Last night I was thinking once again about Donald Trump’s statement that rapists are coming into this country—not about the political ramifications, or about Trump’s candidacy (I’m not a Trump fan), but about the question of whether it’s true, and if so what it means.
Now, on a certain level of course it’s true. It would be bizarre if, among the many millions who have entered this country illegally, there wasn’t a single rapist. So of course there are rapists. The question is “how many?”
Or is it? There’s also the question of what would be an “acceptable” number—is there such a number, or is “one” too many? Another question is: is the percentage of rapists among the population of illegal immigrants higher than it is in the general US population? How do we determine this, and can we?
To do so, the very first statistics you would need are reliable figures on the number of illegal immigrants in the US, and although there are figures (see this, for example), it is the very nature of illegal immigrant numbers to be uncertain, since many of those who enter this country illegally live under the radar, which would make statistics-gathering more difficult.
And although I’ve found figures on the percentages of the prison population that consist of illegal immigrants (more later on that), they tells us little about what violent crimes they are committing—for example, rape, for which there don’t seem to be separate figures on illegal immigrants (at least, I haven’t found any). And how many of their crimes are reported, if they are often committed against other illegal immigrants, who may be loathe to go to the law for help?
You can see the problem. It’s quite substantial. And yet to have an intelligent and informed discussion about illegal immigration, one would need to know these things. What’s more, when you try to find the research and the statistics, you discover (as I have) that often categories are lumped together in a way that makes them less meaningful, such as all “immigrants” (both legal and illegal), all “crimes” (rather than a breakdown of types of crime), and a failure to differentiate among countries of origin. In addition, as with all research of this sort, how the figures are obtained is very important, but in order to find that you have to get the original research rather than summaries or references in the newspapers. It is arduous, time-consuming, and difficult, and sometimes the information is unobtainable by the layperson.
We know a few things, or assume we do. We know that illegal immigration has been growing in the last decade or two not just in absolute numbers but as a percentage of the US population.
We know a few things about illegal immigrants and crime. We know that cartels from Mexico are heavily involved in the drug trade and that they use Hispanic communities to hide in and provide cover and foot soldiers. But are these legal immigrants? Illegal? What percentage of the drug trade do they represent?
Here is a report with some examples of the type of thing I mean; you can see some of the problems demonstrated in the bullet points. As a further demonstration, let’s take this 2007 NY Times article, which is an attempt to debunk statements made by Lou Dobbs in 2003 such as “One-third of the inmates now serving time in federal prisons come from some other country,” and that illegal immigrants were “an increasing part of America’s prison population”:
In 2000, 27 percent of the inmates in federal prisons were noncitizens. Some of these noncitizens were illegal immigrants, and some were in this country legally. In 2001, this percentage dropped to 24 percent, and it continued dropping over the next four years, falling to 20 percent in 2005.
Bottom line: illegal immigrants make up significantly less than a third of the federal prison population, and the share has been falling in recent years.
So Dobbs was wrong in saying in 2003 that a third of the inmates in federal prison were foreign-born (if that is in fact what he said). Depending on what year he was referring to, it was actually either a quarter or a fifth, according to the statistics—both of which are much more than the percentage of foreign-born people in the country, which the article quotes as being 6.9%. So Dobbs was correct in principle but wrong in detail. And the author of the article is mixing apples and oranges when he writes, “Bottom line: illegal immigrants make up significantly less than a third of the federal prison population…” But he’s already said that the Dobbs quote about a third of the federal prison population was about foreign-born people. He’s not quoted as having given any figure for illegal immigrants, just that they are “an increasing part” of the prison population, which they are—although a significant amount of this increase represents the increase in illegal immigration as a whole, and a significant part also represents prisoners who are in prison for immigration violations.
I could go on discussing the Times article and picking on this and that, both in the article and in what Dobbs purportedly said. But my point is that we don’t have anything close to precision or even something definite here about what’s actually going on. That would take time, an interest in accuracy and truth (dream on, you say), and reliable figures, as well as an audience that understands statistics, and how many do? Not many.
The best site I’ve found so far for relevant-seeming statistics on the subject is this one. It’s not perfect, of course, but it contains data from the United States Sentencing Commission’s Monitoring of Federal Criminal Sentences from 1991 through 2007. Go to the link to read the whole thing, but I’ve excerpted some of the more interesting points here [my own comments are in brackets]:
—Hispanics represented 40% of all sentenced federal offenders in 2007, the single largest racial and ethnic group among sentenced federal offenders. [This figure is much higher than those mentioned either by Dobbs or by the Times (which included all foreign-born), but it obviously includes those Hispanics who were born in this country and are citizens, which would naturally make the figure higher than just illegal-immigrant Hispanics or just foreign-born Hispanics.]
—More than seven-in-ten (72%) of Hispanics sentenced in federal courts in 2007 did not hold U.S. citizenship. They accounted for 29% of all federal offenders in 2007.[This answers how many were non-citizens, and it also is close to the one-third that Dobbs mentioned, although he said it a few years earlier, and he was speaking of all foreign-born offenders rather than just Hispanic non-citizens.]
—Between 1991 and 2007, the number of Hispanics sentenced in federal courts nearly quadrupled (270%), rising faster than the number of offenders sentenced in federal courts over this period and accounting for 54% of the growth in the total number of offenders. [If this number rose so dramatically, we can assume that the number of illegal immigrant Hispanics in prison also rose, since they are now such a high percentage of the whole.]
—Among all Hispanics sentenced in federal courts in 2007, 48% were sentenced for immigration offenses, 37% for drug offenses and 15% for other offenses [So it’s not just immigration offenses Hispanics are being sentenced for; over 50% are sentenced for drug and “other offenses.” But the very important question: how many of these Hispanics sentenced for drug and “other offenses” are illegal immigrants? And what are those “other offenses”? How many are violent crimes, how many are rape?]
—Much of the increase in the number of Hispanics sentenced in federal courts has come from a rise in the number of offenders sentenced for immigration offenses between 1991 and 2007. [So there is no question that a significant amount of the increase involved immigration offenses. However, the report ends in 2007, so the statistics don’t cover two very important phenomena: the recent surge of illegal immigration, and the fact that under the Obama administration, many illegal alien criminals are being released from prison.]
There’s lots more, of course. But that will have to do as an illustration of the problem—not just the problem of illegal immigration, but the problem of figuring out how much and in what way it affects crime in this country.

