and a non-discretionary component. Prior to this paper, there had been no systematic evidence bearing on the relative performance of these alternative models at detecting earnings management. This paper evaluates the relative performance of the competing models by comparing the specification and power of commonly used test statistics across the measures of discretionary accruals generated by each model. The specification of the test statistics is evaluated by examining the frequency with which they generate type I errors for a random sample of firm-years and for samples of firm-years with extreme financial performance. We focus on samples with extreme financial performance because the stimuli investigated in previous research are frequently correlated with financial performance. The first sample of firms are targeted by the Securities and Exchange Commission for allegedly overstating annual earnings and the second sample is created by artificially introducing earnings management into a random sample of firms.