The National Interest, July-August 2012
A NARRATIVE is taking root among policy makers and opinion leaders that the illegal-immigration problem has been resolved and further concern over the issue is simply unnecessary. A New York Times op-ed by University of Southern California professor Dowell Myers exemplified this perspective when it began: "The immigration crisis that has roiled American politics for decades has faded into history."
This idea of complete resolution is highly dubious. There is no doubt that the number of people sneaking across the southern U.S. border has declined significantly, and the total illegal population has dropped somewhat from the record high of several years ago. But there is little reason to conclude this is a permanent development ushering in a new migration paradigm for the United States. More likely, it is merely a pause brought on by the U.S. economic slump and other factors on both sides of the border.
Indeed, it could be argued that the current lull in illegal immigration is just the end of the beginning. The United States has made progress toward the first goal of a sensible immigration policy—namely, developing the means to make an enforcement policy stick. But the country has barely begun the process of crafting a comprehensive immigration policy that has a chance of actually working—or even deciding what such a comprehensive policy should be. Thus, the New York Times op-ed had it wrong, and this issue is certain to roil American politics for decades to come.
Those on both sides of the issue have valid points in the ongoing debate, and, of course, they also have distinctive policy prescriptions that flow from their conclusions and outlooks. And while it may be too early to tell how long the current lull will last (assuming it hasn't already ended), assessing these policy prescriptions requires us to examine in some detail the state of illegal immigration, the reasons for the current situation and possible future developments.
THERE'S BROAD agreement that the illegal-immigrant population peaked in 2007 and declined for the next two years. The three main sources of estimates—the Department of Homeland Security, the Pew Hispanic Center and the Center for Immigration Studies—all conclude that in 2007, the illegal population neared or exceeded twelve million (the estimates range from 11.8 to 12.5 million) and declined to roughly eleven million (between 10.8 and 11.1 million) by 2009. The decline stopped after 2009, so until new estimates or new facts are developed, it's safe to say that there are currently about eleven million unauthorized residents of the United States. Dramatically higher numbers sometimes bandied about by commentators and politicians—such as a 2005 estimate by Bear Stearns analysts that the illegal population could be as high as twenty million—are almost certainly incorrect. If the number were that high, it would be reflected in birth and death records, since in a modern society such as ours almost no one is born or dies without that fact being administratively recorded, and we have a pretty good idea of the fertility and mortality rates of the illegal population.
It seems clear that illegal crossings at the border, known formally as entries without inspection, are also down. Although illegal entry is not the only source of illegal immigrants—perhaps 40 percent of the illegal population entered legally on some kind of visa and then overstayed—it is the main source, and there's a lot less of it. The only administrative metric available for illegal crossings is the number of apprehensions made by the Border Patrol. This is an imperfect yardstick; a drop in apprehensions could mean there are fewer illegal immigrants available for agents to catch, or it could mean that the Border Patrol is becoming less efficient at finding them. Likewise, the total apprehensions number includes people who attempt to cross multiple times, though such recidivism appears to have declined slightly. It's possible, then, that a drop in apprehensions, coupled with a drop in recidivism, might not indicate a reduction in the actual number of individuals trying to sneak across the border.
But the decrease in the number of arrests at the border is sufficiently steep that the only plausible explanation is that attempted crossings have declined, and the experience of border residents confirms this. Arrests on the Mexican border exceeded one million almost every year since the early 1980s and totaled almost 1.2 million in 2005. In 2007, arrests were under nine hundred thousand. They dropped to a little over half a million in 2009 and 328,000 in 2011. That 2011 number is the lowest seen in the country since the early 1970s.
The Pew Hispanic Center has estimated that the total inflow of illegal immigrants, both border jumpers and visa overstayers, declined from 850,000 annually during 2000–2005 to just three hundred thousand annually during 2007–2009. The total number of illegal aliens declined despite a continued inflow of new illegal aliens because, over time, many people stop being illegal aliens. Some return home, some launder their status to become legal immigrants and a few die. Though researchers disagree, both statistical and anecdotal evidence suggests the number of illegal immigrants leaving the United States increased during the recession.
With regard to Mexico in particular, Pew has recently reported that net migration has dropped to zero. This includes all immigrants, both legal and illegal (51 percent of recent Mexican immigrants are illegal). Specifically, Pew compares the 2005–2010 period to 1995–2000 and finds that the number of Mexicans arriving in the United States fell by half and the number returning doubled. From 2005–2010, arrivals and departures were roughly equal, with each figure at about 1.4 million. The number of departures is fudged somewhat, since it includes three hundred thousand U.S.-born (and thus U.S. citizen) children of Mexican immigrants, so even Pew finds continued net Mexican immigration, but the increase in departures (most of them voluntary) and the drop in new arrivals is striking.
