Harvard economics professor George J. Borjas writes:
A couple of weeks ago I promised to summarize the evidence on whether high-skill immigration generates the productivity spillovers that would produce large economic gains for natives. I apologize for taking so long to get back to this discussion, but I have so many other projects on my plate these days that it’s hard to keep up.
By far the most convincing studies that attempt to document the existence of spillovers are those that look at natural experiments. In a series of important papers, Fabian Waldinger has looked at what happened in Nazi Germany after Hitler fired all the Jewish professors . In 1933, shortly after it took power, the Nazi regime passed the Law for the Restoration of the Professional Civil Service, which mandated that all civil servants who were not of Aryan descent be immediately dismissed. That meant that Jewish professors like John von Neumann, Richard Courant, and Albert Einstein were fired from their university posts. Many of these stellar scientists found jobs abroad, particularly in the United States.
In his 2010 JPE paper, Waldinger showed what happened to the productivity of the doctoral students stranded behind in German universities after their exceptional mentors were dismissed. The productivity of the students stranded behind in the departments most affected by the dismissals suffered. In his 2012 RESTUD follow-up paper, Waldinger looked at what happened to the productivity of the colleagues of the dismissed scientists. The surprising answer is: Not much.
Here is the trend in the productivity of the graduate students left behind:
And here’s the trend in the productivity of the colleagues left behind:
One lesson I draw is that supply shocks consisting of very high skill workers can certainly generate spillovers, but it seems that there has to be a close personal relationship between the migrants and the recipients of the spillover. There’s also a hint that those who compete for resources may not suffer all that much when their esteemed colleagues are no longer around.
Kirk Doran and I have published a series of papers that look at another natural experiment–the exodus of highly qualified Soviet mathematicians after the collapse of the Soviet Union. The New York Times published a report in 1990 detailing the conflicting forces that such a supply shock would generate: Beneficial productivity spillovers and additional job market competition:
[Harvard mathematician] Dr. Diaconis said he recently asked [Soviet mathematician] Dr. Reshetikhin for help with a problem that had stumped him for 20 years. “I had asked everyone in America who had any chance of knowing”…No one could help. But…Soviet scientists had done a lot of work on such problems. “It was a whole new world I had access to,” Dr. Diaconis said…
American scientists…are being peppered with letters and calls asking for invitations…”I have run across a number of very distinguished Soviet mathematicians who have come here as visitors and spend their time going around the country and looking for a job.”
Because of the long separation between Soviet and Western mathematicians, the two groups had essentially gone their own way during the 20th century, with the Soviets focusing more on “Soviet fields” like partial and ordinary differential equations, and Americans focusing more on “American fields” like statistics and operations research. Kirk Doran and I looked at the publication history of Americans who worked in “Soviet fields” and Americans who worked in “American fields” and found that the Americans who most competed with the Soviets experienced a decline in productivity after the supply shock. Here’s the graph from our 2012 QJE paper:
So what does the Soviet context teach us? Although a supply shock of several hundred exceptional mathematicians can generate spillovers, the law of diminishing returns might also kick in (as the busy job-seeking travelers in the New York Times story had suggested). And this additional competition for resources can offset and reverse whatever benefits the spillovers might have produced in the first place.
My take from all this is very simple: At least in the experimental context, the evidence that high-skill immigrants produce beneficial spillovers is most convincing when the immigrants that make up the supply shock are really, really high-skill; when the number of such exceptional immigrants is sufficiently small relative to the market; and when those immigrants directly interact with the potential recipients of the spillovers.
In a future post I will discuss what (if anything) we learn from the non-experimental evidence.
And in case the reader wants more information about these supply shocks: Here’s a very nice AER paper by Waldinger (with Petra Moeser and Alessandra Voena) looking at what happened after the fired Jewish professors reached the United States. And here’s a 2015 JOLE paper by Doran and I looking at the research agenda of American mathematicians after the Soviet supply shock. And here’s a just-published RESTAT paper looking at the productivity of the Soviets left behind.
