Published — April 30, 2018

How Trump and Sessions cherry-picked data to blame immigrants for lower wages

In this Feb. 28, 2016, file photo, Republican presidential candidate Donald Trump, right, stands next to Sen. Jeff Sessions, R-Ala., as Sessions speaks during a rally in Madison, Ala. John Bazemore

The president uses disputed research to argue for dramatic legal visa cuts


In April of 2015, the nonpartisan Congressional Research Service responded to a confidential request from the U.S. Senate Judiciary Committee. The memo was short, but it ignited a fiery outcry about the perceived threat of immigration from Jeff Sessions, then a Republican senator from Alabama.

Sessions seized on charts in the CRS memo featuring a six-decade timeline estimating average incomes — mostly flat after 1970 — and another showing a rise in the foreign-born population. Exhorting GOP presidential candidates to take these “forbidden facts” seriously, Sessions called on Republicans to fight to slash immigration — legal or illegal. “It is not caring, but callous, to bring in so many workers that there are not enough jobs for them or those already living here,” Sessions, a Judiciary Committee member, co-wrote in a column published by Roll Call, a congressionally focused news outlet.

Far-right media also sprang into action. The Breitbart website, led by future Trump aide Stephen K. Bannon, blasted out posts about the memo, along with TheBlaze and the Washington Examiner. Mark Levin, who hosts the nation’s fourth-most-consumed talk-radio show, read from a Daily Caller story.

“Wages and share of income for the bottom 90 percent of American wage earners declined over the last 40 years, as the foreign-born population increased dramatically,” Levin read with disgust. “Ask the phony economists who play around with the numbers, who try to persuade you that this is a good thing for the economy. These are facts. These are statistics.”

Only they weren’t facts. They were estimates. And they didn’t add up to Levin’s or others’ hyperbolic assumptions.

What Sessions and eager media failed to disclose was that CRS researchers took pains to warn against drawing any conclusions from those two lines of data. “It bears noting,” the researchers cautioned, “that a causal relationship between two variables … cannot be determined through a simple graphic representation.”

But for Sessions, that caveat didn’t seem to matter. The timelines were a simple, powerful tool to support his view that immigration had been suppressing Americans’ wages for decades.

For others, though, Sessions’ use of the memo betrayed a penchant for casting immigration in the most negative light possible — even though the preponderance of economists who research this subject don’t agree with his claims.

“When measured over a period of more than 10 years, the impact of immigration on the wages of natives overall is very small,” said a 2016 report by the National Academy of Sciences. “To the extent that negative wage effects are found,” the report explains, “prior immigrants … are most likely to experience them, followed by native-born high school dropouts.” The survey also notes that a significant body of research has found that immigration causes no negative wage impact, even among lesser-skilled natives — with some research also finding wage gains.

“There’s a very big “if” if it [immigration] does in fact lower wages,” said Francine Blau, a Cornell University economist chaired the National Academy of Sciences’ 14-person panel.

Theresa Cardinal Brown, director of immigration and cross-border policy at the relatively pro-immigration Bipartisan Policy Center in Washington, D.C., said: “There are so many parts to the economy, so many things that can have an impact on wage differentials.”

“Immigrants,” she said, “are not the sole blame, or cause or cure for anything.”

The debate is not just intellectual. Today, Sessions is the attorney general of the United States, and his boss Donald Trump is president, and they’re both crusading to dramatically slash immigration based in part on the jobs and wages argument. Sessions’ spin shows how years of strategic dissemination of cherry-picked and misleading information has helped set the tone for a polarizing, bruising battle over immigration that is splitting the country. And it illuminates how the work of a small group of researchers can become the fulcrum around which policy changes are developed.

“There’s a very big “if” if it [immigration] does in fact lower wages.”

Francine Blau, Cornell University economist

Then and Now

For Sessions, the definitive word on immigrants’ effect on the economy comes from academia’s most prominent immigration skeptic, controversial Harvard University economist George Borjas.