THERE IS little doubt that the annual flow of illegal immigrants has slowed and the total illegal population has shrunk somewhat, although eleven million illegal immigrants is still a very large number. While interesting, this fact on its own doesn't tell us much that can be applied to policy. It's necessary to explore the reasons for the decline. Three explanations have been offered: first, the recession; second, increased enforcement; and third, changes in Mexico, the source of roughly 60 percent of the illegal population.
The economy requires the least explanation. Employment is the main magnet drawing illegal immigrants to the United States, so it makes sense that a tighter job market would lead more illegal immigrants to return home and dissuade some prospective illegal immigrants from embarking on the trek north. The recession officially began in December 2007, when the national unemployment rate was 5 percent. A year later, it was 7.3 percent. Another year after that, it reached 9.9 percent. The increase in the unemployment rate was even greater for illegal immigrants. Using young Hispanic immigrants with no more than a high-school education as a proxy for the illegal population (three-quarters of such people are illegal, and they make up two-thirds of the illegal population), we see that their unemployment rate shot up at the start of the recession from around 8 percent to about 15 percent, according to a Center for Immigration Studies analysis. This jump was especially quick and severe because the recession was kicked off by a housing bust, and illegal workers were disproportionately concentrated in construction and extraction jobs. Some 21 percent of these immigrants worked in such occupations, compared with just 5 percent of native-born American workers. This unemployment, combined with tougher times even for those still employed, led to a drop in remittances to Mexico by immigrants in general (both legal and illegal); the total value of remittances sent home dropped 3.6 percent in 2008 and a further 15.7 percent in 2009.
As bad as economic conditions were for illegal immigrants, the decline in their numbers began even before the start of the recession, so it seems likely that increased enforcement also has played a role. And that's no surprise, given the buildup in enforcement that has been under way since the Clinton administration. In 1995, there were about five thousand agents in the Border Patrol (most, but not all, working on the Mexican border). That number increased to ten thousand in 2002, hit fifteen thousand five years later and now exceeds twenty-one thousand.
Border fencing is also much more formidable, both in quality and quantity. While there is disagreement about the amount of fencing necessary on the border, there's no dispute that some fencing is essential, particularly in high-traffic areas, to slow and redirect potential illegal crossers.
The little fencing that used to exist consisted of welded-together, corrugated metal sheets acquired by the Border Patrol as army surplus. (They were used in Vietnam as portable landing mats for helicopters.) These were poorly maintained and, since they were solid, prevented agents from seeing what was happening on the other side. What's more, the corrugated ridges sometimes ran horizontally, which turned them into ladders for border jumpers.
Although some of this comically inadequate fencing is still in place, it has been replaced and augmented by modern barriers, first south of San Diego and then along much of the Arizona border and smaller sections in Texas. In compliance with the Secure Fence Act of 2006, there is close to seven hundred miles of fencing now along the two-thousand-mile border with Mexico, supported by technology such as ground sensors, pole-mounted remote cameras and even unmanned drones.
Combined with new tactics, the barrier in San Diego, consisting of two rows of fencing separated by a road for Border Patrol vehicles, had a dramatic impact on the flow. While San Diego accounted for more than 40 percent of all apprehensions in 1995, by 2011 it accounted for less than 13 percent of a much smaller total.
The Border Patrol's new tactics were actually pioneered in the mid-1990s in El Paso, the other main crossing point for illegal immigrants at the time, and the decline there has been even more dramatic. In 1993, El Paso accounted for nearly 25 percent of all Border Patrol arrests. By 2011, it accounted for only about 3 percent (again, of a much smaller overall number).
Much of the traffic that originally went through these two locations shifted to Arizona, and the Clinton and Bush administrations were unprepared—or unwilling—to do what was necessary to respond. During much of the first decade of the twenty-first century, Arizona accounted for half of all Border Patrol apprehensions, helping explain the intensity of voter concern over immigration. Most of the Arizona border now has some form of fencing.
