By Jim Shimabukuro (assisted by Perplexity)
Editor
Most recent large‑scale studies continue to find that both lawful and undocumented immigrants in the United States are less likely than U.S.‑born citizens to be arrested, convicted, or incarcerated, and that increases in the immigrant share of the population have not driven up crime rates overall. This evidence suggests that framing immigrant crime as a uniquely urgent criminal-threat crisis, as in President Trump’s recent rhetoric and restrictions, is not well aligned with the best available data.[alexnowrasteh+5]
One way to test the “immigrants cause crime” claim is to look at long‑run trends in crime as the foreign‑born share of the population changes. Using FBI Uniform Crime Reporting data and Census data, the American Immigration Council compared crime and immigrant population shares from 1980 through 2022, the most recent period with complete national data. “In 1980, immigrants made up 6.2% of the U.S. population, and the total crime rate was about 5,900 crimes per 100,000 people.”[americanimmigrationcouncil]
By 2022, the immigrant share had more than doubled to 13.9%, while the overall crime rate fell to about 2,335 crimes per 100,000 people, a decline of 60.4%. Over the same period, violent crime fell by about 34.5% and property crime by about 63.3%. When the Council used beta regression to analyze all 50 states from 2017–2022, it found no statistically significant correlation between a state’s immigrant share and its overall crime rate, meaning states with more immigrants were no more likely to have higher crime than states with fewer. This kind of macro pattern does not prove that immigrants never commit crimes, but it strongly undercuts the thesis that rising immigrant numbers are driving crime waves.[americanimmigrationcouncil]
Beyond aggregate trends, we can compare individual‑level involvement with the criminal justice system across groups. One influential peer‑reviewed study in the Proceedings of the National Academy of Sciences (PNAS) examined arrest data in Texas from 2012–2018, where law enforcement records allow researchers to distinguish U.S.‑born citizens, legal immigrants, and undocumented immigrants. Relative to undocumented immigrants, U.S.‑born citizens in Texas were more than twice as likely to be arrested for violent crimes, about 2.5 times as likely for drug crimes, and more than 4 times as likely for property crimes, on a per‑capita basis. In terms of felony arrests overall, rates were roughly “1,000 per 100,000 among U.S.‑born citizens, 800 per 100,000 among legal immigrants, and 400 per 100,000 among undocumented immigrants,” placing undocumented immigrants at about 40% of the citizen rate.[pnas]
Follow‑up work using Texas conviction data through 2022 yields a similar ranking. A 2024 Cato Institute analysis, again leveraging Texas’s detailed classification of immigration status, reports that over 2013–2022 the overall criminal conviction rate was 1,321 per 100,000 native‑born Americans, 685 per 100,000 undocumented immigrants, and 551 per 100,000 legal immigrants. Put differently, undocumented immigrants were about 48% less likely than natives to be convicted of a crime, and legal immigrants were about 58% less likely. Even for homicide—the most politically salient and emotionally charged offense—the 2022 conviction rates in Texas were 4.9 per 100,000 for native‑born Americans, 3.1 for undocumented immigrants, and 1.8 for legal immigrants.[cato]
Incarceration is another lens on criminal involvement. A 2025 analysis using American Community Survey data from 2011–2023 found that U.S. citizens aged 18–54 had an incarceration rate of about 1.2%, compared with 0.6% for undocumented immigrants and 0.3% for legal immigrants in 2023. After adjusting for demographics (age, race, state, and year), immigrants were roughly 48% less likely than native‑born Americans to enter incarceration in a given year. Looking over the life course, Cato’s cohort analysis of people born in 1990 estimated that by age 33, about 11% of native‑born Americans had been incarcerated at some point, compared with only about 3% of immigrants—a more than three‑fold difference.[cato+1]
These converging lines of evidence—arrests, convictions, and incarceration—point in the same direction: immigrants as a group, including those without legal status, have lower measured rates of serious criminal involvement than U.S.‑born citizens. This does not mean that no immigrant commits a serious crime, but it does mean that using immigrant status as a proxy for criminality is statistically misleading.
President Trump has paired his 2024–2025 campaign and governing agenda with promises of “the largest deportation program of criminals in the history of America” and with new proclamations restricting entry from certain countries, justified in part on public‑safety grounds. A White House fact sheet on his late‑2025 proclamation frames the policy as necessary to protect U.S. national security and public safety by limiting the entry of foreign nationals from countries with high overstay rates, weak cooperation on removals, or elevated terrorist and criminal activity concerns. Implicitly, this narrative suggests that non‑citizens, especially from certain regions, pose an outsized crime threat inside the United States.[whitehouse+1]
When we juxtapose this political framing with the current empirical record, the mismatch is stark. Nationwide, as the foreign‑born share of the population has risen to nearly 14%1, overall crime—including violent crime—has fallen dramatically, and across the 50 states there is no statistically significant link between immigrant concentration and crime rates. At the individual level, both lawful and undocumented immigrants are significantly less likely than U.S.‑born citizens to be arrested, convicted, or incarcerated, even when we focus on serious offenses like homicide or violent felonies, and even in a border state like Texas that is central to immigration politics.[pnas+2]
If the standard for a “justifiable” overwhelming reaction is that a group’s crime rate is unusually high relative to its population share, the evidence does not support portraying immigrants as a uniquely dangerous class. Instead, immigrants—on average—appear to pose lower criminal risks than natives, and periods of growing immigration have coincided with falling crime. This does not settle normative debates about border control, asylum processing, or labor markets, but it does mean that arguments for sweeping restrictions and mass deportations on public‑safety grounds sit on a weak empirical foundation. A more evidence‑aligned conversation would differentiate between isolated, highly publicized tragedies and the overall statistical picture, and would avoid conflating immigration status with criminal propensity.
