What share of violent crimes in the U.S. are committed by undocumented immigrants versus citizens?

1. What the best national and state studies actually measure

Most high-quality work uses arrest or incarceration rates as proxies for crime commission because direct measurement of who "commits" crime is impossible with available data; the Texas study that the National Institute of Justice highlights compared felony arrest rates across groups and used those as proxies for offending ^[1]. That study and its PNAS publication rely on Texas Department of Public Safety records from 2012–2018 and separate people into U.S.-born, legal immigrants, and undocumented immigrants ^{[2] [5]}.

2. Headline numbers from the strongest empirical work

The Texas analysis and its summaries report that undocumented immigrants were arrested at less than half the rate of native-born citizens for violent and drug crimes and about one-quarter the rate for property crimes ^[1]. The PNAS and NIH summaries restate the finding that U.S.-born citizens were over twice as likely to be arrested for violent felonies compared with undocumented immigrants in Texas during the study window ^{[2] [5]}.

3. Broader reviews and replication across methods

Analyses beyond Texas—systematic reviews and fixed-effects regression studies—reach congruent conclusions: undocumented immigration does not increase violent crime, and the relationship is generally neutral or negative in many specifications ^{[3] [4]}. The Migration Policy Institute and American Immigration Council syntheses likewise report that immigrants, including undocumented immigrants, tend to have lower crime and incarceration rates than the U.S.-born at multiple geographic scales ^{[6] [7]}.

4. Limits and important caveats the research acknowledges

Researchers repeatedly note limits: arrest rates are imperfect proxies for actual offending; many studies cannot separate undocumented from documented immigrants except when special administrative data exist (Texas is unusual in that regard) ^{[1] [3]}. Studies may not generalize from one state to the entire U.S., and homicide rates and other rare events can fluctuate because of small numbers ^{[8] [1]}.

5. Variation across crime types and localities

While the overall pattern points to lower felony arrest rates among undocumented immigrants, magnitudes differ by offense: the Texas work shows the biggest disparities for property and drug crimes and smaller—but still large—differences for violent felonies ^{[2] [5]}. Some research flags that specific local contexts or immigrant subgroups could show different patterns, and one older study found unauthorized immigration from particular origins might associate with higher violence in limited specifications ^[3].

6. How policymakers and advocates use these findings

Advocates for immigrant-friendly policies point to these studies to argue that immigration is not a driver of crime and that sanctuary and protection policies do not worsen public safety ^{[7] [4]}. Opponents of lenient immigration policy sometimes emphasize individual high-profile violent crimes or argue that available data are incomplete; the reviewed literature responds by stressing broader patterns in the population-level data ^{[9] [4]}.

7. What the available sources do not settle

Available sources do not mention a definitive national percentage share of all violent crimes attributable to undocumented immigrants across the entire U.S. because most national datasets do not reliably identify immigration status; the Texas study gives comparative rates within Texas but cannot be simply extrapolated into a single nationwide share without additional assumptions ^{[1] [2]}. National-level, immigration-status-specific offending shares are therefore not present in the cited reporting ^[6].

8. Bottom line for readers seeking a short answer

High-quality, peer-reviewed and government-reviewed research consistently finds undocumented immigrants have lower arrest and incarceration rates for violent and other felony crimes than U.S.-born citizens, often less than half the rate for violent offenses in the best state-level data ^{[1] [2]}. However, a single, precise national “share of violent crimes” committed by undocumented immigrants is not supplied in the available sources and would require cautious modeling beyond the studies cited ^{[1] [2]}.