Donald Trump and his allies on the populist right believe they have a compelling argument for why the GOP is the true blue-collar party: Immigration is killing the American worker, and only Trump will put a stop to it. “Kamala Harris’s border invasion is also crushing the jobs and wages of African American workers and Hispanic American workers and also union members,” Trump declared at a recent rally. At other times, he has referred to immigration as “all-out economic warfare” on the working class. It’s a message that the former president repeats in one form or another at just about every one of his public appearances.
The argument carries a certain commonsense logic: Immigration means more workers competing for jobs, which translates to lower wages and employment rates for the native-born. During Tuesday night’s vice-presidential debate, Republican Senator J. D. Vance said that his boss’s proposal to round up and deport millions of undocumented immigrants would “be really good for our workers, who just want to earn a fair wage for doing a good day’s work.”
Mainstream Democrats used to vigorously dispute the notion that immigration hurt native-born workers. No longer. Today, the two major parties are jockeying to convince voters that they are the ones who will truly secure the border. To the extent that liberals still defend immigration, they often do so by arguing that deporting migrants would reduce the labor supply and send prices soaring again—an argument that implicitly accepts the premise that immigrants do in fact depress wages.
This is a tragedy. The effect of immigration on wages is one of the most thoroughly studied topics in empirical economics, and the results are clear: Immigrants do not make native-born workers worse off, and probably make them better off. In many domains, the conventional wisdom among progressives is mistaken, oversimplified, or based on wishful thinking. The economics of immigration is not one of them.
Econ 101 tells us that when the supply of a good, like labor, increases, then the price of that good falls. This is the lens through which economists viewed immigration for much of the 20th century: great for corporations (cheap labor) and consumers (lower prices) but bad for native-born workers. Then a study came along that shattered the consensus.
In 1980, Fidel Castro briefly lifted Cuba’s ban on emigration, leading 125,000 people, most of whom lacked a high-school education, to travel from Mariel Bay to Miami in what became known as the Mariel Boatlift. In a few months, Miami’s workforce expanded by about 25 times as much as the U.S. workforce expands because of immigration in a typical year, creating the perfect conditions for a natural experiment. The economist David Card later realized that if he compared Miami with cities that did not experience the boatlift, he could isolate the effect that immigration had on native-born earning power. If immigrants really did depress wages, then surely the effect would be visible in Miami in the 1980s.
Instead, in a paper published in 1990, Card found that the boatlift had virtually no effect on either the wages or employment prospects of native-born workers in Miami, including those who lacked a college degree. Economists have since used similar natural experiments to study the effect of immigration in countries including Israel and Denmark, arriving at the same conclusion that Card did. (These studies mostly focus on low-skill immigration; high-skill immigration has long been viewed almost universally as economically beneficial.)
[Derek Thompson: Americans are thinking about immigration all wrong]
The simple Econ 101 story turned out to have a blind spot: Immigrants aren’t just workers who compete for jobs; they are also consumers who buy things. They therefore increase not only the supply of labor, which reduces wages, but also the demand for it, which raises them. In the end, the two forces appear to cancel each other out. (The same logic explains why commentators who suggest that immigration is a helpful inflation-fighting tool are probably wrong. I have made a version of this mistake myself.)
Inevitably, not everyone accepted the new consensus. In a paper first circulated in 2015, the Harvard economist George Borjas reanalyzed Card’s data and concluded that even though average wages were indeed unaffected, the wages for natives who lacked a high-school degree—and thus competed most directly with the Marielitos—had fallen as a result of the boatlift. Borjas’s study seemed to back up restrictionist policy with empirical data, and for that reason became a pillar of anti-immigration discourse. In 2017, for example, Stephen Miller cited it when pressed by a New York Times reporter for evidence that immigration hurts American workers.
