Corporate Governance and Firm Value

Scholars have long debated how the governance structure of a firm affects the value of the firm. Confronted with conflicting theoretical arguments, corporate governance scholars turned to empirical evidence. An approach that has become increasingly popular over the past few decades to assess the effect of governance changes over longer time horizons is to measure the effect of these changes on the firm’s Tobin’s Q.  As we show, Q regressions are theoretically unfounded and can lead to biased results. In a within-firm design, Q regressions are inferior to event studies and calendar time portfolio regressions: they suffer from all of their limitations, but are less good in adjusting for market movements, add noise, and pick up treatment effects unrelated to firm value. In a cross-sectional setting, results derived from standard Q regressions shed little, if any, light on corporate governance controversies.