Print
E-MailPeter Haslag is an Assistant Professor of Finance, Berk A. Sensoy is the Hans Stoll Professor of Finance; and Joshua T. White is Assistant Professor of Finance at Vanderbilt University Owen Graduate School of Management. This post is based on their recent paper.
In the modern firm, human capital is a primary source of value. Indeed, an emerging body of research links the stock and flow of rank-and-file employees to firm value and investor returns (e.g., Agrawal, Hacamo, and Hu, 2021). Despite its importance, there is sparse academic literature on what employee information firms disclose and whether it is timely and informative given the SEC’s flexible disclosure rules.
We seek to fill this gap in our research paper, Human Capital Disclosure, using a proprietary dataset of 45 million individuals’ job histories combined with textual analysis of SEC Form 10-K annual reports. Given the recent calls by market participants and regulators for more information on human capital management, the findings of our research have important regulatory implications.
Our analyses first shed light on the supply and demand of human capital disclosure over the past two decades. We extract all paragraphs with employee-linked words from 10-Ks during 2001 to 2021 and use a natural language processing technique to identify topics of human capital information supplied by firms. After removing boilerplate, we augment our analyses to gauge demand based on potentially material dimensions of human capital identified by the Sustainability Accounting Standards Board (SASB).
Our approach identifies eleven key human capital disclosure topics, in order of overall phrase count: Attract & Retain, Integration, Workforce Reductions, Unions, Data Security, Health and Safety, Trade Secrets, Competition, Bribery, Morale, and Diversity. If managers tend to be forthcoming, we expect them to provide more intensive discussions of a topic when it is more likely to be material to its investors. On the other hand, firms may take advantage of the flexible nature of human capital disclosure and withhold potentially relevant information. For example, recent survey evidence shows that 83% of institutional investors say human capital management is the most important topic when asking firms to provide more detailed disclosure (see Morrow Sodali’s Institutional Investor Survey, 2019).
Across each of the eleven topics, we partition the sample based on cross-sectional properties that are likely important to the topic. For example, consider the topic of Unions. We expect managers to provide more frequent discussions of employees and unions when they operate in industries with greater unionization. Indeed, for all eleven topics, we find that firms with higher values of cross-sectional partitioning variables provide more intensive discussions than those with lower values, statistically significantly so in eight of the eleven topics. These results provide initial evidence that firms are, on average, forthcoming in the relevant factors of their underlying human capital.
Across the entire sample of 10-K disclosures, Attract & Retain (“A&R”) is the topic with the most extensive discussions. As noted above, prior studies find that the stock and flow of human capital—which is the essence of A&R discussions—impacts firm value. Accordingly, we further scrutinize firms’ specific disclosures on the stock and flow of employees by comparing A&R discussions in 10-Ks to the actual workforce changes revealed by our dataset of job histories.
We examine which firm characteristics are associated with the intensity of A&R disclosures. A large academic literature points to proprietary disclosure costs as a reason why managers might strategically withhold value-relevant information, lest rival firms gain a competitive advantage (Chen, Hoberg, and Maksimovic, 2021). In this view, managers might withhold human capital information from investors when they operate in competitive industries. However, if A&R disclosure is fulfilling its intended purpose, we expect firms with more workforce competition to inform investors of this risk. We find that proprietary costs do not mitigate workforce disclosure. In fact, firms with both greater product and labor market competition disclose more A&R statements. These findings suggest that strategic considerations outweigh proprietary disclosure costs for this dimension of human capital disclosure.
We also expect managers to be more forthcoming in their disclosures when the information is demanded by key market participants and stakeholders. Consistent with this notion, we find that firms with greater levels of institutional ownership and analyst following provide more A&R disclosures. Thus, managers appear responsive to stakeholder demands for human capital information.
We next use our database of employee job histories to examine whether A&R disclosures accurately reflect the actual flow of the workforce. We compute employee hiring and separation rates for each firm. We then decompose these values into predicted and unpredicted components based on occupational and industry mobility at the local and national levels. Regression analyses reveal a strong positive relation between within-firm employee turnover increases and subsequent A&R disclosure intensity. Moreover, increases in the unpredicted components of both employee hiring and separation tend to be followed by additional firm-level disclosures of A&R information in the subsequent 10-K, especially when a larger fraction of the workforce is classification as high-skill based on their occupation. These results support the notion that managers are forthcoming about A&R risk as disclosure changes reflect actual variation in the underlying workforce.
While flow measures identify how many employees join or leave, they do not gauge which employees are turning over. Thus, to measure the so-called “stock” of human capital, we estimate the tenure of each firm’s employees based on the average length of employment. We find a strong negative relation between employee tenure and subsequent A&R statements. Thus, as the stability of the firm’s stock of human capital grows, managers reduce the intensity of discussions pertaining to A&R risk. This finding is again consistent with informative, as opposed to opportunistic, human-capital disclosures.
In our final set of analyses, we examine how human capital disclosure changes in response to strengthening SEC disclosure requirements. The SEC expanded mandatory risk-factor disclosures in 2005. This rule change required firms to discuss key risks that could disrupt their business in 10-Ks but did not explicitly require firms to disclose human capital information. Despite the generality of the rule, we find a 20% increase in human capital disclosure after 2005. This increase in disclosure is especially strong for firms whose disclosure was previously low, in both absolute terms and relative to our model of predicted disclosure based on firm and employee characteristics.
In November 2020, the SEC again amended its disclosure regime to require firms to provide additional information about their human capital management processes but did not prescribe specific data points. We find that A&R statements increase by 40% following the 2020 SEC rule change. The response is especially strong for firms that were previously under-disclosing based on our model of predicted disclosure. Firms also respond more when they have a larger analyst following and a higher portion of their market value stems from intangible assets. These results suggest the 2020 SEC mandate had a meaningful impact on those firms who might not have otherwise provided this information, especially when there is higher demand for human capital information from market participants.
Taken together, our study advances the academic literature on human capital disclosure and the value-relevance of employees. Our paper also has important and timely regulatory implications. Although the SEC promulgated new human capital disclosure requirements in 2020, some claim their principles-based approach is too flexible. To that end, SEC chair Gary Gensler announced in June 2021 that he had “asked staff to propose recommendations for the Commission’s consideration on human capital disclosure,” and indicated that the SEC might mandate disclosure of specific employee metrics, including workforce turnover. We find that the 2020 SEC rule had a meaningful impact on previously low-disclosing firms and those where human capital is key to value creation. Thus, despite the flexibility afforded to managers, the SEC regulations were effective at increasing human capital disclosure. Our study also identifies key employee data points such as average employee tenure and rates of hiring and separation that would likely benefit investors, but caution that mandating such disclosures could generate a cost for certain firms.
The complete paper is available for download here.
