What the Immigration Debate Gets Wrong
Both parties have built their immigration politics on fictions. The truth, as usual, is more complicated — and more interesting.
The immigration debate in the United States has achieved something remarkable: it has managed to be simultaneously omnipresent and almost entirely devoid of serious empirical content. Both major parties have constructed elaborate narrative architectures around the issue — one built on fear, one on aspiration — and both architectures share the same foundational problem: they are constructed in almost complete defiance of the relevant evidence.
The result is a policy debate that looks, from the outside, like two groups of people arguing passionately about a subject neither has studied carefully. The empirical literature on immigration — its effects on wages, employment, crime, fiscal balance, innovation, and social cohesion — is one of the most robust and contested bodies of research in applied economics. It is also almost entirely absent from the congressional floor, the cable news chyron, and the presidential debate stage.
On Kalshi, as of January 2026, the prediction market on whether comprehensive immigration reform passes before January 2027 was trading at 7 percent. Given that similar reform has been attempted and failed in 1986, 1996, 2006, 2007, 2013, and 2018, this estimate seems, if anything, slightly generous. But what the 7 percent figure obscures is the more important question: even if reform did pass, it would be reform built on a foundation of factual error. That is worth examining regardless of the political odds.
On the economics: what the research actually says
The question of how immigration affects native workers is the most politically charged empirical question in the field, and the honest answer is: it is complicated in ways that make both left and right narratives simultaneously partially correct and substantially misleading.
The most influential work on the subject comes from economists George Borjas and David Card, who have engaged in what is, by academic standards, an unusually public and contentious debate. Borjas, at Harvard, has argued that immigration — particularly of low-skill workers — suppresses wages for low-skill native workers, with disproportionate effects on African Americans. Card, at Berkeley, has argued that the effects are much smaller and frequently positive, with immigrants complementing rather than substituting for native workers.
Both have Nobel-quality methodological rigor. Both have produced findings that have been replicated and challenged. The takeaway is not that the truth lies somewhere between them — it is that the question is genuinely complex, that effects vary enormously by skill level, industry, local labor market conditions, and time horizon, and that anyone offering you a simple answer is selling you something.
What the evidence does show, with reasonable consistency: high-skill immigration has unambiguously positive effects on innovation, productivity, and economic growth. The National Bureau of Economic Research estimates that immigrants account for roughly 25 percent of U.S. patents and a disproportionate share of billion-dollar startup founders. This is not a contested finding. The controversy arises specifically around low-skill immigration, where the distributional effects — who benefits and who bears the costs — are genuinely disputed.
Metaculus maintains a long-term forecast on "U.S. immigration policy effectiveness," defined as whether immigration policy changes lead to measurable improvements in wage growth for the bottom quintile of native workers. The median forecast currently sits at 22 percent probability of success over a five-year horizon for any plausible policy currently under debate. This is partly a statement about economics — the evidence that any specific enforcement regime will produce the promised wage effects is thin — and partly a statement about political sustainability. Policies that restrict immigration tend to generate enforcement costs, labor market distortions, and political backlash that limit their duration.
On crime: the most durable fiction
The relationship between immigration and crime is perhaps the most extensively studied and most consistently misrepresented aspect of the entire debate.
The findings are not subtle. Virtually every rigorous study of the relationship — using individual-level crime data, controlling for age, sex, education, and socioeconomic status — finds that immigrants commit crimes at lower rates than native-born Americans. This is true for both documented and undocumented immigrants. It is true across crime types, geographies, and time periods. It has been replicated so many times, in so many contexts, that it is about as close to a settled empirical question as social science produces.
The mechanism is not mysterious. Immigrants — particularly those who have crossed borders illegally — have enormous personal incentives to avoid police contact. They self-select for risk tolerance and the willingness to defer gratification. They are disproportionately young men, a demographic associated with crime, yet still manage to underperform in criminal statistics because the selection effects overwhelm the demographic ones.