Borjas, 67, has been researching immigration for 30 years, and is a frequent contributor to newspaper op-ed pages. Borjas’ family fled the Cuban Revolution and was admitted to the United States in 1962. He’s repeatedly raised eyebrows by suggesting that today’s immigrants don’t have what it takes to succeed in today’s economy. “Many look back at the melting pot in 20th-century America and assume that history will repeat itself. That’s probably wishful thinking,” he wrote in a New York Times opinion piece in 2017.

When Sessions was in the Senate, Borjas became his go-to economist. Sessions often cited figures from Borjas’ research in speeches and writings before and after becoming chairman of the Senate Judiciary Subcommittee on Immigration and the National Interest in 2015.

Speaking on the Senate floor in 2014, Sessions cited figures from an analysis that Borjas had posted in 2013 on the website of the Center for Immigration Studies, an advocacy group that pushes for cuts to legal immigration.

“Harvard professor George Borjas, who’s perhaps the premiere student of these economic matters, as he’s worked on them for decades, estimates that high immigration flows from 1980 to 2000, based on data, reduced the wages of lower-skilled American workers by 7.4 percent,” Sessions argued. “Professor Borjas estimates that current immigration rates produce a net loss, current rates, of $402 billion [a year] for American workers who compete with foreign labor.”

In January 2015, as the Trump campaign was revving up, Sessions released a booklet that he called an “Immigration Handbook for the New Republican Majority” — featuring Borjas’ $402 billion-a-year wage loss estimate. “We need make no apology in rejecting an extreme policy of sustained mass immigration, which the public repudiates and which the best economic evidence tells us undermines wage growth and economic mobility,” Sessions wrote.

Sessions also presided over hearings he said established that immigration had become a losing proposition for Americans’ wages. Borjas spoke at a March 2016 hearing. Despite the array of contrary research, Sessions opened a June 2016 hearing by saying: “I think very few dispute that the current large flow of labor into the country is suppressing American wages.”

In a pre-election 2016 piece for Politico, Borjas argued that while immigration has been good for the economy overall, “a ‘responsible nationalist’ policy cannot ignore the reality that immigration has made some natives poorer. A policy that keeps them in mind might tax the agricultural and service companies that benefit so much from low-skilled immigrants, and use the money to compensate low-skilled Americans for their losses.”

On his blog, post-election, Borjas expressed confidence in Sessions’ knowledge about immigration and referred to him as a “Southern gentleman.”

The Department of Justice didn’t respond to multiple Center for Public Integrity requests for comment from Sessions. Borjas declined multiple requests for comment as well. He served on the National Academy of Sciences panel that reviewed fiscal-impact research on immigration.

Cornell University economist Blau said, “I have a great deal of respect for Professor Borjas as a scholar. But these are one set of estimates. They are not the consensus” reached by the NAS panel. For politicians or others “to emphasize them to the exclusion of other very well-done research,” Blau said, “I don’t think is appropriate.”

Another panelist, Jennifer Hunt, an economist at Rutgers University, said that Borjas’ estimate that natives lost $402 billion in wages a year is a “meaningless number” because it’s a rough estimate that relies on a theoretical model. Borjas’ own paper concedes his “exercise is a short-run simulation … before the economy adjusts in any way to the immigrant influx.”

Seeds of a Campaign

But among those who picked up on the larger Sessions-Borjas arguments was candidate Donald Trump. An unsavory portrait of immigrants has invigorated pro-Trump rallies, before and after his election.

Last July, in Youngstown, Ohio, Trump disparaged “chain,” or family immigration as a “low-skill system, just a terrible system where anybody comes in” — an exaggeration. Trump wants Congress to eliminate new Americans’ ability to sponsor parents, siblings and adult children — more than 20 percent of all visas — confining family sponsorship only to spouses and minor children.

