To what extent do low-skill immigrants contribute to the US economy, compared to high-skill immigrants? How does immigration affect wages for native workers? Does America have an assimilation problem? How should we think about border security? On a recent episode of Political Economy, I discussed these questions, and many more, with Michael Clemens.
Michael Clemens is a senior
fellow at the Center for Global Development, where he is the Director of
Migration, Displacement and Humanitarian Policy. Michael is also a Research
Fellow at the IZA Institute of Labor Economics in Bonn, Germany. He has also
been published in multiple peer-reviewed academic journals, and his research
has been awarded the Royal Economic Society Prize.
Pethokoukis: What is the benefit of bringing people to this country who are not Nobel Prize winners or computer scientists? Why do we need to bring in people with less than a high school education?
Clemens: Those people may be less visible, but they’re all around us. Every single person who has a college degree depends on a vast army — every hour of every day — of people at all levels of education doing all kinds of tasks. The idea that we only need Silicon Valley engineers or something like that just ignores the fact that if you walk through Silicon Valley, you’ll see an economy that also runs on child care, construction work, security, warehousing, and farmers hand-harvesting the vegetables that some of those immigrant computer scientists are having in their salad.
That’s what the economy of Silicon Valley is, and that’s what the economy of America is: people at all different skill levels doing complementary tasks, benefiting each other and giving each other jobs. The economy is just spectacularly more complex than the superstars that we see like Sergey Brin.
Do you think there is a problem with the assimilation of immigrants today?
It’s difficult to measure. However, Ran Abramitzky and Leah Boustan used census data to track the degree to which descendants of immigrants start to have more American names — a pretty reasonable proxy for the degree to which they’re becoming American. They have studied this phenomenon both in recent waves of immigrants and 100 years ago, and, remarkably, the rate at which they give American sounding names to their kids hasn’t changed over time.
People seem to say, well we used to be a melting pot and that’s all gone now. There’s not a lot of evidence of that. And insofar as some immigrants do take time to assimilate, that partly because of barriers we erect against immigrants — particularly unauthorized immigrants.
What is the impact of immigration on low-skill workers’ wages?
The world’s best economists disagree very strongly about this question, so the issue is not settled. However, two studies come to mind.
First, a study by Giovanni Peri and two coauthors reviews the impact of Puerto Ricans fleeing Hurricane Maria on Orlando, Florida. They find all kinds of different impacts on different parts of the economy. A lot more labor in construction seems to have driven down construction wages a lot. A lot more demand for retail services and hospitality services seems to have driven up wages in those sectors. The average effect, at all skill levels, is nothing in the short term. Puerto Ricans coming into Orlando shift the economy to some degree, but they don’t harm it overall.
As for the other study: Christian Dustmann studies a change in German policy that let people living in the Czech Republic cross the border to work near the border but not actually live in Germany. The study finds important negative effects on German employment, but — as the authors note — this is because the Czech residents were banned from actually living in Germany. So a lot of the things that they could do that would certainly have offset the pure labor market effects — like drive up the demand for housing or drive up demand for local goods and services — weren’t allowed to happen.
At debates on this topic, the anti-immigration expert will often say, “If we throw open the borders, immigrants would flood the country in overwhelming numbers, and that would be the end of America as we know it.” Why is that argument wrong? Wouldn’t everybody come here if they could?
Absolutely not. We’ve had open borders — and a very large initial economic gap — with Puerto Rico for over 100 years now. There’s been substantial migration from Puerto Rico, but certainly not a depopulation of any kind. We also have de facto open borders with Micronesia. A moderate percentage of Micronesians left for Hawaii, since Micronesia is a relatively poor country. But the vast majority of them didn’t.
This is a scare tactic. It was definitely used in the Chinese exclusion debates of the 1870s and 1880s. The same metaphors of floods, tides, dams, and dikes just don’t change one bit over time. There’s just never been any actual evidence of it.
I think it’s vastly more constructive to focus, as
economists would, on marginal analysis. And the evidence is very clear: Current
immigrants create more businesses and commit less crime than natives, and the
same is likely to be true of additional rounds of immigrants. So, at the margin,
there is a big opportunity for the country in bringing in more people.
What are your general thoughts about border security and things like E-Verify?
Border security is enhanced by lawful migration mechanisms. I think some folks who are all about security should consider that those efforts are enhanced when migration is safe, orderly, and regular. That’s just not going to happen without meaningful consideration of the vast economic and demographic pressures that drive migration — and which are not just going to go away. No security is enhanced by cat-and-mouse games at the border. However, I think that it’s absolutely legitimate for countries to know who’s entering.
If the point of E-Verify was to improve employment and wages for workers competing with unauthorized workers, Pia Orrenius and Madeline Zavodny — two leading economists of immigration — have studied what happened to US workers in states that rolled out E-Verify at different times. It’s an interesting natural experiment to see the real impact of this on US workers, and they were not able to detect any effect of E-Verify.