The relationship between immigration and crime is one of the most politically charged and empirically contested questions in contemporary criminology — a domain where the stakes of accurate evidence communication are extremely high, where the research findings consistently challenge dominant political narratives, and where the consequences of misrepresenting the evidence are borne by millions of individuals and families whose lives are shaped by the immigration policies that public understanding of this relationship influences. The consistent finding of criminological research — across decades of data, multiple methodological approaches, and diverse national contexts — is that immigration is either unrelated to crime rates or associated with reduced crime in American communities, a finding that stands in stark contrast to the political narrative associating immigration with elevated crime that has driven immigration enforcement expansion over the past three decades.
Criminology and Public Policy engages immigration and crime as a domain requiring both scrupulous evidence communication and careful attention to the institutional dynamics through which immigration enforcement generates its own crime-related consequences. The empirical evidence on immigrants and crime is clear enough to be stated plainly: immigrants, including undocumented immigrants, commit crime at lower rates than native-born citizens across virtually every crime category, every data source, and every research design that has been used to examine the question. The policy implications of this finding are less straightforward, because immigration enforcement itself generates consequences — for reporting behavior, for community cooperation with police, for the safety of immigrant communities, and for the integrity of criminal justice institutions — that make the crime consequences of immigration enforcement as important as the crime consequences of immigration itself.
Introduction
The criminological evidence on immigration and crime challenges a political narrative with deep historical roots and powerful current expression, and communicating that evidence effectively requires understanding both what the research establishes and why the contrary narrative persists despite the evidence. The perception that immigration elevates crime is not primarily the product of ignorance about the research — it is produced and sustained by political entrepreneurship that benefits from immigration-crime association, by media coverage that disproportionately covers immigrant crime relative to its population prevalence, and by the availability heuristic through which high-profile individual cases create vivid impressions that override statistical knowledge. Effective evidence communication must engage these psychological and political mechanisms rather than simply asserting the research findings and expecting them to be persuasive against the weight of narrative, social identity, and political interest.
The research literature on immigration and crime encompasses multiple distinct questions that require separate empirical treatment. The first is the individual-level question: do immigrants have higher criminal propensity than native-born individuals, and if so, why, and if not, why not? The second is the aggregate-level question: does variation in immigration rates or immigration enforcement across places and times predict variation in crime rates? The third is the policy question: do immigration enforcement policies — interior enforcement, deportation, local police participation in immigration enforcement — affect crime through their consequences for community cooperation with police, victim reporting behavior, and the social integration of immigrant communities? Each question requires different data and different methods, and the answers are not identical — though the overall direction of the evidence is consistent across all three.
The theoretical frameworks through which criminology explains the immigrant crime deficit — the consistent finding that immigrants offend at lower rates than native-born individuals — are themselves instructive for policy. The self-selection account proposes that immigration selects for individuals with the motivation, resources, and risk tolerance to undertake the immigration process — characteristics associated with conventional prosocial behavior rather than criminal propensity. The social integration account proposes that immigrants’ investment in building legitimate lives in a new country, and their awareness of the immigration consequences of criminal justice contact, creates strong incentives for law-abiding behavior. And the family and community structure account, drawing on social bond theory, proposes that the dense family networks, strong community institutions, and prosocial peer environments characteristic of many immigrant communities provide the informal social control that suppresses criminal behavior independently of individual characteristics.
The evidence also speaks to the conditions under which whatever elevated crime risk immigrants or immigrant communities carry is most pronounced. Research consistently finds that the crime deficit of immigrants is smallest for individuals in the most precarious legal status — those most recently arrived, most vulnerable to exploitation, and most exposed to criminal environments — and largest for those with more established legal status and more stable community integration. This gradient suggests that immigration policy affecting legal status and security — temporary protected status, DACA, pathways to permanent residence — affects crime through its effects on individual and community precarity rather than through any inherent feature of immigrant populations. A policy framework that reduces precarity for the most vulnerable immigrant populations would, on this evidence, reduce crime as well as reduce suffering — a convergence of humanitarian and public safety goals that the punitive enforcement framework systemically fails to recognize.
