1.6 Million Self-Deportations? DHS's Unverifiable Claim, Exposed

The Census Bureau's own methodology research cautions that the CPS "shouldn't be used to determine the size of the foreign-born population," because its sample of roughly 60,000 households is a fraction of the size of the American Community Survey (ACS), which samples about 3.5 million households and is the tool researchers normally use to study immigrant populations.^[5][6]

In July 2025 the CPS reported an adult foreign-born population of 48.5 million people with a margin of error of plus or minus 830,000 — nearly as large as the claimed 1.6 million decline.^[2]

Robert Warren of the Center for Migration Studies explained that researchers "generally rely on the bureau's more robust American Community Survey" because "the Current Population Survey's sample size has a large margin of error."^[2]

The Census Bureau separately reported that the United States experienced an unprecedented shift toward negative net migration in 2025: "Currently, the estimates of NIM are trending toward negative net migration. If those trends continue, it would be the first time the United States has seen net negative migration in more than 50 years."^[7] But the Bureau attributed the shift to a mix of factors — slower immigration, more emigration, and a lag between policy change and the data — not to a single clean cause.^[7]

A 60,000-household survey the Census Bureau itself says shouldn't be used this way is the entire foundation of DHS's headline number.

The St. Louis Federal Reserve published an analysis showing that the CPS population decline is substantially explained by decreased immigrant survey participation, not actual emigration.^[8]

Non-naturalized immigrant response rates to the CPS dropped 16.6% from January to November 2025 — roughly 10 percentage points more than the decline among native-born and naturalized citizens.^[8] The Fed concluded that "the CPS population estimate showing a 1.86-million decline in the number of immigrants mostly reflects a decrease in the participation of immigrants in the CPS," driven by non-naturalized immigrants who remain in the country but are "wary of participating in government data collection."^[8]

Julia Gelatt of the Migration Policy Institute called the estimate shaky: "I see the 1.6 million as a pretty shaky estimate, and one that could be explained away by fewer immigrants feeling comfortable answering a government survey."^[5]

A KFF/New York Times survey of immigrants found that 41% worried they or a family member could be detained or deported — up from 26% in 2023 — and 14% said they or a family member had avoided going to church or community events in 2025 because of concerns about drawing attention to their immigration status.^[9]

Jeffrey Passel of Pew Research Center noted: "For ^[20]25, all we have is CPS data and that points to a decline from a January peak of immigrants. How big the decline is could be affected by the decrease in response rates."^[10]

Every economist who has looked closely says the same thing: fewer immigrants answering a government survey is not the same as fewer immigrants in the country.

Even the original researcher behind the 1.6 million figure acknowledged it does not represent purely voluntary self-deportations.

Steven Camarota of CIS told PolitiFact that the "overwhelming majority" of the 1.6 million would have left on their own, but the figure also encompasses "immigrants who, for example, were deported, died, left voluntarily and those whose status changed."^[2]

This directly contradicts DHS's characterization of the figure as representing purely "self-deportations."

CIS's own report flagged the uncertainty in its estimate: "given the recent stepped-up enforcement efforts, it is possible that immigrants have become more reluctant to respond to the CPS" — which would overstate the actual decline — and the researchers' estimate of the legal immigrant population relied on administrative data that "is incomplete, further increasing uncertainty of our estimate of illegal immigrants."^[3]

The think tank whose number DHS is citing warned, in its own report, that the number might be an overstatement — a caveat DHS never mentioned.

The escalation from 1.6 million to 2.2 million represents a 37.5% increase, yet DHS released no new data, methodology, or verification.^[4] The original CIS estimate covered only January-July 2025; DHS's subsequent increases extrapolated beyond the source data's timeframe without explanation.

Austin Kocher, a researcher at the Transactional Records Access Clearinghouse (TRAC), examined a related DHS deportation figure and wrote: "DHS Claims 622,000 Deportations, Here's Why That Number Might Be Meaningless."^[11] Kocher found DHS's public accounting methods so opaque that outside researchers cannot reconstruct how the agency's headline totals are built, and concluded the number "cannot be compared over time to past years and past administrations."^[11]

When a government agency can't explain its own math, and outside researchers can't reproduce it, that's not evidence — it's a press release.