“Decades of record immigration have produced lower wages and higher unemployment for our citizens, especially for African-American and Latino workers,” Trump asserted in his 2016 Republican nomination acceptance speech — a speech that Borjas, in that 2016 piece for Politico, claimed was influenced by his work.

This controversial argument that immigration has suppressed wages for natives has helped Trump justify his vows to force Congress to enact what could be the most restrictive legal immigration policies since the 1920s. The posture reflects Sessions’ influence, and that of a former Sessions aide, Stephen Miller, now a senior Trump adviser.

Last August, Miller added salt to wounds Trump had opened in immigrant communities when he said at a press conference that existing legal immigration “has massively de-skilled the migrant flow in to America” and “exacerbated wealth inequality in this country is a pretty significant way.”

On the left, anti-immigration protest at Civic Center Park on May 1, 2006, in Denver, during a Defend Colorado Now rally. On the right, demonstrators shout slogans during a protest against President Donald Trump’s efforts to crack down on immigration Thursday, Feb. 16, 2017, in San Francisco. (Jack Dempsey / Marcio Jose Sanchez / AP)

Opposing Views

Lost in the rhetoric is a sharply contrasting narrative laid out by other respected economists, most prominent among them David Card of the University of California at Berkeley. Card, 62, himself an immigrant from Canada, is less in the limelight than Borjas, but he continues to contribute to debate on immigration, and he’s widely respected by peers.

Card said theories that an increasing labor supply powered by immigrants automatically leads to lower wages are “completely wrong.”

Immigrants today are 17 percent of the U.S. labor force, and the undocumented — who can be easier to exploit with lower wages — are about 5 percent, a proportion that hasn’t grown since 2009. While numbers have increased since 1970, immigrants as a percentage of the U.S. population are slightly less than they were in 1910.

Card, who was a consultant on the National Academy of Sciences report, said in an interview: “To the best of my knowledge, very few economists subscribe to the Borjas line. George is at the extreme end.”

“He’s a complete ideologue on the topic,” Card added. He noted that Borjas assisted former California GOP Gov. Pete Wilson’s 1994 re-election campaign, providing data portraying undocumented immigrants as a fiscal drain.

For close to three decades now, Borjas and Card and several others have been engaged in what for economists might be considered a duel.

Card, while at Princeton University in 1990, was credited with pioneering a “natural experiment” approach to studying immigration. Using Census survey data and economic modeling, he studied the wage impact of the 1980 influx of tens of thousands of Cuban refugees into Miami. The Mariel Boatlift from Cuba increased Miami’s labor pool by about 7 percent.

For Card, the influx was a unique opportunity to test assumptions that a greater supply of labor would push down wages for those most likely to compete with immigrants. But Card found “virtually no effect” on wages rates of less-skilled non-Cuban workers, and no substantial impact on the wages of Cubans already in the area when new refugees arrived.

“The data analysis suggests a remarkably rapid absorption of the Mariel immigrants into the Miami labor force,” Card wrote.

Borjas came up with his own models. In a 2003 paper published with the National Bureau of Economic Research, Borjas advanced a model to measure immigration impact by dividing workers into more distinct groups than other research, based on education level and experience. He found that immigration influxes between 1980 and 2000 reduced wages for native-born high-school dropouts by about 9 percent and about 3 percent for average wages for natives overall.

The battle continued. In a 2012 Journal of European Economic Association commentary, Card argued that Borjas’ 2003 research was flawed because Borjas assumed a “fixed” amount of capital over his 20-year study period. Without allowing for changes in investment, Card said, Borjas’ models resulted in wage drops. To understand why an economy absorbs labor, Card said in an interview, it helps to think about how technology, machinery, infrastructure and other investments — the “stock of capital” —help utilize labor without lowering wages.

Card also took exception to Borjas’ assumption that immigrant and native workers with the same level of education are perfect substitutes in the workplace. Economist Giovanni Peri at the University of California at Davis has advanced the argument that immigration prompts “task specialization” that lifts lower-skilled natives into complementary higher-ranking jobs—at restaurants or construction sites, for example—because of their English fluency, experience and other comparative advantages.

