The Economics of Immigration: What Research Actually Shows

What Four Decades of Research Has Found

The Mariel Boatlift of 1980, when approximately 125,000 Cuban immigrants arrived in Miami over five months, became one of the most studied natural experiments in labor economics. David Card's 1990 analysis found that the sudden increase in Miami's labor supply — about 7 percent — had virtually no effect on wages or unemployment for native workers. The finding ran counter to simple supply-and-demand intuition and launched an ongoing empirical debate that has produced more than a thousand peer-reviewed studies. After four decades of research, economists have reached broad consensus on some questions and continue to dispute others — a situation that differs substantially from how immigration is discussed in political arenas.

The Labor Market: Complement or Substitute?

The central empirical question is whether immigrant workers compete with or complement native workers. The answer depends heavily on skill levels and the structure of the local labor market.

George Borjas of Harvard has argued consistently that immigration reduces wages for workers who compete in the same skill tier. His 2017 re-analysis of the Mariel Boatlift data, using a narrower reference group of high-school-dropout workers, found substantial wage suppression. Card and others disputed his methodology and found opposing results using different samples. This methodological disagreement has not been fully resolved, though the mainstream of the profession — including the National Academies of Sciences in its comprehensive 2016 report — concluded that wage effects on native workers are small or positive in the long run and concentrated among previous immigrants competing in the same skill categories.

Fiscal Impacts: Costs and Contributions Over Time

The fiscal impact of immigration depends critically on the time horizon and the level of government analyzed. The 2016 National Academies report — the most comprehensive U.S. analysis to date — found that first-generation immigrants typically cost state and local governments more than they contribute, primarily through public education costs for their children. By the second generation, however, the picture reverses: the U.S.-born children of immigrants are among the strongest net fiscal contributors of any demographic group, reflecting their higher educational attainment and earnings relative to third-plus generation Americans.

• The present value of a typical immigrant's net fiscal impact over a 75-year horizon is estimated to be positive $173,000 in the National Academies analysis — though this varies enormously by education level, age of arrival, and whether extended family members are included in the calculation.
• Immigrants with college degrees contribute substantially more in taxes and cost less in social services than immigrants without high school diplomas — a finding that underpins policy debates about merit-based versus family-based immigration selection systems.

Entrepreneurship and Innovation

One of the strongest empirical findings in immigration economics is the outsized contribution of immigrants to entrepreneurship and innovation in the United States. A 2017 National Bureau of Economic Research study by Sari Kerr and William Kerr found that immigrants founded or co-founded more than 40 percent of Fortune 500 companies, including Amazon, Google, Yahoo, eBay, and Yahoo. Immigrants or their children account for about 43 percent of Fortune 500 founders.

Patent data shows similar patterns. Raj Chetty and colleagues have found that children of immigrants file patents at significantly higher rates than children of native-born Americans with equivalent family incomes. This innovation premium is particularly pronounced among immigrants from China, India, and Taiwan — groups with high educational attainment and concentration in technology fields.

Effects on Housing and Local Services

Immigration's effects on housing markets are an emerging area of research. A 2019 study by Albert Saiz at MIT found that a 1 percent increase in population through immigration increases housing rents by approximately 1 percent in the long run, with the effect concentrated in cities with inelastic housing supply (dense, land-constrained cities with restrictive zoning). In cities where construction can respond, the rental effect is smaller because housing supply expands. This means that housing affordability impacts from immigration are largely mediated by local land-use policy — a finding that shifts some of the policy response toward zoning reform rather than immigration restriction.

What the Research Does Not Settle

• The distributional question — whether the gains to some workers and employers outweigh losses to competing workers — depends on value judgments about redistribution, not just empirical measurement.
• The political economy of immigration can affect the quality of democratic institutions, social trust, and policy responsiveness in ways that economic models do not easily capture.
• Climate-induced migration is projected to involve far larger population movements than historical patterns — potentially 200 million people by 2050 according to World Bank estimates — meaning that research based on 20th-century migration flows may not accurately predict the effects of 21st-century immigration at scale.

The economic evidence does not support the most extreme claims on either side of the immigration debate. Large-scale immigration does not systematically destroy native wages or employment, but it does create real distributional effects, particularly for workers in directly competing skill categories. What research consistently shows is that over a generation, the children of immigrants tend to integrate economically and often outperform their native-born peers — a finding that shapes virtually every serious economic policy framework on the subject.