Given how crowded policy commentary is these days, with blogs, articles, e-books, and the like, the surest way to break through and get attention is to write the “man bites dog story.” And that is exactly what we have seen with the issue of science, technology, engineering and mathematics (STEM) education and skills. While the evidence shows that the United States is not producing enough STEM workers, a cottage industry of STEM shortage naysayers has emerged. Most recently, Michael Teitelbaum wrote in The Atlantic that “there is little credible evidence of the claimed widespread shortages in the U.S. science and engineering workforce.” In fact, there is considerable credible evidence as the Information Technology and Innovation Foundation and others have documented (pdf).
Let’s look at Teitelbaum’s claims. First, he, like most of the shortage deniers, argues that “U.S. higher education produces far more science and engineering graduates annually than there are S & E job openings”. But this counts social science and health degrees, which are not really STEM jobs. Moreover, it’s not accurate to count just job openings, you need to count all hires, including ones when a worker retires or leaves to raise a child. When you do both of these adjustments, the ratio is pretty close to one-to-one.
But it’s even worse than that. Many STEM grads work outside of tech-based industries (e.g., the Intels and Mercks of the world) because employers in other industries need them. As Tony Carnivale, director of the Georgetown University Center on Education and the Workforce, writes (pdf), “…the diversion of STEM talent and STEM workers into other occupations results from the increasing value of the competencies that are associated with STEM occupations…In all but two occupational clusters, the rate of growth in demand for core STEM competencies has increased at far greater rates than the growth in employment.”
Teitelbaum would also have us believe that all is well because more students than ever are interested in STEM. If so, why then do 4 times more high school students take the AP Art History test than the AP Computer Science Test? It’s not because of those high wages in art history jobs. Moreover even if they are interested in STEM, a large share switch out to other majors. Seymour and Hewett found that 44% of STEM majors witch out compared to just 30% of humanities majors.
This high switch out rate is also one reason why STEM graduation rates have significantly lagged behind non-STEM graduation. From 2000 to 2007, non-STEM bachelor’s degrees grew 24% compared to just 16% for STEM degrees, while from 1993 to 2007 total Masters’ degrees increased twice as fast as STEM master’s degrees.
Teitelbaum then falls back on the next standard claim: “Were there to be a genuine shortage … there would be evidence of employers raising wage offers.. . But the evidence points in the other direction.” No, the evidence points in the right direction.
As the Brookings Institution’s Jonathan Rothwell shows, the earnings premium for STEM skills (controlling for experience, education and sex) has grown from around 22 percent in 1990 to 30 percent in 2012. Dartmouth’s Matt Slaughter and UC San Diego’s Gordon Hanson found that “the inflation-adjusted wages of major STEM occupations grew over the last decade while real wages for most other U.S. occupations fell.” Hardly evidence of surplus.
So what’s behind the man bites dog STEM stories? The short answer is ideology. Most of the advocates of no-shortage, including people like Ron Hira, Hal Salzman, Richard Freeman, and of course Teitelbaum are focused more on an agenda of redistribution, ensuring higher wages for workers, including STEM workers. Arguing against shortages is part of a strategy to oppose high-skill immigration policies so that shortages increase even more and already well paid STEM workers get paid even more.
Yet policies established to achieve nothing more than an increase in STEM wages by restricting the supply of workers would have two bad effects. First, they would lead to higher prices for products and services that have STEM talent as a significant input. This would be a transfer payment from all consumers, including low income ones, to some workers, many of whom are already very well paid. Restricting the supply STEM workers would also reduce the competitiveness of U.S. establishments that rely on STEM labor, reducing U.S. jobs and economic growth.
The goal of economic policy should be a “larger pie” via increased productivity, innovation and competitiveness, not a smaller pie with slightly higher wages for a few. Ensuring an adequate supply of high-skilled, STEM labor, through both better domestic education and training policies and more liberal high-skill immigration policies is a key factor in achieving this goal.