Five things economists know about immigration
Dylan Matthews of Wongblog at the Washington Post posted this column on January 29:
Few areas of economics have provoked as much fruitful research as immigration, and while disagreements remain, there are at least a few things we can glean from that literature. Here are just a few of them.
1. It’s really good for immigrants
File this one under “duh,” but immigration is a great deal for immigrants, and an even better one than it was during previous eras of mass immigration. Lant Pritchett, a professor at the Harvard Kennedy School and author of Let Their People Come, a great book on the economics of immigration, produced this graph comparing wage gaps between immigrants’ destinations and countries of origins in the 19th century to those gaps in more modern times.
In the 1870s, workers in Ireland could double their wages by coming to the United States. In the 1990s, workers in Guatemala could raise their wages sixfold by coming to the Unted States. In another study, the University of Wisconsin’s John Keenan estimated that completely opening the borders would increase the average developing country worker’s salary from $8,903 to $19,272 — more than double.
2. It’s very good for the economy as a whole
Economists have tried to put a dollar figure on how much the world economy would grow if we just removed all immigration restrictions overnight. The answer: a lot. Angel Aguiar and Terrie Walmsley modeled the effects of three U.S. policy alternatives — full deportation of Mexican immigrants, full legalization and full legalization with increased border control — and found, unsurprisingly, that full deportation reduces gross domestic product and the others would add. Deportation reduces GDP by 0.61 percent, legalization with border control increases it by 0.17 percent and legalization without border control increases it by 0.53 percent.
Pritchett, meanwhile, compared what open borders would do to world GDP, compared to completely free movement of capital and completely free trade with developing countries. It’s not even close. Open borders increase world GDP by $65 trillion. Let me repeat that. $65 trillion — with a ‘t’. The others don’t even come close.
3. It increases innovation
Businessweek’s Charles Kenny, who’s also a fellow at the Center for Global Development, highlighted a slew of studies suggesting that high-skilled immigration is key to innovation in America. Foreign nationals living in the United States accounted for 25.6 percent of all patent applications and founded 26 percent of start-ups, including a majority of Silicon Valley start-ups. In addition, an increase in immigrants with higher education diplomas is associated with an increase in patenting. Charles Lin at Rutgers found that an expansion of high-skilled visas passed in 1998 increased revenue at affected companies by 15 percent.
4. The typical native-born worker probably benefits
There’s a lot of debate on this one. A 2010 white paper by Gianmarco Ottaviano, Giovanni Peri and Greg Wright found that less expensive immigrant labor has a “positive net effect on native employment.” In another paper, Peri found that U.S. immigration from 1990 to 2006 increased real wages by 2.86 percent. Put together, Peri’s research forms the strongest basis for arguing that immigration increases wages for native-born American workers. Patricia Cortes at Unviersity of Chicago has confirmed his findings, Heidi Sheirholz at EPI and Raúl Hinojosa-Ojeda at the Center for American Progress, similarly, have found across-the-board gains from immigration (or, in the latter case, comprehensive immigration reform) to wages.
George Borjas and Lawrence Katz, two Harvard labor economists who tend to be more skeptical of the benefits from immigration, beg to differ. Between 1980 and 2000, U.S. workers saw their wages fall in the short-run by 3.4 percent due to immigration. In the long-run, the economy adjusts such that the overall effect is minimal, but the short term figures are still a cause for concern.
Unsurprisingly, Peri and Ottaviano dispute Borjas and Katz’s methodology. They argue that Borjas and Katz inaccurately assume that U.S. and foreign workers are perfect substitutes. That’s a problematic assumption, since immigrants tend to do a different kind of labor, one which might not even exist in their absence. “[Immigration opponents] say ‘we Americans could do the job!’ but they don’t say ‘we’ll do the job at a significantly higher price at which the job wouldn’t exist,’” said Jagdish Bhagwati, a trade and immigration economist at Columbia and the Council on Foreign Relations. Borjas and Katz also neglect the indirect benefits that immigration provides to all groups through increasing growth.
But even taking Borjas and Katz at face value, the two groups’ estimates aren’t that far off from each other when you look at the long-run, as this chart from Michael Greenstone and Adam Looney at the Hamilton Project shows.
Everyone agrees that high school grads and people with some college benefit in the long run and despite their short-run estimates, even Borjas and Katz show a mildly positive overall effect in the long run. The dispute is about what happens at the low-end.
5. Low-skilled immigrants probably don’t see any effect
That’s what Peri’s findings say above, and they’re confirmed in two notable studies, by David Card and Rachel Friedberg, which found that the Mariel boatlift (which brought upwards of 100,000 immigrants to Miami in 1980) and the early 1990s Russian Jewish migration to Israel, respectively, did not decrease native employment or wages. Both were big events. The boatlift increased Miami’s population by 7 percent, and the Russian migration increased Israel’s population by 12 percent.
The advantage of these studies is that they isolate what economists call a “supply shock” to labor. All of the sudden, for reasons unrelated to other factors in the economy, the supply of labor increased. That makes it easier to determine what that shock’s effects are, because it’s not itself caused by other factors in the economy. This increased Card and Friedberg’s confidence that there really wasn’t an effect on wages from the sudden influx of immigrants. But other studies have found this as well. Peri argues that while low-skilled native workers suffer due to liberalized immigration in the short-run, they aren’t affected in the long-run.
Of course, Borjas, Katz and other skeptics argue that low-skilled immigration very clearly reduces wages and employment for low-skilled American workers. The issue is, as yet, unresolved. But the consensus view among economists is that the effect, even if negative, is negligible.