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.

ABSTRACT We study nonofficer directors’ influence on the accounting conservatism of U.S. public firms. Between 1986 and 2002, all 50 U.S. states enacted laws that limited nonofficer directors’ litigation risk but often left officer directors’ litigation risk unchanged. We find that conditional conservatism decreased after the staggered enactments of the laws, which we attribute to less nonofficer director monitoring of financial reporting in affected firms. Conservatism fell less when shareholder or debtholder power was high, consistent with major stakeholders moderating the influence of nonofficer directors. We verify that our results stem from reductions in the asymmetric timeliness of accruals and, specifically, its current assets components. We also show that affected firms switched away from Big N auditors more often, which reduced these firms’ commitment to conservative financial reports.