Recently, in some OFCCP audits and private litigation, the Agency and plaintiffs’ statistical experts have applied a Human Capital Theory (Becker, 1975) approach when evaluating compensation disparities between different subgroups. According to their arguments, employer or institutional characteristics (e.g. job title) are often tainted by discrimination, and their removal can be justified via Human Capital Theory, which notes that productivity, and inevitably wages, are the result of employee choices and characteristics. Given this perspective on compensation analytics, two questions arise: (1) What is Human Capital Theory and (2) Is it appropriate to apply it to the analysis of pay disparities? A review of the Theory suggests that while it has gained considerable traction, it also has limitations when applied to compensation equity analytics.
Human Capital Theory, an economic theory developed in the latter part of the 20th century, states that individuals determine whether they want to partake in certain beneficial personal investments, such as attending school or exploring training opportunities (Klevmarken & Quigley, 1976). These investments allow workers to develop skills and other characteristics that make them more appealing to employers and more productive, which results in higher pay. This theory is widely applied to the analysis of income inequality, and a compensation analysis (e.g. a regression equation predicting employee pay) derived from it would generally only include variables tied to individuals’ decisions, such as education and experience (e.g. Ang, Slaughter, & Ng, 2002). In other words, predictors related to organizational or institutional factors, such as job level, would be excluded (Ananda & Gilmartin, 1991).
Despite the prominence of the Theory, it has limitations. Specifically, human capital-based analytics that remove institutional variables, such as job title, may be incomplete or misspecified. While the theory focuses on individuals and their characteristics, it assumes that wages vary based on the quality of a job, which represents an adherence to traditional economic concepts. For example, if an organization is offering a dangerous job, it would be willing to pay more than an employer with a relatively less dangerous job in order to offset the increased danger, an idea known as compensating differentials (Flamholz & Lacey, 1981). Put differently, the characteristics of the job itself, and not simply of the employee, are relevant when wage determinations are made. Ultimately, an analyst relying solely on Human Capital Theory must be prepared to determine if a human capital model omits important institutional characteristics that impact pay.
References
Ananda, S., & Gilmartin, K. (1991). Inclusion of Potentially Tainted Variables in Regression Analyses for Employment Discrimination Cases. Industrial Relations Law Journal, 13, 121-152.
Ang, S., Slaughter, S., & Yee Ng, K. (2002). Human capital and institutional determinants of information technology compensation: Modeling multilevel and cross-level interactions. Management Science, 48, 1427-1445.
Becker, G. (1975). Human Capital. New York. NY: National Bureau of Economic Research.
Flamholtz, E. G., & Lacey, J. (1981). The implications of the economic theory of human capital for personnel management. Personnel Review, 10, 30-40.
Klevmarken, A., & Quigley, J. M. (1976). Age, experience, earnings, and investments in human capital. Journal of Political Economy, 84, 47-72.
By Cliff Haimann, Consultant, and Julia Walsh, Consultant, at DCI Consulting
