In the Wall Street Journal, Two Immigration-Boosting Economists Make a False Statement, Duck Debate

By Jason Richwine on January 19, 2016

When academics are faced with criticism, one way for them to respond is to carefully explain why they believe the criticisms are invalid. Another option is to pretend the criticism does not exist. Economists Giovanni Peri and Vasil Yasenov chose the latter approach in today's Wall Street Journal. By ignoring counter-arguments, they have further obscured the debate over immigration and wages.

Some quick background is in order here. Peri and Yasenov are involved in a debate over an historical event called the "Mariel boatlift." During the summer of 1980, some 125,000 Cubans left the port of Mariel for Miami. This large and unexpected wave of newcomers created an intriguing test of how immigration affects wages. Surprisingly, economist David Card was unable to detect any wage impact in his classic study from 1990, and that result has since become a favorite citation for those who want to expand immigration.

Last fall, however, George Borjas re-analyzed the data and found that the boatlift did have a large negative impact on native high-school dropouts in Miami. Borjas's study was quickly followed by a rebuttal paper from Peri and Yasenov. Then Borjas filed a counter-rebuttal.

To help make sense of all of this, I wrote an article for Real Clear Policy that summarizes the debate in plain English. CIS then published an expanded version as a backgrounder. I struck a neutral tone in both pieces, always giving Peri and Yasenov the benefit of the doubt despite our political differences. But Peri and Yasenov are becoming increasingly difficult to defend. Their WSJ op-ed not only includes a false statement, it also ignores Borjas's entire rebuttal. How can the general public ever hope to understand an issue when academics talk past each other?

Peri and Yasenov's false statement is this one: "[Borjas's] sample consisted of only 17 to 24 workers in Miami in each year, a number too small for statistical significance." The negative wage effect observed by Borjas is statistically significant, meaning unlikely to have occurred by chance. In fact, Borjas found statistically significant effects with two different datasets – in the smaller one mentioned by Peri and Yasenov, and also in a sample about three times larger.

Peri and Yasenov also make an interesting claim at the end of their op-ed: "Earlier research by one of us [Peri] has shown that native workers do not suffer the negative impact of arriving immigrants because they take different jobs." There is a relevant back story here. Peri initially published a result that implied immigrants and natives with the same education are not "substitutes" for each other, meaning they do not compete for the same jobs. But Peri had to retract that result when critics discovered he had accidentally included high school kids with part-time jobs as "dropouts." In the revised paper, Peri found only minor departures from the normal assumption that immigrants and natives with the same education are substitutes.

So the first error was fixed, but, as Borjas notes, it inexplicably resurfaced in Peri's Mariel paper, where he again counts high school kids with part-time jobs as "dropouts." The op-ed gives no explanation for this. In fact, as noted above, Borjas's entire rebuttal is left unmentioned. I will not try to rehash the entire debate here – see my backgrounder for a detailed discussion – but Borjas points out significant problems with broadening the sample the way Peri and Yasenov do, even aside from the issue of labeling high school kids as dropouts. As I explained in Real Clear Policy:

What the debate between Borjas and Peri-Yasenov boils down to is which kind of labor market data we should trust more — small samples of a stable workforce, or large samples of an unstable workforce. When Borjas selected his sample of American workers, he limited it to non-Hispanic men ages 25 to 59. These restrictions are common in the literature because they ensure a stable set of workers. Women, for example, are less attached to the labor force than men, and their wages and participation have increased as gender roles have changed over the years. If women's labor-force participation follows a different trend in Miami compared to the control cities, then the wage comparison could be capturing differential social change rather than the effect of the boatlift.


It would be nice to hear a response from Peri and Yasenov on this, but they offered none.

Ordinarily, I can understand the reluctance to address critics in an op-ed. Space in a print newspaper is precious, and the authors should be allowed room to make their own arguments. However, the whole purpose of Peri and Yasenov's paper is to refute Borjas. When academics enter into direct debate, they have a responsibility to confront the other side's arguments head on. Peri and Yasenov failed to do that, and their readers are worse off for it.

Jason Richwine is an independent public policy analyst and contributing writer to National Review.