Daron Acemoglu@DAcemogluMIT
This is a thread about three points that may have been missed in the recent H1B debates.
First point – endogenous technology: One argument I haven’t seen is that sufficiently large flows of skilled immigrants may affect the direction of technology. A general theorem of directed technology is that when the supply of skilled labor increases, technology becomes endogenously more biased towards skilled workers (see, for example, academic.oup.com/restud/article… aeaweb.org/articles?id=10…).
So one question is whether H1B program has contributed to US technology becoming more and more focused on high-education workers and leaving lower-education workers behind. Possibly not, since there has been a lot of unskilled immigration as well, and other factors may have been more important in the direction of technology (and the evidence does not show a total crowd out overall of innovation by natives, see journals.uchicago.edu/doi/full/10.10…).
Nevertheless, the impact of high-skill immigration on the direction of technology is worth exploring in the future. What this argument suggests more generally is that if the US is going to rely on H1B workers, policymakers should pay consider other adjustments so that US technology and corporate strategies do not leave low-education workers completely behind.
Second point – political economy of education: The following argument at first appears watertight: the US has a need for skilled STEM workers, for example in the tech industry, and is not currently able to produce this supply itself. Hence, it seems innocuous and generally beneficial to make up this shortage via the H1B program.
What this argument ignores is the following, however: if it weren’t for the H1B program, the pressure on US institutions to improve the quality of secondary education and the supply of STEM workers would have been much stronger. Put differently, the current system may have made the economic (and political and intellectual) elites care less about the failure of the US education system. (The argument that the elites would like the education system to produce workers that their businesses need goes back to Sam Bowles and Herb Gintis’s classic Schooling in Capitalist America: amazon.com/Schooling-Capi…).
Again, this is just a possibility, and I don’t know of any evidence suggesting that the response from the education system would be large enough to make a difference. Nevertheless, just like in the previous case, this argument suggests that we may want to pay attention to making sure that the H1B program does not contribute to the elites becoming more and more indifferent to the plight of US schools and low-income communities.
Third point – how to make high-skill immigration win-win for source and destination countries: One aspect of the argument in favor of high-skill immigration from the developing world to the United States is that they can be more productive here because they have access to better technology, better institutions and better teams of other skilled workers to work together. (Put differently, better institutions and better technologies are complementary to skills). But if this is all that’s happening, such migration would still be bad for source countries, which will be afflicted by “brain drain”.
The way to avoid this negative would be via the flow of technical expertise and tacit knowledge related to innovation and technology back to the source countries. (There is some evidence suggesting that this happens automatically via ethnic networks; see for example the very nice paper by William Kerr on this: direct.mit.edu/rest/article/9…).
But this raises another problem. If information about new innovations flows too quickly back to the source country, then this could negatively impact the US comparative advantage in the global economy, which is rooted in new innovations. Hence, for this win-win scenario to become a reality, one would need that there is a robust flow of information back to the source country, but not so fast as to be ultimately damaging to the innovating country (in this case the US). In economics, this is captured in models of technology-product cycles, for example, Vernon’s classic model: academic.oup.com/qje/article-ab… Krugman's reformulation thereof: journals.uchicago.edu/doi/abs/10.108…).
Such a win-win scenario would also not take place if nearly all skilled, innovative workers leave source country, since there would not be a critical mass of workers back at home to benefit from these information flows.
Overall, a win-win scenario would require sufficient numbers of skilled workers to remaining in the source savecountry and adequate international protection of property rights so that innovators enjoy the productivity advantage created by their innovations at least for a while.
