The Washington Post Fact Checker takes issue with President Trump’s social media post on sanctuary cities and crime by illegal aliens. The White House sent the Washington Post more than one article to defend its position (including some of our research from Real Clear Investigations), but the only research that the Washington Post discusses is our work. Here is Trump’s post:
“No more Sanctuary Cities! They protect the Criminals, not the Victims. They are disgracing our Country, and are being mocked all over the World. Working on papers to withhold all Federal Funding for any City or State that allows these Death Traps to exist!!!”
— President Donald Trump, in a social media post, April 10
The Fact Checker gives Trump Three Pinocchios for making a false claim and writes: “the available research shows no statistically significant impact of sanctuary policies on crime — and even a possible decrease in property crime.” Below we actually point out one of the studies that the Post cites that despite its flaws provided some evidence that property crime increased.
Here is what the Washington Post wrote up using the information that we provided. The entire information that we provided is shown below that.
But John R. Lott Jr., author of one of the articles sent to us, argues that available data undercounts crimes committed by undocumented immigrants. Using figures released last summer by ICE about the known criminal records of non-detained noncitizens, he estimated that 9 percent of the 7.4 million released noncitizens committed crimes either in the United States or in their home countries, which he says is a conservative estimate but high relative to the rest of the U.S. population.
He reviewed the papers we cited above and said they did not account for inadequacies in the data. FBI reports, for instance, miss many incidents of violent crime and property crime because not all of those crimes are reported.
“None of these papers directly address the empirical problem that criminals tend to commit crimes against people who are similar to themselves,” Lott said. “That means that illegals will commit crimes against other illegals who are living near them. The problem is that crime by illegals is likely to be unreported, and as you have more illegals in the area, the size of that underreporting is likely to increase.”
Fact Checker, “Without evidence, Trump claims sanctuary cities are ‘death traps’,” Washington Post, April 11, 2025.
Here is the full response we sent back to the Washington Post.
Thank you for reaching out. I hope that you are doing well.
My piece showed that 9% of the non-detained illegals (662,566 of 7.4 million) had criminal records, which is high relative to the rest of the US population. But that is an underestimate of the rate at which illegals commit crimes because it doesn’t include gotaways or those that we never saw coming across the border, two groups that presumably had the biggest reasons to avoid coming in contact with border agents. There were 2+ million gotaways during the Biden administration. I suspect the 662,566 number is low and that the Biden administration didn’t accurately identify everyone with a criminal record, even for these non-detained illegals. However, 9% is already a low estimate of the problem. Even if it is a low estimate, it is a hard data estimate of criminal activity by illegals. The question for sanctuary cities is whether they want to have people in their communities who commit crimes at this rate. In my RCI piece, I gave an extremely conservative estimate of the victimization costs from these crimes. When these cities are refusing to coordinate with ICE, it is for those illegals who are engaging in criminal activity.
I have looked at the papers that you sent. I don’t understand how these studies run these regressions without accounting for simple things like policing or prison. Let alone accounting for any other type of laws that could affect crime rates, such as California’s 1994 three strikes law? In many of these studies, a large portion of the sample is from California (e.g., the Gonzalez paper has over a quarter of their sample from that state). As is very common for research in this area, none of these four studies account for this (though the Martínez et al. study doesn’t provide any new empirical work themselves and the Gonzalez paper has only the simplest before and after comparisons for just the sanctuary cities with no attempt to account for other regional crime trends).
None of these papers directly address the empirical problem that criminals tend to commit crimes against people who are similar to themselves. That means that illegals will commit crimes against other illegals who are living near them. The problem is that crime by illegals is likely to be unreported, and as you have more illegals in the area, the size of that underreporting is likely to increase. Papers in this area talk about the illegals who are victims of crime not coming forward, but they don’t then address that in their empirical work and what that means if they are using the FBI UCR data. Please note that the FBI UCR data is already only getting about 40% of violent crime and 30% of property crime that the National Crime Victimization Survey indicates.
In any case, the Martinez study does mention the point of illegal victimss not coming forward to report crime, but that doesn’t apply to what they are discussing because we are dealing with ICE getting help to catch and deport people who have been arrested who are criminals.
If you look at the Hausman study, I am always dubious when people throw out a lot of data (and since he is using month-fixed effects, what difference does it make if a jurisdiction has data for only 10 of the 12 months of a particular year), but putting this aside, his table 4 actually shows some weak evidence that property crime increased. If you divide the 85.145 by 55.6 you get a t-statistic of 1.531, which is statistically significant at the 6 percent level for a one-tailed t-test.
The Politics of Refuge: Sanctuary Cities, Crime, and Undocumented Immigration, Gonzalez et al
Providing Sanctuary or Fostering Crime? A Review of the Research on “Sanctuary Cities” and Crime Daniel E. Martínez et al
Sanctuary cities and crime, Yuki Otsu 2021 Journal of Economic Behavior and Organization
Sanctuary policies reduce deportations without increasing crime David K. Hausman
I could give you more of a discussion, but I didn’t want to take more time. I hope that you find this of some assistance!