Still, the debate wasn’t over. In 2017, Borjas published a “reappraisal” that he claimed “overturned” Card’s 1990 Mariel findings. He found that wages for native-born high school dropouts in Miami fell by 10 to 30 percent after the refugee influx. Trump aide Miller pointed to Borjas’ study as evidence of wage suppression.

But others said Borjas’ dropout sample was so narrow — he removed women and Hispanics and ended up with fewer than 20 people per year — that “a researcher could hypothetically get any positive or negative ‘effect’ of the boatlift,” wrote economist Michael Clemens in Vox. Crucially, Clemens also argued, he and Rutgers economist Hunt discovered that population surveys began abruptly counting more black high-school dropouts in Miami after 1980, which had the effect of lowering average wages.

Borjas defended himself, blogging under the headline “More fake news about Mariel.” He created a chart excluding black dropouts, and still found wage decline. “Using the increase in the relative size of Miami’s black workforce after 1980 to dismiss my Mariel evidence performs the job of obfuscating the debate further, but does little to clarify,” he argued.

For laypeople, it can be challenging to unravel. And for the economists, that brings its own frustrations.

Card said he’s exasperated with media that bolster anti-immigrant positions with cherry-picked studies, as well as neutral media coverage that seeks out negative reports to balance positive studies without explaining nuances. “One competing study just cancels out another study,” he said of coverage.

As for Borjas, his biting blog indicates that he believes his research encounters unfair opposition. “We all know the party line by now: Immigrants do jobs that natives don’t want to do,” he wrote. “As a result, natives do not lose jobs, and natives do not see their wages reduced. And anyone who claims otherwise is obviously a racist xenophobic moron.”

The Academy’s Review

Against that backdrop, the National Academy of Sciences report may provide the broadest overview of work on this subject. The academy is widely considered the nation’s premier arbiter of complex issues, and is frequently relied upon to sort through politically-charged claims.

In September 2016, the academy released its 643-page survey entitled “The Economic and Fiscal Consequences of Immigration.”

In addition to its specific findings on wages, the report concluded that: “Importantly, immigration is integral to the nation’s economic growth.” Panelists found that “immigration supplies workers who have helped the United States to avoid the problems facing stagnant economies created by unfavorable demographics — in particular, an aging … workforce.”

The academy further noted that “if the American economy grows and requires more workers both to replace those who retire and to create new firms and industries, the primary source of labor will be first and second-generation immigrants. This basic fact will hold at all levels, from low-skilled service jobs to professionals with postgraduate degrees.”

The NAS report also concluded that immigrants are a benefit to the federal government coffers but a mixed picture for states — at least at first, primarily because of needed investments in schooling. But as adults, the NAS panel says, “immigrants’ children — the second generation — are among the strongest economic and fiscal contributors in the population,” paying more in taxes than either their parents or other native-born Americans.

Borjas didn’t dissent from the panel’s conclusions, co-panelists said.

“We all agreed high school dropouts are affected. George would say a lot. I would say a little,” said Hunt, the Rutgers economist. “The preponderance of evidence is that native wages are not affected. George agreed with that.” Hunt noted that the biggest cohort of dropouts today is near retirement age; national estimates are that only 6 percent of people between the ages of 16 and 24 were dropouts in 2015 compared to 12 percent in 1990.

Blau noted that Borjas signed the report but “has chosen to simply emphasize his own findings subsequently.”

Source: Department of Homeland Security

Trump makes his moves

For many with immigrant roots, this academic debate is far from academic. It’s affecting real people — people like Zahid Siddiqi of Strongsville, Ohio, a suburb of Cleveland.

