We often hear that illegals commit fewer crimes per capita. Our betters in the media tell us this all the time. The issue is relevant to the immigration question in general, and specifically to Trump’s dopey proposal to send illegals to sanctuary cities (which I have argued endangers the public because the criminal part of that population is more likely to evade deportation). I thought it would be worth devoting a post to questioning some of the flawed reasoning underpinning this claim.

Before I start, let me point out that none of the arguments I advance below depends upon a bigoted view of Mexicans or Central Americans as inferior or naturally prone to crime — views that I do not hold. If you’re among the subset of Trumpists who subscribes to that view, or if you’re a leftist looking for a way to call a conservative racist, move on. You’ll find nothing here to support your nasty preconceptions.

A 2017 Heritage article says:

According to a recent Associated Press article, “multiple studies have concluded that immigrants are less likely to commit crime than native-born U.S. citizens.” But the issue isn’t non-citizens who are in this country legally, and who must abide by the law to avoid having their visas revoked or their application for citizenship refused. The real issue is the crimes committed by illegal aliens. And in that context, the claim is quite misleading, because the “multiple studies” on crimes committed by “immigrants” — including a 2014 study by a professor from the University of Massachusetts, which is the only one cited in the article — combine the crime rates of both citizens and non-citizens, legal and illegal.

That isn’t the only problem with the study. Instead of using official crime data, it uses “self-reported criminal offending and country of birth information.” For obvious reasons, there is little incentive for anyone, let alone criminal aliens, to self-report “delinquent and criminal involvement.” When it comes to self-reporting criminal activity, some respondents will, no doubt, exaggerate. Others will flat out lie. Furthermore, many respondents will likely not disclose if they are a non-citizen out of fear of discovery and deportation.

(Incidentally, Cato, a libertarian outfit of the stripe that prefers open borders to make the businessmen happy, argues that illegals commit less crime but admits that much of the applicable research “combines legal and illegal immigrants to calculate a crime rate for all immigrants.”)

The Heritage piece goes on to cite some “disturbing actual data on crimes committed by criminal aliens” that tends to undercut the conventional wisdom. First, we learn that a GAO report revealed that “criminal aliens (both legal and illegal) make up 27 percent of all federal prisoners” despite making up only about “nine percent of the nation’s adult population.” That certainly seems inconsistent with the claims that they commit fewer crimes than natives. Indeed, the Heritage piece adds: “One 2001 study that does take country of origin and geographic concentration factors into account found that Mexican immigrants ‘commit between 3.5 and 5 times as many crimes as the average native.’”

And if you think this is a small problem, think again. Another GAO report set forth some of the statistics of the total numbers of crimes we are looking at — crimes committed by people who, under our law, don’t belong here in the first place. This GAO report

looked at the criminal histories of 55,322 aliens that “entered the country illegally and were still illegally in the country at the time of their incarceration in federal or state prison or local jail during fiscal year 2003.” Those 55,322 illegal aliens had been arrested 459,614 times, an average of 8.3 arrests per illegal alien, and had committed almost 700,000 criminal offenses, an average of roughly 12.7 offenses per illegal alien.

Out of all of the arrests, 12 percent were for violent crimes such as murder, robbery, assault and sex-related crimes; 15 percent were for burglary, larceny, theft and property damage; 24 percent were for drug offenses; and the remaining offenses were for DUI, fraud, forgery, counterfeiting, weapons, immigration, and obstruction of justice.

That is a lot of unnecessary crime.

Now, for some of my own analysis. To the extent that there remain any valid statistics or studies out there that do show illegal immigrants commit less crime — and perhaps there are — I believe that is a simple function of the fact that we have the ability to deport repeat offenders. Let me go through the logic.

First of all, much crime is committed by repeat offenders. Assume we have two hypothetical groups, A (citizens) and B (illegals). Assume for the same of argument that both groups are composed of the same mix of people with the same characteristics, including the same tendency towards criminality. (As we will see later, the groups are different demographically in terms of age and gender, but for now we are going to assume they are the same.) Also assume that all criminal offenders from group A (citizens) who are sent to jail or prison are returned to the streets once released from custody. Further assume that most criminal offenders from group B (illegals) who are sent to jail or prison are sent out of the country once their sentences are served.

Over a period of years, who will commit fewer crimes in the country: group A (citizens) or group B (illegals)? Likely, in our hypothetical, both will commit equal numbers of crimes overall in the world at large, since we have described the groups as being roughly equal in terms of their tendency to commit crime, for purposes of this (counterfactual) hypothetical. But group B, the group of illegals, will be committing fewer crimes in this country, since they are deported after committing their first crime that gets them incarcerated.

So when you measure the statistics of crimes committed by group A (citizens) and group B (illegals), it will appear that group B is safer overall.

But this is true only as long as you maintain the status quo, and refuse to grant legal status to illegals, and continue to deport them when they commit crimes — at least, as long as sanctuary city and sanctuary state policies are repealed or outlawed, and the machinery of deporting criminals works efficiently. As long as we can deport criminals, there will be a mechanism that tends make the illegals appear less criminal than they are innately.

But the second you start to use that “fact” to justify legalizing illegals and treating them as citizens — because hey, they commit fewer crimes than citizens! they are more desirable! — then the statistics will soon equalize again. Because once you make the illegals into citizens, you will no longer be able to deport the criminals once they get to jail.

As far as I know, any argument that says “illegals commit fewer crimes” fails to take this differential into account.

Further complicating the picture is that the demographics, in terms of gender and age, are actually quite different between citizens and illegals. National Affairs reports that illegal immigrants tend to be younger and more male:

However long they have been here, the undocumented are strikingly young. Pew reports that the median age of undocumented adults is 36.2, compared to 46.1 for legal-immigrant adults and 46.5 for native-born American adults. These numbers reflect the fact that the many risks associated with illegal status — travel through dangerous terrain, larcenous smugglers, unscrupulous employers — are more easily negotiated by the young, and particularly by young men. This is one reason why men significantly outnumber women among the illegal-immigrant population: Of the undocumented immigrants over the age of 18 currently residing in the U.S., there are approximately 5.8 million males, compared to 4.2 million females.

The age and gender profiles of the undocumented translate into a large cohort of young, unattached males — with no spouses, partners, or children, at least in this country. According to Pew, nearly half of illegal-immigrant men are “unpartnered adults without children,” while fewer than one-fifth of illegal-immigrant women are.

Most people are aware that young men commit more crimes on average than people who are not young men. So, given the higher prevalence of young men in the illegal immigrant population, you don’t have to conclude that there is anything bad or inferior about Latinos to conclude that they are more likely to commit crimes.

So I’m not convinced that the folks coming here illegally from Mexico and Central America are innately predisposed to commit less crime than citizens. There are some statistics going either way, and their large demographic cohort of young males tends to support the statistics that suggest they commit more crime. To the extent that the statistics or studies favor the proposition that they commit fewer crimes, some part of that is a function of the fact that they remain illegal and that we retain the ability to deport them after they commit crimes.

That is not an argument that we should legalize them. Moreover, it makes sanctuary city policies especially dangerous — because if the population contains more criminals, it’s that much more important for immigration officials to have access to them in custodial facilities and immediately after they are released.

Trump’s dopey proposed policy makes more illegals subject to these policies, not fewer. And the mayors of the sanctuary cities aren’t deterred. Many of them say they welcome more illegals to their cities.

So to the extent you’re relying on the notion that illegals commit less crime to support (or oppose) Trump’s policy, there’s good reason to question that assumption.

[Cross-posted at The Jury Talks Back.]