Behind the EEO Curtain

The brutal death of George Floyd in May 2020 put the spotlight on U.S. corporations and their role in fostering diversity, equity, and inclusion (DEI) practices. This rising interest in DEI practices has increased calls from investors and other stakeholders for firms to publicly disclose data on DEI metrics. Regulators are also taking notice, as the lack of consistency across existing data has led the SEC to consider “further enhancements” to their rules on human capital disclosures. While U.S. firms are not currently required to publicly disclose data on their workforce diversity, those with more than 100 employees are already required to annually collect these data and confidentially report them to the U.S. government in a standardized EEO-1 report.

Particularly in recent years, there have been calls for firms to share their EEO reports with investors and the public generally. However, firms have typically resisted on the grounds that their workforce composition is akin to a trade secret, and its disclosure could hurt the firm’s competitive position. This behavior has contributed to information scarcity about DEI practices. For example, over two-thirds of U.S. public firms do not disclose a single diversity-related quantitative metric in their 10-K filings.

The universe of publicly available firm diversity-related information expanded significantly with the recent release of EEO-1 reports for over 18,000 federal contractors. Although EEO-1 reports are filed confidentially, firms with federal contracts can be subject to the Freedom of Information Act (FOIA). In 2018, a reporter from The Center for Investigative Reporting submitted FOIA requests for the EEO-1 reports of all firms with federal contracts from 2016–2020. After an extended legal battle, these data were released in April 2023 for all firms which did not object to the disclosure. We use this newly available data to study workforce diversity in our new study.

As a first step, we focus on a constant panel of the 2,500 largest public firms from 2016–2020. Because EEO-1 reports are only available for government contractors, we identify federal contractors in our sample and then determine which of these firms permitted or objected to the release of their EEO-1 reports. We find that EEO-1 forms were released for approximately two-thirds of the government contractors in our sample, indicating that one in three objected to the release of their EEO-1 reports. Using these data, we seek to achieve two objectives.

The first is to paint a detailed picture of workforce diversity among public firms that are federal contractors. EEO-1 forms typically include a breakdown of gender and race by occupation, including executives, managers, technicians, and sales workers. This breakdown is unique as other sources of data (e.g., Bureau of Labor Statistics, U.S. Census) are typically pooled across all employees. We primarily focus on first- and middle managers because prior studies have long argued that firms’ primary diversity challenge is to obtain a diverse group of managers. Our data reveal a lack of diversity in our sample. With respect to gender diversity, we find that women hold 35 percent of the first- and middle-management positions while accounting for 39 percent of the workforce. Racial composition presents more striking patterns. White individuals hold 72 percent of first- and middle-management positions while accounting for only 64 percent of the total workforce. Asian individuals hold 8 percent of the first- and middle-management positions while accounting for 9 percent of the workforce. On the other hand, Hispanic and Black employees account for only 6 percent and 4 percent of first and middle management while respectively accounting for 10 percent and 8 percent of the total workforce in our sample.

We next turn to whether a firm’s voluntary disclosure of its EEO-1 form is associated with the diversity of its workforce. We consider three measures of diversity: (1) the share of managers who belong to a minority group, (2) the share of minority managers relative to the share of minorities across all employee levels, and (3) the firms’ share of minority managers above or below the median across the relevant industry. We use these measures for both women and racial minorities.

Overall, our results suggest that firms’ gender diversity does not seem to correlate with the decision to publicly release its EEO-1 form. Instead, racial performance seems to play a major role. Across all three measures, we find consistent evidence that firms with lower levels of racial diversity are less likely to publicly disclose their EEO-1 reports. Our results are economically important. For example, a firm at the 75th percentile of the share of minority managers to minority employees is over three times more likely to publish its EEO-1 report as a firm at the 25th percentile.

Our results have clear policy implications. Investors increasingly view a lack of diversity as a material risk for public companies and often call for the SEC to mandate disclosure of comprehensive and comparable public data on firms’ DEI practices. Our data provide unique, up-to-date statistics on the distribution of diversity metrics that would be observable to investors should a mandate be enacted and initial evidence of the differences in workforce diversity between firms that have and have not voluntarily released this data. In the short-term, the demand for diversity information by private actors seems unable to achieve full disclosure, allowing firms with less diverse workforces to stay silent. If investors continue to request workforce diversity information, a disclosure mandate seems necessary to satisfy them.

This post comes to us from Thomas Bourveau at Columbia Business School, Rachel W. Flam at London Business School, and Anthony Le at Columbia Business School. It is based on their recent article, “Behind the EEO Curtain,” available here.