Why do women continue to earn less money than men — approximately 20% less, according to some estimates — and what can be done about it?

At least half the pay gap reflects the fact that women tend to work in different kinds of occupations and industries than men, a phenomenon known as “gender segregation.” Understanding the causes of that gender segregation is a key part of any attempt to address the pay differential.

Wharton management professor Matthew Bidwell and Roxana Barbulescu, a management professor at McGill University in Montreal, set out to understand the causes of gender segregation by taking a different approach than studies that typically look at variances in the kinds of jobs that men and women choose, or at the decisions made by employers during the job application process. 

Bidwell and Barbulescu opted instead to look at job applicants themselves to determine whether the decisions they make during their job search process have a significant impact on which offer they accept. Their results are presented in a paper titled, “Do Women Choose Different Jobs from Men? Mechanisms of Application Segregation in the Market for Managerial Workers,” forthcoming in the journal Organization Science.

“Much of the debate over earnings has focused on the idea that there are barriers to women getting certain kinds of jobs, and that a big part of this is due to subtle and not so subtle discrimination on the part of employers,” says Bidwell. “But most of the available data looks at the jobs women end up in, which reflects a series of decisions by both the employee and employer.” The challenge was to separate out data that deal primarily with how women view the employment landscape even before starting the job application process. Do those views, for example, lead women to systematically choose different, and lower-paying, occupations than their male counterparts?

The two researchers analyzed data on 1,255 men and women entering the job market as they were graduating from a large, elite, one-year international MBA program. Such a group is far from representative of the population at large. However, “studying MBA students is particularly valuable for exploring segregation into some of the best-paid and most influential jobs in society, which are the kinds of jobs in which women have traditionally been under-represented,” the authors note in the paper.

Barbulescu surveyed the students about their job interests at the beginning of the MBA program, and then again at the end in order to find out what kinds of jobs they applied for, where they got offers and what jobs they ultimately accepted. 

The researchers’ main finding was that women were significantly less likely to apply to Wall Street-type finance jobs, somewhat less likely to apply to consulting jobs, and more likely to apply to jobs in general management, most notably internal finance and marketing. Not coincidentally, the finance and consulting jobs that women avoided were also the ones that were most highly paid.

No surprises there, but the researchers dug deeper to see what might explain these results. To start, they broke down the different influences on job search decisions into three different factors: applicants’ preferences for specific rewards from their jobs, such as money or flexibility; the ability of applicants to identify with particular kinds of jobs, which often reflects how compatible those jobs are with other ways the applicants see themselves; and the applicants’ expectations that an application could succeed.

The researchers argue that each of those factors might be influenced by gender role socialization, which shapes our basic beliefs about the behaviors that are most appropriate for men versus women, and about the kinds of skills that accompany those behaviors. For example, if women are expected to play different roles in the workplace and at home than men, then they may also look for different rewards from their work, such as pay, intellectual challenge, flexibility, work/life balance and so forth.

Four Nights in a Hotel

Specifically, the researchers looked at expected work/life satisfaction with regard to 19 different job types, and found that women were significantly less likely than men to apply for jobs where work/life satisfaction ranked low. “This explained why women weren’t applying for consulting jobs,” says Bidwell. “The hours are not that much worse than investment banking jobs, but the expectation is that you will be staying in a hotel four nights a week. And that doesn’t change. With investment banking, you might work very hard, but you usually sleep in your own bed, and the hours tend to trail off as you get more seniority.” 

The second decision factor shaping applications is how people identify with different jobs. Bidwell and Barbulescu found that women identified the least with stereotypically masculine jobs, and they tended to apply to industries that usually employ a higher proportion of women. The third decision factor is whether individuals believe their applications for certain jobs will be successful: It may not make sense for applicants to pour a lot of time and effort into applications for jobs they do not expect to get.

Bidwell and Barbulescu found that at the beginning of the MBA program, men and women showed the same level of confidence that they would get an offer for a specific job in most of the fields they might apply to — except investment banking. There are good reasons that women might have lower expectations of job offer success in stereotypically masculine jobs, says Bidwell, and no industry has more of a macho image than investment banking. “Women just didn’t think they would get jobs there, so they didn’t apply,” he notes.

