Changes in the level and dispersion of CEO compensation since the early 1990s have triggered an increasingly heated debate over whether current compensation practices primarily reflect the equilibrium outcome of the CEO labor market or the power of entrenched CEOs. One factor in this debate is the practice of compensation benchmarking in which firms justify their CEO’s compensation by comparing it to the pay packages of a group of companies with highly paid CEOs. Firms rationalize this group by claiming they compete for managerial talent with those selected peer companies. In the paper, Is Disclosure an Effective Cleansing Mechanism? The Dynamics of Compensation Peer Benchmarking, forthcoming in the Review of Financial Studies, we examine the dynamics of the peer benchmarking process. Specifically, we investigate whether the 2006 regulation requiring firms to disclose their compensation peer group members has mitigated opportunistic firm behavior in benchmarking against self-selected peer companies with highly paid CEOs.
Compensation disclosure is intended to make the compensation process at firms more transparent, force the board of directors to effectively monitor management, and prevent managers from setting their own pay. However, disclosed information on the compensation of peer CEOs may actually help justify a CEO’s demand for higher pay. For instance, following the tax legislation in 1992 that capped the corporate income tax deduction of non-performance-related compensation at one million dollars (IRS tax code 162(m)) and the compensation disclosure rules enacted in 1993, dramatic increases in real compensation levels were observed. The effect was a ratcheting up of executive compensation. Kevin Murphy’s 2011 review of the legislative history of executive compensation over the past eighty years finds that for the most part, “the regulations have generally been either ineffective or counterproductive.”
In this paper, we examine the 2006 Securities and Exchange Commission (SEC) rule that requires firms to disclose compensation peer companies if they are used in determining executive compensation. Researchers have documented that even after controlling for characteristics that would likely capture the competitive forces of the managerial labor market, such as industry identification and relative size of the firm and its peer group members, the compensation level of the CEO at a potential peer company had significant incremental power in explaining the company’s peer group memberships (Faulkender and Yang 2010; Bizjak, Lemmon, and Nguyen 2011). In other words, firms appear to be managing the benchmarking process by including companies with highly paid CEOs in their peer groups and omitting comparable companies with lower-paid CEOs. By doing this, firms can rationalize the high level of their CEO pay by claiming that they are paying the median compensation of their peer CEOs. Since the mandatory disclosure of compensation peer companies, investors have been informed about compensation peers, and firms have had ample time to alter their peer companies. Has enhanced disclosure mitigated the opportunistic behavior in peer selection? Or has enhanced disclosure actually increased peer selection biases and created an unintended consequence similar to that of IRS code 162(m)? That is the focus of this study.
The timing of the change in disclosure requirements makes it a useful setting for examining the effects of mandated disclosure. On August 29, 2006, the SEC mandated disclosure of component companies of compensation peer groups. Because the new regulation was enacted well after the peer groups for 2006 had been formed, the first year that we observe should be composed of peer groups that were selected prior to the firm knowing that the names of the compensation peer companies would have to be disclosed. These observations therefore serve as a baseline for assessing firm behavior prior to the disclosure requirement. We follow the peer groups for the subsequent three years, during which boards of directors and their compensation committees could adjust peer companies away from those selected before the commencement of the disclosure rule. Analyzing the difference in peer groups selected prior to and those after enhanced disclosure enables a clean assessment of the effects of mandated disclosure.
Our empirical evidence suggests that firms did not generate less-biased peer groups after enhanced disclosure. Actually, if anything, firms tended to select peer companies with higher CEO pay than those selected in 2006. This was not merely the result of existing peers increasing their CEO’s pay more than unselected potential peers did. We find that firms actively added companies with higher CEO pay to their peer groups and dropped peer companies with lower CEO pay. These findings call into question the effectiveness of disclosure regulation to remedy potential problems in the pay-setting process.
We further examine whether strong corporate governance at firms could counterbalance the labor market pressures for engaging in opportunistic peer selection. First, we examine whether firms that received significant shareholder complaints about their compensation practices selected less-biased peers as a result. Had enhanced disclosure helped mitigate opportunistic activities in peer selection, we would observe some improvement in peer selection at those firms. We did not find that improvement. However, other governance mechanisms were effective. We observe no increase in peer selection bias among firms with high institutional ownership, director ownership, and CEO ownership while firms with low ownership levels showed a significant increase in peer selection bias. We also find that the increased opportunistic behavior in peer benchmarking was limited to firms whose boards of directors were not intensive monitors, firms with directors “busy” serving on the boards of multiple firms, and firms with large boards.
The full paper is available for download here.