The National Association of Colleges and Employers has published a study on starting salaries for female and male college graduates, and it raises more questions than it answers. Matt Yglesias favorably cites an Ann Daly op-ed column on the report. Daly writes:
A woman graduating with a bachelor’s degree last year earned a median starting salary of $36,451. For a man, it was $44,159. When you calculate a lifetime of percentage raises and compound interest, that nearly $8,000 difference is staggering.
As demoralizing as the findings of “Gender and College Recruiting” might be for this year’s female grads, its implications for future generations of women in the workplace are downright alarming. NACE’s analysis, which painstakingly isolates a systematic gender effect by taking into account the differential salary levels among majors and then comparing salaries within the same major, gives lie to the conventional wisdom that paycheck parity will somehow materialize for women with the mere passage of time.
This immediately sends up a number of red flags. Painstaking or no, it would be good to know about the particular jobs graduates are taking. Given the state of the labor market, there are more than a few recent grad math majors who are working as tutors or baristas, etc. Recent cohorts of college graduates are disproportionately female. Of the men who graduate, are there other factors that haven’t been painstakingly isolated? For example, if it is relatively common for men to not make it from high school to college, and relatively common for men who start college to not finish, might there be something distinctive about the men who make it through that makes them more likely to secure remunerative employment than the modal graduate? Might they apply to more jobs, and to jobs in regions located further from home that are more lucrative?
The NACE study breaks down career-oriented majors by the proportion of women and men. How do men and women fare in fields with a big gender disparity? The Health Sciences major is 85.2 percent female. If my intuition is right, this would suggest that the men who major in Health Sciences are pretty determined people who are comfortable working in a female-dominated field, and they might thus fare better in the job market. It turns out that male Health Sciences grads earn considerably more than female Health Sciences grads. Women are only 17.8 percent of Computer Science grads and only 18 percent of Engineering grads. Among Computer Science majors, women are at a disadvantage when it comes to starting salaries while they’re at an advantage among Engineering majors. Are female Computer Science majors concentrated at the top of the heap, are they distributed evenly throughout, etc.? (That is, is the situation analogous to my baseless intuition regarding Health Sciences majors or not?) Call me crazy, but I think we’d need to know the answer to these questions.
Matt preemptively responds to any questions of this kind by writing the following:
Now obviously some people are going to be very resistant to this conclusion. They’ll think that in a competitive labor market with many employers and many workers, employers who discriminate against women in their salary offerings will be at a disadvantage. No firm will want to disadvantage itself in this way, thus the discrimination shouldn’t exist. Consequently, this apparently effect is almost certainly due to some other variable that’s not accounted for. So it’s worth pointing out that by this logic, the gender disparity in employment that existed in 1961 wouldn’t exist either. But obviously it did.
One could just as easily say that some people are going to eagerly embrace this conclusion without subjecting it to scrutiny.
Consider the questions I raised above: one can easily imagine culture playing a role. One can imagine a scenario in which women take family obligations more seriously than men, and that this reflects a kind of gender bias. And so a woman — say a woman of South Asian origin — might be somewhat more inclined to limit her job search to regions that are close to her parents than a man. Gender is definitely playing a role, but it is operating at a somewhat different level. If we assume that cultural practices and assumptions change over time, it is really, really easy to understand the sources of the gender disparity in employment in 1961. Back then, there were also more formal barriers to the employment of women.
I went to the NACE report and found a number of interesting findings:
For all but two of the majors where there are sufficient data to compare male and female salary offers, the median starting salary offered to male graduates was higher than for female graduates. The only two anomalies were engineering and generic liberal arts/humanities majors. Engineering is explained by the fact that the discipline has such a small percentage of women graduates that the women who do graduate from this field are highly sought after “commodities” and command a premium price for their services. The higher salary offers to female graduates in engineering have been showing up for the past couple of years in salary surveys of new college graduates.
Liberal Arts/Humanities majors and Engineering majors are described as anomalies. Another possibility is that a larger number of Engineering majors go into engineering-oriented careers, while large numbers of Visual & Performing Arts majors see greater dispersion, e.g., some wind up in very low-paying professions that are essentially apprenticeships, others leave the field altogether to work as golf pros. The gap between male and female Accounting majors is trivial, which could reflect the fact that Accounting majors tend to enter their chosen profession. The gap between Mathematics majors is strikingly high. Again, some Math majors might become teachers while others work in financial planning.
I’m certainly not suggesting that there isn’t a pay gap. Rather, I’m suggesting that NACE’s data doesn’t give us anything approaching a complete picture, and I’m surprised that Edwin Kroc didn’t underline this fact.