Modeling Earnings Discontinuities: A Maximum Likelihood Approach

Abstract

We develop new distribution discontinuity tests for detecting earnings management and analyzing its determinants. We embed Burgstahler and Dichev’s (1997) intuition on benchmark-driven earnings management in a likelihood-based model that addresses important limitations of the existing distribution discontinuity tests. Our method offers large improvements in test performance relative to both histogram-based tests of the existence of earnings discontinuity and logit-based tests of the determinants of earnings discontinuity, and it changes some of the major findings in the earnings discontinuity literature. Future research on distribution discontinuities could benefit from adopting our likelihood-based tests.