Selection Bias in Audit Firm Tenure Research

This study examines selection bias in audit firm tenure research. We first illustrate conceptually how selection bias can lead researchers to spuriously estimate a causal effect of audit firm tenure on audit quality when none exists. We then empirically demonstrate the existence of self-selection: longer versus shorter duration client-auditor pairs differ systematically in a variety of ways and these differences exist at the outset of relationships. After including client-audit firm relationship fixed effects to control for selection bias, our longitudinal analysis shows no evidence that audit quality changes with tenure over time within client-audit firm relationships. Overall, the combined evidence suggests that the previously documented association between audit firm tenure and audit quality is largely driven by cross-sectional, rather than longitudinal, variation, which is more consistent with a “selection” outcome rather than a “treatment” effect.