If 1.6-2.2 million workers actually left the United States, labor markets should show dramatic disruption. Independent economic analysis found a real but much smaller slowdown.

The number of foreign-born workers in the labor force fell by 1.6 million, about 5%, from March 2025 through the summer.^[12] University of Michigan economist Betsey Stevenson noted the resulting slowdown: "Looking at the last three months, we're creating fewer jobs by quite a lot. Instead of creating 180,000 jobs per month" — roughly the pace during the last two years of the Biden administration — "we're creating something closer to 35,000 jobs-per-month."^[12]

Employment expanded by only 584,000 in all of 2025 — the weakest annual pace since 2020 and the second-weakest since the 2009 financial crisis. A whopping 84% of those gains occurred in the first four months of the year, before immigration enforcement intensified.^[13][14]

The Minneapolis Federal Reserve found that between 40% and 60% of the slowdown in monthly payroll growth is attributable to declining net immigration — a major factor, but one that on its own implies workforce reductions far smaller than 1.6-2.2 million.^[15]

A real slowdown, driven partly by immigration, is not the same as proof that millions of people left the country.

CBP border encounter data for FY 2025-2026 shows unprecedented declines in new arrivals, but this measures who is not crossing in, not who is leaving:

Southwest border apprehensions in FY 2025 (237,538) hit their lowest level since 1970 — an 84% drop from a year earlier — while nationwide encounters fell 79% compared to FY 2024.^[16] October 2025, the first month of FY 2026, recorded just 30,561 total nationwide encounters, the lowest start to a fiscal year in CBP history.^[17] November 2025 fell further still, marking a seventh consecutive month of zero U.S. Border Patrol releases.^[18]

These statistics demonstrate a collapse in new arrivals, not evidence of mass departures. CBP's own nationwide encounters dashboard tracks people entering, not people leaving — it does not track voluntary departures or emigration from the U.S. interior.^[19]

Record-low border crossings are real and well-documented. They say nothing about how many people already inside the country left.

Using Transactional Records Access Clearinghouse data, Brookings researchers found around 40,000 immigration-court voluntary-departure resolutions in 2025, compared to around 10,000 in 2024 — roughly a fourfold increase, but only about 1.8% of DHS's claimed 2.2 million self-deportations.^[20]

That increase is consistent with EOIR's own FY2025 accounting: immigration judges issued 767,398 initial case decisions, of which voluntary departure accounted for about 5% of outcomes — up sharply from roughly 1% the prior fiscal year.^[21]

Attorneys say restricted bond access and falling asylum success rates are driving detained immigrants to give up their cases rather than keep fighting them. Asylum grant rates plummeted from over 50% monthly in 2022-2024 to only 29% by December 2025, and the share of detained removal cases ending in voluntary departure climbed from 28% for all of 2025 to 38% in December alone.^[22]

Voluntary departures really are rising fast. They are also nowhere near DHS's 2.2 million.

The DHS self-deportation claims have become a cornerstone of Republican campaign messaging heading into the November 2026 midterm elections. In January 2026, DHS said it had removed more than 675,000 migrants since Trump took office and claimed an estimated 2.2 million others had "self-deported."^[23]

Senator Katie Britt (R-AL) defended the administration's approach: "President Trump is not going to back away from the mission, the mission that American people said they wanted him to complete, and that is securing our border and making sure that we actually do interior enforcement."^[24]

After federal immigration agents fatally shot Alex Pretti, a Minneapolis nurse, on January 24, 2026, Senate Democrats refused to pass a full-year DHS funding bill without accountability reforms, passing only a two-week stopgap instead; this fed into a partial DHS shutdown that began in mid-February 2026, though ICE and CBP continued enforcement operations using funds carried over from 2025 tax-and-spending legislation.^[25][26]

Rep. Delia Ramirez (D-IL) said it was vital that Democrats "begin building a case against the agency ahead of the midterm elections."^[27]

DHS's chief investigations arm also launched a campaign to investigate naturalized citizens over voting activity that occurred before they became citizens, escalating concerns among federal prosecutors and other law enforcement agents that the effort would frighten legitimate voters away from the polls in the midterms.^[28]

The self-deportation numbers aren't just a statistics dispute — they're the centerpiece of a live campaign message, which raises the stakes on getting them right.

