The Trump administration announced a new immigration plan this week. According to White House adviser Stephen Miller, it’s a “pro-American immigration reform that the American people want, that the American people deserve, and that puts the needs of the working class ahead of the investor class.”
Vox’s Dara Lind explained the details of the plan here. In short, it will cut the number of legal immigrants entering the country in half by 2027 and shift to what the White House calls a “merit-based” system that favors “applicants who can speak English, physically support themselves and their families, and demonstrate skills that will contribute to our economy.”
Ostensibly, the plan rests on a simple economic premise: If we continue letting low-skilled immigrants into the country, working-class Americans will suffer because there will be fewer jobs.
I called up Michael Clemens, an economist at the Center for Global Development, a policy-oriented nonprofit think tank. Clemens’s research focuses on migration and development, economic growth, and economic history. I asked him if immigrants from poor countries really do take jobs from native workers, and whether he thinks the Trump administration’s economic justification for the new immigration plan is sound.
Here’s what he told me.
The central premise of the administration’s new immigration plan is that allowing low-skilled workers into the country will hurt American workers. Is there any evidence to support that claim?
No, not at all. Stephen Miller gave what he described as evidence of that assertion, saying that inequalities have risen during high periods of immigration, but simple correlations like that aren't serious and are certainly not a basis for a sweeping policy shift.
You could also point out that immigration and unemployment have moved inversely for the last 130 years (meaning more immigration has correlated with lower unemployment and vice versa). So if you were looking for such an answer, you could just as easily announce a policy with the exact opposite premise.
But these simple correlations aren't sufficiently explanatory and aren't taken seriously by people who actually study these issues. And it's frightening to see such large policies being contemplated on such flimsy data.
If you're not familiar with the data, it seems intuitively right to argue that allowing low-skilled immigrants into the country means low-skilled American workers will have fewer opportunities. Why is that wrong?
American history shows without a doubt that more low-skill workers doesn’t mean fewer jobs. The American economy and the jobs it creates were built by low-skill workers. As recently as 1940, the fraction of European immigrants to the US who had no high school degree was 88 percent. That includes at least one grandparent of more than 40 percent of Americans, and those people fueled modern economic growth in this country. So something’s wildly wrong with the notion that low-skill workers have harmed American workers.
It is indeed intuitive to people that when there are more workers around, wages drop. But that’s a bad cartoon of how complex the economy is. I think it’s also intuitive to people that workers and firms adjust to changes in the labor supply in myriad ways. US workers respond to there being more workers around by moving to where there’s more opportunity, by going into business for themselves, and by specializing in tasks that use their strengths, like native English ability.
Women in the US respond to the presence of low-skill immigrants by entering the labor force, because immigrants make child care and elder care more accessible. Firms respond to there being more workers around by shifting toward production technologies that use labor more intensively, by moving plants from one place to another, and by producing in new ways that allow natives and immigrants to specialize in different tasks.
These complex adjustments are very intuitive to Americans out there who depend on low-skill immigrants caring for their kids, preparing food in kitchens, guarding buildings, shipping fresh vegetables. So I think when people stop to think about it, the “intuitive” thing is that immigrants fuel the US economy overall, creating more opportunity for US workers than the jobs that they take themselves. And that has been true for many, many generations in this country.
Our status quo approach to immigration is to allow both low- and high-skilled immigrants into the country. What’s the net economic impact of this policy?
Under the current immigration rules, immigrants on average are more educated than Americans. That is, the fraction of recent immigrants who have a college degree is higher than the fraction of Americans with a college degree. That said, the skills of immigrants are polarized: many very high-skill and many very low-skill.
The economic impact of this immigration is that it raises economic growth and the productivity of US firms. That raises the returns to investment in the US, powers the creation and expansion of companies and new jobs, and causes Americans to invest more in education. All of that ends up creating as many jobs as immigrants take, which is why a consensus report of the National Academy of Sciences found that the average effect of immigration on the wages and employment of US workers has been very small or zero.
These economic benefits absolutely do not come from only workers with more education or only workers with less education. They do not come from only workers who arrive speaking English or workers who arrive not speaking English. The economic benefits come from the fact that workers of all skill levels and language abilities specialize in different things and complement each other.
This is true as much today as it was when my own low-skill, non-English-speaking immigrant ancestors came here to help build the US economy.
If we were only allowing high-skilled immigrants into the country, or if we were disproportionately doing so, as the Trump administration clearly prefers, would that be better or worse for the typical American worker?
That would be much worse. High-skilled workers and low-skilled workers complement each other; they depend critically on each other, in fact. The clearest example of that is the Bureau of Labor Statistics forecast for labor demand over the next 10 years.
There's this mythology that the US economy runs on computer engineers and brain surgeons, but in fact the data shows that when you look at the projections for the number of workers that will be needed occupation by occupation in the next decade, if you take the top 20 occupations in terms of absolute number of jobs created, more than half of the number of people needed to fill those positions will not require a high school degree. We're talking about jobs in retail, in personal care, etc.
We need people doing these jobs. If you purged all the low-skilled immigrants from the country — and if you stopped allowing them in — we would not have nearly enough workers to occupy these positions.
Is there any economic justification for the plan currently being proposed by the Trump administration?
The distributional effects of immigration are not zero. It is clear that immigrants do compete with a small number of American workers for jobs, and there are transitory displacement effects for workers. It stands to reason that even though firms can adjust by investing in capital, even though people can move to new geographic areas where there are more opportunities, and even though the economy can adjust in many ways to immigration, these things don't happen automatically and instantly and without costs. There are going to be adjustment costs for small numbers of people for limited periods of time.
The question is, what's the best way to address these challenges? Addressing them by removing millions of people from the United States — which is what this policy will do, as a way to assist a very small fraction of the population that would have temporary adjustment costs to immigration, as has always been the case throughout American history — is like killing a fly with a sledgehammer: You're going to do more damage than good.
The way to assist the small numbers of workers who are temporarily displaced is with social safety nets, with subsidies for job retraining, with earned income tax credits. There are many things that could take the massive productivity gains that immigrants bring and use it to aid the American workers who have to make a transition.
These are real economic challenges that require real economic policies, and we shouldn't be facile and pretend there aren't economic costs no matter what we do. There will be changes and transition costs in any case — that's how innovation works. The choice is between facilitating those changes to unlock their huge economic potential or taking us backward by blocking those changes with draconian policies.
Which is what Stephen Miller outlined this week.