摘要：This article had as an objective to confi rm if deferred tax income analysis increases signifi cantly the detection of earnings management. Data were collected for the mining, steelworks, metallurgy and textile industries for the period 2000-2004, through probit and logit analysis established on the Modifi ed Jones and Kang & Sivaramakrishnan models. The paper found evidences that analysis of deferred tax income does not increment the predictive power of the analyzed models. The divergence can be explained because most of the literature on the subject dealt with American fi rms and industries that supposedly have different incentive factors for earnings management. Further, the differences in Brazilian firms can be originated by the fl exible use of the partial method for recognition of tax deferral.
其他摘要：This article had as an objective to confi rm if deferred tax income analysis increases signifi cantly the detection of earnings management. Data were collected for the mining, steelworks, metallurgy and textile industries for the period 2000-2004, through probit and logit analysis establishedon the Modifi ed Jones and Kang & Sivaramakrishnan models. The paper found evidences that analysis of deferred tax income does not increment the predictive power of the analyzed models. The divergence can be explained because most of the literature on the subject dealt with American fi rms and industries that supposedly have different incentive factors for earnings management. Further, the differences in Brazilian firms can be originated by the fl exible use of the partial method for recognition of tax deferral.
