OP-ED: Study Finds No U.S. Jobs Saved by ICE Raids

The Interdependence of Jobs in a Local Economy

The most common explanation, put forward by labor economists cited in the broader debate on immigration, centers on the interdependence of jobs in a local economy. A restaurant that loses its kitchen staff cuts back on front-of-house positions. A farm that loses its farmworkers reduces its purchases of local equipment and services. In many sectors, immigrant jobs and U.S.-born jobs are not simply “communicating vessels”—they are interconnected.

The areas with high ICE activity documented by the study show no signs of net substitution, where U.S. workers would fill the positions left vacant by arrests. The opposite pattern—a general contraction in local economic activity—is more consistent with the measured data.

An effect that also affects workers without a college degree

The 2.5% decline among male immigrants without a college degree, for every 1,000 arrests, illustrates a direct and expected effect: less available labor in this specific segment of the labor market. This figure is not disputed in the sources consulted; it is the second figure—the effect on U.S.-born workers—that represents the true departure from the official narrative.

Two declines in employment, among two different populations, in the same geographic area, following the same event. This is not the coincidence that the protectionist narrative had promised.

A conservative media outlet calls the study “flawed”

On May 10, 2026, the Washington Times, a conservative media outlet, published a methodological critique of this study, calling it “flawed” (as stated in the original text). This critique is part of the documentation for this post and must be reported with the same rigor as the study’s findings themselves.

None of the sources consulted for this post specify, beyond the general characterization, exactly which specific methodological aspects the Washington Times would dispute—sample size, choice of geographic areas for comparison, or the statistical model used. This limitation must be noted rather than filled in with assumptions.

What the “working paper” status changes—and what it does not

The study is an NBER working paper, which means it has not necessarily undergone peer review in the strict sense of traditional academic journals. This is an important methodological point to note for any reader assessing the robustness of these conclusions.

A working paper is not definitive proof. Nor is it a worthless opinion. The NBER, an academic institution recognized for empirical economic research, publishes working papers that are regularly cited and subject to serious scholarly debate prior to any final publication in a journal.

A disputed figure is not a figure that has been erased

The criticism from the Washington Times exists and must be mentioned. However, given the current state of available sources, it is not sufficient to dismiss or refute the specific figures presented in the study. No counter-study, nor any alternative data with specific figures and sources, was identified in the materials reviewed for this post that would directly contradict the central finding of Cox and East.

This post does not claim to settle an academic methodological debate in a few paragraphs. It reports what the study asserts, what the critique asserts, and leaves the question open where the sources themselves leave it open.

The political significance of the study goes beyond the academic debate

Whether the study is methodologically flawless or flawed, it has already produced a measurable political impact: since it was brought to the forefront by Raw Story on July 29, 2026, it has become a reference point cited in the public debate on the Trump administration’s immigration policy. An academic working paper that bridges the gap between university research and political debate is not a neutral event.

A contested study that nevertheless circulates in the public debate has an impact even before the methodological debate has been resolved.

The Repeated Promise: Raids Protect American Jobs

The Trump administration’s protectionist rhetoric on immigration—repeated since the start of its second term in January 2025—presents ICE raids as a tool to protect jobs for U.S.-born workers. This claim is one of the central rhetorical pillars of current immigration policy.

None of the sources consulted for this post allow for verification of a specific, dated statement by an official administration spokesperson regarding this particular figure from the NBER study. This absence should be noted: this post documents the gap between a repeated general narrative and a specific empirical measure, not a point-by-point response from the administration to this particular study.

This discrepancy adds to other issues documented during the same period

This issue regarding employment is not an isolated case of tension between official rhetoric and measured data during this period. A separate fact-check by Daniel Dale, a CNN reporter, published on July 31, 2026, also concluded that Trump’s claim about a general decline in consumer prices was “absolutely and unequivocally false.” The two cases are distinct but share the same time frame and the same dynamic: an assertive economic narrative, and measured data that contradicts it.

There is no evidence to suggest that these two cases are linked by a common cause. At the very least, they share the same week and the same pattern.

How the study defines these areas

The sources consulted for this post do not specify the exact method used by Cox and East to geographically delineate “areas with high ICE activity.” This post therefore cannot specify whether the study relies on data from counties, cities, or metropolitan areas. This technical limitation warrants mention rather than being filled in with an unverified assumption.

What the data do allow us to state with certainty is that the comparison is made between areas with high and low arrest rates—a standard method in labor economics for isolating the effect of a localized shock.

A standard method applied to a politically charged topic

Comparing areas with high and low intensities of an economic shock—in this case, ICE arrests—is a common approach in empirical research. It is not the method itself that is the subject of the controversy reported by the Washington Times, but its specific application to this particular case, according to the general characterization available in the sources.

A standard method applied to a politically charged topic attracts disproportionate political attention relative to its actual methodological novelty.

No Verdict on Migration Policy as a Whole

This study focuses specifically on the impact on employment measured in areas with high ICE activity. It does not, on its own, constitute a comprehensive assessment of the Trump administration’s immigration policy as a whole—border security, the asylum system, or any other aspect of this policy falls outside the scope of this specific research.

This post is strictly limited to what the study measures: the impact on employment. Extending this conclusion to other dimensions of immigration policy would go beyond what the data support.

No Single, Definitive Causality

Even within its specific domain—employment—the study documents a correlation observed in specific geographic areas during a specific time period. Any empirical economic research of this type has limitations inherent in its methodology; Cox and East themselves, according to the available excerpts, do not present their work as absolute and universal proof of causation.

A measured figure is not a decree. It remains an observation—however robust it may be—subject to the same rule as any scientific data: it may be revised by subsequent research.