Equally interesting, says Bidwell, is that when women did apply to investment banking jobs, they were just as likely to get them as the men who applied.

“Our research shows how hard it is to bring about change,” Bidwell adds.” If you tell employers to stop discriminating, it doesn’t mean you will end up with greater access for women to better, higher-paying jobs. Instead, it’s about changing perceptions of culture. You can imagine that if you have a job that is seen as highly macho and aggressive, and you recruit those kinds of people — mainly men — then these perceptions and stereotypes become self-fulfilling. It’s a much more insidious way in which jobs become gendered.”

The researchers emphasize in their paper that “even when there are no gender differences in the likelihood of receiving a job offer, this does not imply that employers do not influence gender segregation.” Indeed, employer decisions may affect applicant behavior “in ways that we could not detect.” For example, they cite the climate and recent litigation history of some of the sectors they studied, primarily finance, which may have increased the pressure on employers to hire more women, but doesn’t necessarily mean they will promote them into the same senior level positions as men.

The behavior of employers — and the control they often exert over the workplace — can clearly affect whether women apply for jobs with their companies. For example, the researchers write, “practices that reduce conflicts between work and family demands could reduce” segregation, and “interventions in the way that jobs are structured and role behaviors enacted to emphasize either masculine or feminine stereotypical attributes could also” lessen segregation. But that is not an easy sell. For instance, as the researchers note, “workplaces with fewer women face less pressure to adapt their working styles to accommodate family demands” — an example of how segregation becomes self-perpetuating.

At the same time, “addressing these deep-seated organization issues, alongside the more common question of how hiring decisions are made, could be critical for increasing female participation in some of the best-paid jobs in society,” the researchers add.

Token Gestures

According to Bidwell, this research paper is one of the first demonstrations that much of the segregation in the job application and hiring processes “happens because of how people apply for jobs rather than because of employer behavior further down the line. And that, in turn, reflects what jobs women are able to identify with, and where they think they will be hired.”

When he talks to senior managers, particularly HR people, Bidwell says it’s clear that attracting and retaining women are important issues for them, especially because they know that women “are a big part of the talent pool out there, and they also know that customers are becoming more diverse,” reflecting a need for the same diversity in companies’ employees.

“But it’s not just a question of sticking more women in the company brochures, or having more women be part of the company’s on-campus recruiting,” he adds. “It’s a question of trying to change the culture, job perceptions and kinds of behavior that people exhibit. My sense is that everybody talks about this, but I’m not sure how serious managers are about the tradeoffs they would have to make in order to attract and retain more women — such as reorganizing basic work processes to allow for more reasonable work hours, or changing the norms about acceptable behavior. That’s a much harder conversation to have.”

Bidwell and Barbulescu acknowledge that their research is based on a very specific population — highly skilled, highly paid, competitive MBA graduates — and they suggest several areas for future research. Other studies could examine whether these theories hold true for less-skilled and lower-paying occupations and/or for traditionally feminine settings such as nursing schools. Additional research might look at the process by which applicants learn about jobs, given that some recent studies conclude that “differential access to social networks may shape gender differences in job applications.” Researchers could also look more closely at the three broad categories that Bidwell and Barbulescu focus on — finance, consulting and general management — parsing out entrepreneurship, for example, which they say has one of the lowest levels of female applications of any job type.

Coincidentally, Bidwell and another Wharton professor, Ethan Mollick, recently conducted a separate survey of Wharton MBA graduates. Among their findings: On average, people in investment banking and hedge funds 10 to 15 years out were making two to three times the amount of money as people in general management. That gap highlights the tremendous economic consequences of women’s decisions not to apply to investment banking jobs. On the other hand, the investment bankers had the lowest job satisfaction of any other group. “We surveyed them in 2011,” Bidwell says. “It was a particularly miserable time for them, which may have affected the results.”