An argument for this mass crackdown on immigrants is: If you remove any group of people, you remove the crimes that a small fraction of that group would have committed; the mistake is assuming that the net effect on overall crime will be large, unambiguously negative, and achieved without offsetting harms. The best evidence we have instead suggests that broad, high‑intensity deportation/enforcement has not produced meaningful local crime reductions and can plausibly make communities less safe by weakening trust and cooperation with police.[NBER]
The surface logic commits a base‑rate error: it focuses on the fact that some immigrants commit crimes while ignoring that immigrants, on average, have lower crime rates than the U.S.-born population in comparable datasets. If immigrants have lower per‑capita offending, then replacing immigrants with “whoever fills the demographic and labor-market space left behind” does not guarantee a reduction in crime, because the replacement population may have equal or higher per‑capita offending, and because crime is not produced by “headcount” alone but by local conditions, opportunities, and policing capacity.[cato+3]
A second math problem is scale: even a policy that perfectly removed every immigrant who would otherwise offend can only reduce crime by the share of crime attributable to that group, and the available evidence suggests immigrants are not a disproportionate share of offenders. That means the ceiling on crime reduction from mass removal is often smaller than people intuitively assume, while the social and fiscal costs of mass operations can be very large.[pnas+1]
A useful test of the counter‑argument is not “Do deportations remove some offenders?” (they obviously do), but “When deportations increase sharply, do crime rates fall?” Research exploiting the staggered rollout of the federal Secure Communities program—which increased the risk of deportation by tightening local-federal cooperation—finds that increased deportations did not reduce local violent or property crime rates in a meaningful, statistically clear way. In other words, the real‑world, system‑level effect of ramped enforcement looked close to zero on standard crime outcomes, which is hard to reconcile with the idea that broad roundups are a straightforward crime-control lever.[ftp.iza+1]
This result also fits the broader “immigration and crime” literature reviewed by expert bodies: the National Academies’ synthesis chapter on immigration and crime concludes that immigrants are generally less crime-involved than the native born and that higher immigration is not associated with higher crime rates. If immigration is not a driver of crime upward in the first place, then a blanket removal strategy has limited room to “work,” even before considering side effects.[americanimmigrationcouncil+1]
Crime control depends heavily on reporting, witness cooperation, and victims’ willingness to call police. Analyses using the National Crime Victimization Survey indicate immigrant victims report violent crime at higher rates than U.S.-born victims (and noncitizens at higher rates still), suggesting immigrant communities can be important contributors to public safety through cooperation with law enforcement. Policies that heighten deportation fear can suppress that cooperation, meaning violent offenders may be less likely to be identified and stopped, which can raise victimization risk for everyone, including citizens.[cato+1]
There is also a capacity tradeoff: when local police are pulled into immigration enforcement or when jail systems face added detention demands, scarce law-enforcement resources can be diverted from solving and preventing serious crimes. The Secure Communities–based research explicitly investigates police effectiveness mechanisms such as clearance rates and law-enforcement personnel constraints, underscoring that enforcement surges can change how police allocate time and attention, not just who is present in a community.[cato+1]
Targeted removal of noncitizens after serious criminal conviction is a different claim than mass roundup strategies, because it focuses enforcement on the small subset demonstrably involved in harmful offending rather than treating immigration status as the primary risk signal. Even then, the key empirical question remains whether marginal increases in removals produce measurable crime reductions beyond what ordinary policing, prosecution, and incarceration already achieve, and the Secure Communities evidence cautions that simply increasing deportations is not, by itself, a reliable way to lower local crime.[cato+3]
__________
1 “The government’s January 2025 Current Population Survey (CPS) shows the foreign-born or immigrant population (legal and illegal together) hit 53.3 million and 15.8 percent of the total U.S. population in January 2025 — both new record highs” (Steven A. Camarota & Karen Zeigler, “Foreign-Born Number and Share of U.S. Population at All-Time Highs in January 2025: Record increase in last four years driven primarily by illegal immigration,” CIS, 12 March 2025). -js
[End]
Filed under: Uncategorized |
Educational Technology and Change Journal 
Leave a comment