The immigration issue is fraught at the best of times, but especially so right now, with the rhetoric emanating from Donald Trump's presidential campaign.
Yet David Frum, a former George W. Bush speechwriter and senior editor at the Atlantic, has chosen this time to wade into the fray with a long article about the downsides of low-skilled immigration.
Unfortunately, Frum's article has a number of problems, as well as some weak arguments, that make low-skilled immigration appear less beneficial than it really is.
Frum writes: "Donald Trump's noisy complaints that immigration is out of control are literally true. Nobody is making conscious decisions about who is wanted and who is not, about how much immigration to accept and what kind to prioritize-not even for the portion of U.S. migration conducted according to law, much less for the larger portion that is not."
Already there appears to be a major error here -- legal immigration has always been much higher than the illegal kind, and illegal immigration is now close to zero. But Frum's main argument is about legal immigration, and he effectively claims that some groups of immigrants are undesirable.
Now, I too have argued for shifting the U.S. system toward high-skilled immigration (as Canada does). Doing this would increase productivity and living standards. It would also act against inequality, since high-skilled immigrants compete with wealthier native-born workers, while low-skilled immigrants compete with the native-born poor. So for a given amount of immigration, it definitely makes sense to select based on skills.
But that is very different from claiming that low-skilled immigrants are bad for the country. Low-skilled immigration might give gross domestic product a smaller boost per immigrant, but it almost certainly helps the country on net.
As evidence for his claim that not all immigrants are desirable, Frum discusses the situation of Somali immigrants. He notes high rates of unemployment (more than 20 percent in one Minnesota community), high rates of welfare use and some anecdotes of Somali immigrants who went overseas to fight for radical Islamist causes.
But Frum's example is cherry-picked. Given the huge diversity of immigrants who come to this country, it's obviously going to be possible for an immigration skeptic like Frum to look around and find one small subgroup that dramatically underperforms.
But this proves very little, because Frum doesn't propose a way to tell in advance which groups will prosper and which will struggle.
The fact is, most immigrants prosper. According to a 2007 Brookings Institute report, immigrants to the U.S. tend to experience very strong upward mobility between the first and second generations. Second-generation immigrant workers -- that is, the U.S-born children of immigrants -- tend to earn more than non-immigrant workers.
This upward mobility is clearly evident for Mexican immigrants and their children. Mexican immigrants are, on average, low-skilled -- they make about 32 percent less than the native-born. But for their children, the gap is only about 15 percent.
A 2013 Pew Research Center report corroborates this, finding that median income for Hispanic households increases very strongly between first and second generation.
Frum tries to deny this mobility: "Where Americans have more difficulty is offering a path to upward mobility, especially for people born into the poorest one-fifth of the population.
Not all migrants inhabit that bottom one-fifth. But disconcertingly many do-and contra the American Ellis Island myth, their children then stick there."
As evidence for low mobility, he cites a study showing that intergenerational mobility is low for Americans as a whole. But this provides zero empirical support for Frum's argument that poor immigrants are not upwardly mobile.
The Brookings report, the Pew report, a Center for Immigration Studies report, and other data confirm that the income mobility between first- and second-generation immigrants is much higher than for native-born Americans. The Ellis Island "myth" that Frum waves away isn't a myth -- it is a fact.
Although average Hispanic income never converges with the white average, it certainly comes closer to it. There is little doubt that the average first-generation Hispanic American is contributing much more to GDP than he or she extracts in government benefits -- and since the second- and third- generations do even better than the first, it is certain that these immigrants have been a net plus for economic output.
How are we to identify which low-skilled immigrants will prosper and which will struggle? Obviously we can't use education levels, since some low-skilled groups demonstrably prosper.
Yes, there will be a few subgroups who struggle, just as there will be a few individuals who do badly out of any immigrant population. But the net benefit, and the extreme difficulty of predicting success, make it very unlikely that we could pick and choose among low-skilled immigrant groups.
Frum does manage to identify one real and troubling trend among low-skilled immigrants. Although they are assimilating rapidly, low-skilled immigrants are in some ways assimilating "downward."
Frum correctly cites declining marriage and an increase in out-of-wedlock births among second-generation Hispanics.
But while this is surely a concern, it's a problem that goes far beyond immigrant communities.
Immigration restriction won't be nearly enough to solve the cultural problems of the American working class. When white and black and Asian working- class Americans are all suffering family breakdown, keeping out low-skilled immigrants isn't going to put a dent in the problem. The only thing it will dent is economic growth.
In this election-season climate of resentment and anger, it is easy to rail against low-skilled immigrant families. But a more balanced and careful look at the data shows that Frum's negative outlook is misplaced.
Meanwhile, although switching to a Canada-style system that favors high-skilled immigration would be a good move, this is no reason to downplay the real contributions that hard-working less-educated immigrants have made to the U.S. economy.
Noah Smith is a nationally syndicated columnist.