The tendency to associate immigration with crime has been pervasive throughout US history. For example, in 1891, Senator Henry Cabot Lodge advocated closing the border, warning that Italian immigrants were “members of the Mafia, a secret society … using murder as a means of maintaining its discipline.” Indeed, over the past 150 years, congressional speeches about immigration twice as often mentioned words related to crime than did speeches on other topics.

Contrary to this anti-immigrant rhetoric, our research finds that immigrant men have had a lower incarceration rate on average than US-born men for the past 150 years. We used census data to assemble the first nationally representative data series of incarceration rates for immigrants and the US-born between 1870 and the present day. From 1870 to 1950, the incarceration rate of immigrants was slightly lower than that of US-born men. However, starting in 1960, immigrants have become significantly less likely to be incarcerated than the US-born, even though immigrants have become on average younger, less educated, lower income earners, and more likely to be nonwhite— characteristics often associated with involvement in the criminal justice system. Today, immigrants are 60 percent less likely to be incarcerated than all US-born men and 30 percent less likely than US-born white men. Our results are broadly consistent with prior studies documenting incarceration gaps between immigrants and the US-born for specific states and time periods.

Our research also provides the first investigation of incarceration rates by country of origin from 1870 to 2020. The results reveal a substantial decline in incarceration rates relative to the US-born among immigrants from all major sending regions. European immigrants historically had slightly lower incarceration rates than US-born men but have recently experienced far lower rates. Chinese immigrants had similar incarceration rates to the US-born before 1960 but have significantly lower incarceration rates today. Mexican and Central American immigrants had particularly high incarceration rates in the past but have had lower incarceration rates than the US-born since 1960. Since 2005, Mexican and Central American immigrants have been more likely to be incarcerated than US-born white men, though a large portion of this increase is driven by detentions in federal immigration facilities, often for immigration-related offenses. When we exclude from our analysis the areas home to the largest Immigration and Customs Enforcement facilities, the gap relative to US-born white men moderates or disappears in most years.

Our data do not enable us to precisely pinpoint why there has been a sharp relative decline in the immigrant incarceration rate since 1960. Nevertheless, we can rule out three plausible explanations. First, the relative decline in immigrant incarceration is not driven by rising incarceration rates of US-born black Americans; the decline is also apparent when comparing immigrants with US-born white men only. Second, the decline is not driven by changes in immigrants’ observable characteristics—namely, their countries of origin, age, race, marital status, state of residence, or educational attainment. If anything, immigrants’ lower educational attainment in recent decades would predict that they should have higher incarceration rates than they do. Third, the decline is not driven by immigrant offenders becoming more likely to be deported (and thus absent from the incarceration data); the decline is present even among immigrants who are US citizens and thus cannot be deported. Moreover, the timing of the decline is also inconsistent with this explanation; whereas the relative decline in immigrant incarceration emerged in the 1960s, the sharp rise in deportations occurred around 2000.

Finally, our research shows that US-born men (the group that accounts for the vast majority of incarcerated individuals) and less-educated immigrants not only diverged in their incarceration rates in recent decades but also along other dimensions, including their labor force participation and likelihood of marriage. One potential explanation for this broad pattern of divergence is that less-educated immigrants might have remained relatively shielded from changes in the economy—such as globalization and technological change favoring skilled labor—that have negatively affected less-educated US-born men in recent decades. Immigrants are concentrated in manual tasks and service occupations (rather than routine occupations), which have not experienced large wage or employment declines in recent decades. Furthermore, immigrants may be more resilient, as research shows that they generally possess traits such as a greater willingness to move long distances and take on risk, higher adaptability and cognitive ability, and higher levels of entrepreneurship than the US-born.

NOTE
This research brief is based on Ran Abramitzky et al., “Law-Abiding Immigrants: The Incarceration Gap between Immigrants and the US-Born, 1870–2020,” January 2024.