Abstract (EN)

Is earnings management affecting (driving) the measures of earnings conservatism? If discretionary accruals were on average negative, earnings management practices would be enhancing the association between bad news periods earnings and returns, whilst the opposite would occur in good news periods. Consequently, the asymmetry in news recognition in earnings would be driven not only by accountants conservative behaviour (attributable to litigation risk or other causes) but also by negative earnings management. In our study we calculate discretionary accruals and afterwards we use earnings minus discretionary accruals deflated by share price at the beginning of the period as the dependent variable in a Basu (1997) reverse regression to measure earnings conservatism. We expect the asymmetry in news recognition in earnings to be less pronounced when we use true earnings (earnings minus discretionary accruals) with respect to the observed earnings figure. This earnings management effect on earnings conservatism could be affecting the results of comparative studies (i.e. Ball et al. 2000) that analyse the differences in earnings conservatism across accounting regimes.