A policy already being challenged in federal court

This employment study is not the only front on which the administration’s immigration policy has been challenged during this period. On July 30, 2026, the federal appellate courts of the 9th and 7th Circuits each struck down the practice of indefinite detention of migrants without a bail hearing, according to The New York Times. This separate legal development adds, within the same timeframe, to the broader debate over the administration’s immigration policy during its second term.

There is no evidence to establish a causal link between these judicial decisions and the NBER study on employment; these are two separate issues that, taken together, paint a picture of an administration simultaneously facing challenges on several distinct fronts during the same weeks of July and August 2026.

Cumulative pressure, with no evidence of a direct link between the cases

It would be inaccurate to present these issues—the employment study and the court rulings on detention—as components of a single causal narrative. Each issue is based on its own facts, its own sources, and its own timeline. This post mentions them side by side to contextualize the broader climate without ever asserting a proven link between them.

Several distinct fronts, open in the same window, do not automatically constitute a single battle. At the very least, they represent a shared difficult period for the same policy.

No national extrapolation is stated in the available excerpts

The excerpts from the study consulted for this post contain no statements by Cox or East extrapolating their local results to the entire United States in a general and unqualified manner. The results pertain to the areas measured during the period under study.

This post applies the same rigor to the study as to any other source: reporting what is written, pointing out what is missing, and never filling a gap in the evidence with a claim that goes beyond the available text.

A nuanced conclusion, not a slogan

“No evidence of a positive effect” is not the same as “evidence of a widespread economic disaster.” The study’s precise wording—the absence of a measured positive effect, combined with a measured negative effect among U.S.-born workers—deserves to be reported in its exact terms, without exaggeration or downplaying.

Nuance is not a rhetorical weakness. It is the only thing that distinguishes a rigorous analysis from a campaign slogan.

An Academic Debate That Is Not Yet Over

The Washington Times’ methodological critique and the lack, to date, of a counter-study with quantified data mean that this issue remains open from an academic standpoint. Other researchers may, in the coming months, publish complementary or contradictory analyses that would clarify—or more rigorously challenge—Cox and East’s figures.

This post documents the state of the debate at the time of writing, with the caution that this type of evolving academic issue inherently requires.

A study that is already fueling the political debate, regardless of its academic outcome

Whether or not the study withstands long-term scientific scrutiny, it has already entered the political debate on immigration as a cited reference. This political reality—its citation, circulation, and media coverage—exists independently of the final academic verdict, which remains undecided in the available sources as of today.

A figure enters the public debate long before the scientific community has finished testing it. This is the rule, not the exception, for any politically sensitive economic research.

Neither an Acquittal nor a Conviction of Immigration Policy

This post does not seek to exonerate or condemn the Trump administration’s immigration policy as a whole. It reports a specific figure, a specific criticism, and a specific context, within the strict limits of what documented sources allow us to assert.

Any interpretation that would turn this post into an overall verdict on immigration would go beyond what the available factual evidence permits us to conclude.

The figure remains the figure, regardless of how it is interpreted politically

Whether one supports or opposes the administration’s immigration policy, the figure of a -0.54% drop in employment among U.S.-born workers per 1,000 ICE arrests in high-enforcement areas remains what it is: a metric that has been published, criticized, but not refuted by counter-data from the available sources.

Numbers don’t vote and don’t fly flags. They simply wait to be interpreted correctly—or distorted.

A Recurring Gap Between Economic Promises and Empirical Findings

This report illustrates a pattern that extends beyond the specific case of immigration: economic discourse promises a specific outcome, empirical research measures something else, and public debate must then grapple with this discrepancy. This pattern is not limited to immigration policy; it cuts across several economic issues from the second term, documented separately in other analyses.

This post does not claim to generalize this pattern beyond this specific case, but it provides a well-documented and quantified illustration of this discrepancy as it pertains to this particular issue.

The Value of a Contested but Unrefuted Figure

A figure that has been criticized on methodological grounds but never refuted by a counter-figure retains real informational value—provided it is presented with its exact limitations, which this post has sought to do throughout this analysis.

A contested figure that stands its ground is not yet an established fact beyond all doubt. It is, at the very least, a fact that, for the time being, withstands the challenges it has faced.

The Limits of This Post

Due to a lack of available sources, this post cannot specify the exact method used to define the geographical boundaries of the areas studied, nor can it provide a complete breakdown of the Washington Times’ methodological objections beyond the general characterization of the research as “imperfect.” These limitations must be acknowledged, not overcome.

Nor can this post predict the outcome of the upcoming academic debate on this study, nor assess whether future research will confirm or refute its findings.

What, however, remains firmly established

What remains firmly established in the sources consulted: the study exists, it is published by the NBER, it presents specific figures, it has been criticized by an identified conservative media outlet, and no counter-study with specific figures has refuted it to date in the materials available for this post.

There is a clear line between what is uncertain and what is established. This post has attempted to draw that line without blurring it.

What is true: A study by the National Bureau of Economic Research, authored by Elizabeth Cox and Chloe N. East, finds no evidence of a positive effect of ICE raids on the employment of U.S.-born workers; on the contrary, it documents a negative effect measured at 0.54% per 1,000 arrests. What remains debated: the overall methodological robustness of this study, which has been challenged by the Washington Times but not refuted by counterevidence in the available sources.

What this report implies: the repeated claim that raids protect American jobs will now have to contend with published data that suggests the opposite. An economic slogan can survive for a long time without evidence. It is harder to sustain in the face of data that directly contradicts it, published by a recognized research institution.

By Maxime Marquette, columnist