Brookings estimated that reduced immigration will drag on 2025 GDP growth by 0.2 to 0.3 percentage points, and on 2026 GDP growth by 0.1 to 0.3 percentage points, while weakening consumer spending by an estimated $60-$110 billion combined over 2025-2026.^[20]

The Peterson Institute for International Economics modeled two deportation scenarios: removing 1.3 million unauthorized workers (roughly the scale of the 1950s' Operation Wetback) versus 8.3 million (roughly all unauthorized workers). By 2028, the smaller scenario cuts GDP 1.2% and employment 1.1% below baseline; the larger scenario cuts GDP 7.4% and employment 7.0% below baseline, and raises consumer prices well above what they would otherwise be.^[29] A related analysis found deporting 500,000 immigrants costs roughly 44,000 U.S.-born workers their jobs.^[30]

By every independent economic model, mass removals this large would be impossible to miss in the data. The labor-market damage that is visible is real — but it is a fraction of what 1.6-2.2 million departures would produce.

No previous administration has claimed self-deportation numbers on the scale of the Trump administration's 2025-2026 claims. The closest historical parallel is Operation Wetback (1954).^[31][32]

Operation Wetback was a program of mass apprehensions and deportations targeting undocumented Mexican nationals, carried out by the U.S. Border Patrol under President Eisenhower.^[33] A key aspect was using fear to induce self-deportation.

Alexander Aviña, associate professor of history at Arizona State University, noted: "There is this element of spectacle that both administrations use. They try to induce fear and terror in immigrant populations to get them to self-deport, and that's something they tried to do with Operation Wetback."^[33]

The historical record on Operation Wetback's scale is heavily contested. The Eisenhower administration officially claimed more than 1 million "returns."^[34] But historian Kelly Lytle Hernandez's own review of Border Patrol's fiscal-year data found the agency apprehended, at most, fewer than 300,000 people during the actual 1954 campaign — the "one million" figure came from double- and triple-counting people caught more than once.^[35]

Doris Meissner, former INS Commissioner, said of the operation: "It was lawless; it was arbitrary; it was based on a lot of xenophobia, and it resulted in sizable large-scale violations of people's rights, including the forced deportation of U.S. citizens."^[36]

The one precedent DHS implicitly invokes was itself built on an inflated headline number — history repeating with a very on-the-nose irony.

The government launched the CBP Home app with cash incentives to encourage voluntary departures — starting at $1,000, tripled to $3,000 for those who departed by the end of 2025, then settled at a $2,600 standing stipend from January 2026 onward.^[37][38]

Actual, government-confirmed participation is a small fraction of the claim: DHS itself said in January 2026 that "there have already been nearly 100,000 users of the CBP Home app."^[38]

Even taking the highest figures at face value, that verified program has served about 100,000 people — 1/16th of DHS's claimed 1.6 million self-deportations and 4.5% of the claimed 2.2 million.

DHS Claim: 2.2 million self-deportations by January 2026

Verified CBP Home participants: ~100,000 (4.5% of claim)^[38]

Court-ordered voluntary departures: ~40,000 in 2025 (1.8% of claim)^[20]

Brookings' modeled estimate of voluntary departures: 210,000-405,000 (9.5%-18% of claim)^[20]

Discrepancy: Every verifiable measure of actual voluntary departures falls 4x-10x short of DHS's claim.

