The Age of Mass Migration and the 1920 Border Closure with Leah Boustan

Episode Summary

Executive Summary: Leah Boustan discusses research on the 1920s U.S. immigration quota laws, which sharply reduced European immigration and created a natural experiment for studying labor-market adjustment. The paper finds that local wage effects were not driven by simple labor-supply changes: cities largely replaced lost immigrants with other workers, while rural areas substituted toward mechanization and capital. The episode highlights how firms and workers adapt when immigration is restricted.

Main Topics: The 1920s immigration quota as a natural experiment (Priority: 5/5): The paper uses the U.S. shift from near-open European immigration to strict nationality-based quotas as the largest immigration policy change in U.S. history, enabling clean before-and-after analysis. Why some local labor markets were more exposed than others (Priority: 5/5): Because quotas targeted Southern and Eastern European sending countries, U.S. places with immigrant communities from those regions were hit harder than places with mostly German, British, Irish, or Canadian immigrants. Unexpected earnings declines among U.S.-born workers (Priority: 5/5): Prior work found that U.S.-born workers in more exposed places saw earnings fall after immigration was cut off, a result that motivated the paper’s focus on labor-market adjustment mechanisms. Urban adjustment through worker substitution (Priority: 5/5): In cities, lost immigrant inflows were offset almost one-for-one by inflows of other workers, including U.S.-born movers and some migrants from Mexico and Canada, which helps explain why wages did not rise. Rural adjustment through capital deepening and mechanization (Priority: 4/5): In rural areas, reduced immigrant labor was not replaced by U.S.-born labor; instead, farms appear to have shifted toward mechanization, including tractors and more capital-intensive crops. Mexico as an overlooked source of labor supply (Priority: 4/5): Mexico remained outside the quota system, so some of the lost labor inflow was replaced by Mexican workers, especially in rural areas; policymakers had not fully anticipated this substitution channel. Broader lessons from the age of mass migration (Priority: 4/5): Boustan places the study in the larger historical period when U.S. immigration was very high, linking it to work on immigrant selection, assimilation, and intergenerational mobility.

Key Arguments: The 1920s quota system is ideal for causal analysis because it was sudden, large, and targeted at specific sending countries rather than being an across-the-board restriction. Local exposure depended on immigrant composition: areas with many Southern/Eastern Europeans were hit harder than areas with mostly Western/Northern Europeans. The puzzle motivating the paper is that earnings of U.S.-born workers fell in exposed areas, contrary to basic labor-supply intuition. In urban areas, there was effectively no net labor-supply reduction because other workers moved in to replace the lost immigrants. The replacement workers were higher skilled on average, changing the composition of labor even where the number of workers stayed roughly the same. In rural areas, employers substituted away from immigrant labor toward capital, especially mechanization and tractor use, rather than toward U.S.-born workers. Mexican migration mattered substantially because Mexico was not quota-restricted in the 1920s, and roughly 30% of replacement inflows came from Mexico. The policy’s welfare effects are ambiguous for Americans already in the country because some groups gain and others lose, even if migrants themselves are clearly harmed by exclusion. The broader lesson is that restricting immigration does not simply ‘free up’ jobs for natives; firms and farms can respond through migration from elsewhere, mechanization, or other substitution channels. Historical debates about immigration closely mirror today’s debates, so the 1920s are a useful warning about unintended consequences.

Data Points: Annual immigrant inflow before quotas: around 1 million per year - U.S. immigration arrivals from Europe before the early 1920s quotas U.S. population at the time: a little under 100 million - Used to contextualize how large pre-quota immigration flows were Annual immigrant inflow after quotas: 150,000 per year - Quota system reduced immigration over the course of the 1920s Reduction in immigration flow: almost an order of magnitude - Comparison of pre-quota and post-quota immigration levels Share foreign-born in the U.S. today: 15% - Presented as the highest since 1910 Share foreign-born during the age of mass migration: about 15% for over 50 years - Approximate foreign-born share from around 1860 to 1920 Total entrants to the U.S. during the age of mass migration: 35 million - Broader research program on migration to the U.S. Estimated lifetime earnings gain from migrating historically: 70–75% increase - Boustan’s work on the economic return to migration in the mass migration era Estimated lifetime earnings gain from migrating today: 3–400% increase - Used as comparison to historical migration returns Mexican entrants in the 1920s: around 500,000 - Mexico remained outside the quota system during the 1920s Mexican deportations in the 1930s: around 400,000 - Referenced as a major reason Mexican settlement patterns were disrupted Bracero program entrants: around half a million entrants per year - Guest-worker program from 1942 to 1965, mostly agricultural labor

Pivotal Quotes: "It is the largest change in immigration policy in U.S. history." — Leah Boustan: Explaining why the 1920s quota law was chosen for the study "We see that the local areas that initially had larger clusters from Southern and Eastern Europe that ends up having greater declines in their immigrant inflow in the 1920s." — Leah Boustan: Describing the empirical exposure measure used in the paper "Never discount the ingenuity of employers and firms in finding other sources of factors." — Leah Boustan: Summarizing the main policy lesson about substitution away from immigrant labor

Implications: Immigration restrictions can trigger substitution rather than simple native wage gains: workers may move, firms may mechanize, and labor may come from elsewhere. Policymakers should expect adaptation and unintended consequences, not just a direct one-for-one job transfer to natives.