Back in December, journalists were surprised when President Obama called exclusively on female reporters during a press conference. The president is not the only one explicitly focusing on getting women to have more of a say. Google’s technical employees recently found this message in their inbox: “I wanted to update everyone on our efforts to encourage women to self-nominate for promotion… We know that small biases—about ourselves and others—add up over time and overcoming them takes a conscious effort.” What is this about? Can women not speak up for themselves?
The answer is complicated. Much research suggests that women who “lean in” too much are likely to experience backlash. The Yale social psychologist Victoria Brescoll, for example, found that when a group of professional men and women was asked to evaluate the competence of male and female chief executives, they appraised “speaking up” differently depending on the gender of the speaker. Male executives were rewarded with higher ratings for speaking up while their female counterparts were punished for such assertiveness and received lower competence ratings than their quieter peers. Women do not prefer saying less: They anticipate the treatment they will receive when they say more.
field experiment, they observed that women were more likely to respond to a job advertisement for an administrative assistant position when the ad made it clear that wages were negotiable. In contrast, men preferred the ambiguous version of the ad, which did not indicate whether salary was negotiable. And it did not hurt them: Men were substantially more likely than women to negotiate their pay in jobs that left the negotiation of the wage ambiguous.
Removing uncertainty from situations like this can help women’s compensation catch up to men’s. In a survey of recent MBA-graduates, my colleagues Hannah Riley Bowles, Linda Babcock, and Kathleen McGinn found that in fields with high degrees of ambiguity around salary negotiation, the gender gap in initial salaries reached almost $10,000 per year, favoring men, even when controlling for a wide range of other salary predictors such as job function, pre-MBA work experience, job market activity, geographic location and job preferences. In contrast, in low-ambiguity fields, no significant gender pay gap was observable.
Transparency around the acceptability of negotiations is just one sort of transparency. Bringing transparency to other aspects of professional life would also help. For example, transparency has helped increase the fraction of women on corporate boards of the United Kingdom’s FTSE100 companies, from 12.5 percent in 2011 to more than 25 percent at the beginning of 2016. Today, there are no companies with all-male boards left among the FTSE100, a first in the history of the London Stock Exchange. Tracking numbers, setting goals and showing the progress, or lack thereof, helped move the needle without having to use quotas.
The use of big data in human resource management, often referred to as “people analytics,” is another way that greater information can bring about more fairness, as it enables companies to track biases much more precisely. Relying on data, MIT was one of the first universities to discover gender differences in space, resources, awards and salary. The data told the tale of what later was coined “performance-support bias” where female faculty did not receive the same kind of “support” as their male counterparts. Data being a “very MIT thing,” as the dean of the School of Science remarked, allowed the university to point to what was broken and made it easier to fix.
The Harvard Kennedy School, where I work, has similarly developed an elaborate measurement system to make the invisible visible and compensate accordingly. Too often, the provision of public goods that makes everyone better off goes unnoticed. Someone has to organize the speaker series, go the extra mile and support our students in their job searches, or serve on search and promotion committees. These services for the community matter—not just in academia but in pretty much any organization. Most partners in law or consulting firms would rather have someone else mentor their summer associates, even if everyone benefits from great future colleagues. It turns out that women are more likely to provide such services.
A clever experiment by Lise Vesterlund, Linda Babcock, Maria Recalde, and Laurie Weingart brought this to the forefront. Motivated by their own experiences as female faculty members, they wanted to explore whether women had a harder time saying “no” when asked for a favor, or what they referred to as “nonpromotable tasks.” They designed an experiment to find out why some people do the tasks everybody benefits from, but bring little to no credit to the people who actually do the work. It turns out, in groups made up of either men or women only, men and women both stepped up at about the same rate. But in groups made up of both men and women, the pattern changed. Women stepped up far more often. Nobody was surprised by this. It conformed to both men’s and women’s expectations of what the right thing to do was. Belonging to a social category expected to be more “caring” had turned into a liability for women.
By measuring and compensating accordingly, organizations can bring such inequities out into the open and assign value to tasks that are important to them. As the saying goes: What does not get measured, does not count. As a design, this is almost a pure win-win. True, not everything is quantifiable, which is why every few years the HKS faculty has a discussion about whether our workload system measures and rewards effort adequately at the Kennedy School.
Nobody is immune from self-stereotyping. Seeing is believing, and as long as people see more women in care-giving and more men in bread-winning roles, they are more likely to associate care with women and work with men. And of course, women’s reluctance to speak up can also turn into a self-fulfilling prophecy where people associate “speaking” with men and “listening” with women. To break the vicious cycle, those in power need to design systems that promote fairness—by calling on women, increasing transparency, and measuring and compensating all contributions. It works.