The H-1B visa program, which allows a limited number of high-skilled workers to work in the United States temporarily, is controversial because some claim that it lowers wages for high-skilled workers. However, a new paper by Peri, Shih, and Sparber of UC Davis and Colgate University shows just the opposite—that additional H-1B visa recipients raise wages in cities where they come to work.
The authors’ interest in immigration is a side effect of their interest in a more general labor market question: they use fluctuations in high-skill immigration due to the H-1B visa program as an instrument to examine whether the supply of STEM workers affects productivity growth. In essence the question is, what happens when you increase the amount of high-skill STEM workers: do wages fall as one would assume in a standard supply/demand framework? Or do they increase because of the effect that the high-skill workers have on productivity, demand for innovative workers, and economic growth?
To answer this question you can’t simply look at the amount of STEM workers in a city and average wages for those workers, because you can’t tell which way causality is going. The amount of workers may be driving wages up or down; or wages may be attracting or repelling STEM workers. This indeterminate causality is a fundamental difficulty in economics and lots of blood, sweat, and tears have gone into getting around it. Peri, Shih, and Sparber use an instrumental variable technique that takes regulatory changes in the amount of H-1B visas and assumes that the resulting shifts in the supply of STEM workers are independent of any demand-side influence. This isolates the causal effect of changes in supply from changes in demand.
They find that a 1 percent point rise in foreign STEM workers relative to the total workforce caused a 7-8 percentage point rise in the wage of college-educated workers in that city. Moreover, in support of the “rising tide lifts all boat” hypothesis, they find that wages of non-college educated workers rose 3-4 percentage points. Housing costs also rose, but only by about half as much as the rise in wages.
In other words, if 1,000 foreign STEM workers on H-1B visas moved into a city of 1 million people, the average wages of a college graduate making $50,000 per year would increase by $350-400. Or the average wages of a non-college graduate starting with the same amount would increase by $175-200. If 10,000 stem workers moved into the city, wages would theoretically increase by $3,500-4000 for college educated workers.
This paper is good news for advocates of increasing high skill immigration and the H-1B visa program because it challenges the idea that additional high-skill workers push wages down for other workers (high or low skilled). In U.S. cities where more high-skill immigration occurred it caused the exact opposite to happen: wages rose relative to cities with fewer high-skill immigrants. Thus the market for STEM workers doesn’t appear to follow the laws of supply and demand. Instead, increases in supply raise productivity and therefore increase STEM wages (with no discernable effect on employment).