There have been significant improvements in enforcement away from the border as well. One of the most important elements of the expansive 1996 immigration law was the 287(g) program, which established a formal mechanism for integrating selected state and local police into federal immigration enforcement. An even broader program known as Secure Communities is on track to connect all police and sheriff's departments in the nation to immigration databases so that every time a suspect is booked, his fingerprints will be checked against both the FBI's records and those of the Department of Homeland Security. The target date for full national coverage is the end of 2013, but two-thirds of the nation's local jurisdictions are already online, including all in the four border states and many others in the West, Midwest and South.
One result of these improvements has been an increase in the number of deportations—or "removals," as they are now labeled. (Removals are distinct from "returns," most of which involve Mexicans caught at the border and sent back across by the Border Patrol.) Although a comparison with past years isn't exact because the Obama administration has added some categories of people to returns for public-relations purposes, the increase is still dramatic. In 1995, about fifty-one thousand people were deported (most of them illegal immigrants, but some legal immigrants who rendered themselves deportable by committing crimes). By 1998, the total had more than tripled to 175,000, and then it more than doubled again by 2008 to 360,000. It has remained close to four hundred thousand for the past several years.
Another vital area in which enforcement has improved is employment. Only in 1986 did it become unlawful for employers to hire illegal immigrants; before then, they were specifically exempted from the definition of "harboring" illegal immigrants by what came to be known as the Texas proviso. But even after 1986, the paper-based enforcement system was easily gamed and had little impact on illegal employment.
That began to change, very slowly, with the enactment of the 1996 law, which required the development of pilot programs to enable employers to electronically verify the information provided by new hires. The current iteration of these programs is called E-Verify, and employers use it to check the name, date of birth and Social Security number of new hires. While still voluntary, the system is having an increasingly large impact. More than three hundred thousand employers currently use it to screen new employees, and last year about a third of all hires nationwide were run through the system. Although the system is improving daily, it is still possible to fool it with a fully developed stolen identity. But most illegal immigrants don't have the resources for that, and employers in industries with lots of illegal workers are finding that when they inform prospective employees they use E-Verify, many simply leave and look elsewhere.
Other enforcement improvements also merit attention. State driver's license systems (including the nondriver identification cards issued by the state motor-vehicles agencies) are America's national ID system—a decentralized one, to be sure, but a national system nonetheless. The weaknesses of that system became apparent a decade ago when the 9/11 Commission revealed that the nineteen hijackers had between them thirty state-issued pieces of identification, which they used to smoothly navigate American society. The result was the REAL ID Act of 2005, which established minimum standards for state IDs to be accepted for federal purposes, such as boarding airplanes.
Although the bill's focus was preventing terrorist access to American society, any weakness can be exploited. Before 9/11, not even half the states checked a license applicant's legal status. Today, only New Mexico and Washington State still issue regular driver's licenses to illegal aliens, while Utah provides a special illegal-alien license. Although the REAL ID Act's deadlines have been repeatedly extended to accommodate state objections, the genuine improvements in identification (including other innovations such as digitized birth records) have made it harder for illegal immigrants to live undetected in the United States.
Finally, states have passed a variety of measures that have contributed to the decline in illegal immigration. These include requirements that landlords verify the legal status of those applying for a lease and making the presence of illegal aliens a state offense. Even where these measures have been held up by the courts, the hyperbolic coverage of them in Spanish-language media outlets effectively persuaded some illegal immigrants to leave and deterred some prospective entrants from making the trip.
One state measure that survived court challenge has been shown to have made a notable difference. In 2007, Arizona passed a law requiring employers in the state to use the federal E-Verify system for hiring as a condition of retaining a business license. It was challenged by the U.S. Chamber of Commerce and various liberal and civil-rights groups, but it was finally upheld by the Supreme Court last year. Research by the Public Policy Institute of California found that in its first two years, the new law reduced Arizona's illegal population by 17 percent—nearly one hundred thousand persons.
A weak U.S. economy and stronger enforcement are joined by a third possible reason for today's relative lull in illegal immigration: changes within Mexico itself. The key development is the trend toward smaller families, thus slowing population growth and easing pressures for emigration.
The UN reports that Mexico's total fertility rate (TFR), the number of children the average woman would have during her lifetime, was 6.7 in the early 1950s; in 1950, the population was about twenty-eight million. Over the next fifty years, that high fertility rate caused Mexico's population to nearly quadruple to one hundred million at the turn of the century and to about 114 million now.
