Here we study the economic effects of a large-scale experiment in the United States: the natural, nationwide, firm-level randomization of restrictions on the employment of immigrants for low-skilled jobs. The United States has a primary work visa for low-skilled labor in the nonfarm economy: the H-2B visa. Access to this visa for US employers is limited by a quota and awarded in part via a random lottery conducted by the federal government. This exogenous variation in immigrant employment restrictions allows for unusually transparent and policy-relevant estimates of how American firms and workers are adjusting. After publicly engaging in our hypothesis testing and predicting treatment effects with a pre-analysis design, we collected data from 2021 H-2B visa lottery winners and losers in a new survey of companies. This allows predefined tests of basic theoretical predictions about the magnitude and heterogeneity of the effect of restrictions on the immigration of low-skilled people. It also allows estimation of the “combined” immigrant-native elasticity of substitution at the firm level (Hicks 1936).
We find that exogenous permission to employ immigrants for low-skilled labor causes the marginal firm to increase its output. In other words, the exogenous restrictions on the employment of the number of immigrants maximizing the profit for low-skilled labor lead to the contraction of the marginal firm. These restrictions lead to a large and statistically significant drop in income and investment. The restrictions lead to no increase, or decrease, in the employment of low-skilled indigenous workers and the rate of profit. Losing the lottery reduces the employment of low-skilled immigrants in companies by 56%. This decline causes businesses to contract, shrinking operations with an elasticity of +0.164 for revenue and +1.03 for investment (statistically distinguishable from zero at conventional levels), and with an elasticity of +0.102 for low-skilled American employment, and +0.100 for the rate of profit (statistically indistinguishable from zero at conventional levels).
This is by Michael A. Clemens & Ethan G. Lewis,”The Effect of Low-Skilled Immigration Restrictions on American Firms and Workers: Evidence from a Random Lottery“, NBER Working Paper No. 30589, October 22, 2022.
The methodology is quite clever. Because lucky employers are chosen by lottery, there is no selection bias. This means that Clemens and Lewis can examine job changes for employers who won the lottery and job changes for employers who lost.
The authors put it more succinctly in their abstract:
Firms exogenously allowed to employ more immigrants significantly increase output (elasticity +0.16) with no decrease or increase in US employment (elasticity +0.10, statistically imprecise) in multiple subsamples pre-recorded. The results imply very low substitutability of domestic labor with foreign labor in policy-relevant occupations.
In short, low-skilled Americans are not losing their jobs to low-skilled immigrants.
HT2 Tyler Cowen.