Does Economics 101 Apply to Immigration?
It's basic economics: When supply expands, prices fall. So one might think that when immigrants increase the supply of labor, the price of labor -- that is, employee compensation -- will decline. But it's "suprisingly difficult" to demonstrate that this actually happens, according to the famed Harvard labor economist George Borjas.
Borjas's new book, Immigration Economics, spans a mere 215 pages (excluding appendices and notes), and yet it stands as a testament to the immensity of this challenge. Reading Borjas -- or, more accurately, wading through his intricately woven thicket of economic models and the theoretical justifications for them -- one can't help but suspect that the entire enterprise is doomed.
While discussing one approach to measuring immigration's impact on wages, for instance, Borjas notes that different ways of grouping people by education level "can make the estimated wage effects as small or as large as one would like." Discussing another method, he shows that focusing on different types of geographical areas -- states, metropolitan areas, etc. -- can also dramatically change the results, for reasons we don't fully understand. (Part of it seems to be that immigration to a metropolitan area causes natives to move elsewhere, spreading out the impact.) He even reveals flaws in the basic datasets that national governments make available to researchers.
Given the malleability of the numbers and the politically charged nature of the topic, it should hardly be a surprise that experts disagree. For his part, Borjas has for years been arguing that the Economics 101 story has a lot of truth to it -- immigration is certainly good in many ways, but it drives down wages for competing native workers. Others, however, argue that immigrant workers "complement" natives instead of competing with them, and thus higher immigration will improve the wellbeing of just about everyone.
This book won't be the last word, obviously. But it does serve as a terrific introduction to the technical side of this topic -- at least for those willing to cope with multiple equations per page -- as well as a careful argument that immigration may not be all it's cracked up to be.
Studying immigration's effect on wages might seem simple: When immigrants come to a new place, do natives' earnings rise or fall? But there's much more to it than that. Immigrants do not choose their destinations at random -- they specifically head to places with jobs suited for their skills -- so it's very difficult to compare high-immigration and low-immigration areas. And as already mentioned, researchers need to make important decisions about how they sort their data into geographical regions and educational groups. A poorly conducted study can fail to pick up on key phenomena.
In decades of well-regarded research summarized in Immigration Economics, however, Borjas has built an empirical case that there are "distributional consequences" to immigration -- newcomers reduce the wages of the natives they compete with directly, but make others better off. In the end, the overall economic benefits of immigration to natives are quite small: In the various models presented here, a 15 percent expansion of the labor supply can't provide a "surplus" for natives amounting to even 1 percent of GDP. And high-school dropouts -- about 10 percent of the native U.S. population and 30 percent of immigrants -- often take a hit to their wages of several percent.
Does the complementarity argument fare better for higher-skilled workers? There are reasons to think it might -- even if you find the idea that immigrant high-school dropouts "complement" native high-school dropouts somewhat absurd, it's hard to deny that highly skilled workers often learn from each other. But Borjas presents some important data that undermines this notion as well, much of it drawn from academia, where the benefits of smart peers should be especially pronounced.
For example, analyzing supply shocks in various doctoral fields, Borjas concludes that a 1 percent increase in the supply of labor reduces earnings by 0.24 percent. Fluctuations in the cap on H-1B visas, meanwhile, in one study didn't seem to help (or hurt) high-skilled native workers in the metro areas where the immigrants went. When mathematicians left the Soviet Union in droves, American mathematicians in Soviet-dominated fields were able to publish less. And when Nazi Germany expelled Jewish academics from its universities, their departure harmed their students' careers but not their peers'.
As Borjas notes, this doesn't mean there are no benefits to complementarity. It just suggests that, from the perspective of an individual facing an influx of peers, the benefits don't outweigh (and often don't equal) the costs of new competition. Even if Borjas is correct here, the case for high-skill immigration remains pretty strong -- it's hard to be too upset when wealthy people encounter stiffer competition and as a result offer their services to the rest of us at a lower price. But if wage-boosting "complementarity" is hard to detect among the highly skilled, we should be rather skeptical of claims that it benefits the poor.
