Some executives and lobbyists representing science-based industry have argued that American workers lack the training or interest in STEM (science, technology, engineering, and mathematics) fields needed to fill the hundreds of thousands of positions they claim are available. Consequently, they propose, industry must be granted access to more high-skill "guest workers." (For the purposes of this story, "guest workers" means foreign workers in the United States who have temporary employment-based visas, such as the H-1B visa for "specialty occupations".) Several recent legislative proposals—including the recent "Gang of Eight" proposal in the United States Senate—aim to meet this perceived need by raising the cap on available H-1B visas and making it much easier for foreign workers with STEM degrees to receive green cards.

Earlier this week, Hal Salzman—a labor economist at Rutgers University in New Brunswick, New Jersey—and colleagues released a report considering the role of guest workers in the U.S. high-skill labor market and predicting the effects of the proposed changes to national policy. (Read Beryl Benderly's pre-publication analysis in her April Taken for Granted column.) Salzman's data indicate that, contrary to industry claims, U.S. students' interest in STEM careers has been very responsive to market signals. It's not the case that domestic students aren't interested in STEM careers, the report notes; rather, students look at the field and see the relatively poor job opportunities and middling salaries and seek work elsewhere. Guest workers, the report argues, still seek those jobs because they come from countries where job opportunities are worse.

The report points to a "natural experiment" that contradicts the idea that U.S. workers aren't willing to fill STEM jobs: petroleum engineering. Over the past decade, demand for petroleum engineers increased dramatically. Wages shot up 71% in 5 years, and the number of Americans graduating with petroleum engineering degrees doubled.

Then why do industry leaders and policymakers continue to argue that there is a domestic shortage of STEM talent in the face of numerous reports to the contrary? Science Careers asked Salzman to speculate.

Q: Why do companies say that they can't find the domestic STEM talent they supposedly need so badly?

H.S.: There are multiple answers. First, the current thinking goes, how can you not find workers with all this unemployment? What I've found is that during a recession, everybody thinks exactly that: "With all these workers out here, I should be able to find the perfect worker." And it's not really a pressing need to hire right away. They can extend the search time. When it's a tight labor market, they'll say, "I'll take somebody who's good and train them." When it's a slack labor market, they say, "I'd rather not train so I'll just hold out for the perfect candidate." They get pickier. And some studies were done—I think [Burt] Barnow did this—he said, "Well, let's look at how they're recruiting." And interestingly they found that recruitment efforts are less intense, they're not doing much advertising. It's not indicative of a shortage.

And part of it is certainly a price issue. When I talk to folks in industry, I ask, what are the particulars of the job they're offering, what are they paying, what are they looking for? And a consistent problem is the wages being offered. One example from a few years ago, a manager for a materials science division was looking for a materials science engineer. He wanted a Ph.D. and he was offering about $65,000. So I went and looked it up and a bachelor's degree in materials science averages about $57,000. So I said, "You're asking someone who's now spent 6, 7 years in a Ph.D. and you're going to pay them $7000 or $8000 more than they could have gotten with a bachelor's? I think you've got a problem there."

Q: We've heard companies say that they can't find the workers they want at any price.

H.S.: That is a "not getting a date to the prom" problem. So, with Microsoft—not to single them out, but I mean the big companies like Microsoft—in 1998, 1999, Microsoft could recruit almost anybody they wanted because they were telling anybody who came there, A.), There was new product development, and B.) They'd be millionaires in 5 years. What does Microsoft promise people today? They've cut back on options. You get a good salary but you're not going to be a millionaire 5 years from now, and you're going to spend your time figuring out how to rearrange buttons in Word. There's no significant product development that I can tell, not the kind you'd find elsewhere. The top 10% of the class says, "Hey, I'm going to start up my own idea and see what happens. Five years from now it's a bust? Then I'll go work at Microsoft."

Q: Can guest workers fulfill companies' STEM needs just as well as domestic workers?

H.S.: [Companies] are often looking for these skill combinations that don't exist or are very tough to find. But you're not going to find them in guest workers. Guest workers don't get those levels of skills coming from a foreign university. Not that that they're not bright and motivated, just that they don't often have the talent among the faculty. So you'll hear they aren't getting the people they want.

Q: What has happened to on-the-job training?

H.S.: In IT, larger, mature companies know they have to train. Now, they've been cutting back in recent years, in the last 5 years pretty significantly, but they understand you get somebody who's good, who has capability, and then you have to invest in training. But since the dot-com era, especially in Silicon Valley, the trend has been toward smaller IT companies. And most smaller IT companies don't want to invest in training; they want someone who's going to be productive on day two, if not hour one. You're just not going to find many people like that. It's not realistic. The irony of it is, when they go hire guest workers, they're getting people they have to invest significantly in because they're lacking not just skills, but also the context.

Q: Why has petroleum engineering been successful in hiring domestic workers but IT hasn't?

H.S.: Small numbers. They've got a lot of money, they needed people quickly, and they needed to be highly skilled. There's relatively little cost—and I mean relatively, because they'll rightly say that there's a cost to having people who aren't good and don't perform well—in the scheme of things, to poor-quality software compared to, say, something going wrong in the oil industry. So for oil companies, paying somebody a little bit more is cheap in the long run. And they have billions and billions of dollars; it doesn't really affect their bottom line.

Q: Are companies intentionally hiring more guest workers to keep wages low?

H.S.: I don't know how intended it is, but certainly they see what happens. They observe that this keeps wages down, and I suspect that is a nontrivial part of the lobbying effort. I don't know if a hiring manager says, "I want more guest workers because they keep wages down," but certainly at the strategy level, folks understand that. They all know Econ 101. And the irony is, in their product market, they don't want government intervention. They'll say they shouldn't be regulating this or that. But when it comes to them being on the other side of the market, the buying side, for some reason they think there's a role for government to play in increasing supply so wages don't go up. And maybe that's what they call a shortage: having to pay more than you want.

The dilemma for them is that all the evidence lines up to say there's no shortage. There are no data to indicate shortage, however you cut it. You just don't see it. If you go listen to the testimony, or you listen to discussions I've had with industry, and you ask, "Where's the evidence of a shortage?" they say, "We can't find workers and we need them." They repeat the same things over and over. Lobbying is a solution to lack of evidence.

Q: What do you think will happen to the STEM job market if Congress passes the Senate's current proposed bill?

H.S.: Immigration is a social policy for social diversity and a lot of other reasons. The guest worker policy is a labor market policy. Nobody wins by undercutting the quality of jobs by flooding the market with guest workers. The outcome, like you've seen in the report, is flat wages for 14 years. And they're claiming a shortage; it just doesn't add up. There's certainly a lot of good immigrant-founded companies and we want to have a diverse immigration policy, but the guest worker policy is not the way to do it.

Michael Price is a staff writer for Science Careers.

10.1126/science.caredit.a1300088