The tumultuous Mariel refugee influx of 1980 is back in the news — this time at the core of a roiling debate about whether immigrants hurt less-educated native-born workers.
The heated arguments focus on the new work of a Cuban-born Harvard professor, George Borjas, who concludes that Mariel caused a drastic drop in pay among native-born Miami high school dropouts — the majority of whom were black.
His work has sparked an uproar among fellow economists and many policymakers because, for years, the academic world has believed that the sudden arrival of 125,000 Cubans via the boatlift had no effect on the wages of those already living in Miami — the conclusion of a widely respected 1990 research paper by David Card, then a Princeton professor.
Card’s study has long been cited worldwide by policymakers and the media as a major reason why people shouldn’t be afraid of immigrants — including Mexicans entering the United States and Syrians arriving in Europe. More recently, Attorney General Jeff Sessions has pointed to Borjas’ studies as a reason why immigrants should be kept out and President Donald Trump cited a Borjas conclusion in his acceptance speech at the 2016 Republican convention.
Borjas has become so important that last year Politico dubbed him the 17th most influential thinker “transforming American politics.”
The Wall Street Journal, the Economist and many other publications have recently revisited the Mariel issue because of Borjas’ studies. The New Yorker praised his “dogged research” for putting into doubt the previous consensus on immigration.
“I have been quite surprised by the never-ending brouhaha over this one paper,” Borjas says. He believes many of his fellow academics are “in incredible denial” about basic economic principles. “The narrative of immigration is that nobody gets hurt. I’d love that to be true, but in any economic policy there are winners and losers.”
Giovanni Peri, a California economist who’s a leading Borjas critic, disagrees. He calls some of Borjas’ conclusions “peculiar” and others “quite misleading.” Peri says his recent research re-affirms Card’s 1990 no-harm conclusion.
For decades, Mariel “has held a special place in the minds of the American economists as … an unique example of sudden, unexpected and large refugee inflow,” as Peri and Vasil Yasenov phrased it in a 2016 paper. That’s because too often waves of immigration are blurred by many factors, including uncertain time frames and new arrivals often flocking to places where the most jobs are available, meaning their effects on native-born Americans are minimized.
With Mariel, time and location were limited. In April 1980, Fidel Castro announced that virtually anyone could leave through the port of Mariel. Boats rushed to Cuba from Key West to pick up friends and relatives. In the first six weeks, according to the Coast Guard, about 100,000 Cubans made the trip — most of them going to Miami, where the workforce suddenly swelled by about 60,000. About 60 percent of the new arrivals had less than a high school education.
Basic supply-and-demand theory states that if you increase the number of people who want to be janitors, for example, the wages of janitors tend to go down.
Card upended this concept when he examined U.S. Census and federal labor statistics, then compared those numbers with four other cities to see what was happening elsewhere in the American economy. He concluded: “The Mariel influx appears to have had virtually no effect on the wages or unemployment rates of less-skilled workers.”
Card’s work was mostly unchallenged until two years ago when Borjas decided to write an economics book for a general audience combining his theories with his own experience as a Cuban immigrant.
Growing up in Marianao, he had seen his parents’ garment factory confiscated by the communists. His father died in 1961. A year later, when he was 12, he and his mother moved to Miami for about two years, then to New Jersey.
When Borjas showed a first draft of “We Wanted Workers” to a few people, “a family member told me, this is too difficult for a general audience.”
Deciding he needed a concrete example, Borjas re-visited the boatlift, focusing on those native-born Americans most likely to face competition from the new arrivals — male non-Hispanic high school dropouts aged 25-59. More than half of that group in Miami were black.
The wages for this group, he found, “dropped dramatically, by 10 to 30 percent.” Using four other cities that he thought made for more accurate comparisons than the ones Card used, he concluded Mariel also “substantially worsened the relative economic status of the typical African American in Miami relative to his counterpart in the [other] cities.”
When he published a first draft of his research in late 2015, other academics quickly slammed his work. The biggest attack came from Peri and Yasenov at the University of California at Davis.
They criticized his choice of comparison cities and his use of one set of labor statistics versus another. In particular, they decried his “very drastic choices” to narrow his study to a high school dropout group that excluded Hispanics and women — two-thirds of the dropout population in Miami.
What this meant, the Californians reported, was that, from a pool of 120,000 total high school dropouts in Miami, Borjas was using labor surveys that produced an extremely small sample size — 17 to 24 individuals per year.
“That’s only one-sixth of the total sample,” Peri said in a telephone interview. Such a small sample meant a huge margin of error, he said, making Borjas’ conclusions untrustworthy.
What’s more, Peri said, looking back at Miami wage levels to 1972, he found that there was a “very long, downward trend” in wages among less-educated native Miami workers that continued through Mariel and the 1980s.
Peri believes it made sense that Mariel didn’t harm native workers in Miami: “There was already a large Cuban population there” and it’s quite possible Cuban-American entrepreneurs expanded investments to take advantage of the new workers. What’s more, 60,000 new immigrants needed much more services — from supermarket clerks to auto dealers — that helped boost the job market.
Borjas rejected all criticisms. “I admitted it from the very first draft” having a small sample size, he said in a telephone interview. He compensated by taking three-year rolling averages to expand his sample — and the results were so striking, he said, that they couldn’t have been the result of statistical error.
The California economists hadn’t been dealing with a huge sample themselves — about 140 people annually to represent 120,000 people. “That’s the nature of the beast,” Borjas said because of the limits of the two surveys of the Bureau of Labor Statistics that all economists use.
He said he didn’t measure Hispanic dropouts because in Miami 1980 most of them were immigrants themselves and not the native-born workers he sought to examine. He didn’t include women because in 1980 their situation was in flux, their workforce participation increasing and their wages rising.
Borjas says the larger the sample, the more likely a finding that immigrants do no harm. In the largest sense, “the influx of immigrants can potentially be a net good for the nation, increasing the total wealth of the population,” he wrote for Politico. It’s when he looks at smaller segments that he sees certain natives being harmed.
Who’s right? Last year, the National Academies of Sciences, Engineering, and Medicine issued a 642-page report on immigration. It acknowledged Borjas’ main point — that native groups with similar skills to new immigrants “may experience a wage reduction,” particularly those with “low-education, low-skills.”
But the report noted “one sobering reality: Wage and employment impacts created by flows of foreign-born workers into labor markets are complex and difficult to measure” because they’re hard to isolate from all other factors, including changes in technology.
“That said, quantitatively significant labor supply shocks do occur, especially in localized markets, such as that which accompanied the 1980 Mariel boatlift in Miami,” the report said. “Even then, the wage impacts may be difficult to detect.”
Though economists continue to embrace Mariel as a valuable “natural experiment,” all qualify their conclusions with “all other things being equal.”
Borjas and critics acknowledge that many other factors were playing out in Miami at the time of Mariel, including a major Haitian influx and a drug trade that heated the economy.
Most striking was the May 1980 riots that left 18 dead and $100 million in property damage, as African Americans protested the acquittal of four police officers accused of murdering a black motorcyclist. Businesses in black areas declined, both because of the damage and non-blacks fearing to go into riot areas — and some speculate that perhaps these factors caused blacks to lose jobs or settle for lower wages.
“A lot was going on,” Borjas acknowledges, It’s “a really big question” whether the Mariel experience “really tells you what’s happening elsewhere. … Miami was a really peculiar place. If there was a Mariel situation tomorrow in Peoria, would that be the same? I don’t know if it [Mariel] tells you that much, to be honest with you.”
With this, his critic, Peri, agrees: If a Mariel situation happened elsewhere, “yes, that obviously would create a different dynamic.”