“Accruals” Discricionários: o Erro de Estimação Induzido pelo Conservantismo

This paper discusses the sign of the expected measurement error in discretionary accruals (DAC) estimates when accrual models do not control for the asymmetric treatment of gains and losses underlying conservatism. I find that DAC in firms with “bad news” are expected to be understated (positive measurement error), while those in “good news” firms will be overstated (negative measurement error). Based on this original result and using graphical analysis I discuss an empirical illustration, which corroborates the expectations.