Abstract:

This study discusses the impact of financing changes (banks, shareholders loans, equity increases and other equity instruments) on accruals models, and discretionary accruals estimates. It pursues a threefold objective. Firstly, to show analytically how the occurrence of such changes affects discretionary accruals estimation. Secondly, to analyse empirically whether different accruals models - Jones (1991), Dechow and Dichev (2002), and McNichols (2002) - reflect in a similar way the impact of changes in corporate financing. Thirdly, compare for the Portuguese context the efficiency of the proposed methodology to Shan, Taylor and Walter (2013), in order to assess the relative performance of each one. Empirical evidence shows that the measurement error induced by not well-specified accruals models is affected by the sign of financing changes, being different for positive and negative changes; all models reflect in a similar way the impact of changes in corporate financing; and for the Portuguese context, the matched-firm approach on financing changes, intended to mitigate the problem caused by such changes on discretionary accruals, does not work well.