Trump’s De Facto H-1B Ban Will Boost the Trade Deficit, Reduce US Competitiveness, and Have No Impact on Unemployment

MAGA Republicans have long criticized the H-1B visa program that lets U.S. organizations hire high-skilled foreign workers on a temporary basis, arguing it takes jobs away from Americans. So, it wasn’t surprising to see President Trump issue a proclamation on Friday requiring any organization petitioning to employ an H-1B worker to pay a $100,000 fee.

The White House says the fee will have to be paid when companies submit petitions for new H-1B visas—which are typically good for a renewable three-year term—while Commerce Secretary Howard Lutnick described it as an annual fee at the announcement ceremony. But either way, let’s be clear: This is a de-facto ban, as few organizations will be able to afford it. And the result will be that traded-sector companies will hire more workers overseas, while the other workers who would have come to America will instead work for foreign competitors of U.S. companies. And to the extent any companies do pay the fee, their cost structure will go up relative to foreign competitors. As a result, domestic U.S. technology capabilities and competitiveness will decline.

The H-1B program enables employers facing a shortage of domestic talent in specialty occupations to hire foreign workers, granting them temporary work authorization. The number of these visas that are available is capped at 85,000 per year, and Congress allows the U.S. Citizenship and Immigration Services (USCIS) to impose a modest administrative fee to cover the cost of processing them. (As such, Trump’s proclamation is probably illegal and could be overturned in court.)

If the president is worried about some organizations abusing the H-1B program and hiring skilled foreign workers when they could be hiring skilled American workers, then instead of trying to impose a fee on visa petitions a better solution would be for Congress to allow USCIS to auction off the 85,000 visa slots to the highest bidders. Or Congress could impose a base fee of something like $25,000 and have the cap on the number of visas issued per year be unlimited. Organizations that truly need them will bid up the price and win the auctions, and organizations that can find skilled American workers to hire for less than the auction price plus salary and benefits will do that instead.

The current lottery system leaves too much to chance. If a U.S. company has 10 positions it needs to fill with foreign workers, it would have to submit applications for two or three times as many visas, knowing that many applicants will not be approved. The lottery system also means that U.S. companies cannot ensure they are always hiring the best of the best foreign workers—instead, they are just getting the luck of the draw. Creating an auction for H1-B visas would provide companies with more certainty.

Critics complain that employers only use the program to undercut American workers’ wages. But consider that the average wage for H-1B workers is $149,000, far above the U.S. median wage. And the wages also differ considerably, with U.S. high-tech firms paying considerably more. For example, Google’s average wage is $179,000, Meta’s is $199,000, and Apple’s is $200,000.

Indeed, a scholarly, peer reviewed study of the H-1B program in 2010 shed new light on the debate about its impacts on the labor market. The study found IT professionals with an H-1B or other work visa earn on average 6.8 percent more annually than IT professionals who have U.S. citizenship. When the authors controlled for the state in which the IT professionals work and for job titles, the premium declines, but is still significant at 2.6 percent.

Outside of the IT sector, the Milwaukee Board of School Directors employs 351 workers with H-1B visas, paying them $78,000. Evidently, that includes 275 international teachers. The program is also used to bring in nurses and similar human services occupations, which can be a lose-lose proposition. Some of these visas take scarce and valuable workers who could be providing education and nursing services in developing nations that need them, and substitute them for Americans when there would be no real shortage of teachers or nurses if not for the modest salaries. Lots of students can get degrees in these fields. It doesn’t require Calculus 2. A visa auction program would lead these organizations to hire Americans and apply market pressure to raise their salaries.

Also, a significant share of H-1B visas go to IT offshoring firms that do most of their work overseas and need some workers here, including to interact with clients. Again, an auction program would likely reduce their share.

There are, however, legitimate shortages of U.S. STEM workers, and putting a de-facto ban on H-1B visas would make several things more likely to happen. First, these American tech firms competing in global markets would instead just hire these workers overseas, either as remote workers (much easier after Covid) or in a foreign affiliate office. The result would be fewer American jobs and a worse trade balance. The reason it would mean fewer jobs is that H-1B visa workers spend their salaries domestically at restaurants and movie theaters, buying cars, etc., which creates U.S. labor demand. If they are overseas, that U.S. labor demand is not created. Meanwhile, the reason trade balance would be harmed is that these workers will be overseas, but still paid by American companies, leading to increased imports (in this case, foreign labor services).

