The Employment Effects of Immigration Enforcement: An Initial Assessment

“Unprecedented.” That description comes up repeatedly in describing the size of the immigrant population that President Trump inherited from President Biden. During Biden’s four years, the estimated number of immigrants in the United States, both legal and illegal, rose from 45.0 million to 53.3 million. The increase was so rapid that the U.S. Census Bureau had trouble keeping up.¹ Initially, the Bureau reported that net international migration (people entering the United States minus people leaving) was a combined two million in 2022 and 2023. It later revised that estimate up to four million. In 2024 alone, net migration hit 2.8 million, a number that is quite literally unprecedented: it is more than double the highest annual net migration figure recorded in U.S. history before the Biden presidency.

The historical high point for immigration in the United States had been the “Great Wave” era at the turn of the twentieth century, when the U.S. population was 14.8 percent immigrant in 1890 and 14.7 percent in 1910. By January of 2025, however, immigrants constituted a record 15.8 percent of the population. Just two years earlier, the Census Bureau had projected that this percentage would not be reached until the year 2042.²

Due to a pre-election tightening of border policy, illegal immigration slowed in the last months of the Biden presidency. Then it came to a near halt when Trump took office. Operational control of the border, as well as deportations, voluntary departures, natural mortality, and a slowdown in legal entries appear to have caused the immigrant population to decline in 2025. Rarely do demographic trends undergo such whiplash.

The sudden decline represents an opportunity for U.S. workers, who now face less competition from immigrant labor. At the same time, the volatility of immigration policy may limit the benefits it can offer. Below, we discuss how immigration can put downward pressure on the wages, employment opportunities, and working conditions of competing American workers. Both low- and high-skill natives are susceptible, but our greater focus will be on the bottom end of the skills spectrum, where immigration impacts the most vulnerable Americans.

Based on our view of the data, the theorized benefits of reduced immigration, such as drawing more low-skill natives back into the labor force, may be beginning to emerge. But in order to fully realize these gains, a sustained policy of low immigration will be necessary over the long term.

What the Numbers Say

We begin with the fraught question of just how many immigrants are in the United States now compared to the beginning of Trump’s term. Estimating the size of the decline is essential to gauging the potential benefits to American workers. Throughout this discussion, we will use the terms immigrant and foreign-born interchangeably to refer to any person residing in the United States who was not a U.S. citizen at birth. Such persons include naturalized citizens, legal permanent residents, temporary workers, foreign students, and illegal aliens. By contrast, we will refer to natural-born citizens as “native” or “U.S.-born.”

Due to its large sample size, the Census Bureau’s annual American Community Survey (ACS) is the most reliable source for estimating the size and characteristics of the foreign-born. Collecting information from roughly nine million participants, however, is a time-consuming task, leaving the ACS perpetually a year or two behind. The most recent ACS data covers the U.S. population as of July 2024, which is too long ago to measure the effect on recent enforcement efforts. Fortunately, there is also the Current Population Survey (CPS), informally called the household survey. Though much smaller than the ACS, the CPS is conducted monthly, with new data released as little as a week after collection ends. The CPS is the source of monthly labor force statistics, including the closely watched unemployment rate.

The January 2025 CPS showed that the total immigrant population when Trump took office was a record 53.3 million. By July of 2025, the CPS showed that the total number of foreign-born had declined by 2.2 million. Our organization, the Center for Immigration Studies, estimated that more than two-thirds of the decline occurred among illegal immigrants. The total number of immigrants working was down one million, with the rest of the decline occurring among children and nonworking adults.

On their face, the CPS data seem to indicate that the Trump administration’s enforcement efforts caused fewer people to enter and significantly more to exit. The margin of error around these numbers, however, is large and the point estimates have fluctuated since July. By February of 2026, the most recent data available at the time of this writing, the foreign-born population is 1.3 million lower than in January 2025. This is not as large as the decline recorded in the July 2025 CPS, but it is still very notable.

