Immigration Crime Data: A Tale of Different Numbers

Migrant crime has become one of the most talked-about issues this year as the Trump administration ramps up deportations and tightens border enforcement. But depending on which dataset you examine, you’ll get starkly different pictures of what migrant-related crime actually looks like in the United States.

A recent Department of Homeland Security press release ignited a political firestorm after DHS reported that New York released nearly 7,000 individuals with active Immigration and Customs Enforcement (ICE) detainers between January and early December 2025. ICE had specifically requested the state to hold these individuals for federal custody, but New York declined to honor those requests.

According to DHS, the individuals in those detainer cases had been charged with or convicted of 29 homicides, 2,509 assaults, 300 weapons offenses, and 207 sexual predatory offenses.

These figures come directly from local and state arrest or conviction records, the same records DHS uses to flag individuals with pending detainers. However, that distinction is crucial because local arrest data is not equivalent to measuring crime rates among immigrants overall. It simply reflects cases that happened to intersect with New York’s criminal justice system.

When examining federal data that tracks nationwide noncitizen convictions, the picture shifts dramatically. This is primarily because only a small number of crimes become federal offenses, mostly illegal entry/reentry, federal drug trafficking, and federal weapons crimes.

The Department of Homeland Security’s Criminal Alien statistics show that illegal entry and re-entry remain the most common convictions for noncitizens, typically 8,000 to 10,000 cases per year. DUIs represent the second-largest category, while drug and theft convictions fall in the hundreds to low thousands annually.

This federal dataset doesn’t capture every crime an immigrant may commit; it only includes federal convictions. This limitation helps explain why federal numbers often appear much lower than the local data DHS highlights in press releases.

To understand broader crime trends, the CATO Institute, a libertarian think tank, analyzed more than a decade of data from the U.S. Census and the American Community Survey. Their 2023 incarceration rate findings present a different narrative: native-born Americans have a rate of 1,221 per 100,000, illegal immigrants show 613 per 100,000, and legal immigrants have the lowest rate at 319 per 100,000.

CATO’s conclusion is straightforward: illegal immigrants are about half as likely to be incarcerated as native-born Americans, and legal immigrants are the least likely of all groups. These trends have remained consistent every year from 2010 through 2023.

State-level data provides additional context, though only two states—Texas and Georgia—track immigration status in arrest, conviction, and incarceration records. CATO’s analysis of these states’ data revealed similar patterns to the national figures.

In Texas, illegal immigrants had lower conviction and arrest rates than native-born Texans—even for serious offenses like homicide. Georgia’s data showed that illegal immigrants had lower incarceration rates than U.S.-born residents.

The disparities between these datasets highlight the challenges in forming a complete picture of immigrant crime in America. DHS detainer data focuses on specific local cases that intersect with immigration enforcement, while broader statistical analyses from organizations like CATO examine overall incarceration rates by population groups.

The conflicting narratives underscore a broader issue in immigration policy debates: the data used often shapes the conclusions drawn. This has significant implications for policy decisions at both federal and state levels.

CATO has urged federal and state governments to collect better data, noting: “The states and federal government should collect better incarceration, conviction, and arrest data by immigration status so that the public and policymakers can more accurately understand how immigrants affect crime in the United States.”

As immigration remains a divisive political issue, the question of how to accurately measure and report on immigrant crime will likely continue to be contentious, with different stakeholders emphasizing the datasets that best support their policy positions.