AMERICAN ACCOUNTING ASSOCIATION
CONTACT: Ben Haimowitz (718-398-7642)
Long-awaited Dodd-Frank regulation may very well turn out to be effective in restraining CEO pay, study suggests
In an effort to get companies to rein in galloping CEO pay, enforcement of Section 953(b) of the Dodd-Frank act, scheduled to begin next January, will require firms to publish the ratio of the CEO’s compensation to the median pay of company employees. Will this long-awaited mandate of the 2010 bill have its intended effect? An experiment reported in the current issue of a scholarly journal suggests it just may.
The study, in the spring issue of the Journal of Management Accounting Research, published by the American Accounting Association, acknowledges the failure of present regulations that seek to curb pay excesses by requiring companies to disclose executive-compensation levels and explain how they were arrived at. Adding CEO-to-employee pay ratios to the disclosure mix, the new research suggests, is likely to be significantly more effective than the current requirements.
In the words of the paper, by Khim Kelly of the University of Waterloo and Jean Lin Seow of Singapore Management University, "incrementally disclosing a higher-than-industry pay ratio (versus disclosing only higher-than-industry CEO pay) significantly decreases perceived CEO pay fairness...and has a significant indirect negative effect on perceived investment potential."
The authors add: "Given that companies are concerned about whether investors, employees, and the public perceive their CEO pay and employee pay to be fair, our results are consistent with the argument...that pay-ratio disclosures may be better able than current CEO pay disclosures to shame companies into restraining CEO pay."
Prof. Kelly comments that "whereas CEO-to-employee pay ratios in the largest U.S. companies were about 20-1 fifty years ago and about 80-1 twenty-five years ago, estimates nowadays range from about 200-1 to 300-1. It stands to reason that confronting investors with ratios approaching these will make an impression, and our research suggests it does."
The finding emerges in an experiment in which MBA students were asked to make judgments about a hypothetical company. Those to whom high CEO-to-employee pay ratios were disclosed were significantly more likely than others to regard the CEO's pay as unfair, and the more unfair they regarded it, the less likely they were to view the firm as a worthy investment. This latter finding is particularly noteworthy, the professors believe, because one of the criticisms leveled against Section 953(b) is that it engenders overreach by the SEC, requiring it to meddle in issues of income inequality rather than stick to its intended job of protecting investors.
Says Prof. Kelly: "Here were men and women with an average of six years' work experience and a fair degree of sophistication about business and accounting, and clearly disclosure of a lopsided CEO-to-employee pay ratio turned out to be relevant to an investment decision. Actually it would be surprising if it didn't. As a recent nationwide survey conducted by Stanford University suggests, the whole issue of CEO pay is in bad odor. Seventy-four percent of the respondents to that survey said CEOs were overpaid relative to the average worker; yet, ironically, when asked what they believed the pay of Fortune-500 CEOs to be, they vastly underestimated the amount. In this climate of opinion, how could perceiving the CEO-to-employee pay ratio as unfair not be relevant to investment decisions?"
The 75 participants in the experiment were asked to assume they worked in the investment department of a corporation and were assigned to assess the potential of a fictional firm in which their company was considering making a medium- to long-term investment. They were informed that the firm was in the semiconductor industry and was regarded a leader in market share and innovation. The subjects were provided with company financial data and information on CEO compensation, both of which were adapted from an actual NASDAQ-listed semiconductor firm.
One third of the subjects were informed that the CEO's total compensation in the most recent year was about $4.3 million and that this amount was roughly the average for CEOs in the company's industry.
A second group was informed that the CEO's total compensation was about $7.4 million, and that this amount was higher than the pay of three fourths of the CEOs in the industry.
A third group was provided the same information as the second group plus the fact that the median pay for employees of the company (other than the CEO) was about $45,000 and, additionally, that this meant the CEO-to-employee pay ratio was about 162-1, well above the average industry ratio of about 96-1.
Asked their opinion of the CEO's level of compensation, on a scale of -7 (very unfair) to +7 (very fair), the groups offered significantly different views, as follows:
■ The first group produced an average positive fairness rating of +1.12.
■ The second group, on average, bestowed a neutral fairness rating of +0.12.
■ The third group rated fairness negatively at -1.32.
Statistical analysis indicated opinion of CEO pay fairness to be significantly related to assessment of the company's investment worthiness. In other words, the less fair the chief executive’s compensation was perceived to be, the less potential the company was judged to have as an investment.
Why didn't comparatively high CEO pay by itself push fairness ratings into negative territory? One important reason, the researchers found, was that participants viewed high compensation levels as beneficial in attracting executive talent. But this belief is evidently outweighed by the sense of unfairness created by the additional disclosure of a lopsided CEO-to-employee pay ratio.
Further research by Profs. Kelly and Seow has resulted in still another intriguing finding from a similar experiment, this one involving 100 MBA students who were asked to assess a hypothetical company in the restaurant industry. The experiment, reported in a working paper, tests not only the three conditions described in the JMAR study but a fourth condition in which both the CEO's compensation ($4.3 million) and the CEO-to-employee pay ratio (96-1) are average for the industry. The professors find that the combination of the lower CEO compensation level and lower pay ratio evokes a fairness rating that is just as negative as the one produced by above-industry averages of $7.4 million and 162-1. In short, in the words of the study, "pay-ratio disclosures, even when they reveal less extreme CEO-to-employee pay multiples that are comparable to those in peer companies, can result in negative perceptions of CEO pay fairness and workplace climate that indirectly reduce perceived investment potential of companies."
Prof. Kelly sums up: "High CEO-to-employee pay ratios, whether 100-1 or 200-1, clearly made a special impression on the participants in our experiments. It seems eminently plausible that high ratios will also impress actual investors, with concomitant effects on companies, when they stare out from proxy statements."
The Journal of Management Accounting Research is published twice yearly by the American Accounting Association, a worldwide organization devoted to excellence in accounting education, research, and practice. Other journals published by the AAA and its specialty sections include The Accounting Review, Auditing: A Journal of Practice and Theory, Accounting Horizons, Issues in Accounting Education, Behavioral Research in Accounting, Journal of Information Systems, and The Journal of the American Taxation Association.