STEM: Still No Shortage

On a myth that just won’t die

Freddie deBoer
I. M. H. O.
Published in
10 min readNov 27, 2013


This morning, I read a deeply depressing story by the formidable Rebecca Schuman, detailing how two colleges are opting to cut academic programs rather than the administrative or facilities costs that actually drive college costs. (Schuman’s recent work for Slate has been excellent, in general.) The decision is curious on a number of levels. Faculty salaries have been more or less stagnant for decades, with much of the growth attributable to the few star professors in prestige fields who generate a great deal of grant money. (This is without even discussing the deplorable, flatly immoral treatment of adjuncts and contingent faculty, whose low pay and lack of benefits are to my mind the great moral failing of the American university system today.) The spending on dorms, gyms, and food courts in a stagnant economy is, well, insane. Administrative costs have absolutely skyrocketed, as universities stuff more and more deans, assistant deans, provosts, and all manner of directors and coordinators into their buildings. Athletics departments, despite reputations as money makers, are enormous financial drains on the system as a whole. And yet academic departments— you know, the purpose of a university— are on the chopping block. Why?

I can’t help but think that a big contributor to this phenomenon is the continued perception that the Science, Technology, Engineering, and Mathematics fields are facing a shortage of graduates, and that the reason to cut other programs is because they fail to produce the kind of job opportunities these practical majors do. The problem is that there’s no STEM shortage. In fact, there’s evidence of a STEM surplus. I’ve been arguing this point for years, and have been seeing it get more and more traction culturally, and yet the idea endures. I don’t think I go a day without seeing the notion of a STEM or computer science or technology shortage asserted without evidence. The facts simply say otherwise.

One of the most comprehensive and rigorous considerations of the idea of a STEM shortage comes in the form of a research paper, released this April, from the Economic Policy Institute. Among their findings:

The flow of U.S. students (citizens and permanent residents) into STEM fields has been strong over the past decade, and the number of U.S. graduates with STEM majors appears to be responsive to changes in employment levels and wages.

For every two students that U.S. colleges graduate with STEM degrees, only one is hired into a STEM job.

In computer and information science and in engineering, U.S. colleges graduate 50 percent more students than are hired into those fields each year; of the computer science graduates not entering the IT workforce, 32 percent say it is because IT jobs are unavailable, and 53 percent say they found better job opportunities outside of IT occupations. These responses suggest that the supply of graduates is substantially larger than the demand for them in industry….

Analyzing new data, drawing on a number of our prior analyses, and reviewing other studies of wages and employment in the STEM and IT industries, we find that industry trends are strikingly consistent:

Over the past decade IT employment has gradually increased, but it only recovered to its 2000–2001 peak level by the end of the decade.

Wages have remained flat, with real wages hovering around their late 1990s levels."

In 2011, Ron Hira, a professor of public policy at the Rochester Institute of Technology, gave Congressional testimony in which he disputed the idea of a STEM major shortage.

At present, "there are too many skilled workers chasing too few jobs."

Focusing specifically on computer and mathematical occupations, "a field where Mr. Smith argues there's a shortage of workers," Hira also finds "unemployment rates...much higher than we would expect at full-employment." These two fields, which constitute "the largest of all STEM occupations," suffered "unemployment rates of 5.2% in 2009 and 2010,...more than twice the levels at full-employment" based on historical data. In fact, in 2010, the unemployment rate for computer and mathematical workers exceeded that of all college graduates by half a percentage point. The unemployment rates for electrical and electronic engineers and for medical scientists in 2010 were 5.4% and 4.1%, respectively, Hira writes. Again, he finds that, instead of any "broad-based shortage" in these fields, "there are too few jobs for those skilled workers."

Also in 2011, Vivek Wadhwa, who holds a large number of academic and corporate appointments in these fields, wrote an open letter to President Obama, asking him to stop claiming that there is a STEM shortage. Among Wadhwa’s purposes is to dispute the idea that China and India are churning out an unstoppable army of engineers who will render the United States incapable of competing. As Wadhwa points out, many of the professionals identified as engineers in these stories would in fact not be labeled as such here in the United States, with many of them in fields like automotive repair and HVAC. Wadhwa also relays the fact that 94% of the people who are offered jobs at Microsoft take them. That’s great for them, but a 94% acceptance rate indicates a field where applicants lack bargaining power. Wadhwa’s discomfort with claims about a STEM shortage, given that he’s a true insider in this domain, is in keeping with a lot of my own investigations on this issue, which I’ve been pursuing for several years. When I talk to professors in these fields, they are often quite forthcoming in saying that many undergraduates expecting to find an easy job market are in for a rude awakening. (They are also, naturally, protective of the perceived value of their fields, and rightfully so.)

If you prefer your data in graph form, you’ve got plenty to choose from. My friend Alex Waller, the Abstracted Engineer, pointed to this chart from the Bureau of Labor Statistics. (Alex, a biomedical engineer, also reports anecdotally that he is constantly approached online by recent engineering graduates who are desperate for a job.)

Similarly, Jordan Weissmann of The Atlantic has been covering this issue for awhile. He points to these charts from the National Center for Education Statistics.

None of these would indicate that getting a STEM major is some sort of economic panacea. And there’s a lot more charts and graphs where that came from.

