What HR Managers Can Learn from Nobel Prize Winner Claudia Goldin’s Explanation of the Gender Pay Gap
What did this year’s Nobel Prize in Economics say about the gender pay gap?
The 2023 Nobel Prize in Economics was awarded to Dr. Claudia Goldin for research about women’s earnings and labor force participation. A major focus of her research is about the magnitude and causes of the Gender Pay Gap. Her research shows that while there is a Gender Pay Gap in the group of women she studied, much of the Gender Pay Gap is explained by the tendency of companies “to disproportionately reward individuals who labored long hours and worked particular hours.”[1] Goldin finds that both women and men who work disproportionately long hours and who do not request/require “leaves of absence” receive higher compensation per hour than other employees in similar occupations, and it happens to be that men are willing to work these types of high reward schedules more often than women, resulting in the Gender Pay Gap. Goldin finds that some types of jobs benefit greatly from work delivered during particular hours to provide, for example, continuous attention of an employee over extended periods or during irregular peaks labor demands, and these jobs have the greatest Gender Pay Gap.
What can employers take from these findings?
For Pay Equity Analysis: Dr. Goldin’s analysis reminds HR managers that identification of the right explanatory factors is critical to a valid pay equity analysis. Dr. Golden found that unusually long hours may be an important and valid explanatory factor for the Gender Pay Gap in some jobs, but would do little to explain the Gap in others. HR managers should consider including a measure of hours each employee worked and/or the employee’s flexibility to work unexpected or long hours in pay equity analyses. Whether an employee of any demographic background is willing to work particular, difficult or extended hours may be considered a valid, gender-neutral explanatory variable for differences in compensation. You likely have data about hours worked for hourly workers. Of course, overtime wage requirements will cause overtime hours to be paid a higher hourly wage. If you are using regression analysis[2] to control for factors that explain differences in women’s and men’s compensation, consider adding a separate variable for the number of overtime hours each hourly employee works.
In some industries, such as legal, accounting, and consulting among others, salaried employees might submit timesheets with hours worked. Again, consider including both the number of hours worked up to the company’s standard hours per work week, e.g., 35 or 40 hours, and a second, separate variable indicating the number of additional hours worked over that weekly standard as variables in the regression. This will allow the regression to measure whether employees that work a more extreme number of hours are rewarded at a higher rate per hour than employes that work less extreme hours, as Dr. Goldin suggests is the case in some jobs. And it may be an important explanatory variable for the difference in compensation between men and women.
In other companies, where salaried employees do not track their hours, it may be possible to find other objective measures of work above and beyond the standard effort. This could include a measure of productivity, such as sales, units produced or perhaps even headcount managed, indicating an unusually high level of effort. Here again, consider including a measure of this productivity for each employee in your regression along with a second variable indicating the level of the employee’s performance considered above normal on these metrics. These additional variables for extra productivity and performance may go a long way to explaining the Gender Pay Gap in some jobs, according to Dr. Goldin.
For HR Management: Dr. Goldin’s work shows that some workers, predominantly women, are willing to accept lower wages in exchange for job requirements that fit their lives better. Broadly interpreted, Dr. Goldin’s work demonstrates that employers can reduce payroll costs by configuring jobs in ways that are more convenient, acceptable, and comfortable for workers. To reduce payroll costs, companies should configure jobs in ways that employees find less burdensome, but which at the same time have limited effects on overall employee productivity. Dr. Goldin’s research recognizes that this may be more easily done with some jobs than others. As an example, Dr. Goldin points out that pharmacists are able to limit their work hours without a reduction in productivity to the pharmacy because pharmacists are comparatively interchangeable over the course of a day. As long as the pharmacist completes a prescription, the replacement pharmacist can take over the job with limited downtime or loss of productivity to the pharmacy. Pharmacists’ hours are relatively exchangeable (“fungible”) across pharmacists. Therefore, allowing individual pharmacists to limit hours or “share” a job may be desirable to the employee and have little impact on productivity of the company. There may be jobs in your company that have similar characteristics where “extreme” hours are not required or perhaps where specific work schedules are not needed as long as the quantity of work is accomplish in a given timeframe. In contrast, other jobs like the legal/litigation group at your firm may require employees to pay close attention to their work for extended periods and during inconvenient days. Here you may not be able to benefit your company by allowing too many members of this team to work on limited, highly scheduled hours without degrading the effectiveness of the group. Look for these differences in jobs for opportunities to identify jobs that can be made more convenient and comfortable where it is efficient to do so. It may reduce your payroll burden.
Dr. Golden’s Nobel Prize winning work reminds us that many workers, predominantly women in her study, will accept lower wages for job requirements that fit their non-work lives better. More broadly, Dr. Golden’s work reminds us that the more we understand about the explanatory factors behind pay equity, the better we can configure jobs that fit our workforce and the more productivity we can get for our payroll.
[1] Claudia Goldin, “A Grand Gender Convergence: Its Last Chapter”, American Economic Review 2014, 104(4): 1091–1119
[2] https://www.dol.gov/agencies/ofccp/directives/2022-01-Revision1; Regression analysis for pay equity analysis can be performed with EquityPath software. See http://www.equitypath.com