Should we believe the CPS data? There are legitimate reasons to be skeptical of the month-to-month estimates. After all, the method used by the Census Bureau to measure net international immigration, which is incorporated into CPS data, is not designed to capture large changes over such short periods of time. At the start of each year, however, the government readjusts the data based on the Census Bureau’s updated estimates of births, deaths, and migration. The fact that the readjusted data from February 2026 still show a substantial foreign-born decline is evidence that the decline has been real all along, although possibly overstated in July.

Another potential issue with the CPS data is self-reporting. The CPS has always relied on the willingness of immigrants to identify themselves as such. Skeptics have alleged that self-identification may have declined, given stepped-up enforcement starting last year. Yet there is no clear indication of this. While the CPS response rate (the share of people agreeing to participate) has been declining for years, monthly fluctuation in response rates since the start of 2025 does not correlate with changes in the size of the foreign-born. There also has not been a significant increase in the share of respondents in the CPS who refuse to answer specific questions about citizenship or country of birth in 2025 and 2026. Furthermore, employers themselves report little growth in their workforces, which is what we would expect if fewer immigrants are coming and more have left. While it is possible that immigrants in general and illegal immigrants in particular have become less willing to take part in the CPS, as best we can tell, this does not seem to be the case.³

Researchers will probably be debating the exact change in the foreign-born population across 2025 for a long time, but here is what we can say right now: the CPS shows the number of immigrants in the country is down substantially since President Trump took office, and although the estimates are imprecise, allegations that they are invalid or worthless are not supported by evidence.

Amid the broad decline in the total foreign-born population, a special focus on illegal immigration is important for the policy insights it can generate. Put simply, illegal immigration is a choice, not an inevitability. A reduction in the illegal immigrant population proves that controlling the border is indeed possible. By contrast, advocates on the progressive left and libertarian right have argued for decades that little can be done to curtail illegal immigration. The border is practically uncontrollable, they claimed. The illegal immigrants already here are never going home. The only real solution, they insisted, is amnesty for the existing illegal population and new avenues for legal immigration. Typical of this genre was a 2023 op-ed in the New York Times, titled “Biden Can’t Stop Immigration. Time to Embrace It.”⁴

The experience of Trump’s first term already hinted that this view was incorrect. New arrivals were lower prior to the pandemic, even as the economy expanded, inflation remained low, and low-skill workers made real wage gains.⁵ “The tightest labor market in more than half a century is finally lifting the wages of the least-skilled workers on the bottom rung of the labor force, bucking years of stagnation,” the New York Times reported in 2019.⁶

That the second Trump administration has succeeded not only in controlling the border but also in reducing the number of illegal immigrants undermines the inevitability narrative. It is understandable that opponents would be reluctant to offer this concession, but the data make it hard for them to avoid.

Can we believe the data specifically on illegal immigrants? Perhaps surprisingly, the majority of illegal immigrants do seem to respond to government surveys.⁷ The most common way to estimate the size of the illegal immigrant population is to subtract the number of legal immigrants in the country, which we know reasonably well, from the total foreign-born recorded in the ACS or CPS. The difference or “residual” population gives us a baseline estimate of illegal immigration, which can then be adjusted upward based on estimates of nonresponse rates.

We believe the illegal population hit 15.8 million in January of 2025, up from 10.2 million at the start of the Biden administration. Importantly, the increase of 5.6 million is not the number of new arrivals. It is the net increase, meaning new additions minus those who “exited” the illegal population over this time period as a result of deaths, outmigration (voluntary or otherwise), and legalizations. The total number of new illegal immigrants who arrived during Biden’s four years could be eight million.

The flow began to reverse under Trump. In July 2025, we estimated that the illegal population had dropped to 14.2 million, down 1.6 million from its peak. As with estimates of the total foreign-born discussed above, similar caveats apply. Given the month-to-month fluctuations and the revised numbers in 2026, the decline has probably not been as large as 1.6 million, but it remains substantial. Corroborating evidence comes from the decline in the CPS of noncitizens from Latin America, who are a common proxy for illegal immigrants. Outside of survey data, we know that border encounters have fallen dramatically since President Trump took office.⁸ Finally, the media have circulated numerous stories of illegal immigrants leaving the country. These are mere anecdotes, of course, but they are consistent with what the other evidence is telling us.

