Blue and red shirts made of paper with piles of coins in between. The concept of gender employment and wage gap in the enterprise.

How companies can close the gender pay gap as efficiently as possible and what that may mean for pay at their firms, according to UF Warrington research

GAINESVILLE, Fla. –  With women making up almost half of the American workforce, it’s no surprise that more and more companies are looking for a way to respond to disparities in the pay between their male and female employees. Add in increased pressure from the government, media and employees, and companies are under growing scrutiny when it comes to not only to closing the gender wage gap, but also to ensuring they do so fairly.

David Gaddis Ross

R. Perry Frankland Professor David Gaddis Ross

“After decades of progress, firms still seem to have trouble ensuring gender equity despite the benefits to firm performance from doing so,” writes Warrington’s R. Perry Frankland Professor David Gaddis Ross, along with colleagues David Anderson of Villanova University and Margrét V. Bjarnadóttir and Cristian L. Dezső of the University of Maryland Robert H. Smith School of Business. “Nowhere is this more in evidence than with the ‘gender pay gap’ — that is, the empirical regularity that women are paid less than men despite equivalent qualifications and job responsibilities.”

In response, many companies are hiring pay consultants and HR experts to determine whether they have a gender pay gap and, if so, address it by strategically allocating raises to eliminate the gap without ballooning their wage bill. As part of a broader research program on how firms can identify and address their gender pay gaps as fairly and efficiently as possible, Ross and colleagues have developed a mathematical algorithm for allocating raises to close a gender pay gap at minimum cost. They have also explored what kinds of employees are likely to get a raise if firms focus on efficiency rather than fairness in closing or avoiding gender pay gaps.

A crucial piece of this research is a new statistic called “differential influence,” which Ross and colleagues derived. Differential influence determines which employees have the biggest impact on a firm’s gender pay gap and are therefore candidates for raises. Ross and colleagues use this statistic to derive a number of results, some of which are counterintuitive.

First, they find it is more efficient to give raises to employees at the bottom of a firm’s salary distribution, especially low-wage women. In consequence, at firms that are trying to close their gender pay gaps efficiently, the pay gap tends to be narrower among low-paying jobs than high-paying jobs, a widely observed phenomenon.

Second, and counterintuitively, their research shows it can increase the pay gap if you give certain women raises and decrease the pay gap if you give certain men raises. Yep, you read that right: you can actually make women appear to be paid more fairly relative to men by paying certain men more! The reason is that the statistical analysis used to measure a gender pay gap tries to explain each employee’s compensation as a function of job attributes (e.g., functional area), personal attributes (e.g., years of schooling), and of course gender.

Having a gender pay gap essentially means that gender is doing a lot of work to explain pay beyond what the job and personal attributes explain. It turns out that in the statistical analysis, some men have a big influence on how much job and personal attributes contribute to explaining pay and a small influence on how much gender explains pay. If you give such a man a raise, it increases the explanatory power of these attributes, and that necessarily makes gender explain less.

Despite these anomalous consequences, it is understandable why some employers focus on cost in rectifying their gender pay gaps. In a word, they do it to save money, as Ross and colleagues explain.

“We show that by strategically allocating raises to its employees, the employer could eliminate its gender pay gap for less than half the cost of doing so by giving equal percentage raises to all its female employees or by using a standard wage equilibration model.”

Ross said that the current focus of their ongoing research program, which includes working with companies on addressing gender pay equity, is on balancing cost and fairness.

This research, titled “On a Firm’s Optimal Response to Pressure for Gender Pay Equity,” is forthcoming in Organization Science.

Interested in learning more about this research? Read about how it can help boost inclusivity in your organization in this article from Forbes, check out the researchers’ website PayAnalytics, and follow Ross on Twitter @GaddisRoss for additional insights.