I just came across this report from the Immigration Policy Center in 2008.
Numerous studies by independent researchers and government commissions over the past 100 years repeatedly and consistently have found that immigrants are less likely to commit crimes or be behind bars than the nativeborn. In the early decades of the 20th century, during the previous era of large-scale immigration, various federal commissions found lower levels of crime among the foreign-born than the native-born. More recently, the U.S. Commission on Immigration Reform reached a similar conclusion in a 1994 report, as have academic researchers using data from the 1980, 1990, and 2000 Census; the National Longitudinal Study of Adolescent Health; and the results of community studies in Chicago, San Diego, El Paso, and Miami.
The problem of crime in the United States is not “caused” or even aggravated by immigrants, regardless of their legal status. This is hardly surprising since immigrants come to the United States to pursue economic and educational opportunities not available in their home countries and to build better lives for themselves and their families. As a result, they have little to gain and much to lose by breaking the law. Undocumented immigrants in particular have even more reason to not run afoul of the law given the risk of deportation that their lack of legal status entails.
In 2000, among men age 18-39 (who comprise the vast majority of the U.S. prison population), the incarceration rate for the native-born (3.5%) was five times higher than the rate for immigrants (0.7%).
In stereotyping immigrants as criminals, some anti-immigrant activists have pointed to estimates by the U.S. Government Accountability Office (GAO) that one quarter of all federal prisoners in the United States are “criminal aliens.”
However, these estimates are highly misleading for two reasons:
Only about 8% of the 2.2 million persons behind bars in the United States at the end of 2005 were in federal prisons. The majority of inmates are in state prisons (57%) or local jails (34%).
Undocumented immigrants are likely to be transferred into the much smaller federal prison system simply on the basis of their immigration status even if they have not committed a criminal offense, or have committed an offense that is relatively minor.
I still think that completely open borders probably would cause a spike in crime. Immigration => poverty => crime is the chain of causation in my mind. That’s one of the only arguments against open borders that’s at all creditable, even if there’s no real evidence for it; crude theories are sometimes right, and sometimes the best guides we have to guess the effects of policies that step well outside the range of experience. I still support open borders, of course, but I think crime risks are a good reason to open the borders carefully and in a somewhat gradual manner. But we should always bear in mind that nothing remotely like an immigration-induced crime wave has actually happened. On the contrary, immigrants are more law-abiding than natives. The easiest way to keep crime down is probably to let in a few more of these law-abiding foreigners to restrain unruly natives.
UPDATE: I had planned to make this a simple “utility” post with a link to an interesting study, but I’m afraid I added more verbiage than I planned, and may have caused confusion. The last statement “let in a few more of these law-abiding foreigners to restrain unruly natives” may have come across as flippant. It actually conflates an obvious point with a more subtle, speculative point. The obvious point is that if immigrants have lower crime rates than natives, they’ll bring down average crime rates even if they don’t affect crime rates among natives. In that case, though, one couldn’t say except as a joke that they were “restraining unruly natives.”
The subtle, speculative point, which I might develop further in a future post, is that immigrants might counter the tendency for natives to sort themselves out in ways that deprive low-class natives of opportunities to interact with conscientious people. Here’s the hypothesis in a stylized form, which is fairly consistent with some big facts, but I don’t know how true it is overall. First, native-born Americans have a strong tendency, especially in recent decades, to sort and stratify themselves, in a fairly meritocratic fashion, with some talented and conscientious people rising very far, while people with less intelligence and/or responsible lifestyles are unable to capture the opportunities American society offers and sink into the lower classes. Educated people work with educated people, live in neighborhoods and even cities where other educated people concentrated, and generally secede from most contexts where they’d have to interact with the uneducated. Again, it’s just a hypothesis, don’t take offense. It’s been advocated with plenty of statistical backing in Charles Murray’s Coming Apart, and elsewhere in various forms, but perhaps it’s a mischaracterization.
At any rate, when you add immigrants into the mix, you add in a group of people among whom skill/education/social class is much less correlated with conscientiousness. Immigrants often work in places where smart, conscientious natives would rarely end up working: being supermarket checkout counters, in fast-food restaurants, as janitors, taxi drivers, etc., and often live in neighborhoods where smart, conscientious natives would be unlikely to live. It’s not that there’s anything wrong with working in a fast food restaurant or living in a slum. Rather, those jobs and neighborhoods are relatively undesirable, and American natives are given a whole lot of opportunities from childhood on up, so that even the relatively disadvantaged, if they’re smart and disciplined, can usually find their way to better jobs and homes. Those who are left behind by this sorting process include most of the criminal elements of the population– people not inclined to follow the rules, people with low impulse control– along with relatively little of the conscientious elements of the population who would be more inclined to condemn crime, urge obedience to the law, and/or inform the police. The hypothesis is not, of course, that most lower-class people are criminals, but rather that there are few of the strong law-abiding figures who would exert an effective moral and organizational influence against crime. Immigrants who work those jobs and live in those places generally didn’t have the opportunities that natives did, and their social class does not reflect a failure to take advantage of opportunities, but rather, poor English, lack of access to education, lack of connections, etc. They may have better impulse control, more inclination to follow the rules and to despise lawbreakers, be more likely to resist and/or report crime, and so forth. In short, they may “restrain unruly natives.” As I said, this is all rather speculative, though for what it’s worth, my guess is that there’s some truth in it. I’ll keep my eye out for evidence one way or the other.
There’s a lot more about immigration and crime on the page: Hispanic crime and illegal immigration in the United States. It seems to corroborate the study linked above.