Corporate Human Capital Disclosures: Early Evidence from the SEC’s Disclosure Mandate

Elizabeth Demers is Professor of Accounting at the University of Waterloo School of Accounting & Finance; Victor Xiaoqi Wang is Assistant Professor of Accountancy at California State University Long Beach; and Kean Wu is Associate Professor of Accounting at the Rochester Institute of Technology Saunders College of Business. This post is based on their recent paper.

Related research from the Program on Corporate Governance includes The Illusory Promise of Stakeholder Governance (discussed on the Forum here) by Lucian A. Bebchuk and Roberto Tallarita; For Whom Corporate Leaders Bargain (discussed on the Forum here) and Stakeholder Capitalism in the Time of COVID (discussed on the Forum here), both by Lucian Bebchuk, Kobi Kastiel, and Roberto Tallarita; and Restoration: The Role Stakeholder Governance Must Play in Recreating a Fair and Sustainable American Economy—A Reply to Professor Rock (discussed on the Forum here) by Leo E. Strine, Jr.

Human capital is a critically important source of corporate value creation in the modern economy, yet disclosures related to what executives commonly refer to as their “most important asset” have been extremely limited relative to those of other asset classes. This was supposed to change in November 2020 when the SEC’s amendment to Regulation S-K took effect. The new rules require that filers provide expanded discussion related to the firm’s human capital (HC) as part of Item 1 (i.e., the business description section) of their 10-K filing. The new rules are principles-based, however, so they allow for a tremendous amount of discretion without stipulating any specifics as to what companies should disclose. Early critics expressed concern that this approach would lead to too much heterogeneity in HC disclosures, that it was fraught with the potential for “greenwashing,” and that it would otherwise not yield the comparable quantitative data that investors need to properly assess corporate performance.

Our study provides the first comprehensive descriptive evidence required to assess the efficacy of the new regulation that has been subject to widescale criticisms in the investment community. We use textual analysis to extract the linguistic features and numerical intensity of HC disclosures for more than 3,000 unique public companies (i.e., all 10-K filers with corresponding financial data available), each reporting for the first time under the new regulation.

A number of interesting, stylized facts emerge from our analysis. First, consistent with anecdotal accounts, we provide systematic evidence that there is tremendous cross-sectional variation in the amount, numerical intensity, tone, readability, and similarity of HC disclosures both in absolute terms, and when benchmarked against the rest of the contents of the firm’s Item 1 disclosures. Although this may seem like good news from a regulatory perspective to the extent that it suggests that firms are not providing totally uninformative boilerplate disclosures, a less sanguine interpretation of the evidence is that the low level of similarity—even for firms within the same industry—will make it harder for investors to compare HC performance across firms.

Second, HC disclosures are generally more readable and considerably more positively-toned than the remainder of Item 1 disclosures. The highly optimistic tone suggests that companies are either very satisfied with their own HC performance, or that they would like investors to believe this to be the case (i.e., they are greenwashing).

Third, 18% of sample firms provide disclosures consisting of less than 100 words. Although these firms tend to be younger, smaller, and have fewer employees (i.e., they may have less to discuss), such abbreviated disclosures are unlikely to be adequate for investors who wish to gain an understanding of the firm’s HC management practices. As such, the disclosures arguably do not comply with the spirit of the new regulation.

We next investigate the determinants of these various disclosure attributes. We first find that, with few exceptions, the firm’s HC disclosures tend to inherit similar properties to its Item 1 disclosures—e.g., if the firm offers higher specificity or numerical intensity in their other Item 1 disclosures, they tend to do so in their HC disclosures as well. In other words, filers tend to be of a corporate communication or disclosure “type,” and this carries over to their HC disclosures.

Further analyses reveal that although contemporaneous firm performance (ROA), firm size, the firm’s asset structure and growth prospects (captured by the ratio of PPE to total assets and the book-to-market ratio), and the competitiveness of the industry in which the firm operates are each significant determinants of at least some of the examined disclosure attributes, the observed relations are not always consistent with expectations. For example, firms with better financial performance provide more specific and more quantitative disclosures, but their disclosures exhibit a more negative tone. Alternatively stated, firms that are performing less well financially provide less specific (i.e., more boilerplate) and less numerically intensive disclosures, and their disclosures tend to be more positively-toned. This latter finding is suggestive of greenwashing wherein financially underperforming firms talk-up their HC performance, but without backing this up with hard quantitative metrics.

We also find that institutional investor ownership is associated with longer and more positively-toned disclosures, but not with other desirable disclosure attributes such as specificity, numerical intensity, and readability. This finding is suggestive of firms superficially responding to investor demand, but without offering shareholders the kinds of informative disclosures that are required for meaningful analysis and investment decision-making.

We extend our descriptive analyses to test two specific hypotheses. First, we hypothesize and find that the characteristics of HC disclosures change over the course of the first year of the regulation’s implementation. Specifically, HC disclosures become longer, more similar, and more positively toned over the course of the first year, while decreasing in specificity and numerical intensity. These findings suggest that firms learn from the disclosures of earlier filers to increase the length and inflate the tone of their HC reporting, but not to improve the informativeness of their disclosures. Second, we hypothesize that firms that are more reliant on human capital to create value will provide superior disclosures given that this information is particularly important to investors’ assessments of the firm’s historical performance and future prospects. Although firms for which HC is more strategically important provide significantly more numerically intense, more readable, more similar, and more positive HC disclosures, their disclosures are neither longer nor more specific.

Overall, our results confirm that, in the absence of detailed guidance, corporate HC disclosures are extremely heterogeneous in terms of their length, numerical intensity, tone, readability, and similarity with peer firms. The discretion afforded to filers leads to disclosures that are generally not very numerically intensive, that are overwhelmingly positively toned, and that tend to inherit many of the properties of the firm’s other Item 1 disclosures. While the firm’s financial performance, the strategic importance of HC to its success, and the level of institutional investor ownership in the firm are each associated with some desirable disclosure attributes, none of these factors helps to induce disclosures that are superior along all, or even most, dimensions. Finally, the documented time trends are concerning as they suggest that firms have learned over the first year of the non-directive regulation to provide HC disclosures that are longer and more optimistic, but less informative (i.e., more similar or boilerplate, less specific, and less numerically intensive). Overall, our comprehensive descriptive evidence suggests that, consistent with widespread criticism, the SEC’s new principles-based rule has generated HC disclosures that are likely to be highly insufficient for investors’ needs and that some further guidance or overhauling of the rules may be required.

The complete paper is available for download here.

Both comments and trackbacks are currently closed.

One Comment

  1. Kiers
    Posted Thursday, August 4, 2022 at 3:55 pm | Permalink

    Firms must be made to disclose straight payroll labor expense in every income statement. Cost of goods sold should be broken out into labor and material. All other indirect categories should also break out labor expense in each category (SG&A, R&D, etc.)