What Are the Causal Effects of Low-Skill Immigration Restrictions on U.S. Firms and Workers?

Clemens and Lewis (2025) ask whether restricting low-skill immigration through the H-2B visa lottery affects U.S. firms and domestic workers. They use survey data on 472 firms that either won or lost a randomized visa lottery in 2021–2022, combined with administrative petition data. They find that firms allowed to hire more H-2B workers increase revenue (elasticity about 0.20–0.22), investment (about 1.5–2.1), and profits (about 0.15). The elasticity of substitution between immigrant and U.S. workers is low (0.8–2.2), and effects on U.S. employment are generally zero or positive, especially in rural areas.

Why This Article Was Selected for The Policy Scientist

This article addresses a central policy question with broad economic relevance: how restrictions on low-skill immigration shape firm behavior, local labor markets, and overall economic activity. The topic is especially timely given ongoing debates over labor shortages, supply chain resilience, and immigration policy reform in the United States and other advanced economies. Clemens and Lewis have contributed extensively to this literature, and this study advances prior work by directly examining firm-level responses rather than relying on aggregate correlations. The randomized visa lottery provides unusually strong causal identification, improving on much of the existing literature. The dataset is carefully constructed but limited in size, which may constrain external validity. Generalizability is strongest for similar labor markets with binding quotas.

Full Citation and Link to Article

Clemens, M. A., & Lewis, E. G. (2026). The effect of low-skill immigration restrictions on U.S. firms and workers: Evidence from a randomized lottery. American Economic Journal: Applied Economics. https://www.aeaweb.org/articles?id=10.1257/app.20250049 

Central Research Question

This study investigates whether restricting firms’ access to low-skill immigrant labor—specifically through the H-2B visa cap and its randomized allocation—affects firm-level economic outcomes and the employment of U.S. workers. The central question is whether low-skill immigrant labor acts as a substitute for, or a complement to, domestic labor, and how these relationships translate into changes in firm production, investment, profitability, and hiring. By exploiting randomized variation in visa access, the authors aim to identify the causal impact of immigration restrictions on both firms and native workers, addressing a long-standing empirical and policy debate with unusually transparent identification.

Previous Literature

The study engages a large and contested literature on the labor market effects of immigration, particularly the degree of substitutability between immigrant and native workers. Foundational contributions, such as Card (1990) and Borjas (2003), rely on geographic or skill-cell variation to estimate impacts, often producing divergent conclusions due to differing assumptions and identification strategies. More recent work, including Ottaviano and Peri (2012), emphasizes complementarities, especially among high-skill workers, while the evidence for low-skill immigration remains mixed. A key limitation in this literature is reliance on shift-share instrumental variables, which may suffer from endogeneity if historical settlement patterns correlate with unobserved economic trends. This paper contributes by using policy-driven randomization at the firm level, avoiding many of these concerns. It also complements a growing body of firm-level research, which has primarily focused on high-skill immigration and innovation, by examining low-skill labor demand responses within firms.

Data

The authors construct a novel dataset combining administrative records from the H-2B visa program with original survey data collected from U.S. firms. The sample includes 472 firms that participated in the H-2B visa lottery in 2021 and 2022, spanning industries such as landscaping, forestry, hospitality, seafood processing, and construction. These firms represent a meaningful share of the universe of H-2B petitioners and are broadly representative in terms of lottery outcomes and industry composition. The survey captures firm-level outcomes including revenue, employment (both immigrant and U.S. workers), investment expenditures, and profit changes. A key strength of the dataset is its linkage to a natural experiment: firms are exposed to exogenous variation in their ability to hire immigrant workers due to randomized processing order in the visa allocation system. However, the sample size is modest, and outcomes are observed over a relatively short time horizon, limiting the ability to assess long-run effects or firm survival dynamics.

Methods

The empirical strategy centers on causal inference using a naturally occurring randomized lottery embedded in the H-2B visa allocation process. Firms are effectively assigned different probabilities of hiring immigrant workers based on randomized processing order, which determines whether they secure visas before the quota is exhausted. The authors exploit this variation using both reduced-form (intent-to-treat) and instrumental variables (two-stage least squares) approaches. The primary instrument is an indicator for whether a firm “wins” the lottery (i.e., receives early processing), with an alternative continuous instrument based on expected visa success rates by lottery category. These methods allow estimation of the causal effect of immigrant labor on firm outcomes for compliers—firms whose hiring decisions are affected by the lottery. The use of pre-analysis plans strengthens credibility by reducing concerns about specification searching. Compared to much of the prior literature, which relies on observational variation and multivariate regression, this design provides stronger internal validity and more policy-relevant estimates.

Findings/Size Effects

The results indicate that access to low-skill immigrant labor has substantial positive effects on firm performance. Firms that gain additional H-2B workers experience increases in revenue with an elasticity of approximately 0.20–0.22, implying that a 10% increase in immigrant labor raises revenue by about 2%. Investment effects are notably larger, with elasticities ranging from 1.5 to 2.1, suggesting that firms significantly expand capital expenditures when labor constraints are relaxed. Profitability also increases, with an estimated elasticity of approximately 0.15. These findings indicate that immigration restrictions impose meaningful constraints on firm growth and economic activity.

With respect to U.S. workers, the results do not support the view that immigrant labor displaces domestic labor. The estimated elasticity of substitution between H-2B and U.S. workers is low, between 0.8 and 2.2, indicating limited substitutability. Consequently, the overall effect on U.S. employment is either neutral or modestly positive. In some specifications, the elasticity of U.S. employment with respect to immigrant labor is positive but not statistically significant. However, in rural subsamples, the effects are larger and statistically meaningful, with an elasticity of approximately 0.61, suggesting that immigrant labor may facilitate complementary job creation in less densely populated labor markets. These patterns are consistent with a model in which immigrant labor enables firms to expand production, thereby increasing demand for complementary domestic labor inputs.

The reduced-form results reinforce these conclusions, showing that firms losing the lottery contract operations, reduce investment, and may reduce hiring. The magnitude and consistency of the estimates across specifications, along with robustness checks addressing potential biases, strengthen the credibility of the findings. Importantly, the study also provides evidence that alternative labor sources, including informal or black-market labor, are not sufficient substitutes for legal immigrant labor at the margin.

Conclusion

This study provides clear causal evidence that restricting low-skill immigration reduces firm growth, investment, and, in some contexts, domestic employment. By leveraging randomized variation in visa allocation, the authors overcome key identification challenges that have limited prior research. The findings contribute to a more precise understanding of how labor market policies affect firm behavior and economic outcomes, particularly in sectors reliant on low-skill labor. While the data are limited to a specific institutional setting and short-run outcomes, the results are likely informative for other contexts where binding labor supply constraints affect production decisions. The study’s methodological approach represents a significant advance in the literature, demonstrating the value of quasi-experimental designs in resolving longstanding empirical debates about immigration and labor markets.