June 24, 2014 | Research Feature

Has “Say on Pay” Worked?

Shareholder voting on executive compensation has done little to curb CEO pay — but has brought transparency to governance.


The rise of dot-coms in the late 1990s pushed CEO compensation up so high that just before that industry’s bubble burst in 2000 the average CEO-to-worker pay ratio in the United States, across all industries,was 400 to 1. For context, consider that in 1965 the ratio was just 18 to 1. Then, in 2001, the Enron and WorldCom accounting and governance scandals came to light. The increasingly popular practice of offering stock options as compensation to upper management created perverse incentives for executives to doctor financial reports in hopes of inflating stock prices. Stock options also played a key role in the compensation packages of dot-com executives and may have influenced valuations of those firms.

As a result, large institutional investors (such as pension funds) pushed for shareholder votes on CEO compensation policies, and many nations, including the United States and the UK, have mandated some form of so-called Say-on-Pay (SOP) practices. While such SOP votes are nonbinding, campaigns to implement the measures often brought unwanted scrutiny to targeted firms and their boards.

How successful have these nonbinding votes been in curbing CEO pay, and what are the other economic consequences of SOP for firms? Professor Fabrizio Ferri looked at the last decade and a half of data and research on SOP in the United States and the UK to learn how effective such measures have been.

Across countries and firms, none of the data or studies Ferri analyzed showed that SOP votes had much effect on reigning in CEO compensation levels. However, they did affect pay for performance sensitivity: He found that at firms with negative SOP votes — where a majority of shareholders voted against a board’s executive compensation proposal — CEOs did face a greater penalty for poor performance than CEOs at other firms (perhaps unsurprisingly, because many SOP votes have been motivated by shareholders’ requests to link pay to performance, especially on the downside). Ferri also found that many boards sat down with institutional investors in the lead up to SOP votes to negotiate key policies such as severance contracts and equity grants, and that almost all boards changed compensation contracts after negative SOP votes. So while SOP doesn’t appear to lower CEO compensation on average (except in cases of poor performance), it does prompt boards to be more involved and transparent with shareholders in the compensation policy process.

To determine what happens to firm valuation when SOP is on the table, Ferri analyzed event studies around SOP legislation, which, for example, consider the market reaction after a government announces a proposal to make SOP votes mandatory. “If the market reacts positively to this and similar announcements, it is consistent with the idea that it expects Say on Pay to create value overall by reducing excess compensation and providing better incentives, which should result in a higher value,” Ferri explains. “Or, if the market thinks that shareholder involvement in setting executive pay is just going to create distraction and be destructive, that should result in a lower value.” The results were mixed but suggest that the overall effect is probably positive, albeit modest — there’s some evidence of positive market reactions, much evidence of neutral reactions, and no evidence of negative, value-destroying reactions.

Ferri notes that SOP votes against board-proposed compensation packages are actually quite rare, although they make for good press and tend to get outsized attention. “That suggests that institutional investors feel that compensation packages overall are OK and there are a few instances when it’s important to raise their voice,” Ferri notes. “And in those cases very often the compensation is a symbol of governance more generally. Shareholders don’t necessarily want to micro-manage compensation contracts, but they do want to play a role in how policies are shaped.”

While there are some differences between US and UK in terms of regulations and relationships between firms and institutional investors, Ferri found that the effect of SOP has been similar in both countries: little effect on pay levels, more pay-for-poor-performance sensitivity, and the elimination of controversial features of compensation contracts after rare negative votes. Because other nations’ SOP mandates share key features of those in the US and UK, it’s likely that they will experience similar outcomes.

Overall, says Ferri, SOP works, if not exactly in the way its architects intended. “SOP per se is just a tool. It worked, in the sense that boards heard what shareholders had to say on pay. But by and large, shareholders had little to say on compensation levels and more to say on specific contractual features viewed as affecting the relation between pay and performance. Hence, overall SOP has had no significant effect on the level of executive compensation but has resulted in greater use of certain features, such as performance-based vesting of equity grants, and the elimination of other features such as excise tax gross ups,” he says. “Even when it has not affected compensation contracts, it has led to better dialogue with shareholders, and increased board transparency over the pay-setting process.”

Fabrizio Ferri is associate professor of accounting at Columbia Business School.

Read the Research

Fabrizio Ferri

"Say on Pay"

View abstract/citation  Download PDF  

Have You Read Columbia Business?

Columbia Business School’s alumni magazine connects alumni with each other and the School; celebrates alumni milestones and accomplishments; and chronicles the impact of Columbia Business School alumni, faculty members, and students on the global business landscape.

Read Columbia Business>