What do state-level studies (e.g., Texas) reveal about homicide rates among undocumented immigrants compared with U.S.-born residents?
Executive summary
State-level research—most notably studies using Texas Department of Public Safety arrest data—consistently finds that undocumented immigrants have lower homicide arrest and conviction rates than U.S.-born residents; the influential PNAS study and related National Institute of Justice summaries report undocumented homicide arrest rates in Texas at roughly half (or less) those of native-born Texans [1] [2]. These findings are robust across multiple analyses but come with important methodological caveats about measurement, rarity of homicide, and limits to generalizing from Texas to every state [3] [4].
1. Texas evidence: a rare, data-rich window
Texas is unusually valuable to researchers because its arrest records include immigration-status checks, enabling separation of undocumented, legal immigrant, and U.S.-born arrestees; analyses using that dataset—most prominently the PNAS paper by Light et al.—report considerably lower felony and homicide arrest rates for undocumented immigrants than for U.S.-born citizens [1] [3]. The National Institute of Justice summarized those results, noting undocumented immigrants were arrested for homicide at less than half the rate of native-born Texans across the study period 2012–2018 [2].
2. What the numbers look like in plain terms
In the Texas analyses, overall felony arrest rates were about 1,000 per 100,000 for U.S.-born citizens, ~800 per 100,000 for legal immigrants, and ~400 per 100,000 for undocumented immigrants, with homicide arrests specifically reported at roughly 1.9 per 100,000 for undocumented people versus 4.8 per 100,000 for U.S.-born citizens in one published estimate [3] [5]. Comparable work by other researchers and think tanks has produced similar directional findings—Cato’s analysis found lower conviction rates for undocumented people compared with native-born Texans—though point estimates vary by study and time frame [6] [5].
3. Measurement matters: arrests, convictions, victims, and rare events
Researchers commonly use arrest or conviction rates as proxies for offending, which carries limits: homicides are relatively rare and more likely to be detected than other crimes, but many homicides remain unsolved or have multiple suspects, and arrest records reflect policing and prosecutorial practices as much as underlying offending [2] [4]. Studies also note potential undercounts of undocumented populations and rely on population estimation methods that can change rate calculations; the Texas work attempts to adjust for these issues but scholars caution about residual uncertainty [1] [4].
4. Broader state and local evidence: consistency and variation
Beyond Texas, state- and city-level research generally finds no positive relationship between immigration levels and violent crime—indeed, several studies report an “immigration effect” tied to stable or declining homicide rates as immigrant populations grow—supporting the Texas findings as part of a broader pattern that immigrants are less likely to be offenders or victims compared with U.S.-born residents [7] [8] [9]. At the same time, differences in data availability (Texas is exceptional) and local dynamics mean conclusions drawn from Texas should be extended to other states with care [5].
5. Politics, narratives, and implicit agendas in the literature
The topic is politically charged: advocacy groups, think tanks, and policymakers often emphasize different aspects—some highlight the Texas data to argue against claims of a “migrant crime wave,” while others point to isolated violent incidents to press for stricter immigration enforcement; the scholarly literature and fact-checkers note both the empirical pattern of lower offending among immigrants and the frequent politicization of single cases or worst-case anecdotes [10] [5]. Analysts warn to separate methodologically rigorous, state-level evidence from rhetorical claims that overgeneralize or use incomplete denominators [4] [5].
6. Bottom line: credible pattern plus guarded generalization
State-level studies that can distinguish immigration status—most notably Texas—consistently show undocumented immigrants have lower homicide arrest and conviction rates than U.S.-born residents, often by substantial margins [1] [2] [3]; however, those findings rest on arrest/conviction proxies, special data circumstances in Texas, and well-documented statistical caveats, so while the pattern is credible and echoed in broader research, it is not an absolute proof that every locality will display the same differences [4] [8].