At the same time, the increased difficulty in hiring STEM workers would mean that U.S. companies would at the margin become less competitive than foreign counterparts, including Chinese companies.

Critics of the H-1B program miss a broader and deeper point: Too many American workers are not high-caliber due to the poor quality of the U.S. education system, including colleges. H-1B opponents assume that a worker is a worker is a worker, and that if companies need a worker for a particular task, they can just hire an American. Secretary Lutnick made this point at the signing ceremony, saying: “If you’re going to train somebody, you’re going to train one of the recent graduates from one of the great universities across our land. Train Americans, stop bringing in people that take our jobs.” Ah, if only it was that simple.

First, comparatively few Americans get STEM degrees—and many who try don’t finish. Indeed, according to the Department of Education, only 41 percent of students who had begun STEM majors in higher education ended up obtaining a STEM degree of any kind in 2009. Part of the explanation is likely inadequate preparation in high school. As of 2016, more high school students in California were taking ceramics than computers science, according to state data. In 2022, 15-year-old U.S. students ranked 28th out of 37 OECD member countries in math.

In the 2011–2012 school year 40 percent of first-year undergraduate students at four-year institutions in the United States took at least one remedial course. The reality is that most graduate students in STEM—72 percent of computer science graduate students, and 74 percent of electrical engineering students—are foreign-born.

More broadly, American students don’t perform very well compared to foreign peers. During his 2024 run for the Republican presidential nomination, Vivek Ramaswamy got into hot water for saying the obvious:

The reason top tech companies often hire foreign-born and first-generation engineers over ‘native’ Americans isn't because of an innate American IQ deficit. A key part of it comes down to the c-word: culture.

He went on to point out: “Our American culture has venerated mediocrity over excellence,” and, “A culture that celebrates the prom queen over the math Olympiad champ, or the jock over the valedictorian, will not produce the best engineers.” But you are not supposed to say that, so he was vilified.

While China’s team won this year’s world math Olympiad, America earned second place thanks to a team comprised of students from Asian families. Similarly, most of the students at specialty math and science academies like Thomas Jefferson in Northern Virginia were from Asian families—at least until parents protested and the School Board eliminated the admissions test.

In 2006, 20 percent of U.S. college students completing four-year degrees—and 30 percent of students earning two-year degrees—had only basic quantitative literacy skills, meaning they were unable to estimate if their car had enough gasoline to get to the next gas station, or calculate the total cost of ordering office supplies.

Literacy among those college graduates is also quite low. In 2006, among seniors at four-year U.S. colleges, more than 50 percent of students did not score at the proficient level of literacy. In 2005, only 31 percent of college graduates could read a complex book and extrapolate from it. As the saying in college goes: The students pretend to learn, and the professors pretend to teach. The students pay good money to demand and receive “A’s” in response. Saying America has great universities is a good sound bite, but saying it doesn’t make it true.

Of course, the response to this reality for decades has been to push for improving U.S. education. But if anything, education is getting worse, especially as universities compete intensely for fee-paying consumers (er, students). But even if we could waive a magic wand and improve education, employers wouldn’t realize the benefits for another two decades.

But there is one more factor, intelligence. Anyone who says intelligence does not matter for STEM jobs and knowledge jobs more broadly has not worked in these fields. One does not need to be an Einstein, but workers with above-average intelligence are, let’s face it, on average more valuable to an employer than workers with below-average intelligence.

The distribution of IQs in the population is on a bell curve, so 16 percent of people are at or above one standard deviation from the mean (that is, above 115, with 100 being average), and just 2 percent are gifted at 130 or above. That means that there are around 34 million working age Americans above 115, and 4 million above 130. Contrast that to the rest of the world which has 846 million above 115 and 106 million above 130.

So, if you are an employer and you want smarter employees, it is natural to be open to hiring foreign individuals, especially because the unemployment rate for high-IQ individuals already in the United States is likely extremely low.

So, does anyone really think that America would not be vastly better off with 85,000 highly skilled foreign workers moving here per year? The United States would be stronger, more innovative and more competitive, especially against our adversary China, with more overall high-paying jobs.

So, it’s time for real debate about this issue. We can turn inward and protect a relatively small number of American workers at the cost of weakening republic, or we can allow U.S. employers to recruit the best and brightest from around the world and have a more innovative and stronger republic. As a patriot, I choose the latter.

Image credit: Mathematical Association of America (MAA)