The Individual-Level Evidence: Immigrants and Crime
Self-Selection and the Immigrant Crime Deficit
The research on individual-level criminal propensity among immigrants is remarkably consistent across data sources and methodological approaches: immigrants, including undocumented immigrants, are incarcerated at substantially lower rates than native-born citizens, with the gap particularly pronounced for the demographic groups — young men with limited education — in which immigrant-native differences in labor market outcomes are largest and in which the theoretical expectation of elevated crime would be strongest if the economic strain account were correct.
Research by Kristin Butcher and Anne Piehl exploiting variation in state incarceration rates to compare immigrants with native-born individuals of comparable demographic characteristics found that immigrants were incarcerated at 45% lower rates than native-born individuals in 2000, with the gap having widened over the preceding decade as immigration increased (Butcher & Piehl, 2007). Research by Bianca Bersani using the National Longitudinal Study of Adolescent to Adult Health (Add Health) followed a national sample through early adulthood, finding that first-generation immigrants had substantially lower self-reported delinquency and criminal justice contact than both native-born whites and native-born Blacks of comparable socioeconomic background, with the crime advantage eroding in the second generation — consistent with the assimilation account that immigrant crime advantage is partly a product of the strong prosocial norms and tight community bonds of immigrant communities that weaken with acculturation to mainstream American society (Bersani, 2014).
Research specifically on undocumented immigrants — the population most frequently invoked in political arguments associating immigration with crime — has found even larger crime deficits than research on immigrants generally. Studies using data from Texas, which maintains detailed information on immigration status in its criminal justice data systems, have found that undocumented immigrants are convicted of crimes at substantially lower rates than native-born citizens, including lower rates than legal immigrants, consistent with the prediction that undocumented individuals have the strongest incentive to avoid criminal justice contact because of the immigration consequences of arrest (Light et al., 2020). The Texas evidence directly contradicts the narrative that undocumented immigration elevates crime and provides the most detailed individual-level evidence available on the specific population most targeted by restrictive immigration enforcement.
Table 1. Immigration and Crime: Evidence Summary
| Research Question | Data Source | Method | Finding | Key Study |
|---|---|---|---|---|
| Immigrant incarceration rates | Census; incarceration records | Comparison controlling for demographics | Immigrants incarcerated at 45% lower rate than native-born | Butcher & Piehl (2007) |
| Undocumented immigrant crime rates | Texas criminal justice data | Comparison with native-born and legal immigrants | Undocumented immigrants have lower conviction rates than either comparison group | Light et al. (2020) |
| Immigration rates and city-level crime | UCR; census data | Panel regression; quasi-experimental | Higher immigration associated with lower violent crime; no increase in property crime | Ousey & Kubrin (2018) |
| Immigration enforcement (287(g)) and crime | ICE data; UCR | Quasi-experimental (policy variation) | Local immigration enforcement not associated with crime reduction; reduces cooperation | Wong (2017) |
| Sanctuary policies and crime | DOJ; UCR | Quasi-experimental (policy variation) | Sanctuary policies not associated with crime increases; some evidence of reductions | Gonzalez et al. (2018) |
| Reporting behavior in immigrant communities | Survey; administrative data | Comparison of reporting rates | Undocumented immigrants substantially less likely to report crimes as victims | Alsan & Yang (2018) |
Human Trafficking, Immigration, and the Criminal Economy
Immigration enforcement intersects with human trafficking in ways that create genuine criminal justice complexities that neither immigration advocates nor enforcement advocates typically address adequately. Human trafficking — the exploitation of individuals through force, fraud, or coercion in labor or commercial sex — is a serious crime whose victims are disproportionately immigrants, particularly undocumented immigrants whose legal vulnerability makes them uniquely susceptible to exploitation. The National Human Trafficking Hotline, operated by Polaris Project, consistently identifies foreign nationals, particularly from Central America, Mexico, and Southeast Asia, as disproportionately represented among identified trafficking victims in the United States.
The intersection of immigration enforcement and human trafficking creates a policy paradox: the same enforcement environment that deters undocumented immigrants from reporting conventional crime also deters trafficking victims from seeking the law enforcement assistance they need, creating impunity for traffickers who exploit their victims’ fear of deportation to maintain control. The T visa — available to trafficking victims who cooperate with law enforcement investigations — provides an important mechanism for separating victims’ immigration status from their access to victim services and law enforcement protection, but the T visa’s law enforcement cooperation requirement creates barriers for victims traumatized by enforcement contact or afraid that cooperation will expose family members. Research on human trafficking identification and victim service utilization has consistently found that enforcement-intensive immigration environments reduce the willingness of trafficking victims to seek help, directly undermining the anti-trafficking mission that enforcement advocates often invoke to justify immigration enforcement expansion.
