By Darrell Dow
Homeland Security Secretary Kristi Noem has said the immigration crackdown is working, and that 1.6 million illegal aliens have self-deported since President Trump took office. “This is massive,” Noem said. “This new data shows illegal aliens are hearing our message.”
But Noem’s claim mightn’t be true.
Noem’s number comes from a report by the Center for Immigration Studies, which analyzed federal labor force data. On the surface, the math looks convincing: Fewer immigrants showing up in the survey means fewer immigrants in the country. Yet scratch the surface and you see how fragile the evidence is.
Start with the basics. The federal government knows very little about the day-to-day size of the illegal-alien population. Deportations are carefully tracked — but only about 150,000 have taken place since Trump’s inauguration. The much larger 1.6 million figure comes not from enforcement records but from the Current Population Survey, a monthly survey of about 60,000 households conducted by the Census Bureau and the Bureau of Labor Statistics.
The CPS is invaluable for broad labor market trends. But using it to measure the number of illegal aliens is like using a bathroom scale to weigh a truck. The tool simply wasn’t built for the job.
Here’s why:
First, immigrants — especially those here illegally — are less likely to respond to government surveys. If that reluctance grows because of stepped-up enforcement or fear, the survey doesn’t show fewer immigrants living in the U.S.; it shows fewer willing to answer the phone. The result is a phantom “decline.”
Second, the survey is adjusted by response rates, or the lack thereof, which creates an illusion because changing “nativity” affects the weighting of the survey. CPS doesn’t measure the U.S. population directly. Instead, it’s statistically modeled to match Census Bureau projections, called “population controls.” These controls lock in the total population, as well as age, sex, and race/ethnicity counts, based on estimates made the year before.
Imagine the CPS as a pie chart that always has 100 slices. The Census Bureau decides how many slices go to men vs. women, young vs. old, but not to U.S.-born vs. foreign-born. That part depends entirely on survey responses. If fewer immigrants answer, their slices don’t vanish — they are reassigned to native-born Americans. On paper, the native-born population grows while the foreign-born shrinks, even if nothing has changed.
Third, the sample size is small. A shift of just a few hundred immigrant respondents in the survey can be magnified into hundreds of thousands in the national estimate. Indeed, the official confidence interval for the July 2025 immigrant employment number is plus or minus 720,000. That’s nearly the same size as the supposed decline.
None of this means immigration enforcement has had no effect. Industries that rely heavily on immigrant labor — construction, restaurants, hospitality, agriculture — have seen slowed rates of employment growth. States with more immigration arrests, like Texas and Florida, show weaker labor force growth than states with fewer. Clearly, something is happening.
But the evidence that 1.6 million people have packed up and left is far thinner than Noem suggests.
Other labor market indicators flip the story. Unemployment among native-born workers is higher than among immigrants. Native workers today are more likely to be unemployed than they were a year ago. If immigrants were truly vanishing in droves, we’d expect natives to be snapping up the jobs they left behind. But that’s not happening.
Truth is, we don’t know how many illegal aliens are leaving because the surveys we rely on aren’t designed to tell us. What we do know is that fear, politics, and methodology quirks can make it look like millions have “self-deported” when the truth is murkier.
So whether the crackdown is “working” depends on what one means. Enforcement is creating uncertainty in various industries and states. But the big number of self deportations — 1.6 million — rests on shaky ground. It might be less a measure of people leaving America than of immigrants disappearing from the survey data.
That might be good politics. But it isn’t solid evidence.
Darrell Dow writes for American Remnant.
