What do state-level studies (like Texas) reveal about arrest and conviction rates by immigration status?
Executive summary
State-level studies centered on Texas—the only state that systematically links arrest and conviction records to DHS immigration-status responses—consistently find that undocumented immigrants are arrested and convicted at lower rates than native-born U.S. citizens for major categories of crime, with undocumented offending rates often reported as less than half for violent and drug crimes and substantially lower for property crimes [1] [2] [3]. Those findings are robust across peer‑reviewed work and federal summaries, but they are contested by advocacy and policy groups that point to classification delays, undercounts, and varying metrics (arrests versus later-identified convictions), producing a contested evidence landscape [1] [4] [5].
1. The Texas data: what it is and what major studies report
Researchers used Texas Department of Public Safety arrest and conviction records matched to DHS fingerprints to separate U.S.-born, documented, and undocumented offenders, and multiple analyses of these linked records (notably a peer‑reviewed PNAS study and an NIJ‑funded report) conclude undocumented immigrants have substantially lower arrest and conviction rates than native-born citizens across violent, drug, property, and traffic offenses during roughly 2012–2018 [1] [3] [2]. Federal writeups and migration policy summaries synthesize that undocumented Texans were arrested at less than half the rate of U.S.-born citizens for violent and drug crimes and roughly a quarter the rate for property crimes, and that the share of arrests involving undocumented immigrants was stable or declining over the study period [2] [6] [3].
2. Arrests, convictions and measurement choices that shape conclusions
The major studies explicitly examine both arrest and conviction measures and report that their core conclusions hold under alternative specifications, but they also note statistical adjustments (for example log transforms) and sensitivity tests to compensate for skewed distributions and population-estimate uncertainty [1] [3]. At the same time, critics and some analysts emphasize that Texas’s immigration-status flags can move from “unknown” to “noncitizen” only after later DHS processing, meaning some noncitizens are initially misclassified—an issue that can bias short‑term arrest tallies and complicate conviction-rate calculations unless longitudinal linkage and timing are carefully handled [4] [5].
3. Contrasting interpretations: peer‑reviewed, think tanks, and watchdogs
Peer‑reviewed and NIJ‑sponsored work (PNAS, NIJ) reach the broad finding of lower rates among undocumented immigrants [1] [3] [2], while policy analysts produce divergent takes: libertarian‑leaning Cato updates find lower homicide-conviction rates among illegal immigrants through 2022 and reiterate lower overall conviction rates, but caution readers about data limits for offenses other than homicide [7]. Conversely, the Center for Immigration Studies (CIS) argues Texas data undercounts illegal-immigrant convictions for certain serious crimes because of delayed identification and claims higher conviction rates for homicide and sexual assault when reinterpreting the timing of classification [4] [5]. These disagreements reflect methodological choices about timing, numerator/denominator construction, and which offenses are most informative [7] [5].
4. What the state‑level findings imply for policy debates and their limits
The Texas evidence undermines simple claims that increased removals or broader immigration enforcement will reliably reduce overall crime, since several rigorous analyses show undocumented immigrants commit fewer crimes per capita than natives [1] [3]. Yet the data do not answer every policy question: Texas’s system is unique, results may not generalize to other states without identical data linkage, and unresolved classification delays mean estimates for some offenses or short time windows carry uncertainty [8] [5]. Moreover, different actors bring explicit agendas—advocacy groups emphasizing public‑safety risks, and immigration‑rights scholars highlighting integration benefits—so interpreting the numbers requires transparency about methods and motives [4] [7].
5. Bottom line
The best available state‑level research from Texas, corroborated by federal summaries, shows undocumented immigrants in Texas have lower arrest and conviction rates than native‑born Americans for broad crime categories during the years analyzed, a pattern robust to many alternative specifications but sensitive to classification timing and certain offense types; competing analyses point out those sensitivities and reinterpret specific rates for severe crimes like homicide or sexual assault, so the conclusion is strong at the aggregate level but not unassailable at the margins or for every offense [1] [2] [5] [7].