Of course, there's another side to the coin: benefits to immigrants. People disagree as to what role these benefits should play in making immigration policy, but hardly anyone would say they are entirely irrelevant. And considering that much of humankind suffers a level of poverty unknown in the First World, the gains offered by liberal immigration laws might be substantial indeed. Some economists consider open borders a "trillion dollar bill" for humanity just sitting there on the sidewalk.
The problem is that human beings tend to love the place where they grew up, and to establish networks of friends there. Borjas notes that when Puerto Rico became part of the United States, only a third of the population actually migrated, despite the huge economic opportunity available. This suggests that the psychic cost of uprooting oneself is quite high -- for most Puerto Ricans, the substantial benefits of moving to America weren't even worth the cost, and the Puerto Ricans who did move presumably benefited far less than the raw increase in their wages would suggest.
Studies within the U.S. find that moving from one state to another can cost hundreds of thousands of dollars once all this is taken into account. Borjas calculates, using some figures he concedes are at best ballpark estimates, that if we could magically move everyone in the world to the economically optimal place, these psychic costs might well outweigh the benefits. (He includes the present value of future generations' benefits through the use of a discount rate.)
The economist Bryan Caplan, a strong open-borders advocate, has offered a serious critique of this particular exercise. Most importantly, no one is advocating forcing people to move, and the people who willingly chose to immigrate under an open-borders regime would almost by definition be people who would profit from it.
But Borjas makes some less speculative points about the fate of immigrants in America as well, complicating the rosy narrative that open-borders advocates often present. One significant problem seems to be that "economic assimilation," the tendency of immigrants to start out behind natives but steadily catch up as they remain in the country, has slowed in recent decades. Immigrants' children often fail to catch up as well, thanks in part to segregated neighborhoods that transmit "ethnic capital" to each new cohort, keeping some ethnic groups from reaping the full benefits of immigration.
This clustering of ethnic groups in specific neighborhoods, which are continually replenished with fresh immigrants, also raises questions about balkanization. In a note, Borjas presents evidence that English acquisition among immigrants is slowing along with their overall economic assimilation.
And that brings us to the even bigger issue, which Borjas brings up at various points but is never able to address fully: At what point does immigration defeat its own purpose, recreating the problems immigrants are fleeing in a new place? As Borjas writes of fully open borders, it's a big assumption indeed that "whatever it is that makes workers much more productive in the developed countries [will remain] intact after the influx of hundreds of millions, if not billions, of immigrants." Regardless of how you parcel out the blame for the poverty of many nations, at least some relevant factors, most prominently culture, can travel right along with immigrants when they leave.
Borjas isn't being alarmist when he suggests that open borders could move an incredibly large number of people, either. Again, Puerto Rico lost only a third of its population when its residents gained the option of moving to the U.S. -- but a whole lot of people live in poor countries. Only about 15 percent of the world's population lives in developed nations, and billions of people are estimated to live on a few dollars a day.
Looking just at the Americas, whose residents would be most likely to head to the U.S. (total population 314 million, with 10 million illegal immigrants already), one-third of Mexico is 40 million people; Central America, 14 million; South America, around 100 million. It's not unreasonable to fear that with enough immigration, coupled with low levels of assimilation, America and other First World nations could become merely an amalgam, both economically and culturally, of the countries that so many were trying to escape. And the neighborhood trends Borjas has studied are not an encouraging sign.
Where does that leave us policy-wise? One shouldn't make too much of all this, considering that Borjas's findings are entirely open to dispute -- and that arguments about open borders are entirely theoretical. And even if one accepts Borjas's version of the facts, it's not too difficult to argue that the potential benefits of mass immigration outweigh the costs -- especially if you take a cosmopolitan approach, weighting benefits to natives and immigrants equally, and downplay the risks of balkanization and ethnic tensions.
But at the very least, Borjas provides the key elements of an argument against mass low-skill immigration: He lays out evidence that it undermines the employment prospects of the Americans who need decent jobs the most, and that it comes with risks of broader negative consequences.
Robert VerBruggen is editor of RealClearPolicy. Twitter: @RAVerBruggen