An engineer who retired from the Ford Motor Company, Siddiqi immigrated from India in 1969, after a 1965 U.S. reform ended national-origins quotas dating from 1924 that favored Northern European immigrants. Siddiqi was sponsored by a professor, under a more relaxed job entry program. And after becoming a U.S. citizen, he sponsored two siblings whose families, like his, have thrived professionally.

“They’re deporting people who’ve been here 40 years. I’m horrified at what’s going to happen,” the 78-year-old grandfather said. “Not only for immigrants. But it’s the community, it’s America itself, that’s going to suffer.”

But regardless of what immigrants or economists think, Trump is pushing ahead, trying to force Congress to adopt dramatic immigration cuts. Sen. Tom Cotton, an Arkansas Republican, has emerged as the Senate’s new GOP immigration hawk and last year he introduced the RAISE Act — a name suggesting that with fewer immigrants, Americans’ wages would rise. As recently as last month, Cotton praised Borjas’ work on the subject during a hearing.

The bill’s only co-sponsor so far is Sen. David Purdue of Georgia. But Trump endorsed the proposals in August, with a tweet claiming that they would “reduce poverty, increase wages & save taxpayers billions of dollars.”

The proposal would cut the 5 percent of green cards that go, by lottery, to people from low-immigration countries. It would end all family immigration except for spouses and minor children, and create a new “merit” system to award visas. The White House wants premium points for higher education degrees in STEM fields, “extraordinary” talent and wealth to invest.

Trump claimed in a radio address last December that, under the current system, “foreign nationals can come to the United States and bring in unlimited numbers of foreign relatives.” In truth, family sponsorship can take many years. Per-country quotas keep siblings and adult children from some countries waiting a decade or even more than 20 years.

It doesn’t escape immigrants that Trump’s own history stands in contrast to new rules he wants.

Trump’s Slovenian parents-in-law were able to obtain green cards, for example, although the couple’s lawyer wouldn’t disclose how to The Washington Post. Bill Stock, an attorney with the American Immigration Lawyers Association, told the Center for Public Integrity that of the handful of routes possible for immigration — unique talent is one — the most likely for Trump’s in-laws would be sponsorship by an adult child, the same sort of “chain migration” that Trump has disparaged.

Trump has long hired immigrant workers at construction sites and hotels. Trump businesses continue to hire seasonal foreign “guest workers” at Mar-a-Lago and at son Eric’s Virginia winery — a practice that Sessions, in the past, attacked as bad for American workers.

But the issue doesn’t break down along traditional lines. For years, the U.S. Chamber of Commerce and the Service Employees International Union, hardly the best of friends, have shared in calling for legalization of some undocumented workers. They’ve also called for reforms to the current system that provides only 5,000 immigrant visas a year to fill shortages in lower-skilled but essential work like elder care.

In April of last year, 1,470 economists, Democrats and Republicans, signed a letter to Trump urging caution: “We write to express our broad consensus that immigration is one of America’s significant competitive advantages in the global economy. With the proper and necessary safeguards in place, immigration represents an opportunity rather than a threat to our economy and to American workers.”

In January, Trump made his first attempt to extract visa cuts from Congress. In exchange for endorsing a proposal to open a 12-year path toward green cards for 1.8 million “Dreamers” — young people brought here as undocumented children — Trump demanded that legislators approve a $25 billion trust fund for a southern border wall and deep cuts to family visas. The Senate deadlocked.

With midterm elections looming, though, immigration is likely to be a contentious issue on the campaign trail. During his 2016 campaign, Trump would often combine lambasting immigrants and trade agreements, while vowing to bring back jobs at manufacturing plants. But the reality, said economist Alex Nowrasteh of the libertarian Cato Institute in Washington, D.C., is that immigration played no role in factory shutdowns in the Midwest.

“It’s just easy to blame immigrants,” said Nowrasteh. “You can always rely on nationalism and cultural arguments that you couldn’t make if an American took your job.”

Read more in Inequality, Opportunity and Poverty

Share this article

Join the conversation

Show Comments

Notify of
Inline Feedbacks
View all comments