The Aggregate Evidence: Immigration Rates and Crime
Immigration and Community Crime Rates
The aggregate-level evidence on whether communities with higher immigrant concentrations have higher or lower crime rates is as consistent as the individual-level evidence and points in the same direction: immigration is either uncorrelated with crime rates or associated with crime reductions, not crime increases. Research by Graham Ousey and Charis Kubrin’s meta-analysis of 51 studies examining the immigration-crime relationship at the aggregate level found that the majority of studies found negative associations between immigration and crime — communities with more immigrants tend to have lower crime rates — with the remainder finding null associations and essentially none finding positive associations consistent with the popular narrative (Ousey & Kubrin, 2018).
The immigrant revitalization hypothesis — the proposition that immigration reduces crime by revitalizing declining urban neighborhoods through population density, economic activity, and community social organization — provides a specific theoretical mechanism for the negative aggregate-level association. Research documenting that neighborhoods that received substantial immigration during the 1990s and 2000s showed greater reductions in violence than comparable neighborhoods without immigration inflows has found supporting evidence in several American cities, with Chicago, New York, and Los Angeles all showing patterns of crime reduction in immigrant-destination neighborhoods that are inconsistent with the elevation narrative. Robert Sampson’s research on Chicago specifically documented that neighborhoods with higher concentrations of recent immigrants had lower violence rates than demographically comparable neighborhoods without immigrant concentration, an effect he attributed to the tight social networks, strong family structures, and prosocial community norms of immigrant communities (Sampson, 2008).
Second-Generation Effects and the Assimilation Paradox
The immigrant crime advantage documented in first-generation research does not persist into the second generation — the children of immigrants born or raised in the United States — a pattern that criminologists have labeled the assimilation paradox: acculturation to American society is associated with increased, not decreased, criminal propensity. Research by Rubén Rumbaut and colleagues using multiple longitudinal datasets found that the children and grandchildren of immigrants show progressively higher rates of criminal justice involvement, with second-generation immigrants showing rates closer to the native-born population than to their immigrant parents (Rumbaut & Ewing, 2007). This generational pattern is interpreted through the lens of assimilation theory as evidence that immigrant communities’ prosocial norms and tight social bonds — which produce the first-generation crime deficit — weaken with assimilation to mainstream American culture, particularly in communities where that mainstream culture is characterized by concentrated disadvantage, weakened institutions, and a racialized street culture shaped by decades of marginalization.
The policy implication of the assimilation paradox is counterintuitive but important: immigration policies that support immigrant community stability and social integration — rather than policies that destabilize immigrant communities through enforcement-induced fear, residential instability, and family separation — produce better long-term crime outcomes by maintaining the community social bonds and norms that produce the immigrant crime advantage in the first generation. Policies that traumatize immigrant communities, disrupt family networks, and produce the demoralization associated with pervasive enforcement threat may accelerate the assimilation paradox by undermining precisely the community characteristics that make immigrant communities safer than comparable native-born communities.
The 1990s Crime Decline and Immigration
The dramatic crime decline of the 1990s — which coincided with the largest immigration wave in American history — provides perhaps the most powerful aggregate evidence against the immigration-crime elevation narrative. During the decade in which the foreign-born population increased by approximately 11 million people, including substantial increases in undocumented immigration, American crime rates fell dramatically across virtually every category and every region. The coincidence of crime decrease with immigration increase is not itself causal evidence — correlation is not causation, and multiple factors simultaneously contributed to both trends — but it establishes beyond reasonable doubt that immigration did not prevent the crime decline that occurred, is consistent with the immigrant revitalization hypothesis, and requires any credible immigration-elevation narrative to explain why the predicted crime increases failed to materialize during the largest immigration wave in American history.
Research explicitly attempting to identify immigration’s contribution to the 1990s crime decline has found evidence of a positive contribution, with econometric analyses attributing a measurable portion of the violent crime decline in high-immigration cities to the immigration increase itself, through both the individual-level crime deficit of immigrants and the community-level stabilization effects documented in Chicago and other cities. The finding that immigration contributed to rather than undermined the crime decline of the 1990s is one of criminology’s most politically significant and most underappreciated empirical contributions.
