In March 2020, President Donald Trump’s administration sealed U.S. borders in what was billed as a measure to fight the spread of COVID-19. The estimated number of people illegally crossing the southern border plummeted, and most of those who did cross were immediately returned to Mexico. Soon after, the U.S. erupted in a surge of violence. The homicide rate rose 29% in 2020, by far the biggest one-year gain on record.

In May 2023, the Biden administration finally lifted the Trump-era border measure, known as Title 42. Illegal border crossings had already been surging; now most of those apprehended were allowed to stay and apply for asylum rather than being sent back. Meanwhile, the U.S. homicide rate fell by more than 10% in 2023; it’s down an additional 18% so far this year in the 262 cities tracked by AH Datalytics.

Looking further back, the flow of illegal immigrants into the U.S. appears to have peaked relative to population in the mid-to-late 1990s, which of course is when the U.S. murder rate (and crime in general) began a long decline. In the 2010s, the estimated number of illegal immigrants in the U.S. actually shrank — and murders began rising at mid-decade.

Now, I don’t really believe that increases in illegal immigration cause the murder rate to decline and that decreases cause it to increase. But given how frequently one hears these days that rising illegal immigration is causing a crime wave, it does seem worth pointing out that there’s at least as much evidence for the opposite claim.

This isn’t to say that illegal immigration is a good thing, or that leaky borders don’t encourage certain kinds of criminal activity. It’s just that — not counting the crime of entering the U.S. illegally (the most prosecuted of federal offenses) — illegal immigration simply doesn’t seem to influence crime rates.

Immigrants overall appear to be much less likely to commit crimes than native-born Americans. In a recent National Bureau of Economic Research working paper based on 150 years of U.S. Census data, economists Ran Abramitzky, Leah Platt Boustan, Elisa Jácome, Santiago Pérez and Juan David Torres found that foreign-born men were about as likely to be as incarcerated as native-born men in 1870, slightly less likely from 1880 to 1950, and much less likely since 1960, with their incarceration rate 60% lower in 2020.

A paper published last year in the American Journal of Sociology by Michael T. Light, Jason P. Robey and Jungmyung Kim — at the time all at the University of Wisconsin-Madison — found that noncitizens in California were more likely to be convicted and incarcerated than U.S. citizens arrested for the same crimes, so the incarceration disparity may if anything understate the difference in criminality.

With illegal immigrants, it’s much harder to make such comparisons. Census surveys (which include the monthly Current Population Survey and annual American Community Survey as well the decennial census) generally ask whether respondents are foreign-born, but citizenship questions have been asked inconsistently over the past half century, and there are no questions aimed at sussing out whether people are here legally or not — because this would discourage illegal immigrants, already wary of such surveys, from participating. (National and local estimates of the undocumented population generally start with the foreign-born numbers from the ACS, then try to subtract those here legally using visa-issuance statistics and other administrative records.)

Meanwhile, crime statistics seldom include immigration status, with some states and cities explicitly refusing in recent years to share such information with immigration authorities. By moving in the opposite direction, and attempting to check and record the immigration status of every arrestee in the state, Texas has become the main focus of research on this topic.

Alex Nowrasteh of the libertarian — and, yes, dovish on illegal immigration — Cato Institute was the first to take advantage of these statistics, finding in 2018 that illegal immigrants were less than half as likely (relative to their estimated population) to be convicted of crimes in Texas in 2016 as native-born citizens. A subsequent paper by sociologists Light, Robey and Jingying He, published in the Proceedings of National Academy of Sciences in 2020, reported similar results based on arrest rates for 2012 to 2018.

In 2022, Sean Kennedy, Jason Richwine and Steven A. Camarota of the hawkish-on-illegal-immigration Center for Immigration Studies published a brief critique arguing that these studies ignored the fact that “it can take years for Texas to identify convicts as illegal immigrants while they are in custody” and that when one included those identified as such after they were already in prison, illegal-immigrant crime rates in Texas were moderately higher than overall crime rates.

Earlier this year, Nowrasteh wrote that the three made some “good points” but also some calculation errors, offering a revised finding that illegal immigrants in Texas committed homicides (which he chose to focus on “because government checks of convicted murderers are more thorough and will identify more illegal immigrants after conviction”) at a rate about 15% lower than US-born Texans.

Other recent research includes a 2018 study by economist John R. Lott (known for his much-debated research linking increased gun ownership to reduced crime) that found illegal immigrants to be greatly overrepresented among those sentenced to prison in Arizona (which Nowrasteh said relied on faulty data and in any case didn’t quite show that illegal immigrants were more likely to commit crimes), and a 2019 one by economist Christian Gunadi that used a new method of identifying illegal immigrants in Census data to conclude that they were 33% less likely to be in prison than native-born Americans.

My takeaway from all of this is that while illegal immigrants are probably a little less crime-prone than native-born Americans, it’s really hard to say for sure. (It may also vary depending on where they live, and where they came from.) As for the somewhat different question of whether the presence of illegal immigrants leads to increases or decreases in overall crime rates, two recent analyses — one by sociologists Light and Ty Miller on violent crime, the other by economist Gunadi on violent and property crime — found negative relationships between estimated illegal immigrant populations and crime rates (that is, more illegal immigrants meant less crime) but not statistically significant ones.

So why all the attention paid to illegal immigrants and crime? Clearly a lot of it is driven by bigotry and political expediency, and because nothing I write here will have any impact on those motivated by such considerations, I’m not going to dwell on them. But there are couple of other reasons that deserve to be taken more seriously.

One is that when someone who is in the U.S. illegally commits a high-profile crime, politicians can be held responsible for it in a way that doesn’t really apply when crimes are committed by native-born Americans. It’s “if only the government hadn’t let that guy in, my daughter would still be alive” versus “if only that guy hadn’t been born.” That at least feels like a legitimate reason to be choosy about who gets to enter the U.S.

Another is that “crime” often serves as shorthand for concerns about something else. Slate’s Henry Grabar made this point a couple of years ago with regard to homelessness, which like illegal immigration is a disorderly product of government policy failure that seems to bear little relation to crime rates — “They are not meaningfully correlated; they do not share causes; they do not share solutions” — yet is often discussed in the same breath.

How to end this confusion? Well, end homelessness and end illegal immigration. This probably won’t have any impact on crime, and my solutions may look a lot different from yours, but both seem like goals worth striving toward.

Justin Fox is a Bloomberg Opinion columnist covering business, economics and other topics involving charts. A former editorial director of the Harvard Business Review, he is author of “The Myth of the Rational Market.”