How much “voice” should shareholders have in a firm? How valuable are shareholder’s views on overall firm performance? Are non-binding shareholder votes useful to aggregate shareholder’s information? These are important current questions corporate governance that, more broadly, speak to the value of direct democracy mechanisms in modern corporations.
One important channel through which shareholders can voice their views are Say-on-Pay votes asking for the shareholders’ approval of executive. Since shareholders are not merely voting on the level of pay but whether that level reflects the value that the CEO adds to the firm, Say-o-Pay is an explicit vote of confidence which aggregates the opinions of all shareholders into a simple, highly visible metric.
In Say Pays! Shareholder Voice and Firm Performance, Vicente Cuñat, Maria Guadalupe and Mireia Gine explore the effect on shareholder value, firm performance and CEO compensation of Say-on-Pay votes.
Adopting Say-on-Pay is correlated with multiple firm attributes that also affect performance and hence is highly endogenous. To overcome this problem the paper uses a regression discontinuity design on shareholder votes to implement Say-on-Pay between 2006 and 2010. The intuition of the identification strategy is that the characteristics of firms are similar for those that closely pass or reject a vote – yet this small difference will have an impact on the probability of Say-on-Pay being implemented.
The results show that Say-on-Pay significantly increases shareholder value. On the day of the vote, a Say-on-Pay proposal that passes yields an abnormal stock market return of 1.8% to 2.7% relative to one that fails. Since the outcome of the vote is not binding, the market reaction should only account for the increase in the probability that the proposal will be implemented after a positive shareholder vote (which is around 40% to 50%). This implies that Say-on-Pay delivers an average increase in shareholder value of about 5%.
The paper also shows that Say-on-Pay has a positive impact on firms’ accounting and operational performance in the years following the vote. However, there is no systematic change in the level or structure of CEO compensation, or the probability of the CEO leaving the firm after a positive Say-on-Pay vote. While there are significant changes in the composition of pay, these are not consistent across measures or over time, although the lack of an average effect on the level or structure of compensation may mask the possibility that different firms adjust compensation along different (and maybe opposing) dimensions.
Overall, the paper suggests that Say-on-Pay serves to monitor and incentivize CEOs to deliver better firm performance by providing a clear mechanism for shareholders to voice their opinions, as confirmed by major improvements in shareholder value and firm performance. This improvement in performance cannot be ascribed to a particular change in compensation or firm policies common to all firms, implying that “one size does not fit all” and that optimal responses may vary from one firm to another.