None of this appears to have made any impression on public discourse. Kalshi ran a contract in December 2025 asking whether a sitting U.S. Senator would make a demonstrably false statistical claim about immigrant crime rates in a floor speech in the subsequent 90 days. It was trading at 87 percent when it was listed. It resolved Yes within three weeks.
This is not, as is sometimes suggested, simply a matter of politicians cynically exploiting voter ignorance. Many of the politicians who make these claims appear to believe them. This is, in some ways, worse.
On fiscal effects: the most misunderstood complexity
The fiscal question — do immigrants cost more in government services than they contribute in taxes? — is the one where the honest answer most confounds both partisan narratives.
The Congressional Budget Office and the National Academies of Sciences have both produced careful analyses of immigration's fiscal effects, and both have reached the same basic conclusion: the answer depends almost entirely on which immigrants you're studying and over what time horizon.
Recent low-skill immigrants, particularly those with children in public schools and without employer-sponsored health insurance, are fiscal costs in the short to medium term. This is a finding that immigration advocates have been reluctant to acknowledge, but it is well-supported by the evidence.
Over a longer time horizon — 30 to 75 years, the full lifetime of an immigrant and their children — the picture reverses. The children and grandchildren of immigrants have higher educational attainment, higher incomes, and higher tax contributions than native-born Americans of similar initial socioeconomic backgrounds. The NAS study estimates that the average immigrant contributes a net fiscal positive of $173,000 over a 75-year horizon.
The political debate, of course, operates on the time horizon of the next election. From that vantage point, recent low-skill immigration does impose genuine fiscal costs on states and localities — costs that are felt most acutely in the communities that host large immigrant populations and least acutely by the federal government, which captures most of the long-term benefits. This mismatch between who bears the costs and who receives the benefits is a genuine policy problem that neither party has seriously addressed.
Polymarket's contract on whether federal fiscal transfer payments to high-immigration localities will increase meaningfully before 2028 sits at 11 percent. The probability is low because it would require Congress to acknowledge the mismatch — and acknowledgment would require stepping off the narrative scaffolding both parties have constructed.
What a serious debate would actually address
A serious immigration debate would start from roughly this set of propositions, each of which is supported by substantial evidence:
The United States needs more high-skill immigration than its current system allows. The H-1B cap, set at 65,000, is absurdly insufficient for an economy that generates roughly 500,000 high-skill job openings in STEM fields annually. Countries that compete for global talent — Canada, Germany, Australia — have moved toward points-based systems that prioritize skills and education. The United States has maintained a family-reunification-centric system that, whatever its humanitarian merits, does not optimize for economic benefit.
The asylum system is in genuine crisis, but not for the reasons usually cited. The backlog in immigration courts — now exceeding 3.5 million cases — is primarily the result of chronic underfunding of the immigration judiciary. Cases that should take 18 months take 8 years. The uncertainty created by this backlog makes it rational for migrants with weak cases to apply anyway. Fixing the backlog through resources rather than restriction would do more to reduce irregular migration than any enforcement measure currently under debate.
Enforcement without legal pathways is not sustainable. The demand for low-skill labor in American agriculture, food processing, hospitality, and construction has not declined. When legal pathways are restricted, the work gets done anyway — through unauthorized migration, through visa overstays, through the tolerated illegality that has characterized large sectors of the American labor market for decades. The enforcement-only model has been tried repeatedly and has repeatedly failed to achieve its stated goals, not because enforcement is impossible, but because enforcement against a labor market that genuinely wants the workers produces political and economic costs that ultimately reverse it.
None of this is what you will hear in the next Senate debate. The Kalshi contract on immigration reform sits at 7 percent, and in this case, the markets are probably the most honest participants in the conversation.
Priya Nandakumar is a staff writer at The Auguro covering American politics and policy. She has reported from the US-Mexico border and from immigration courts in Chicago, New York, and Los Angeles.