The growing compensation gap between CEOs and rank-and-file employees has generated considerable debate about potential adverse consequences at both the firm and societal levels. Despite interest in the topic, assessing vertical pay disparity has been difficult due to the lack of public disclosure about employee compensation.

While companies have long reported top-executive pay, transparency on employee compensation was recently enhanced when the SEC adopted the CEO Pay Ratio Rule requiring most reporting companies to provide new disclosures of the median employee’s pay and a ratio comparing the CEO’s compensation with this value.[1] For example, if the CEO and median employee earn $2,500,000 and $50,000, respectively, the firm would report a pay ratio of 50-to-1.

Though passed by Congress as part of the Dodd-Frank Act in 2010, the rule was heavily debated and not effective until the 2018 proxy season. Proponents of the rule argue the disclosure highlights within-firm pay issues that could be important to investors, customers, and employees.[2] Critics contend, however, that the ratio provides little value due to managerial discretion in identifying median employees and computing their pay.[3] Moreover, organizational differences could limit the comparability of the ratio across companies, including those within the same industry.[4]

We analyze the initial disclosure of CEO pay ratios in our paper, “Spinning the CEO pay ratio disclosure.” Using data from the first two proxy seasons in which the rule has been in effect, we shed light on these disclosures and potential consequences for stakeholders across a broad sample of companies in the Russell 3000 index.

We begin by modeling the economic determinants of the ratio’s two components, CEO and median worker compensation. We find that industry factors account for the largest portion of the variation in the model of median worker pay and the pay ratio. This finding provides evidence that there is some comparability across companies in the same industry. We caution, however, that organization form or other differences in the median worker computation can still mitigate the usefulness of comparisons.

To address this latter point, we analyze the actual pay-ratio disclosure narratives from proxy statements. We find that managers use significant discretion in constructing the ratios that influence the reported median pay. Companies with a “high” pay ratio, which we define along several dimensions, are more likely to use exemptions and choose a median employee at a date other than the fiscal year end. These choices likely have the effect of excluding lower income or temporary workers, which can deflate the pay ratio by increasing the reported median pay.

In addition to flexibility in constructing the ratio, the SEC rule also gives managers discretion in generating the narrative of the pay ratio disclosure in the proxy statement. We use textual analysis to examine the number of words, which is a proxy for both the complexity of the ratio calculation and whether the firm provides more context for the reported numbers. We also study the use of certain language we denote as “spin.” For our purposes, spin involves words that managers might utilize to influence how external parties receive the pay ratio or its components.[5] We find that companies with a high pay ratio have lengthier narratives and a greater propensity to use spin in describing its construction. These findings suggest that managers attempt to assuage negative perceptions of the reported ratio through the description of the ratio and its components.

Next, we examine the reception of the initial pay ratio disclosure by studying certain stakeholder outcomes. We focus on three constituencies – the business press, shareholders, and employees – by measuring changes in the tone of media coverage around the disclosure, shareholder voting on say-on-pay issues, and labor productivity as a gauge of employee morale. Regressions show that higher pay ratios are associated with more negative sentiment from media articles just after the pay ratio disclosure, higher voting dissent on say-on-pay proposals, and lower labor productivity gains. These negative stakeholder outcomes are present even when we control for the levels of CEO and median employee pay, which suggests that the pay ratio contains uniquely informative content. Moreover, we find that the stakeholder response stems more from the unexpected component of the pay ratio and is stronger when companies report a higher pay ratio than those reported by industry or geographic peers.

Our next test examines whether the use of spin in the pay ratio disclosure attenuates these stakeholder outcomes. On the one hand, providing additional narrative could impart information on why firms might have a higher pay ratio. On the other hand, some consultants warned that over-describing the pay ratio might attract undesirable attention.[6] We do not find that the use of spin or lengthier disclosures attenuates the negative stakeholder outcomes. In fact, the tests suggest that spin slightly exacerbates the negative media sentiment, say-on-pay voting dissent, and relative declines in labor productivity when the company reports an unexpectedly high pay ratio combined with spin.

Collectively, these results are consistent with the notion that disclosure of higher pay ratios resulting from high CEO pay, low median worker pay, or both could attract more negative attention. Immediately, such effects can be seen in news articles, but shareholders also appear to respond negatively to the ratio, with more voting against CEO compensation packages even after accounting for the level of CEO pay.

