Canada Poaches Talent that Competes with American Workers

As I wrote earlier, Canada is poaching talent that should be American! A new working paper by Agostina Brinatti (on the job market) and Xing Guo studies this in detail. The paper first documents that Canada attracted more immigrants when America shut the door but then it traces the consequences through the growth in Canadian firms and on through international trade to American firms and workers. Computer scientists were especially affected by the US policy and the bottom line is that when the US started to deny many more H-1B visas computer scientists went to Canada, increasing the creation of Canadian businesses, jobs and exports. American computer scientists gained because they had less competition but they gained less than the direct effect because they faced more Canadian imports. Moreover, other lower-skilled Americans were harmed because they had fewer higher-skilled workers to work with.

By the end of 2018, there was a decrease of 140,000 H-1B approvals (relative to trend) and an unprecedented spike in H-1B denial rates. Denial rates increased from about 6% in 2016 to 16% in 2018….Immediately following this policy change, Canada experienced a surge in the number of skilled immigrant admissions, equivalent to 76,000 additional admissions in the period between 2018 and 2019. This inflow represents 3.5% of the stock of college-educated immigrants in Canada, or about 2% of all workers in the high-skilled service sector.

…Our event-study estimates imply that a 10 percentage point increase in H-1B denial rates increases Canadian applications by 30%. A back of-the-envelope calculation suggests that for every four forgone H-1B visas, there is an associated increase of one Canadian application.

[the inflow was especially large in computer science]….This inflow decreases the welfare of Canadian computer scientists because they are relatively close substitutes to the incoming immigrants. However, the inflow increases the welfare of workers in other occupations because Canadian sectors expand, especially high-skilled service sectors. For instance, in these sectors, the welfare of computer scientists decreases by 2.9% and that of lower-skilled workers increases by 0.9% approximately.

In the U.S., immigrant labor decreases by 1.6% and is particularly pronounced among computer scientists. As a result, we find that the drop in U.S. approval rates benefits primarily American computer scientists but tend to harm American workers employed in other occupations, especially if their sector contract. For instance, computer scientists in high-skilled service sectors experience a 0.7% welfare increase, while lower-skilled workers experience a 0.3% welfare decrease. These effects on American workers include both direct and indirect effects.

Addendum: Tyler reminds us that in the long-run, immigrants who get Canadian citizenship may immigrate to America! The basic point that Canada and the US are in a more or less free trade area is well taken and so we can thank Canada for making US immigration policy less harmful than it might otherwise be. Still, I’d rather see fewer artificial barriers distorting the allocation of talent.