Before moving beyond the topic of illegal immigration, we need to correct a common misunderstanding. While the growth in the illegal population was the most rapid ever from 2021 to 2024, the total did not reach the twenty or thirty million mark, as border hawks sometimes allege. Most experts, including us, do not believe the illegal population is that large because birth records and school enrollment are not dramatically out of line with what the government surveys show. It is theoretically possible that illegal immigrants have very low fertility or do not put their kids in public school, but this is unlikely given everything else we know about the populations from which they are drawn.

While wrong as a technical matter, the sense that there are twenty to thirty million illegal immigrants in the country is understandable. If we take the actual illegal population and add their estimated five to six million U.S.-born minor children, the sum could be around twenty million. There are also several million U.S.-born sons and daughters of illegal immigrants who are now adults. Furthermore, the amnesty granted by the 1986 Immigration Reform and Control Act legalized nearly three million people, and millions of other former illegal immigrants have received legal status in recent decades through asylum grants, marriages to Americans, visa lottery wins, and so on. Putting these numbers together, the total population of illegal immigrants, their offspring, and former illegal immigrants could easily amount to thirty million. Therefore, the sense that the population impact of illegal immigration is much larger than the oft-cited current population size is not entirely misplaced.

How Immigration Affects American Workers

Having established that the foreign-born population has indeed shrunk in the past year, we can now discuss some of the labor market effects to expect if the downturn is sustained. First, wages should rise in the most affected industries. That immigration puts downward pressure on wages follows directly from economic theory. As the supply of labor goes up, its price (the market wage) will go down.

The empirical research generally confirms the theory. Nevertheless, a remarkable disconnect exists between the research itself and the popular discourse about it. Journalists looking to dismiss concerns about immigration often claim that a “majority of economists” or even a “consensus among economists” finds little or no wage effects.⁹ When pressed for detail, advocates usually cite a 1990 study about an influx of Cubans to Miami (the “Mariel boatlift”) as the definitive proof. When asked about contrary studies that may have appeared in the ensuing three decades, they characterize them as outliers.

Consider what the literature actually says. In 2017, the National Academies of Sciences, Engineering, and Medicine published a book-length report on the economics of immigration. It includes a table listing twenty-two different wage effects on different groups of natives from a variety of studies. Eighteen of the twenty-two effects are negative; the table is part of the National Academies’ full chapter discussing an extensive and complex wage literature. No one could read it and come away believing that the effects of immigration are trivial or nonexistent.¹⁰

Since then, economists have continued to publish evidence of negative wage impacts. A 2019 review of over fifty studies, including international evidence, finds that “immigration can create winners and losers among the native-born workers,” with the losers being those who are in competition with immigrants.¹¹ A new historical study shows how immigration in the early twentieth century made skilled natives the “winners” as their wages rose. By contrast, the lower-skill natives were the “losers.” Their wages declined, and they tended to move away from high-immigration areas.¹² A paper published earlier this year shows how immigrants can lower wages even more than expected based on their population size alone, due to various barriers to employment mobility that prevent them from bargaining for higher pay.¹³ A 2024 study describes how the Biden immigration surge “might reduce competition and bidding for workers, thereby easing wage pressures.”¹⁴

That last paper’s result is ironic in that some immigration advocates have actually cheered on the effects it documented. After years of minimizing the wage impacts of immigration, advocates responded to the high inflation of the pandemic years by arguing that more immigrants could lower inflation by holding down wages!

