When Robert Nardelli lost his job as CEO of Home Depot in January 2007, he walked away with cash, stock options and other benefits valued at $210 million — and left a trail of angry shareholders in his wake. Nardelli is one of many CEOs facing increased scrutiny as investors question ballooning executive compensation. “Shareholders today are concerned because executive compensation doesn’t seem to reflect underlying performance,” says Professor Sudhakar Balachandran.
One idea asserted repeatedly is that giving shareholders more power in determining a CEO’s pay structure may rein in soaring salaries. Would Nardelli, for example, have gotten the same windfall if shareholders were able to vote on his compensation package? While it seems an obvious no, Balachandran’s new research shows a more ambiguous conclusion.
To examine the effect of shareholder voting on executive compensation, sometimes called say on pay, Balachandran and coresearchers Fabrizio Ferri and David Maber of Harvard Business School turned to a 2003 UK regulation that required publicly traded firms to submit an executive remuneration report to a nonbinding shareholder vote at their annual meeting. The researchers aimed to find out how the regulation, called the Directors’ Remuneration Report (DRR), affected the sensitivity (or correlation) of cash and total CEO pay to performance at UK companies.
The researchers examined average compensation and pay-to-performance sensitivity in a large sample of UK firms both before and after regulation (2000–02 and 2003–05, respectively), with mixed results. They found no overall change in pay-to-performance sensitivity, but when they examined differences depending on whether stock returns and return on assets were positive or negative, an interesting trend emerged. Compensation after the DRR became more sensitive to negative return on assets — good news for those unhappy about executives receiving high pay despite poor performance. In other words, after the rule, compensation decreased as performance became more negative, a pattern not found before the rule.
However, the researchers found weaker evidence of another effect. When a specific industry garnered positive stock returns, CEOs of individual companies within that industry received higher pay than they did before the DRR, even when their firms had not outperformed the industry overall. Therefore, executives may receive greater compensation during times of positive performance — when shareholders are likely less skeptical about compensation — even if the performance in question is the result of simply being lucky enough to be part of an industry trend.
These findings suggest that the DRR did not have the same effect in all firms but rather had different effects in different circumstances. Therefore, supporters of say on pay should not view it as a panacea for executive compensation problems, says Balachandran. The researchers further caution that a number of other variables — including the overall changes in financial disclosure requirements brought about by the DRR — may have come into play, clouding the true impact of the regulation.
To benchmark their findings, the researchers compared the UK data to executive compensation at U.S. companies of comparable size during the same time periods to see if parallel compensation dynamics occurred in a country that did not enact say on pay legislation. The data showed similar trends in total compensation in the two countries.
Factors other than say on pay — that were specific to U.S., but not UK, companies — may have caused similar effects on executive compensation in the United States as the DRR, notes Balachandran. “During the same time as the DRR in the UK, U.S. companies were faced with the Sarbanes-Oxley Act and a more attentive mood overall by shareholders and directors,” he says.
The issue of shareholder voting’s influence on executive compensation is not likely to disappear anytime soon. A bill seeking to give shareholders the right to an annual advisory vote is working its way through Congress. In addition, many U.S. firms — including Verizon, Aflac and Blockbuster — have recently been subjects of shareholder proposals requesting the adoption of say on pay.
But shareholders who win the right to an advisory vote may get less than they hope for. “If shareholders vote no on a compensation package, it is a nonbinding vote; a company can still go ahead and do what it wants or needs to do,” Balachandran says.
Balachandran, Sudhakar, Fabrizio Ferri and David Maber. “Solving the Executive Compensation Problem Through Shareholder Votes? Evidence from the UK.” Working paper, Columbia Business School, October 2007.
For a copy of this paper, please e-mail Professor Balachandran, svb34columbia.edu
Sudhakar Balachandran is assistant professor of accounting at Columbia Business School.