The Effect of Firm Performance on Modeling Discretionary Accruals: An Evaluation of Accrual Models

Abstract

Previous studies detected earnings management behaviour through various methods. Since the middle 1980s, discretionary accruals have become the primarily focus in detecting earnings management. While the Jones and Modified Jones models attempt to control for contemporaneous performance, empirical assessments of these models suggest that estimated discretionary accruals are significantly influenced by a firm's contemporaneous and past performance (Kothari, 2005). In this paper, we formally derived the relation between firm performance and accruals. We show that the evolvement of different models and demonstrate why firm performance should be controlled when estimating discretionary accruals. Using a sample of ASX listed firms with 5,947 firm-year observations from the period of 1999 to 2006, we estimate discretionary accruals based on Jones Model, Modified Jones Model, Cash Flow Modified Jones Model and Performance Adjusted Technique. The results show that Performance Adjusted Technique tends to adjust the effect of performance on estimated discretionary accruals by removing the measurement in discretionary accruals that correlated with earnings performance and therefore improve the reliability of further detection of earnings management.