With respect to the famous Mariel boatlift, the labor market lessons of this incident are often overstated in conventional media presentations. When Fidel Castro announced in 1980 that Cubans were free to leave via the port of Mariel, 125,000 immigrants arrived in Miami over a six-month period. Despite the sudden influx, economist David Card was unable to find any wage impacts in his 1990 study of the event.¹⁵ As interesting as the study was at the time, the research has clearly moved beyond it. For one thing, technical criticisms related to defining the native workforce and selecting control cities has revealed the fragility of the result.¹⁶ Second, a more robust data set covering employment listings (not workers) finds a drop in vacancies for low-skill jobs immediately after the boatlift, an indication that the new immigrants helped forestall wage increases.¹⁷

A third criticism of the Mariel study is that immigration affects the internal migration of natives. When immigrants moved to Miami, some natives moved out because the wage went down. Just as importantly, some natives who would have otherwise moved to Miami chose not to do so because the wages there were no longer as high as expected. As a result, the Miami labor market might have produced the appearance of a quick adjustment to the new arrivals without a change in wage levels—but only because native workers have been pushed away.¹⁸

This “crowd-out” effect is a burgeoning area of the literature. It shows that immigrants absorb wage growth in areas with high labor demand, limiting the benefit that natives would experience by moving to those places themselves. The economist Michael Amior has formalized this model in the Journal of Labor Economics. Investigating why internal migration seems to have declined even as average wage disparities persist across regions, Amior observes that immigrants have established enclaves in high-growth areas, crowding out native migration to those same places.¹⁹ Recent studies also point to how immigration reduction influences native mobility. After the Johnson-Reed Act of 1924 sharply curtailed the flow of new immigrants, manufacturers replaced the lost labor with white and black Americans who moved from rural areas to major industrial centers.²⁰ The movement implies that immigration had been discouraging natives from taking these jobs prior to restriction. Afterward, more Americans had access to them.

Therefore, despite the claims made by immigration advocates and allied media, there is certainly no “consensus” in their favor. In fact, the evidence that immigrants do depress the wages of competing natives is strong. Still, the studies are hardly unanimous—few issues are in academia—and one can easily get bogged down in methodological disputes while missing the big picture.

A useful reality check comes from the on-the-ground experiences of American workers in low-skill jobs. The Center for Immigration Studies examined Equal Employment Opportunity Commission (EEOC) cases involving immigration-related claims, and it found the same story over and over. For low-skill manual labor jobs, employers systematically prefer Hispanic (usually immigrant) workers over white and black native workers.²¹

The bias is not subtle. “All you Americans are fired” is an actual quote from the manager of a fruit and vegetable farm during harvest season.²² Employers have used separate employment lines, fake sign-in sheets, unnecessary Spanish-language requirements, and word-of-mouth hiring techniques to purge their companies of native workers and replace them with immigrants. They are often successful. One Houston-area bakery maintained a workforce that was less than 1 percent black, despite being located in a 30-percent-black area.²³

These are anecdotes, to be sure, but there are a lot of them, and they are remarkably consistent. In each EEOC case we examined involving Hispanic immigrants and black Americans, the former were preferred over the latter for low-skill work. When Hispanics do file EEOC claims, it is not to complain that they are being passed over for jobs. Instead, they complain about low pay, dangerous conditions, harassment, and so on. All of these cases are consistent with employers importing foreign labor in order to depress wages and working conditions. As we wrote in the report, “Once we become aware of such behavior, claims that low-skill natives are unharmed by foreign labor seem out of touch with reality.”²⁴

We have focused on low-skill immigration because of the wage and employment competition it creates for Americans who are already disadvantaged. One might assume that the dynamics of high-skill immigration are different, given the entrepreneurship and innovation it may foster. While it is true that genuine “Einstein”-type immigration can create positive externalities, most immigrants who arrive under the high-skill banner are ordinary workers who substitute for rather than supplement native labor.

Take the H-1B program. Elon Musk has praised it for bringing in “critical people,”²⁵ but is that the program’s main purpose? In 2014, Disney infamously laid off about 250 employees and provided some of them severance only on the condition that they train their H-1B replacements.²⁶ The incident should not be surprising. A new NBER study finds that H-1B workers are paid 15 percent less than comparable natives.²⁷ Furthermore, a 2022 analysis shows that when H-1Bs are randomly distributed to firms via lottery, the winning firms do not increase their employment levels compared to losing firms. Instead, the visa-holders appear to crowd out native workers.²⁸ As with low-skill immigrants, many employers prefer “high-skill” immigrants over natives because they are cheaper and more controllable, not because they possess rare talents.

