Just decades ago, few women pursued MBAs. Today, in some schools, women make up almost half of those earning management credentials. Nonetheless, it has been widely reported that men graduating with an MBA degree earn up to a ﬁve-ﬁgure premium in starting salary over women graduating with the same degree. Why is this the case?
Is it because women and men choose different aspects of managerial careers? Different industries or companies? Different jobs? What causes this gap? “It’s not unequivocal,” says Adina Sterling, an assistant professor of organizational behavior at Stanford Graduate School of Business. “In thinking about this problem, internships were a nice setting to look for one possible answer.”
With Roberto Fernandez at MIT, Sterling analyzed salary data from two consecutive business school classes: what they earned before school, what they earned during a summer internship, and what they earned after graduation. The researchers split this information along gender lines and, critically, compared the arcs of men and women who went on to work for the companies where they interned against those who worked at a different company. The researchers also collected detailed information on the students’ GMAT scores, grade point averages, chosen industries, and job functions, along with aspects of their careers prior to the MBA, like salaries and years of work experience.
“Lo and behold, we find some evidence that internships have a leveling effect on initial managerial salaries,” says Sterling.
A Clever Test to Reveal Employer Bias
A simple logic underpinned the study and its findings. “If the way employers are assessing the value of people is accurate at the point of hire, then there should be no difference in the relative salaries they set for men and women after a period of observation,” says Sterling. If companies have an accurate, quality-based reason to reward men more than women, then the gender gap shouldn’t change after watching interns perform for several weeks. “But if, after observation, there are adjustments for one group more than another, then perhaps companies are relying on stereotypes or biased thinking when they hire outside of internships.”
Adjustments are exactly what Sterling and Fernandez found. (Sterling added that they could not and do not measure stereotypes or bias in the study; but they do observe behaviors that would be consistent with the notion that the salaries of female managers are discounted relative to male managers in the absence of direct information.)
Prior to business school, women and men made, respectively, $68,685 and $75,560 annually. During their internships, women and men earned annualized salaries of $76,272 and $83,676, respectively. After graduation, women received offers of $100,649 on average next to men’s $108,364. However, women and men who worked for companies where they held internships received offers of $108,600 and $108,196, respectively, meaning women earned $400 more on average (though not a statistical difference).
And while Sterling and Fernandez were not able to provide a definitive explanation for this leveling effect, they marshal evidence that internships serve as tryouts in a number of ways. They find evidence that these tryouts help companies make more informed appraisals of candidates and, therefore, more informed salary offers.
“Our minds make quick judgments all day long; in milliseconds we assess people,” Sterling says. “But social psychological evidence suggests what tends to happen is that the more information people have, the more they are willing to depart from stereotypical beliefs and update their opinions.” In the context of employee hires, internships provide this information.
The Promise, and Limits, of Internships
Fortunately, implicit in the diagnosis of the problem — employers may be biased in their salary offers — sits a solution: Offering paid internships that channel interns toward full-time managerial positions can help close the gender wage gap. Importantly, narrowing this gap just as people are launching their careers provides long-term benefits. “If you’re first hired at a lower salary,” says Sterling, “then that figure can follow you over a lifetime.”
And while internships present a possible solution to reducing the gender wage gap, Sterling called for more research on the issue and emphasized that internships likely must be paid, otherwise they would heighten economic inequities in the employment marketplace. She also indicated two key limits to the findings. First, she and Fernandez studied a relatively rarefied population: MBA students—a setting in which 97% of individuals held internships. Outside of this context, how often are internship opportunities available? And second, if they are available, is there bias in the broader market that sorts prospective interns by gender? “Our findings are a moot point if there is some screening process in an earlier stage,” Sterling says.
These limitations aside, the main results both spotlight possible bias in the marketplace and offer a path around this bias. “If you can get people in the door and give employers time to assess them, then this study suggests it’s likely you’re going to see some differences in outcomes,” Sterling says.