But Borjas’s debunking of Card, such as it was, has itself been debunked. The data underlying his argument turned out to be extremely suspect. Borjas had excluded women, Hispanic people, and workers who weren’t “prime age” from his analysis, arguing that the remaining group represented the workers most vulnerable to immigrant competition. As the economist Michael Clemens has pointed out, Borjas ended up with an absurdly tiny sample of just 17 workers a year, making it impossible to distinguish a legitimate finding from pure statistical noise. Another study looking at the same data, but for all native-born workers without a high-school degree, found no negative impact on wages. Subsequent natural experiment studies have yielded similar conclusions. “Economic models have long predicted that low-skill immigration would hurt the wages of low-skill workers,” Leah Boustan, an economist at Princeton University, told me. “But that turns out not to be true when we actually look at what happens in the real world.”
On paper, immigrants and natives without a high-school education might look like easily substitutable workers. In reality, they aren’t. Take the restaurant industry. New immigrants may disproportionately get hired as fry cooks, which, in turn, depresses wages for native-born fry cooks. But by lowering costs and generating lots of new demand, those same immigrants enable more restaurants to open that need not just fry cooks but also servers and hosts and bartenders. Native-born workers have an edge at getting those jobs, because, unlike new immigrants, they have the English skills and tacit cultural knowledge required to perform them.
This dynamic helps explain why many efforts to deport immigrants have hurt native-born workers. From 2008 to 2014, the Department of Homeland Security deported about half a million undocumented immigrants through its “Secure Communities” program. Because the initiative was rolled out in different counties at different times, researchers were able to compare how workers fared in places where mass deportation was under way against outcomes for those in as-yet unaffected places. They found that for every 100 migrant workers who were deported, nine fewer jobs existed for natives; native workers’ wages also fell slightly. Other studies of immigration crackdowns throughout American history have reached similar conclusions. When a community loses immigrant workers, the result isn’t higher-paid natives; it’s fewer child-care services provided, fewer meals prepared, and fewer homes built.
Low-skill immigration does have some economic costs. Most studies find that the income of other immigrants takes a hit when a new wave of migrants arrives. Low-skill immigration also tends to slightly exacerbate inequality because it increases demand for college-educated professionals such as doctors, managers, and lawyers, resulting in even larger wage gains for that group. But these complications don’t mean that immigration is crushing the American working class.
Hold on, immigration’s critics say: Natural experiments can only tell you so much. You must instead look at the broad sweep of American history. As the liberal New York Times columnist David Leonhardt has pointed out, the decades in which American workers experienced their fastest income gains—the 1940s, ’50s, and ’60s—occurred when immigration was near historic lows; since the ’70s, immigration has surged while wages for the median worker have stagnated. “The trajectory of American history tells a very clear story,” Oren Cass, the chief economist at American Compass, a conservative think tank, told me. “High levels of immigration are correlated with poor outcomes for workers.”
The problem with relying on history is that correlations also only tell you so much. Some readers will recall that quite a few things have changed since the 1970s; most relevant for our purposes, these include the loosening of trade policy, the weakening of labor unions, and the enormous rise in corporate concentration. All of these trends have been more persuasively linked to the declining fortunes of the working class. Without some evidence of causation, the co-incidence of stagnating wages and rising immigration really does look like just that: a coincidence.
[Michael Podhorzer: The paradox of the American labor movement]
Two data points are instructive here. First, the parts of the country that have received the largest numbers of immigrants in recent decades—Texas, Florida, the D.C.-to-Boston corridor—are those that have experienced the least wage stagnation. Second, since the onset of the coronavirus pandemic, the U.S. has experienced both a huge surge in illegal immigration and perhaps the most significant reduction of wage inequality since the 1940s. That doesn’t mean high levels of immigration caused the spike in wages at the bottom. But that’s exactly the point: Historical trends don’t necessarily imply neat causal relationships.