What the Early Signs Indicate

It remains too early to fully assess whether the immigration decline in 2025 is helping to improve the labor market prospects for natives, as the theory described above predicts. For one thing, the recent decline came after a historic surge, and there is no indication yet that the size of the foreign-born population is even back down to where we would expect based on the pre-Biden trend line. Furthermore, we know that periods of relatively low immigration are consistent with a booming economy and rising low-skill wages, as the first three years of the first Trump administration showed. It is difficult, however, to separate current labor market conditions with exogenous shocks to the economy, such as a spike in oil prices.

That said, there are some positive signs. Comparing January of 2025 to February 2026 shows that the number of employed native workers is up 412,000, while the number of employed immigrants is down 606,000 over this period. Given the way the CPS is conducted, a decline in the number of immigrant workers will tend to be accompanied by an increase in the number of U.S.-born workers, so it is important not to overstate this evidence, but it is encouraging nonetheless.

The unemployment rate for natives at 4.7 percent is slightly higher in February than the 4.3 percent in January 2025. The rise in the number of natives working, however, means that the increase in the unemployment rate was not caused by a net loss in jobs among natives; rather, it was due to more natives entering the labor force looking for work. (People who are not looking for work are not counted as “unemployed.”) Most encouragingly, this return to the labor force was concentrated among prime age (twenty-five to fifty-four-year-old) U.S.-born men without a bachelor’s degree. The share of this group not in the labor force (neither working nor looking for work) declined from 15.7 percent to 15.0 percent between January 2025 and February 2026.

There are always short-term fluctuations in the data, but over the last six decades, the share of U.S.-born men without a college degree who are out of the labor force has increased dramatically. In 1960, only 4.2 percent were not in the labor force. As recently as 2000, the figure was still “only” 10.5 percent. This long-term decline in work is not due to more men being incarcerated. If men in prisons and jails were counted as not working, the deterioration would appear even worse. In any case, the decline in work is a social disaster. There is a large body of research linking long-term unemployment to a host of social pathologies, from crime and welfare dependency to overdose deaths and political alienation.²⁹ Even a modest reduction in the share of these men on the economic sidelines is good news.

These positive signs could easily be missed if commentators conflate immigrants and natives. It is true that job growth slowed in 2025 and the first part of 2026. The slowdown must be interpreted in the context of a fall-off in the size of the illegal immigrant population, however. At the start of 2025 there were perhaps ten to eleven million illegal immigrants working in the United States; the rest were children, the elderly, and working-age people not in the labor force (mostly women with young children). Employed illegal immigrants account for 6 to 7 percent of all workers.³⁰ Any successful effort to reduce illegal immigration will reduce the number of illegal immigrant workers and, hence, the number of workers firms employ, especially in the short run.

The overall effect of immigration reduction on employment is large because much of the increase in employment in recent years has gone to the foreign-born. The household survey shows that the number of employed immigrants increased by 2.2 million from February of 2023 to February of 2025, accounting for three-fourths of total employment growth.³¹ As already noted, the number of employed immigrants in February 2026 is lower than in February 2025, while native employment has gone up. Enforcing immigration means there will be less job growth. This is entirely expected, and even desirable, if the overall population is growing less and natives have more job opportunities.

Toward a Low Immigration Economy

Unfortunately, a single year of decline in the foreign-born, especially coming on the heels of record-high increases, will not be enough to produce large benefits for American workers. What is required is a long-term reduction in immigration that is consistent and predictable. Only after employers, politicians, and the workers themselves come to expect a low-immigration environment will behaviors change fundamentally.

Among the three groups listed above, the response of employers to persistently low immigration should be the quickest and most straightforward. Right now, employers constantly claim that there are not enough native workers, but what they actually mean is not enough natives will work at the (low) wage they would like to pay.

