The Marshall Project and the New York Times today jointly published a piece claiming that Secure Communities, a program that flags illegal immigrants for possible deportation when they’re arrested in the U.S., is ineffective at reducing crime. The piece is based on an academic paper that crunched the numbers on more than a thousand areas of the U.S.
The paper is interesting, but the results are not that surprising, whatever you think of deporting criminal aliens. Secure Communities is a targeted at a pretty narrow subset of criminals, and yet the paper looks for the effect of this policy in areas’ overall violent- and property-crime rates — which of course fluctuate for all sorts of reasons besides deportations. It’s inherently unlikely you’ll be able to measure the benefit or harm of removing a relative handful of illegal immigrants from most places this way, even if those illegal immigrants are unusually crime-prone.
A table near the end of the paper indicates that just 0.02 percent of the overall working-age population was deported annually, on average. More than half the areas had deportation rates of zero. And while the Marshall Project piece says that half a percent of the overall working-age population was deported in some places, the paper reports 0.2 percent as the 99th percentile, so numbers that high were apparently quite rare.
Further, as the piece notes, areas varied widely in how they implemented the program, with some using it as a sort of dragnet that didn’t target the worst criminals. Per the paper, “nearly half of deportees under SC had only minor offenses” — pot possession, traffic violations, etc. — “as their most serious criminal conviction (MSCC) or no offense at all.”
That’s a fair criticism of the program in itself, perhaps, but I do not trust the methods of social science to detect the effects of deporting a small fraction of 1 percent of the population in such a fashion. I’m going to go ahead and continue to support the deportation of criminals, especially the worst ones.