The other problem is that you can just as easily make the circumstantial case that the natural-experiment literature underestimates the economic benefits of immigration. The aforementioned Denmark study tracked every single individual across the country (something that isn’t possible in the U.S. because of data constraints) over a 20-year period and found that low-skill natives who were most exposed to immigration responded by pursuing higher levels of education and moving to higher-paying occupations. Ultimately they achieved higher earnings than their peers who weren’t exposed to immigration. A study in the U.S. found that immigrants were 80 percent more likely than native-born Americans to start a business, and that the rate of entrepreneurship was just as high for immigrants from low-income countries as those from high-income countries. “Immigrants to the U.S. create so many successful businesses that they ultimately appear to create more jobs as founders than they fill as workers,” Benjamin F. Jones, one of the authors, wrote in The Atlantic last year. Immigrants, he noted, are inherently risk-takers. “We should not be surprised that they are exceptionally entrepreneurial once they arrive.”
I admit to being partial to this view for personal reasons. My grandfather came to the U.S. in the 1960s as an undocumented immigrant from Lebanon, having never finished high school and speaking very little English. Within a few months, he landed a job as a car mechanic at a local gas station, leaving for work each morning before his kids woke up and returning after they were asleep at night. An economic study might find that he helped depress the wages of native-born mechanics, which might have been balanced out by his spending in other areas. What it probably wouldn’t capture is what happened next: He opened up his own station, and then another, and then another, employing dozens of mostly native-born mechanics, attendants, and cashiers. Along the way, he became a darling of his community, bringing a little bit of Arab hospitality to a mostly white suburb of New Jersey. His life was its own kind of natural experiment.
The appeal of restricting immigration has, to put it lightly, never been primarily about economics. Surveys of public opinion generally find that people’s feelings about immigration are driven less by material concerns than they are by cultural anxieties about crime, social norms, and national identity. Anti-immigrant sentiment is much higher among older Americans (many of whom are retired) living in rural areas that contain few immigrants than it is among working-age Americans in immigrant-heavy cities such as New York and Los Angeles.
Even if conservative policy wonks sincerely believe that limiting immigration would help the American worker, the guy at the top of the Republican ticket clearly has other things on his mind. In his debate against Kamala Harris, Trump, who has accused immigrants of “poisoning the blood of our country,” mentioned the supposed economic impact of migration exactly once. He spent much more time portraying undocumented immigrants as a marauding horde of psychopathic murderers “pouring into our country from prisons and jails, from mental institutions and insane asylums.” At one now-infamous moment, he even claimed that immigrants were eating pets in Springfield, Ohio. In Trump’s hands, the economic case against immigration is a fig leaf that barely obscures a much larger and more nakedly bigoted body of work.
[Gilad Edelman: Donald Trump’s theory of everything]
The example of Springfield is a revealing one. In the past few years, thousands of Haitian immigrants—overwhelmingly with legal status—have settled in the town of 58,000. This has led to some problems. Housing prices rose quickly. The health-care and education systems have come under stress. And relations between longtime residents and the new arrivals have at times been contentious, especially after a traffic accident caused by a Haitian immigrant last year resulted in the death of an 11-year-old boy.
But after decades of dwindling population and shrinking job opportunities, Springfield has also experienced a jolt of economic energy. The immigrants have helped auto factories stay in operation, filled shortages at distribution centers, and enabled new restaurants and small businesses to open. Wage growth in the city took off during the migration wave and stayed above 6 percent for two years, though it has since slowed down. And the flip side of strain on the housing, education, and health-care systems is that there are now more jobs available for construction workers, teachers, and nurses to meet that increased demand. “What the companies tell us is that they are very good workers,” Ohio Governor Mike DeWine, a Republican, said in a recent interview, referring to the Haitian immigrants. “They’re very happy to have them there, and frankly, that’s helped the economy.”
For DeWine and other public officials, this is a trade that is well worth making: Immigrants might cause some social tensions, but overall they make the place better off. Others, of course, disagree. According to Gallup, 2024 is the first year in nearly two decades that a majority of the public wants less immigration to the U.S. In the past year alone, the desire to reduce the amount of immigration has jumped by 10 points for Democrats and 15 points for Republicans. No matter who wins in November, we will likely see more restrictive immigration policy in years to come. If that is the will of the voters, so be it. Just don’t expect it to do anything to help the working class.
theatlantic.com