There’s another side to these STEM shortage arguments, and they are straightforwardly moralizing: the reason for our continued employment crisis is that too many students took “impractical” majors and are suffering as a result. As Virginia Postrel pointed out last year, this narrative simply is not supportable. We don’t, actually, graduate a ton of people in the supposedly impractical arts or humanities. While participation in the humanities is stable, the number of students who pursue humanities majors is low, around 12%-15%. (Incorrect claims that the humanities are in a crisis of plummeting enrollment somehow coexist with arguments that too many students are taking them as majors.) These majors are also disproportionately concentrated in elite colleges, whose graduates enjoy far better economic outcomes than the median graduate, whether through quality of education, selection bias, or some combination of factors. A quick glance at the actual data shows that the notion of an army of deluded dreamers taking supposedly impractical majors is simply not supportable. What sticks out, more than anything, is the relentless rise of the Business major, by far the largest and one which now produces a mind-blowing 350,000 BAs or so a year. (I’d be very interested to see the economic outcomes for graduates of this eminently “practical” major.)

The notion that recent graduates are facing such steep economic challenges because of choices they made in major or career is not supportable. We face a stagnant job market and a crisis for the long-term unemployed because of inadequate aggregate demand, not because too many people decided to study French poetry. (French poetry, somehow, has become the go-to for “frivolous major.” I suppose it just speaks to people’s biases about cultural elitism.) As both Paul Krugman and Ben Bernanke have argued, the idea of a skills mismatch is not supportable from evidence. I think the point of claiming one is to blame broad macroeconomic problems on individuals, and to make our problems seem easier to solve than they really are.

I could go on. We could talk about changes in the pharmaceutical and chemistry industries that have caused them to hemorrhage jobs. We could point out that the notion that H-1B visas, used to pull highly educated workers from abroad, have no relationship with the underlying economy, despite the common claims that the need for such visas demonstrate a domestic STEM shortage. Earlier this year, the Institute of Electrical and Electronic Engineers sounded a warning about soaring unemployment in electrical engineering. Given our media’s crude assumption that “STEM jobs = jobs in consumer technology,” a soaring unemployment rate in electrical engineering cuts directly across the prevailing narrative. An earlier Rutgers study (PDF), from 2009, found the idea of a STEM shortage unsupportable. Among its findings was that computer scientists actually lost jobs at a higher rate during the financial crisis than the national average, speaking to the idea not only of unemployment rates but how susceptible individual fields are to economic fluctuations.

Like I said— I could go on.

The irony of all this is that the typical argument for the superiority of STEM disciplines would probably focus on these as fact-based disciplines, but the notion of a STEM shortage has almost no facts in its support. The notion of a STEM shortage is based on hype, cultural resentment against the arts and humanities, and an unshakeable American faith in technology as the deliverance from all of our problems. I genuinely believe that the biggest part of the belief in a STEM shortage results from our cultural obsession with technology and our perpetual belief that it will cure all of our ills. This discussion echoes another one of my hobby horses, the notion that technology will solve our education woes in K-12. Again and again and again in education research, rigorous independent studies find little or no statistically significant gains from using new technologies in the classroom. (I’ve often joked that I could make a small fortune simply by betting on the null hypothesis being proven in forthcoming ed studies.) Yet in the media and our broader discussion, the assumption that technology is the key to fixing education is an orthodoxy. We see major political and economic problems, and see how communicative and entertainment technologies have advanced so quickly, and desperately want to believe that we can solve the former with the latter. But it turns out political and economic questions are more vexing than the problem of how to squeeze more pixels into a screen.

The sad/funny thing is that the reverse may be true: that it might be exactly these new technologies that are gradually and relentlessly undermining labor and destroying the bargaining power of workers. The notion of capital-biased technological change is controversial, as is how to solve it. But as more and more jobs are innovated out from under more and more Americans, we’ll need to have a conversation about how to capture the very real value of automation and new technology for a larger slice of the American people.

I was talking about this issue with a friend of mine, a brilliant PhD student in Electrical and Computer Engineering. We were walking by the quad during one of the big tech job fairs here at Purdue University, where some of the most powerful and profitable companies in science and technology come to entice Purdue students to apply for jobs. Purdue is a good school generally, but its reputation largely comes from its top-flight engineering and computer science programs. Looking at all of these billion-dollar companies spending time, money, and energy on developing elaborate booths, all to attract applications for employment, it was hard not to believe in the notion of a STEM shortage.

When I mentioned that point to my friend, he laughed and said, “These companies are all trying to get the same 50 students.” This, more than anything, may be the source of the persistent STEM shortage myth: the inarguable value of being a star in a STEM field. There’s little doubt that people at the top of the food chain in computer science or electrical engineering or biomedical engineering, etc., often enjoy fantastic material and economic gain. But this is a banal point: it’s good to be a star. It’s good to be a star engineer in the same way it’s good to be a star musician or a star psychologist or a star writer. What public policy and politics demand is that we pay attention not to stars but to the median person. And the median American is facing a world of stagnant wages, the arbitrary nature of the employment market, and the constant fear of our financial system’s boom and bust cycle. The problem is that by definition, very few people get to be stars. I don’t doubt that the median Purdue STEM graduate is doing well. But Purdue is a top-flight STEM school, and half of our graduates will be below the median, and many who start those majors fail out of them, and the country is filled with schools who graduate STEM students who can’t get jobs. Basing our perception of the employment market on the outcomes of those 50 star students is pure folly.

None of this is to question the legitimacy or value of the STEM disciplines. Indeed, they are absolutely central to our experience of the good life. But then again, so are the arts, and in neither case does their inherent value say anything about the employment conditions for those fields. The elementary problem here is in part the notion that college exists for the uncomplicated process of training workers in a particular field. Universities are an essential part of our society, but they were never meant to solve all of our macroeconomic problems. Indeed, this whole conversation elides the large majority of Americans without a college degree at all, who suffer far worse outcomes on average than those who have one and who are struggling simply to stay afloat. What’s required is not to blame individual students for the failures on the job market, but to take a long, hard look at the future of employment, our winner-take-all economy, and the basic American social contract.