The Facts About Immigration
Border Patrol agents pulled over a minivan carrying immigrants in Pharr, Texas, last year.CreditPHOTOGRAPH BY PETER VAN AGTMAEL / MAGNUM
During a conference at the Brookings Institution last week, the Princeton economists Anne Case and Angus Deaton presented their latest paper on the rising mortality rates of white working-class Americans, which received, and is still receiving, a huge amount of attention. That’s understandable. Even taking account of the critiques of the paper that have been presented, some of which the economics blogger Noah Smith counter-critiqued on Wednesday, the issues raised by Case and Deaton’s work are profound.
The subject here, though, is another article that was presented at the Brookings conference, which got nowhere near as much media attention. Co-authored by Gordon Hanson, Chen Liu, and Craig McIntosh—three economists at the University of San Diego—it addressed immigration, particularly immigration by low-skilled workers from Mexico and other countries in Latin America.
The basic message of the paper was that the political discussion about building a wall across the U.S.-Mexico border is at least a decade out of date. In the past ten years, the flow of undocumented immigrants entering the U.S. has slowed dramatically. And because many undocumented immigrants are either deported or move home every year, the total number living in the United States is currently falling at rate of about a hundred and sixty thousand a year. Consequently, the competitive pressure being placed on the wages of low-skilled American workers, who do similar jobs to low-skilled immigrant workers, is declining, the paper says. Indeed, many industries that employ a lot of low-skilled immigrant workers—such as agriculture, construction, and food services—are facing a potential shortage of labor.
“The current debate about U.S. immigration policy—with its discussion of walls at the border and mass deportations of undocumented residents—thus has something of an anachronistic feel to it,” Hanson, Liu, and McIntosh write. “The dilemma facing the United States is not so much how to arrest massive increases in the supply of foreign labor, but rather how to prepare for a lower-immigration future.”
To people who’ve kept up with the latest immigration trends documented by organizations like the Census Bureau and the Pew Research Center, it’s not news that the undocumented population is falling, or that more people from Mexico—the largest source of low-skilled immigrant workers—are now leaving the United States than are coming in. Indeed, as far back as the summer of 2015, when Donald Trump launched his Presidential campaign by accusing Mexico of sending America its criminals and rapists, Bill Clinton pointed out, “Basically between 2010 and 2014, there was no net in-migration from Mexico.”
One reason these facts haven’t had much impact on the political debate is a widespread presumption that the decline in low-skilled immigrant workers could easily be reversed, especially if the U.S. economy picks up. After 2007, the U.S. construction industry, which had been a major employer of undocumented workers, went into a prolonged slump. If construction picks up—which is already happening in some parts of the country—won’t many more undocumented immigrants cross the border in search of work?
One of the most important findings in Hanson, Liu, and McIntosh’s new paper is that this is unlikely to happen. The authors explain that demand for unskilled workers is only one of the factors that drive undocumented immigration. Others factors include both the potential supply of immigrants and immigration policy, such as border-security measures. In these last areas, Hanson, Liu, and McIntosh point out, the past decade or so has seen changes that suggest the recent fall in low-skilled immigration will be permanent.
On the supply side, a drop in fertility rates in Latin American countries and an improvement in employment prospects have made young people there less eager to leave their home countries. Since Mexico accounts for more than half the low-skilled foreign-born workers in the United States, developments there are particularly important.
In the nineteen-sixties, Mexican women had almost seven children each, on average. Today, the figure is just over two, not much different than the U.S. fertility rate. Fewer Mexican children being born in 1995 translates to fewer twenty-two-year-olds looking for work today. And with the Mexican economy having expanded over the past couple of decades, the number of jobs available there has risen, too.
To determine the relative importance of labor supply versus labor demand, the authors estimated a statistical model of Mexican emigration, and they found that changes in labor supply could explain up to four-fifths of the recent fall. Since Latin American fertility rates are unlikely to rebound, these results suggest a good part of the decline in undocumented immigration is permanent. “Because U.S. neighbors to the south are today experiencing much slower labor-supply growth, the future immigration of young low-skilled labor looks set to decline rapidly, whether or not more draconian policies to control U.S. immigration are implemented,” the paper says.
It also reminds us that, despite Trump’s rhetoric about lax immigration policy, the Bush and Obama Administrations devoted an enormous amount of resources and manpower to beefing up border security, and they also deported large numbers of people. Between 2000 and 2010, the number of federal agents policing the U.S.-Mexico border rose from eight thousand six hundred to more than seventeen thousand. Hundreds of miles of fencing have been erected, and border agents have also been equipped with high-tech equipment, such as surveillance drones and movement sensors.
“In effect,” the paper says, “the United States already has a wall in place, with hundreds of miles of new fencing, the rollout of technologically sophisticated border surveillance, a near quintupling of Border Patrol agents since the early 1990s, and the criminalization of illegal border crossings since the late 2000s.”
Meanwhile, immigration enforcement inside the United States has become a lot more draconian. In 2001, the U.S. government deported a hundred and sixteen thousand non-criminal aliens, many of whom had been picked up in routine traffic stops or for other minor offenses. Between 2007 and 2015, the government deported almost twice as many non-criminal aliens each year: two hundred and twenty six thousand. Deportation of criminal aliens has also risen sharply: from seventy three thousand in 2001 to an average of a hundred and fifty six thousand a year between 2007 and 2015.
Given all these developments, the makeup of the undocumented population in the U.S. has changed a lot. It is no longer young and itinerant. “By 2015, three quarters (75.1 per cent) of low-skilled immigrant workers had resided in the United States for 11 or more years,” the paper says. It also notes that in 1980, the modal Mexican-born resident of the United States was twenty years old. (“Modal” refers to the age that is most common.) Today, the modal age of Mexican-born U.S. residents is about forty years old, and in 2040 it will be almost seventy.
At the end of their analysis, Hanson, Liu, and McIntosh pose an interesting question: “Why build a wall to stop an immigration surge that has largely already occurred?” One possibility they mention is that “voters may be upset by the laxity of past enforcement and willing to reward politicians who are seen as atoning for these transgressions.” A second possibility: voters want to “forestall future claims on public resources . . . by increasing deportations today—when many low-skilled immigrant workers are approaching middle age—the United States may avoid demands on social spending in the future.”
I suspect that may be apportioning too much foresight to the supporters of the wall. When I e-mailed George Borjas, a Harvard economist who studies immigration, to ask what he thought of the new paper, he suggested that there could also be a precautionary motive for building the wall. “We know that illegal immigration from Mexico responds dramatically to changes in economic conditions,” Borjas wrote. “Does anybody know what will happen to the economies of Central America and Mexico in the next 5-10 years? What if there were another economic crisis? Wouldn’t there be a valid argument that building a wall today as insurance against future illegal immigration? None of this should be interpreted in terms of whether I support or do not support the wall. But these are questions that somebody who wanted to think through the various possibilities would need to ask and answer.”
I put what Borjas had said to Hanson, and he replied: “George is making a perfectly valid point. The question is whether the wall passes the cost-benefit test. With close to 20k border patrol officers already on the ground and 650 miles of border barriers already in place, it seems that the extra deterrence effect of the wall would be low while its cost would be enormous.”
