Abstract

Authors: Emiliano Catan, Marcel Kahan

For decades, scholars have turned to empirical evidence to resolve theoretical debates about the impact of corporate governance on firm value. Initially, the most common empirical technique to evaluate corporate governance was the short-term event study. More recently, short-term event studies have been supplemented, if not replaced, by three techniques – longer-term event studies, calendar time portfolio regressions, and Q (and other accounting-based ratio) regressions – that examine the effect of corporate governance changes over a longer term. We argue that one should apply a fair amount of skepticism in evaluating empirical studies that purport to discern the effect of corporate governance on firm value. Q regressions are theoretically unfounded and can often lead to biased results. Results derived from standard Q regressions thus shed no light on corporate governance controversies. While event studies and calendar time portfolio regressions are greatly superior to Q regressions, these methodologies have their own shortcomings, which need to be kept in mind in interpreting their results.