For companies reporting a high pay ratio, the relative decline in labor productivity could stem from workers being less motivated when they perceive their pay to be inadequate versus the workers in the same industry or geographic area. Indeed, anecdotal evidence suggests that these ratios are reported prominently by local media near the company’s headquarters, thereby emphasizing pay disparity to employees.[7] High pay ratios are also highlighted in labor disputes, such as the recent strike by union workers at General Motors.[8]

In additional tests, we show that some companies exhibit significant variation in their pay ratios over the first two reporting years. In particular, companies with higher ratios in the first reporting year are more likely to report a lower ratio in the second reporting year. This reduction stems from high pay-ratio companies having a greater likelihood of reducing overall CEO pay and increasing the median worker pay.

The increase in median worker pay prompts the question of whether companies are boosting aggregate pay or simply utilizing discretion in the SEC’s rule to either select a new median employee or retain the same median employee to maximize the reported wage. The SEC only requires the selection of a new median employee if the circumstances have changed in a material way. We find that two-thirds of companies select a new median employee. Companies reporting a high pay ratio in reporting year one tend to retain the same median employee for reporting year two at a greater rate. This choice is associated with reporting increased employee pay, which in turn deflates the pay ratio.

Overall, our study provides a comprehensive examination of CEO pay ratio disclosures. The evidence shows that companies with a high pay ratio use discretion afforded by the SEC rule to deflate the reported figure and add narrative to the disclosure that ostensibly attempts to influence the perception of vertical pay disparity. The disclosure of an unexpectedly high pay ratio overall and within an industry is associated with negative stakeholder outcomes via media coverage, shareholder voting, and employee productivity. We find little evidence that spinning the CEO pay ratio disclosure mitigates the negative consequences of reporting a high pay ratio.

ENDNOTES

[1] See U.S. Securities and Exchange Commission, Pay Ratio Disclosure, Release No. 33-9877 (Aug. 5, 2015) [80 FR 50103 (Aug. 18, 2015)] (“CEO Pay Ratio rule”). All SEC-reporting issuers, except for those meeting the definition of (i) Emerging Growth Companies; (ii) Smaller Reporting Companies; (iii) Foreign-Private Issuers, or (iv) Canadian issuers filing annual reports on Form 40-F, are required to disclose the pay ratio annually for fiscal years beginning on or after January 1, 2017.

[2] Loh, J. (2017). Could the pay ratio disclosure backfire? Examining the effects of the SEC’s pay ratio disclosure rule, Texas A&M Law Review, 4, 417-448.

[3] Bank, S. A., and G. S. Georgiev (2019). Securities disclosure as soundbite: The case of CEO pay ratios. Boston College Law Review, 60, 1123-1203.

[4] Edmans, A. (2017). Why we need to stop obsessing over CEO pay ratios. Harvard Business Review, February 23, 2017.

[5] For example, we categorize the word “talented” as spin in the context of a pay ratio disclosure. Wynn Resorts reports a fiscal year 2017 CEO pay ratio of 909-to-1, which is above the 90^th percentile of the sample. They note in their pay ratio disclosure that, “Our talented and dedicated employees play an integral role in our overall success and we place great emphasis on creating an environment for our employees to excel and advance. We are committed to the development, health and well-being of our workforce through various programs, benefits and amenities.” See Wynn Resorts, Limited, SEC Form DEF 14A, April 18, 2018, available at https://www.sec.gov/Archives/edgar/data/1174922/000119312518121082/d572661ddef14a.htm.

[6] Lifshey, D. (2018). CEO Pay Ratio Disclosure Round Two: Top 10 Things to Worry About, https://www.pearlmeyer.com/blog/ceo-pay-ratio-disclosure-round-two-top-ten-things-to-worry-about.

[7] For example, media in Cleveland highlighted local CEO pay ratios in Ohio. See Serino, D., “CEOs at some of Ohio’s largest employers make up to 300 times what their typical workers earn.” WKYC.com, September 13, 2018. https://www.wkyc.com/article/money/ceos-at-some-of-ohios-largest-employers-make-up-to-300-times-what-their-typical-workers-earn/95-594587342.

[8] Buss, D., Rise of CEO Pay Issue Is Underscored by Unionists’ Concerns in GM Strike. ChiefExecutive.net, September 26, 2019. https://chiefexecutive.net/rise-of-ceo-pay-issue-is-underscored-by-unionists-concerns-in-gm-strike/. Noting that the 240-to-1 pay ratio “resonates as a genuine sticking point” to union workers.

This post comes to us from Audra Boone, professor of finance at Neeley School of Business, Texas Christian University; Austin Starkweather, postdoctoral fellow at Owen Graduate School of Business, Vanderbilt University; and Joshua T. White, assistant professor of finance at Owen Graduate School of Business, Vanderbilt University. It is based on their paper, “Spinning the CEO Pay Ratio Disclosure,” available here.