The Government Accountability Office (GAO) identified systematic data quality problems with ICE deportation reporting in a comprehensive audit published in July 2024.^[39]

GAO found that ICE systematically understates the total number of people it detains in its public reports. ICE excludes people who were first booked into certain temporary facilities before being transferred to immigration detention facilities — an exclusion amounting to tens of thousands of people.^[39]

In FY 2022 alone, ICE's reporting excluded detentions of 136,706 individuals. ICE does not fully explain its rationale for these exclusions or its methodology.^[39]

GAO concluded: "Congress and the public may not have a complete understanding of how to interpret the data."^[39] A separate 2025 GAO report found that DHS and ICE also lack clear performance goals and measures for immigration detention facility inspection programs.^[40]

Deficiencies related to ICE's 14 core detention standards — including environmental health and safety, medical care, and food service — accounted for approximately 74% of the total deficiencies identified across all inspections GAO reviewed.^[40]

DHS's basic detention and deportation numbers have an independently documented accuracy problem — before you even get to the self-deportation claim.

DHS Secretary Kristi Noem testified before Congress multiple times in 2025-2026, making escalating claims about deportation numbers while contradicting herself on key facts.

On May 20, 2025, Noem testified before the Senate Homeland Security Committee that the Trump administration's FY 2026 budget request included over $175 billion for DHS "to implement the President's mass removal campaign."^[41] She said ICE arrests of suspected gang members represented a "105 percent increase" compared to the same period in FY 2024, and arrests of known or suspected terrorists a "700 percent increase."^[41]

In a contentious hearing on December 11, 2025, Rep. Seth Magaziner (D-RI) asked Noem directly whether DHS had deported any military veterans. "Sir, we have not deported U.S. citizens or military veterans," Noem responded — before an aide displayed a Zoom call with Purple Heart recipient Sae Joon Park, a green-card-holding Army veteran shot twice while serving in Panama in 1989, who had self-deported to South Korea earlier that year after being given the choice between detention or leaving voluntarily.^[42]

The same day, Congressman Seth Moulton separately released a September 2, 2025 letter from Secretary Noem that directly contradicted her testimony. The letter read, in part: "Regarding your question on the number of veterans that have been removed since January 20, 2025, ICE has removed eight veterans."^[43]

A Washington Post analysis of ICE arrest data found that the administration is increasingly targeting immigrants with no criminal record, and that more than half of those removed from the country since January 20, 2025 do not have a criminal conviction — contradicting the administration's framing that enforcement targets only the "worst of the worst."^[44]

A Secretary who testified under oath that no veterans had been deported was contradicted, on the same day, by her own department's paper trail.

The Department of Homeland Security's claim that 1.6-2.2 million undocumented immigrants "voluntarily self-deported" in 2025 represents one of the most significant instances of government statistical overreach in recent immigration enforcement history. The claim fails basic methodological scrutiny on every level:

Independent estimates of actual voluntary departures — from Brookings' economic model (210,000-405,000) to TRAC's immigration-court data (roughly 40,000 court-ordered departures) to DHS's own verified CBP Home app enrollment (roughly 100,000) — cluster far below DHS's claim, at somewhere between one-twentieth and one-fifth of the 2.2 million figure.^[20]

The Peterson Institute for International Economics separately concluded that the underlying 2.2 million CPS-based population estimate is "implausibly large, uncorroborated, and implies absurd gyrations in the labor market."^[45] Brookings went further, writing of DHS's own December 2025 combined total that it "should not be considered a serious source" for an estimate of net migration.^[20]

The most defensible reading of the evidence: aggressive enforcement genuinely pushed some immigrants to leave and drove others "deeper into the shadows," reducing their willingness to respond to government surveys. DHS then combined a shaky, self-admittedly caveated third-party population estimate with its own removal count, presented the sum as a verified enforcement outcome, and kept inflating it for six months without producing any new evidence.

As the St. Louis Federal Reserve concluded, the real question is "whether the decline in surveyed immigrants indeed means an actual drop in net immigration or simply fewer immigrants choosing to participate in the survey... the CPS population estimate showing a 1.86-million decline in the number of immigrants mostly reflects a decrease in the participation of immigrants in the CPS."^[8]