Out of necessity, if low immigration becomes a permanent, enforceable economic reality, employers will put more effort into recruiting and retaining workers. We already have small examples of this from previous years. For instance, the Wall Street Journal reported that the farming and construction industries in Arizona suffered a “shortage” of workers after the state passed new enforcement laws back in 2016. According to one contractor, “Now you have to put out feelers, buy ads, go on Craigslist, tap job agencies just to get a few men.” Wages increased as a result.³² Similarly, the period of relatively low immigration during the first Trump administration caused employers to recruit from marginalized groups. The low unemployment rate “has brought record numbers of people with disabilities into the workforce,” the Washington Post reported at the time. In addition, “big companies such as Walmart and Koch Industries . . . are turning to an underutilized source of labor: inmates and the formerly incarcerated.”³³ This happens only when immigration is restricted.

Despite the clear examples of firms successfully recruiting natives when immigrant labor is less plentiful, employers continue to insist that they have done all they can and that disastrous “shortages” will result from reductions in immigration. The wage data do not bear that out. While the food service and farming sectors have experienced real (inflation-adjusted) annual wage gains of about 3 percent, wages in most immigrant-heavy occupations have been rising far less rapidly. In each of the occupational categories that the government labels “construction, transportation, and moving,” “building cleaning,” “grounds keeping,” and “installation and repair occupations,” real wages were about the same in the fourth quarter of 2025 as they were in 2019 before the pandemic.³⁴ That is not what we would expect if employers had exhausted every option in trying to find native workers.

There is, of course, a limit to the number of out-of-work Americans who can be quickly absorbed back into the labor market. Some have personal problems such as poor work ethic, drug abuse, and welfare dependency that cannot be solved overnight. Here is where employers need to petition politicians to address these problems directly. We will not try here to weigh the relative merits of restricting welfare access (particularly disability), tying benefits to work, mandating harsher punishments for illegal drugs, increasing the minimum wage, or promoting the trades. The point is that long-term immigration reductions generate the incentive to try all of these ideas and more. By contrast, when foreign labor flows freely, employers care little about the status of low-skill natives, and politicians feel much less pressure to address the social problems those natives experience.

As for low-skill natives themselves, especially those with a tenuous connection to the labor force, attitudes toward work should gradually change. When the political scientist Charles Murray called for a moratorium on low-skill immigration, despite his libertarian leanings, he noted the value of cultural feedback loops: when men have fewer excuses to avoid work, their growing interest could spur other men to work, and perhaps group expectations could change in the long run.³⁵

Skilled natives, although not our primary focus, also could see new opportunities from a low-immigration environment. As of 2023, 7.3 million natives had a degree in science, technology, engineering, or mathematics (STEM) but worked in a non-STEM job.³⁶ If “high-skill” immigration were reduced to about ten thousand genuinely exceptional candidates per year, forgoing the ordinary college graduates that programs such as H-1B import, then native workers may be able to fill the void in the STEM labor market.

Regardless of skill level, a significant component of worker adaptation must come from geographic mobility. As noted earlier, much of the wage depression caused by immigration is a result of immigrants absorbing wage gains from areas experiencing high labor demand. Most natives will not move to these places in response to a small and temporary uptick in wages. Instead, a sustained period of low immigration will be necessary to assure natives that it is in their long-term interest to move to those places where the wage is higher.

Is a long-term immigration reduction feasible politically? Our hope is that a focus on bringing natives back into the labor force can start to alter the current political dynamics around immigration. Progressives have become so uniformly in favor of more immigration that even the Economic Policy Institute, the labor unions’ think tank, aligned itself almost completely with the Biden administration’s approach.³⁷

Meanwhile, despite President Trump’s working-class base, Republicans remain susceptible to business talking points about labor shortages. Look no further than Texas, which still does not mandate universal E-Verify, the system that allows employers to check the work eligibility of their employees. We are not naïve about the political difficulties surrounding immigration, but we nonetheless wish to underscore the plight and wellbeing of disadvantaged native workers, a group that both parties claim to have some affinity for, either now or in the recent past; recognizing labor market security as an essential factor in advancing their economic future should be the way forward for American policymakers of all stripes.

Although 2025 was a great start, undoing the damage to the American labor market caused by mass immigration will take years of effort. Both Congress and the president should focus on ways to sustain the progress made and avoid regressions back to the destabilizing policies of the past.

This article is an American Affairs online exclusive, published May 20, 2026.

---

Notes