Skip to Content

New Evidence on Immigrants and Jobs

18 Jan

With millions fleeing Syria and other war zones, the U.S has committed to increasing the refugees it will admit from 70,000 to 85,000 in 2016 and 100,000 in 2017. A majority of Americans oppose admitting any refugees from Syria, their opposition driven by concern over terrorism. But there is another question that may shed light on the immigration controversy. What would be the economic impact on American workers if the U.S. were to admit tens of thousands of Syrian refugees?

Immigrants come to this country mostly to get jobs when the economy is growing, and immigration, including illegal immigration, is mostly a gradual and economy-driven process. Refugees, however, often arrive suddenly, may settle in areas where the economy is not booming, and can compete legally for jobs. These features let economists study the impact of a purely supply-driven inflow of arrivals on the wages and employment of native workers.

A well-known episode took place after April 20, 1980, when Fidel Castro opened the port of Mariel, enabling anyone to freely leave the island. More than 125,000 Cubans fled to the U.S. until the Mariel boatlift, as it was called, ended in September. More than half of these refugees settled in Miami. Most were low-skill—which meant that the supply of workers without a high-school diploma in the city increased between 12% and 15%.

Economist David Card analyzed how the wages and employment rate of native workers in Miami changed from 1979 (before the inflow) to 1981-82 (after the inflow). His influential study, published in 1990, compared Miami with Atlanta, Houston, Los Angeles and Tampa-St. Petersburg, a control group of cities with similar demographic and labor-market characteristics during the 1970s.

The results were striking: The 1979-1981 wage and employment changes in Miami were not much different than in the other cities. The evidence, he concluded, was that a sudden increase in the supply of low-skill workers had no significant negative effect on native laborers with similar schooling levels.

While there have been some criticisms of Mr. Card’s study over the years, a recent paper by Harvard economist George Borjas rejected the study’s conclusion. According to Mr. Borjas, the wages of one subgroup in Miami—non-Hispanic males older than 25 with no high-school diploma—did decline after the Marielitos arrived.

But his study has serious limitations. Mr. Borjas’s subgroup consisted only of males ages 25-59, and it ignored non-Cuban Hispanics. His sample consisted of only 17 to 24 workers in Miami in each year, a number too small for statistical significance. The four cities Mr. Borjas chose as a control group reflected the similarity in their employment growth rate for the years 1977-79. That’s too short a period to be confident that the market for low-skill workers in these cities was similar to Miami’s.

We have reanalyzed the Mariel episode using the largest and most representative annual sample of high-school dropouts from the May/ORG Current Population Survey. It includes 44 cities among which a recently developed statistical methodology allows the researcher to identify those whose labor markets behaved as closely as possible to Miami’s between 1972 and 1979. We then compared the average wages and employment rates of low-skill workers in Miami with such a control group after 1979.

Our results—released as National Bureau of Economic Research Working Paper No. 21801 on Dec. 15—confirm Mr. Card’s original study. There is no evidence that Miami’s low-skill workers experienced wage or employment decline relative to those in our control group of cities in 1980, 1981 or 1982. We also analyzed different subgroups—males, females, Hispanics and non-Hispanics—and did not find any significant wage effect in Miami after 1979.

This result suggests that the common belief that more immigrant workers depress native workers’ wages or employment is not a good representation of what happens. Earlier research by one of us has shown that native workers do not suffer the negative impact of arriving immigrants because they take different jobs. Moreover, their arrival stimulates productivity and growth in the economy.

Miami’s experience after the Mariel boatlift suggests that an influx of refugees from Syria to the U.S. would have no significant economic impact on American workers.