Modeling the Determinants of Meet-or-Just-Beat Behavior in Distribution Discontinuity Tests

Abstract

We develop new distribution discontinuity tests conditional on multiple explanatory variables for analyzing meet-or-just-beat behavior around benchmarks. These tests combine Burgstahler and Dichev’s (1997) meet-or-just-beat intuition with a flexible statistical model that addresses important limitations of the existing tests. Our method considerably outperforms logit-based tests of distribution discontinuity determinants and changes the interpretation of a major finding in the earnings discontinuity literature. As a secondary benefit, it also has slightly higher statistical power than histogram-based tests of distribution discontinuity existence. Our method is robust, easy to implement using our publicly available Stata command, and could benefit researchers in many fields.

Keywords: standardized difference test; performance benchmark; bright-line rule; smooth distribution; non-linear interpolation; conditional distribution

JEL Classification: M41; C20; C25