Critics have argued that the rule requiring companies to disclose the ratio of CEO compensation to employee pay is too expensive and time consuming, with the U.S. Chamber of Commerce estimating the cost to U.S. companies at more than $700 million per year. The Securities and Exchange Commission, however, has put the annual cost at about $73 million[1], and in 2015 its commissioners voted 3-2 to adopt the rule. Since then, income inequality has reached unprecedented levels, and the CEO-to-employee pay ratio has skyrocketed. According to the Economic Policy Institute, the ratio was on average 21-to-1 in 1965 but grew to 320-to-1 in 2019. An important question is whether a high CEO-to-employee pay ratio is evidence of rent extraction by managers or of the firm’s efforts to secure a top-quality CEO. Prior examination of these competing hypotheses was limited in academic literature, because the average employee pay was not publicly available until the passage of this rule and subsequent disclosure of the pay ratio starting in 2017.
In a new paper, we study these questions in the context of the association between pay ratios and corporate borrowing costs. Existing literature offers different predictions about the relationship between CEO-to-employee pay disparity and the cost of debt. For instance, a high pay ratio may be associated with lower perceived risk and corporate borrowing costs, because high executive compensation can serve to secure high-quality CEOs (talent assignment) and motivate optimal CEO effort (incentive provision). Conversely, a high pay ratio may correspond with a higher cost of debt, as pay disparity can reflect excessive compensation that hinders firm profitability (rent extraction) and perceived inequities can contribute to employees’ shirking their duties (inequity aversion).
We find a significant negative relation between industry-adjusted CEO-to-employee pay ratio and yield spreads while controlling for covariates and endogeneity. The association is both statistically and economically significant: A one standard deviation increase in the adjusted pay ratio corresponds with up to a 38 basis point reduction in bond yield spreads. To determine whether incentive-provision or talent assignment explains these findings, we examine the relationship between CEO pay ratios and the cost of debt in firms with varying levels of financial constraints, labor versus capital intensity, and firm size. A more pronounced effect among financially constrained and labor-intensive firms is consistent with the incentive-provision hypothesis, as an efficient compensation structure that motivates optimal effort should play a larger role in such firms. The talent assignment hypothesis predicts a more pronounced effect among the largest firms, which are more likely to attract superstar CEOs. Our results are consistent with the incentive provision explanation, as the negative relation between pay disparity and the cost of debt is strongest among financially constrained, labor-intensive, and smaller firms.
The incentive-provision hypothesis also predicts that increases in CEO compensation will be associated with reductions in corporate borrowing costs, because CEO pay often includes substantial bonuses for meeting performance targets that motivate additional executive effort. By contrast, the talent assignment hypothesis predicts no relation, given that innate CEO ability is relatively stable from year to year. Our empirical analysis shows that increases in CEO-to-employee pay ratios are associated with reductions in borrowing costs, adding further support to the incentive-provision explanation. Altogether, our findings suggest that high CEO-to-employee pay ratios reflect efficient compensation practices that motivate optimal effort leading to reductions in perceived risk and corporate borrowing costs.
Previous research used self-reported labor costs to construct an estimate of median employee pay. Our research finds that this approach yields an incorrect conclusion about the association between CEO-to-employee pay ratio and the cost of debt. We find that the relation is positive for a small sub-sample of firms that self-report labor costs, in contrast to our findings of a negative relation across the full sample of firms following required pay ratio disclosure. This shows the usefulness of the reported CEO-to-employee pay ratio for understanding its implications.
ENDNOTE
This post comes to us from professors Katsiaryna Salavei Bardos, Steven Kozlowski, and Michael Puleo at Fairfield University. It is based on their recent article, “Entrenchment or Efficiency? CEO-to-Employee Pay Ratio and the Cost of Debt,” available here.