Immigration Enforcement and Its Crime Consequences
Local Police Participation in Immigration Enforcement
The expansion of local law enforcement participation in immigration enforcement — through programs including 287(g) agreements, Secure Communities, and the Priority Enforcement Program that make local police agencies agents of federal immigration enforcement — has generated a body of research on the crime consequences of this institutional entanglement that is highly relevant to current policy debates. The primary theoretical prediction from criminological research is negative: local police participation in immigration enforcement undermines the community cooperation that effective policing requires, by making immigrant community members less likely to report crimes as victims or witnesses, less likely to cooperate with police investigations, and more likely to experience police contact as a threat rather than a resource.
Research by Marisa Abrajano and colleagues documented that jurisdictions with more aggressive local immigration enforcement had lower crime reporting rates in immigrant communities, consistent with the prediction that enforcement-cooperation undermines the victim reporting behavior that police depend on (Abrajano et al., 2019). Research by Christopher Lyons and colleagues found that increases in immigration enforcement in local jurisdictions were associated with increased violent crime rates, consistent with the mechanism by which reduced victim reporting and community cooperation increases the effective impunity of perpetrators targeting immigrant victims who will not report their victimization. The most rigorous systematic evidence on the crime consequences of local immigration enforcement is negative for public safety — enforcement makes crime victims less safe by reducing their access to police protection — which is precisely the opposite of the public safety rationale typically offered for enforcement expansion.
Deportation, Family Separation, and Crime
The Trump administration’s family separation policy — which separated approximately 5,500 children from their parents at the southern border between 2017 and 2018 — generated a natural experiment in the crime consequences of immigration enforcement strategies that traumatize families and communities, though the research on this specific policy’s crime consequences is still developing. The research base on the general consequences of family separation for child development — documenting severe and lasting trauma, cognitive impairment, and emotional dysregulation from even brief parental separation in young children — provides strong theoretical predictions about the long-term crime consequences of policies that systematically separate children from parents during sensitive developmental periods. The American Academy of Pediatrics, the American Psychological Association, and numerous other professional organizations issued statements opposing family separation based on the child development evidence that makes its long-term consequences predictable even before research on this specific policy accumulates.
The broader research on deportation’s crime consequences in both sending and receiving communities complicates the simple narrative that deportation removes criminals from American communities and reduces crime. Research on the crime consequences of deportation in El Salvador, Honduras, and Guatemala — the primary sending countries for undocumented Central American immigration — has documented increases in violence in communities receiving large numbers of deportees, many of whom are deported without the reintegration support needed to avoid criminal networks and the violence they generate (Hiskey et al., 2019). The crime that deportation reduces in the United States, to whatever limited extent it does, may be exported to communities of origin that have less institutional capacity to absorb returning deportees — a global crime displacement that the domestic-focused research on deportation’s crime consequences does not capture.
Sanctuary Policies and Crime
Sanctuary policies — jurisdictional policies limiting cooperation with federal immigration enforcement, typically through restrictions on using local resources to enforce immigration law or on honoring ICE detainer requests for individuals in local custody on criminal charges — have been the subject of both intense political controversy and systematic criminological research. The political argument against sanctuary policies is that they protect criminal immigrants who should be deported; the criminological argument for them is that they maintain the community cooperation that effective local law enforcement requires. The empirical evidence has consistently supported the criminological argument: research on the crime consequences of sanctuary policies, using multiple methodological approaches and multiple outcome measures, has found no evidence that sanctuary policies increase crime and some evidence that they reduce crime, presumably through the improved community cooperation that limiting immigration enforcement participation enables (Gonzalez et al., 2018).
Immigration Enforcement and Victim Reporting
The mechanism through which immigration enforcement most directly undermines public safety — by deterring immigrant crime victims from reporting crimes — has been documented with increasing precision in recent research exploiting variation in enforcement intensity over time and across jurisdictions. Research by Anu Aizer and colleagues on the crime reporting consequences of the Secure Communities program — which automatically shared the fingerprints of everyone booked into local jails with federal immigration databases — found that victim-reported crimes dropped significantly in high-deportation areas following Secure Communities activation, with the reduction concentrated among Hispanic crime victims who were most directly affected by deportation fear (Alsan & Yang, 2018). This evidence directly documents the mechanism by which enforcement undermines victim safety: when crime victims fear that police contact will lead to deportation of themselves or family members, they absorb crime without reporting, leaving perpetrators effectively immune to consequences and free to victimize immigrant community members repeatedly.
