Footnote:
Nach Informationen von SSRN wurde die ursprüngliche Fassung des Dokuments July 11, 2023 erstellt
Description:
We propose a novel discretionary accruals (DA) measure in the unexplored dimension of non-current accruals. We show that our measure (DA-NC) can improve the forecasting performance of fraud prediction models that rely on conventional current accruals-based DA measures. Analysing accounting frauds in a sample of Accounting and Auditing Enforcement Releases (AAERs), we find that DA-NC provides a good advanced indicator of fraud. However, among the existing DA measures, only accrual estimation error (Dechow and Dichev, 2002) provides an early indication of frauds. While the accrual estimation error measure predicts frauds based on current account misstatements, only DA-NC predicts frauds based on non-current account misstatements. Our results explain the limited success in predicting accounting frauds using existing current accruals-based DA measures. We recommend that a comprehensive predictive model of accounting frauds should incorporate both current and non-current accruals-based DA measures