To understand the future of the immigration debate, one must first understand the H-1B visa program. Though not very well known among ordinary Americans, the program, first established in 1990, has grown enormously, and at least some lawmakers are hoping to expand it further still.
What exactly is the H-1B visa program? Essentially, it is a guest-worker program for the college-educated set. H-1B-visa holders are foreign nationals who agree to come to the United States to work for three years and to return to their native countries when this period draws to a close, unless they renew their H-1B visas for one (and only one) additional three-year period. There are other types of visas for other categories of guest workers, from agricultural workers to workers “of extraordinary ability or achievement,” a category that famously smoothes the way for film starlets and other glamorous people to settle in Williamsburg and West L.A. Yet it is the H-1B, for workers in “specialized occupations,” that attracts the most attention, since this is the visa program that Silicon Valley and Wall Street see as absolutely essential to their financial health.
This is no small thing, as technology and finance employers have a great deal of political influence. This influence stems not just from political contributions, though there is that, but also from their prestige. If you’re an elected official, you might aspire to work one day for a Goldman Sachs or a Google. You look up to their senior executives, who represent the meritocratic elite of American business. You are inclined to believe what they say.
And, by and large, employers in these sectors and their advocates claim that they are facing a grave shortage of skilled workers and that their only hope for expanding their operations is to hire foreign workers. There is, however, another explanation for why high-tech employers are so keen to expand the H-1B-visa program: Hiring skilled workers is a cost of doing business, and employers would like to keep that cost as low as possible.
Earlier this year, a bipartisan gang of six — Senators Orrin Hatch (R., Utah), Marco Rubio (R., Fla.), Jeff Flake (R., Ariz.), Amy Klobuchar (D., Minn.), Chris Coons (D., Del.), and Richard Blumenthal (D., Conn.) — introduced the Immigration Innovation Act of 2015, which they’ve dubbed, in a cutesy homage to tech-speak, the “I-squared bill.” Among other things, the bill lifts the H-1B cap on workers with advanced degrees in science, technology, engineering, and mathematics (the “STEM” fields, as they’re sometimes known) and greatly increases the cap for all other H-1Bs.
It’s easy to see why this new gang is calling for a higher cap. In April, U.S. companies filed 233,000 petitions for the 85,000 H-1Bs available this year, a sharp increase over the 172,500 petitions filed last year. These numbers are slightly misleading, as it is not uncommon for companies to file multiple petitions on behalf of a single worker, and some workers have more than one company filing petitions on their behalf. To have multiple petitions filed on one’s behalf would be very flattering indeed, as the legal fees involved are notably high. Each petition sets a company back somewhere between $2,000 and $4,000, and of course there is no guarantee that filing a petition will help the company land its chosen candidate. There is a voracious appetite for H-1Bs. But does that necessarily mean that Congress ought to greatly increase their number, as I-squared would have us do?
On the surface, the I-squared bill has much to recommend it. Instead of trying to address unauthorized immigrants, border security, and everything else related to immigration all at once, I-squared tries to tackle one small but important piece of the immigration debate in a neat little package. If you believe, as I do, that there is a case for restricting less-skilled immigration and increasing high-skilled immigration, I-squared is, in theory, a perfectly reasonable approach. Scratch beneath the surface, however, and it becomes clear that I-squared is a bad bet. If we really want an immigration policy that serves America’s national interest, we ought to go in a very different direction.
First, it’s not at all obvious that there is in fact a grave shortage of tech workers. If anything, there is evidence that the opposite is true. In Learning by Doing, the economist James Bessen reports that U.S. colleges and universities graduate twice as many scientists and engineers each year as the number actually hired. To be sure, it could be true that this doubling still hasn’t been enough to meet the insatiable demand for skilled tech workers — except that wages for computer and IT jobs have remained flat over a decade. If employers aren’t willing to offer higher wages to fill these positions, they probably aren’t so desperate after all.
Peter Cappelli, a management professor at the Wharton School of Business, has observed that, for a variety of reasons, employers have grown more reluctant to train workers. One could argue that instead of recruiting a larger number of H-1Bs, employers ought to be more willing to train U.S. workers and to pay them attractive wages. Of course, many employers fear that if they invest heavily in training, they will find themselves at a disadvantage when their employees then leave for other jobs. Governments could partner with employers to expand apprenticeship programs, which would allow employers to pay somewhat lower wages while young workers gain experience.