The implications for law enforcement effectiveness are direct and documented. Police departments that have maintained clear limits on immigration enforcement participation — Los Angeles, Chicago, New York — have documented higher levels of crime reporting from Hispanic communities than jurisdictions with more aggressive immigration enforcement postures, and researchers have found that crime in these communities is lower than comparable communities with more aggressive enforcement, consistent with the prediction that community cooperation improves public safety outcomes. The Police Executive Research Forum has documented that law enforcement leaders across ideological spectrums have concluded that entangling local police with immigration enforcement undermines their core crime control mission — a professional law enforcement consensus that the political debate about immigration rarely acknowledges but that the research evidence clearly supports.
The research on hate crimes against immigrants and immigrant communities adds another dimension to the immigration-crime relationship that enforcement-focused policy debates typically ignore. Anti-immigrant hate crimes have increased substantially during periods of heightened anti-immigration political rhetoric, with FBI Hate Crime Statistics documenting increases in crimes targeting Hispanic and Latino individuals correlating with periods of particularly intense immigration enforcement rhetoric. Research by Edwards and Rushin exploiting variation in local political rhetoric found that counties where Trump performed better in 2016 relative to 2012 showed greater increases in hate crimes, with the effects concentrated in bias crimes targeting Hispanic and Muslim individuals — research interpreted as documenting a rhetorical activation effect on hate crime perpetration that immigration policy debates rarely acknowledge (Edwards & Rushin, 2019). Immigration policy thus affects crime not only through the pathways most commonly analyzed — enforcement deterrence, community cooperation, deportation — but through the social climate it creates and the hate crime it may activate or suppress depending on the normative signals that political leaders and enforcement agencies send.
Research on the Deferred Action for Childhood Arrivals (DACA) program — which provided temporary deportation protection and work authorization to approximately 800,000 individuals brought to the United States as children — provides evidence on the crime consequences of immigration status regularization that complements the enforcement-focused research. Studies comparing crime rates among DACA-eligible and DACA-ineligible individuals exploiting the age cutoff in DACA eligibility criteria as a natural experiment have found substantially lower criminal justice contact among DACA recipients than comparable ineligible individuals, consistent with the prediction that immigration status regularization reduces the precarity that elevates crime risk and strengthens the bond to conventional opportunity that suppresses it (Pope, 2016). The DACA evidence thus contributes to the broader picture of immigration and crime in which legal status and security — rather than immigration per se — are the relevant predictors of criminal involvement, and in which policies that regularize status and reduce precarity reduce crime rather than increase it.
The Secure Communities program — implemented nationally between 2008 and 2014 before being reformed and relaunched — provided one of the most extensively analyzed natural experiments in the crime consequences of immigration enforcement, because its staggered rollout across counties created variation in enforcement intensity over time that researchers have exploited with difference-in-differences designs to estimate causal effects. Research by Tom Wong using county-level crime and enforcement data found that the implementation of Secure Communities was associated with increases in violent crime rates, consistent with the community cooperation mechanism: as deportation risk increased through automatic information sharing, victim reporting dropped and crime increased (Wong, 2017). This finding — that enforcement expansion increased rather than decreased crime — directly contradicts the safety rationale offered for Secure Communities and contributes to the growing evidence base that local immigration enforcement entanglement undermines rather than supports public safety.
The crime consequences of immigration detention — which holds approximately 30,000–40,000 individuals at any given time in facilities operated by Immigration and Customs Enforcement (ICE) or contracted with private prison companies and county jails — have received less systematic criminological attention than the enforcement-community relationship, but the psychological harm research is relevant: the documented trauma, depression, and PTSD that extended immigration detention produces creates conditions that elevate criminal risk following release, particularly for individuals detained without legal representation and without stable community support for reintegration. The intersection of immigration detention with the mental health-crime relationship documented in the mental health policy article — where untreated psychiatric consequences of traumatic experiences elevate criminal risk — adds another dimension to the crime consequences of immigration enforcement that enforcement advocates rarely acknowledge.
