In the paper, Which Skills Matter in the Market for CEOs? Evidence from Pay for CEO Credentials, which was recently made publicly available on SSRN, we show that boards’ compensation decisions reward several reputational, career, and educational credentials of CEOs using a panel of S&P 1,500 firms between 1993 and 2005.
Our study is motivated by anecdotal accounts of executive search consultants, recent empirical evidence, and a growing theoretical literature that point to an increased importance of the labor market for CEOs over the last two decades. The central message of these studies is that there are fundamental differences in CEOs’ skill sets and that these differences are an increasingly important determinant of CEO pay. However, we still have scant direct evidence on whether differences in CEO skills matter for their pay. Even less is known about which CEO skills actually carry a premium in CEO pay and whether skill pay premia can help to explain key stylized facts of CEO pay, such as its dramatic rising trend and the increasing gap between the most and the least paid CEOs. In order to fill this gap, we use new hand-collected biographical data on a large sample of CEOs to examine whether there is a credentials premium in CEO pay—i.e., do firms make inferences about otherwise difficult to observe CEO skills from readily available facts that can be gathered from CEO resumes and professional career track records? If so, is the relation between CEO pay and credentials consistent with market-based theories?
We construct three measures of CEO credentials that summarize a broad array of publicly available information at the time of CEO pay decisions for a panel of nearly 4,000 CEOs of S&P 1,500 firms over the 1993 to 2005 period. These include information about their industry reputation, labor market status, and educational pedigree. First, the CEO reputational signal measures outside perceptions of CEO skills and is constructed by counting the number of major business newspapers articles containing the CEO’s name in the prior year. The basic idea is to test whether previous recognition of the CEO by the business press is perceived by firms as a signal of the CEO’s skills. Second, the labor market signal measures the quality of the CEO’s career record and is defined as a function of the age at which the executive first took a CEO job. Intuitively, if the market for CEOs is meritocratic, then the younger an executive is when she gets her first CEO job, the more positive is the signal of her skills. Third, the schooling signal measures the quality of the CEO’s educational background and is constructed using Barron’s rankings of college selectivity. Based on signaling models of education, we expect attendance at more selective colleges to also be a signal of CEO skills.
Our main finding is that newly-appointed CEOs earn up to a 5 percent total pay premium for each decile improvement in the distributional ranking of these three credentials. The credentials pay premium is economically significant and measures in at about $280,000 per credential decile. Consistent with boards using credentials as publicly observable signals of otherwise hard to gauge CEO skills, we show that the credentials pay premium displays key cross-sectional features predicted by market-based theories: it is even higher for those CEOs that have better credentials and for those that run larger firms. The premium is robust to tests that address selectivity and endogeneity concerns by using instrumental variables and controlling for firm and CEO fixed effects, as well as for other market signals, such as the pay of the CEO in his previous job. It is also robust to controlling for CEO lifetime work experience and social ties, which helps to distinguish our results from alternative stories based on general human capital and entrenchment.
Finally, we show that the CEO credentials premium helps to explain some key stylized facts of CEO pay, thus supporting market-based explanations of the overall rise in CEO pay and the increasing inequality between the most highly and the least highly paid CEOs. In particular, we show that the credentials pay premium has significantly risen over time in the last two decades. The upward trend in CEO pay over the 1990s and 2000s was about twice as large in magnitude for CEOs at the top of the credentials ladder relative to those at the bottom. Strikingly, there is no significant upward trend for the total pay of those recently appointed CEOs that have the lowest credentials. Thus, a rising premium for CEO credentials can help to explain the overall upward trend in CEO pay, especially for newly appointed CEOs. The rising premium does a particularly good job at explaining the overall trend among outside hires and at the very top of the distribution of pay, suggesting that a rising credentials premium may have also contributed to the increasing pay gap between the most and the least paid CEOs. Finally, when we repeat the analysis by broad industry groups, we see that a rising credentials premium is especially relevant for understanding the stylized developments in CEO pay for the manufacturing, services, and hi-tech sectors.
The full paper is available for download here.