On a Firm’s Optimal Response to Pressure for Gender Pay Equity

Abstract

We present a theory of how a rational, profit-maximizing firm would respond to pressure for gender pay equity by strategically distributing raises to reduce the pay gap between its female and male employees at minimum cost. Using formal analysis and pay data from a real employer, we show that (1) employees in low-paying jobs and whose pay-related observables are similar to those of men at the firm are most likely to get raises; (2) counterintuitively, some men may get raises, and giving raises to certain women would increase the pay gap; and (3) a firm can reduce the gender pay gap as measured by a much larger percentage than the overall increase in pay to women at the firm. We also identify the conditions under which a firm could “explain away” a gender pay gap using other pay-related observables, such as job category, as well as the conditions under which this strategy would backfire. Our paper helps explain some empirical puzzles, such as the tendency for some men to get raises after gender equity pay reviews, and yields a rich set of implications for empirical research and practice. The online appendix is available at https://doi.org/10.1287/orsc.2018.1248 .

Female CEOs and the compensation of other top managers.

“…Total compensation is the natural log transformation of a top manager's total compensation, including inter alia, salary, bonus, and grants of stock and options. The log transformation reduces skewness and is standard in regressions using compensation as the dependent variable (Anderson et al, 2019). Note that the log transformation means that coefficients on independent variables can be interpreted as approximate percentage changes (see Dranove, 2012 for a detailed explanation).…”

“…This methodology produces identical coefficient estimates to running a separate regression for female managers and a separate regression for male managers, while allowing us to cluster standard errors at the firm level using a common covariance matrix. Our methodology is also statistically equivalent to interacting every variable in the regression (including the firm fixed effects) with a gender indicator variable (Anderson et al, 2019) but, we believe, is easier to read than that alternative. 6 We conducted several robustness checks related to the control variables, including regression permutations that only used firm and year effects, excluded job categories, only used job, firm, and year effects, and excluded year effects.…”

The gender equity gap: A multistudy investigation of within-job inequality in equity-based awards.

“…we provided evidence about one area-equity-based awardswith substantial impact on workers' wealth. Thus, our findings suggest an area that practitioners seeking to reduce gender inequality can focus on, with a potentially large effect (Anderson, Bjarnado ˇttir, Dezsõ, & Ross, 2019;Banks et al, 2016).…”

“…For example, transparency (Arnold & Fulmer, 2018; Smit & Montag-Smit, 2019) and worker unionization (Rios-Avila & Hirsch, 2013) have been argued to be effective means of reducing inequality in traditional forms of pay. Likewise, Anderson and colleagues (2019) suggested firms might optimally allocate raises to strategically reduce gender pay gaps at minimum cost (which can help address inequality in the near term). Examining how these, as well as other practices, affect the gender equity gap is an interesting avenue for future scholars to pursue.…”

Optimal data-driven hiring with equity for underrepresented groups

“…This is the criterion adopted in our paper. It has been criticized in the literature on the grounds that it can be easily gamed (Dwork et al, 2012;Kearns et al, 2018), for example by hiring at random from the protected group, or through other manipulations that reduce population-wide measures of disparity (see Anderson et al, 2019 for some examples in the context of the gender pay gap). In our setting, however, these issues are overcome by the presence of a decision problem with an economic objective function: among all policies that are fair in this sense, we wish to find the best-performing one with respect to the employer's interests.…”