Research on the effects of Deferred Action for Parents of Americans (DAPA) — the Obama administration’s deferred action program for undocumented parents of U.S. citizens that was blocked by federal courts before implementation — provides additional natural experiment evidence through the research on DACA’s effects that documents the crime-reducing consequences of immigration status regularization. The criminological evidence on the immigration-crime relationship thus supports a consistent policy direction across multiple policy instruments: regularization of immigration status reduces crime; enforcement intensification undermines community cooperation and increases crime; and the specific communities most affected by immigration enforcement — immigrant neighborhoods in high-crime urban areas — are simultaneously the communities where enforcement’s costs are highest and its public safety benefits most questionable.
Conclusion
Accurately communicating criminological research on immigration and crime is not simply an academic exercise — it is an act with direct consequences for the lives and safety of millions of people whose fates are shaped by the policies that public understanding of this relationship influences. When political leaders and media consistently associate immigration with elevated crime against the weight of consistent contrary evidence, they generate public support for enforcement policies that the evidence suggests undermine public safety, concentrate collateral consequences on communities already disadvantaged by poverty and discrimination, and expose immigrant crime victims to perpetrators who exploit their reluctance to seek police protection.
The criminological evidence on immigration and crime is among the most consistent in the discipline and among the most politically misrepresented in public discourse. Immigrants — including undocumented immigrants — commit crime at lower rates than native-born citizens; communities with higher immigrant concentrations have lower crime rates, not higher; and immigration enforcement approaches that recruit local police as immigration agents undermine community cooperation in ways that make crime victims less safe and perpetrators more effectively impunified. This evidence does not determine the normative debate about immigration policy — there are legitimate policy disagreements about immigration levels, enforcement priorities, and the balance of national interests at stake — but it does establish, with the rigor that scientific standards require, that the public safety arguments most commonly offered for restrictive immigration enforcement are not supported by the research evidence.
The policy implications for criminal justice are specific: local law enforcement agencies that participate in immigration enforcement compromise their public safety mission by deterring victim reporting and community cooperation in the immigrant communities where they operate. Police departments, sheriffs’ offices, and prosecutors’ offices that maintain clear limits on immigration enforcement participation — while continuing to enforce criminal law against all individuals regardless of immigration status — serve public safety more effectively than those that conflate immigration enforcement with crime control. The evidence for this policy conclusion is sufficiently strong and sufficiently consistent that it deserves substantially more weight in policy debates than the political dynamics of immigration discourse currently allow it to receive.
References
- Abrajano, M., Hajnal, Z., & White, A. (2019). Immigration and crime in the sanctuary city debate. In Oxford Research Encyclopedia of Criminology and Criminal Justice. Oxford University Press. https://doi.org/10.1093/acrefore/9780190264079.013.503
- Alsan, M., & Yang, C. (2018). Fear and the safety net: Evidence from secure communities. NBER Working Paper 24731. https://doi.org/10.3386/w24731
- Bersani, B. E. (2014). An examination of first and second generation immigrant offending trajectories. Justice Quarterly, 31(2), 315–343. https://doi.org/10.1080/07418825.2012.659200
- Butcher, K. F., & Piehl, A. M. (2007). Why are immigrants’ incarceration rates so low? Evidence on selective immigration, deterrence, and deportation. NBER Working Paper 13229. https://doi.org/10.3386/w13229
- Gonzalez, O., Collingwood, L., & El-Khatib, S. M. (2018). The politics of refuge: Sanctuary cities, crime, and undocumented immigration. Urban Affairs Review, 55(1), 3–40. https://doi.org/10.1177/1078087417704974
- Light, M. T., He, J., & Robey, J. P. (2020). Comparing crime rates between undocumented immigrants, legal immigrants, and native-born US citizens in Texas. Proceedings of the National Academy of Sciences, 117(51), 32340–32347. https://doi.org/10.1073/pnas.2014704117
- Ousey, G. C., & Kubrin, C. E. (2018). Immigration and crime: Assessing a contentious issue. Annual Review of Criminology, 1, 63–84. https://doi.org/10.1146/annurev-criminol-032317-092026
- Sampson, R. J. (2008). Rethinking crime and immigration. Contexts, 7(1), 28–33. https://doi.org/10.1525/ctx.2008.7.1.28