The Agenda

NRO’s domestic-policy blog, by Reihan Salam.

Debating Immigration with Cosmopolitans


Michael Clemens and Robert Lynch have written a defense of less-skilled immigration. It is very carefully written. Consider the following passages:

The best, new research we have now is exemplified by the work of Economists Gianmarco Ottaviano and Giovanni Peri. Their research finds that immigration has had essentially zero net effects on the wages of low-wage U.S. workers. To the extent that it has had any effect on wages at all, the most thorough recent research suggests that immigration has had a small positive impact on the wages of lesser skilled native born Americans.

Ottaviano and Peri assess the impact of all immigration, not just less-skilled immigration. In an earlier iteration of the paper, released in 2006, the authors find that immigration has “a positive effect on wages of U.S.-born workers with at least a high school degree and a small negative effect on wages of U.S.-born workers with no high school degree.” One assumes that this finding has been superseded in their more recent research. Yet the question remains: would substituting skilled for less-skilled immigrants have a more substantially positive impact on the wages of U.S.-born workers, and in particular U.S.-born workers without a high school diploma?

Judis claims that legalizing the 11 million unauthorized immigrants living in the U.S. would “thrust [them] into the mainstream labor market, where they will compete with native-born workers.” But research indicates that legalizing these aspiring Americans would help our economy. Legalization would allow unauthorized workers to find jobs that best match their skills, increasing the productivity of our workforce and increasing the earnings of all Americans. Specifically, this research found that legalization adds a cumulative $832 billion to the U.S. economy over a decade, increases the earning of all Americans by $470 billion, generates an additional $109 billion in new tax revenue, and creates an average of 121,000 new jobs each year, benefits that accrue to all Americans.

The crucial question is which Americans benefit. Judis seems to be primarily interested in the impact of immigration reform on native-born U.S. workers. A category that encompasses “all Americans” is by definition not relevant to this concern, as no one doubts that unauthorized immigrants will benefit from legalization. Had Clemens and Lynch separated out the benefits to the native-born population, we’d have a clear picture of what is the best course of action. And of course one could favor legalization without also favoring a further increase in less-skilled immigration. 

In a forthcoming article, Scott Winship makes the following observation:

As the number of immigrants has increased—most of them Hispanic—so too has the share of the workforce with low education levels. These workers are undoubtedly better off than they would be in their home country, but if we add more and more below-median earners to the economy, we necessarily end up with a lower median.

It is impossible to identify immigrants in the CPS prior to 1993, but we can identify Hispanics as early as 1970. Assuming the 1969 to 1970 change was the same as for men in general, median earnings among non-Hispanic men declined not by 12 percent from 1969 to 2011 but by 8 percent. From 1969 to 2007, they rose by 2 percent. Among Hispanic men, earnings fell by 24 percent through 2011, but this decline is simply a more dramatic demonstration of how rising immigration pulls the median downward over time. Among non-Hispanics, the trend for blacks was stronger than the trend for whites, so this is not simply a story about non-Hispanic whites doing better than everyone else. [Emphasis added]

If we have it in our power to decide whether the future U.S. workforce will have a higher or lower share of skilled workers, I would suggest that we ought to choose one with a higher share of skilled workers. Immigration policy represents just such an opportunity. Cosmopolitans like Clemens will object on the grounds that the benefits that accrue to immigrants should outweigh any concerns about, for example, rising wage dispersion in the United States or rising expenditures on means-tested benefits, and that is their right. But those of us who believe that partiality towards the native-born and foreign-born individuals who already reside in the U.S. is morally sound have good reason to object. The benefits that flow from less-skilled immigration are matched and exceeded by skilled immigration, and the U.S. is in a position to shift the composition of its immigrant influx from the former to the latter.

There are reasons to favor less-skilled over skilled immigrants — if we assume that the U.S. will accept some number of immigrants that is lower than the number of skilled immigrants willing to settle in the U.S. in any given year, admitting any number of less-skilled immigrants above zero effectively represents a decision to prioritize a less-skilled worker over a skilled worker — beyond the fact that doing so will tend to alleviate global poverty. As Clemens and Lynch observe, increasing the pool of less-skilled immigrants lowers the cost of outsourcing child-rearing, meal preparation, and other household tasks, and this in turn tends to increase labor force participation among the most productive female workers. It is not obvious, however, that admitting large numbers of less-skilled immigrants is the most cost-effective strategy for increasing female labor force particpation, if that is our goal. We might, for example, reduce marginal tax rates, a strategy that would increase the after-tax returns to increased work hours, thus making it worthwhile for skilled workers to pay caregivers higher wages. To be sure, this strategy might lead to a deterioration of the net fiscal position of the government. Yet the same might be true of increasing the share of the workforce with low education levels. 


Sign up for free NRO e-mails today:

Subscribe to National Review