When COVID first started spreading more widely in the US, I began worrying that this would lead to an upsurge of anti-immigrant sentiment. I worried that people would draw the wrong lesson from this experience and return to the isolationism of the 1920s, closing our borders on a more permanent basis to both people and goods. This would slow economic growth and lead to a poorer nation. It seems particularly ironic that just now Americans are becoming the unwelcome foreign visitors abroad, particularly from my home states of Texas and California.

Nowrasteh and Forrester at CATO discuss some papers by Giuliano and Tabellini on the question of if increased immigration moves the median voter to the left. They also add a few suggestive regressions of their own. Their summary is interesting and nuanced. First, they find that closing the borders to immigrants in the 1920s encouraged much greater government spending (as a percent of GDP) while allowing more immigrants in the 1960s has slowed the growth of government spending. This effect seems to work both ways: American voters are more willing to vote for welfare benefits, etc, when there are fewer immigrants getting them, and the larger the welfare state is, the more concerned voters are about allowing immigrants into the country. So it may not be so much that adding immigrants from more left-leaning countries shifts the median voter as much as it moves native voters further to the right? (See also Rosenthal and Eibner 2005, who also conclude that “a voter of a given income is less eager to redistribute given that redistribution has to be shared with the non-citizen poor.”)

This suggests an interesting line of argument, that these feedback effects can work in the opposite direction as well. I imagine a friend who is very concerned about the size of government and also would prefer to have fewer immigrants. To that friend, I would suggest that allowing more immigrants can help slow or even reverse the growth of government and the welfare state. Using their aversion to one issue could potentially reduce their aversion to the other. <epistemic status: highly speculative>