How women managers close the gender wage gap

Alex Bryson, John Forth and Nikos Theodoropoulos 

In the last few decades there has been what has been called ‘a grand gender convergence‘. Women now outperform men in educational attainment and are closing the gap in labour market experience. 

These trends are common across much of the developed world. Yet a gender wage gap persists, with women earning substantially less than men. While the gap has been closing the rate of convergence is slow. Britain’s 1970 Equal Pay Act came into force in 1975 yet in 2017 the raw gender wage gap in median hourly pay was 18.4 per cent.

What can be done? In our study we ask: 

Does the share of female managers in the workplace affect the size of the gender wage gap? 

Workplace managers are the ones in positions of authority who make decisions affecting the pay of men and women, either directly through wage offers, or indirectly through promotions and the like. The question is interesting because, as Chart 1 shows, there is substantial variation in employees’ exposure to female managers in British workplaces.

Chart 1. Distribution of Employees by Female Share Among Managers in their Workforce, by Gender of Employee

We suspect that the share of female managers at a workplace might influence the gender pay gap in a number of ways. 

Perhaps when a greater proportion of those in authority are women:

•  they are more likely to take women’s interests into consideration when it comes to access to training slots; 
• they may make additional efforts to accommodate preferences for work-life balance;
•  they may be more scrupulous in implementing equal treatment policies; and 
• they may act as mentors or role models for women coming through the ranks.

We tested this proposition in a newly published paper. We used data from 3,236 workplaces and 39,966 of their employees, collected in the Workplace Employment Relations Surveys of 2004 and 2011. 

The results are striking. As indicated in Panel A of Chart 2 men’s mean hourly pay falls with the rising share of female managers, while women’s rises, consistent with a reallocation of limited resources from men to women when the share female managers rises. Panel B shows the net effect where the size of the gender wage gap declines as the share of female managers rises. The gap is no longer statistically significant when 90 percent of workplace managers are women, a scenario that obtains in 12 percent of workplaces.

Chart 2. The Impact of Share Female Managers on the Gender Wage Gap

Note: Vertical lines are 95 percent confidence intervals. 

In analysis shown above, we define managers as those occupying senior managerial positions in the workplace (general managers, finance managers, production managers and so on). However, we obtain similar findings when we expand our definition to include line management and supervisory positions.

The association between the share of female managers and the size of the gender wage gap is not only apparent when examining workplaces at a specific point in time. If we follow workplaces over the period 2004-2011, we also find that the gender wage gap falls as the share of female managers rises in a workplace (and vice versa).

But is the association actually caused by the impact of female managers on the gender wage gap? Our first piece of evidence uses a statistical technique called instrumental variables (IV) estimation, commonly used in economics to identify causal effects in non-experimental data. The analysis accounts for non-random exposure of workers to female managers and the results suggest that raising the share of female managers in a workplace does have a causal effect in reducing the gender wage gap.   

Our second piece of causal evidence is the identification of a potential mechanism by which female managers affect the wages of men and women: the use of individual performance pay systems. Panel A of Chart 3 shows the influence of the share of female managers on the gender wage gap for employees paid for performance. We see that the convergence in the gender wage gap is achieved when the share of female managers is about 40 percent. It begins to flip in women’s favour when the share of female managers is about 60 percent – although this gap in favour of women does not reach statistical significance. For workers in fixed term pay contracts (Panel B) the closure of the wage gap is not so striking. It is eliminated when the share of female managers is around 50%, but a gap in favour of women does not emerge beyond this threshold.   

This is evidence consistent with the proposition that women are more likely to be paid equitably when managers have discretion in the way they reward performance and those managers are women.

Chart 3. The Closure of the Gender Wage Gap for Performance Pay and Fixed Pay Workers

Note. Vertical lines are 95 percent confidence intervals. 

Policy analysts may be tempted to conclude that one way to close the gender wage gap is to enforce increases in the share of females in managerial roles through quotas. But the current empirical literature on quotas for top executives cautions against simple policy responses, since studies find few positive spill-overs of executive quotas on female representation and female wages lower down the corporate hierarchy. Such initiatives may have limited effects without corresponding action on the mechanisms through which managers actually affect wage outcomes for their employees. 

Note: Alex Bryson would like to thank the Economic and Social Research Council for funding (grant number ES/S012583/1)