Immigrant Crime Is Good, Actually

If you want to kill someone, try it in a rich country.

In 2024, Americans voted to dismantle the Constitution and fuck up the economy because of immigrant crime. A lot of them now regret it, but not as many as you might think.

Some people, like the brilliant Alex Nowrasteh, have pointed out that immigrants are less likely to commit crimes than native-born Americans, regardless of their legal status. This is likely correct, but it also misses the bigger picture.

The correct position on immigrant crime is that it’s a good thing, or at least plausibly a good thing compared to the alternative, which is immigrants committing crimes in their home countries.

Why? Because the United States is better at deterring and incapacitating criminals than a typical sender country, so more criminals moving from sender countries to the United States is going to result in less crime.

The case for this is rather straightforward. Some people are always going to try to commit crimes, but whether they get away with it depends in large part on the social and political environment of the country in which they live. If you accept that crime doesn’t simply go away when people aren’t in the United States, you’re left with two options. Either immigrants will try to commit crimes in their home countries (where they’ll sometimes get away with it), or they’ll try to commit crimes in the United States (where they’ll also sometimes get away with it but typically less often).

You can see below how homicide rates in the United States compare to Mexico and Central America. In 2023, the U.S. homicide rate was less than one-fourth as high as any of the sender countries for which we have data. As long as you think at least one-third of the difference in homicide rates is explained by environmental variables (for example, social attitudes toward lawbreaking or the effectiveness of law enforcement), then for every person who’s victimized by an immigrant criminal in the United States, there’s at least one person who’s spared from crime in the immigrant’s home country.¹

So what have we ought to do? For one, be much more indiscriminate about who we let in the country. Ideally, there should be thousands of illegal criminals poisoning the blood of our country every day. Give me your tired and poor? Sure, but we’re much more interested in the wretched refuse — criminals, rapists, and some, I assume, are good people. We’re looking for folks like the late, great Hannibal Lecter.

Next, devote significant resources to deterring and incapacitating the criminals. This doesn’t look like more immigration enforcement, which is very bad at catching actual criminals, but likely more money and authority for police and whatever other programs effectively stop crime. Maybe it means more welfare (or less welfare), or looser restrictions on racial profiling, or gun control. Whatever works without unacceptable tradeoffs.

Taken together, these policies should result in a lot fewer people being victimized by crime. All good, right?

You might object that we shouldn’t care about crimes committed against foreigners because they’re not from our country or part of our social contract. Maybe we can care about the rest of humanity once we have our own house in order, but we’re Americans first and we ought to care about crimes committed against Americans.

This makes intuitive sense to a lot of people, but it doesn’t jibe with why most people think crime is bad. If you asked most people, they’d say crime is bad because it’s bad for the people who experience it and have to suffer being crime victims. But the fact of suffering doesn’t change if the victim is a different nationality or lives in a different country than the observer.

You might also say that a crime committed in the United States matters more than a crime committed in a sender country because the United States is more productive and crime is more socially costly.

This is a much better objection, but the difference in costliness of crime in the United States and sender countries probably isn’t enough to change the equation. The Value of Statistical Life in the United States (the amount people are willing to pay to avoid a probabilistic death) is about three times as high as it is in Mexico. However, we ought to take this as an upper bound for the comparative cost of crime, since VSL is only designed to capture the economic costs of risks to life (constrained by people’s ability to pay), rather than moral value. Among other things, VSL may struggle to capture whatever value we ought to place on our conceptions of human rights and universal human dignity.

Still, even if you accept VSL as a good coefficient for the comparative moral cost of crime between countries, the case for immigrant crime remains plausible. The homicide rate in the United States is at least four times lower than it is in Mexico, so as long as you think most of the difference is environmental, there’s a strong case for bringing more Mexican criminals into the United States.

Ultimately, you can sanity check the case for immigrant crime with a simple thought experiment. In the trolley problem below, if the trolley was headed toward the one American, would you redirect it to kill the three Mexicans? This is what you’d be doing if you tried to prevent homicides in the United States by blocking Mexican immigration. Most people would say no to switching the trolley, and that’s the correct answer.

1. Why should we think at least one-third of the difference can be explained this way? Take a look at El Salvador’s homicide rate before and after it implemented changes to its criminal policy. It managed to close 99% of the gap between itself and the United States.