But let’s assume that Silicon Valley and Wall Street are right and the U.S. badly needs more skilled workers. Why increase the number of H-1B visas rather than simply grant more green cards (which confer permanent-resident status) to people deemed sufficiently qualified? One thing to keep in mind about the H-1B program is that its recipients are allowed to bring their spouses and children, which is why the 85,000 H-1Bs granted in a given year can account for tens of thousands of additional immigrants who receive H-4 visas, for dependents. (Green-card holders can also do this, but they have to file a petition and the process can take years.)
And not all H-1B-visa holders intend to return home. Though no one can definitively say how many H-1Bs see their temporary status as a stepping stone to something more permanent, I’d hazard that it’s more than a few. Though the H-1B visa has a six-year maximum, any time spent overseas is not counted against the maximum, nor is time spent in another visa category. So a couple could, in theory, trade off. Say a wife is residing in the U.S. on an H-1B while her husband is residing on an H-4. Eventually, the husband could get hired on an H-1B, at which point his wife would switch over to an H-4. There are a few other visas to choose from too. And hopping from visa to visa isn’t the only way to extend your “temporary” stay. Some H-1Bs wind up having U.S.-born children, or marrying locals. Why continue with the charade that H-1Bs are guest workers who will eventually return home?
One plausible reason is that guest workers are less likely to press for wage increases than their U.S. counterparts. Why would that be? Part of the answer is that guest workers are less likely to defect to other employers, because not all employers would be willing to take on the cost associated with filing a new H-1B petition and because guest workers tend to have more limited social networks than U.S. workers or green-card immigrants. Recently, the economists Kirk Doran, Alexander Gelber, and Adam Isen released a study that attempted to measure the impact of hiring a capped H-1B worker. To that end, they compared two groups of employers: those that applied for H-1B visas and managed to secure them, and those that applied for them but failed to secure them. Doran et al. compared these two groups of firms along several different dimensions, e.g., the number of patents they issued, as a crude measure of innovative activity; whether they hired more or fewer U.S. workers; and overall profits. They found that H-1Bs don’t appear to make firms more innovative, but they do lead them to hire fewer U.S. workers than they would have otherwise, and they do make them more profitable.
How can this be, when under current law employers are required to pay H-1Bs the “prevailing wage” for a given occupation in a given location? It turns out that employers routinely circumvent this provision. For example, Ron Hira, a fellow at the left-of-center Economic Policy Institute, reports that in 2013, the average wages for H-1Bs employed by Infosys and Tata, two leading offshoring firms, were $70,882 and $65,565 respectively, far lower than the average annual wage of $91,990 for a computer-systems analyst based in Los Angeles that year.
All of these findings are compatible with the notion that H-1Bs and non-H-1Bs are substitutes, and that the chief motivation of employers seeking H-1Bs is to restrain growth in labor costs. I can see why this is excellent news for employers. But why should our elected representatives place a higher priority on expanding the H-1B program than on, say, expanding apprenticeships?
None of this is to suggest that we shouldn’t welcome extraordinary immigrants who can make outsized contributions to U.S. society. Yet the H-1B program is not limited to extraordinary people — if it were, it would not be used chiefly as a means to lower wage costs.
So I have a straightforward proposal for reforming employer-based immigration: Let’s lift the cap on H-1Bs, but let’s also limit the program to workers who will be paid an annual wage of at least $155,817, or three times the median household income in the United States. Think of this as a simplification of the H-1B program’s current prevailing-wage provision. Granted, $155,817 would be a higher-than-average wage. But this is in keeping with the idea that we as a country should roll out the welcome mat for the best of the best — not workers who are just average in their fields. This H-1B visa would last for five years, not three, and those workers who manage to earn an average of $155,817 or more over this entire period would then be invited to apply for permanent-resident status.
True, some of the low-wage tech workers who now work in the U.S. as H-1Bs will remain in their native countries, where they might staff outsourced back-office operations for U.S. tech firms. I see no problem with that. Just as garment manufacturing has gone overseas, so will a fair amount of basic IT work, whether or not we expand the H-1B program. The question is whether U.S. taxpayers want to pick up the tab for the safety-net benefits that low-wage tech workers will require when offshoring and automation render them less employable. My guess is that the answer is no. Better to recruit tech workers who will pay much more in taxes than they’ll ever receive in transfers.
There is another downside to my approach, and it’s a big one: Limiting H-1Bs to high-wage workers would keep out younger, less experienced foreign workers who might eventually become superstars. Fortunately, there are many younger, less experienced American workers, both native- and foreign-born, who would happily take their place.