Do CEO Pay Cuts Really Work?

Boards sometimes cut a CEO’s pay following poor performance. This study examines whether such CEO paycuts really work. We identify 1,496 instances of large CEO paycuts during the period 1994–2013. We then create a propensity-score-matched control group of firms that did not cut their CEOs’ pay and employ a difference-in-differences approach to examine the consequences of paycuts. Our results show that, following a paycut, CEOs are likely to engage in earnings management in an attempt to accelerate improvement in the reported performance and to achieve a speedier restoration of their pay to pre-cut levels. Further, we find that improvement in long-term performance after a paycut occurs only for those firms with lower levels of earnings management after the paycut. Finally, we show that paycuts are more likely to lead to unintended value-destroying consequences in the absence of high institutional ownership or when the CEO is sufficiently entrenched, thereby impairing the effectiveness of internal monitoring by boards.