But as Mexico's population exploded, the TFR began to decline. It is now estimated by the UN to be just over 2.2 (the CIA estimates 2.3). This is close to the U.S. rate of about 2.1, which is the "replacement rate" that a country needs for long-term population stability. What's more, the UN projects Mexico's fertility rate will fall to 1.7 by midcentury (about the current rate for Denmark and Finland), at which time Mexico's population is expected to start declining.
OF THESE three factors—the U.S. economy, an enhanced enforcement regime and Mexico's demographic shifts—the first is obviously the most changeable. The unemployment rate has already begun to decline, dropping from 9.1 percent in August 2011 to 8.2 percent in June of this year. GDP growth resumed in 2010 at an annual rate of 2.8 percent (though the 2011 rate was lower).
What's more, there's evidence that immigrants are capturing a disproportionate share of whatever new jobs are being created. A Center for Immigration Studies report looking specifically at Texas found that, from 2007 through the second quarter of 2011, 81 percent of job growth went to newly arrived immigrants, half of them illegal aliens.
The vicissitudes of the economy obviously can't be relied on to limit illegal immigration. So to the degree that the economy is the cause for the current lull, it would seem to be temporary. That raises a question: Even with the return of strong job growth, would the new enforcement measures continue to blunt renewed pressure for illegal immigration?
President Obama's answer would appear to be yes. Speaking in El Paso last year, he outlined various improvements in border enforcement, then mocked those still dissatisfied:
But even though we've answered these concerns, I've got to say I suspect there are still going to be some who are trying to move the goal posts on us one more time. . . . You know, they said we needed to triple the Border Patrol. Or now they're going to say we need to quadruple the Border Patrol. Or they'll want a higher fence. Maybe they'll need a moat. Maybe they want alligators in the moat. They'll never be satisfied. And I understand that. That's politics.
Obama's mockery implies the country has done everything a nation can possibly do to control immigration, that the recent improvements complete the infrastructure necessary for border control and that it's time to move on to other things—namely, amnesty for the illegal immigrants already here and increases in future legal immigration.
Unfortunately, this is not true. Much needs to be done before the United States has the enforcement arrangements necessary to permanently reduce illegal immigration to a nuisance rather than an ongoing crisis. Start at the border. The increases in the Border Patrol over the past fifteen years have been real, but even at a staff level of twenty-one thousand, the agency—responsible for more than 7,500 miles of our land frontiers—is smaller than the New York City Police Department, which has 34,500 uniformed officers. Furthermore, the improvements in fencing are often exaggerated. Of the nearly seven hundred miles of physical barriers along our southern border, a large portion are Normandy barriers, designed to stop vehicular incursions across the border but of no use in stopping people on foot. What's more, when Congress passed the Secure Fence Act, it imagined double fencing of the kind south of San Diego. In fact, only about 1 percent of the border has a double layer.
This is why the Government Accountability Office reported last year that only 44 percent of the border is under "operational control," with only 15 percent actually "controlled" (the tightest level of security). That 44 percent figure is triple what it was in 2005, but it's hard to say a task is complete when it's not even half done.
The high level of deportations is likewise deceptive. The administration likes to boast of "record" deportations, but the other half of the story is that the growth in deportations has stopped. The reason annual deportations have been just under four hundred thousand during this administration is that the White House refuses to ask Congress for the funds to increase it further. Like the child on trial for killing his parents who then pleads for mercy as an orphan, the administration has created the very resource constraints it points to as the reason for not increasing deportations further.
Beyond that, some key enforcement tools remain unused. The E-Verify system is working well, but its effectiveness is necessarily limited until it is required for all new hires. Regrettably, the president is holding that change hostage in exchange for amnesty for illegal immigrants. In other words, he (and many in both parties who share his perspective) will accept the tool needed to exclude illegal immigrants from the workforce only after amnesty has ensured there are no more illegal aliens left in the workforce.
Another potentially valuable tool that is still not fully functional is a system to track all entries and exits by foreign visitors. As noted above, a large share of the illegal population was admitted legally in some sort of temporary status and never left. But despite a congressional mandate passed in the 1996 immigration-reform bill and reiterated by the 9/11 Commission, we still have no fully functional exit-tracking program and no plans to build one. The lack of such an elementary capability makes a mockery of the claim that our immigration infrastructure is complete.
And then there's the problem of political will. Even the improvements we've put in place over the past fifteen years are of little benefit if they are not used properly. And the current administration has announced that, as a matter of policy, it will seek to arrest and deport only those illegal immigrants who have committed serious, nonimmigration offenses. In a series of memos encouraging agents in the field to exercise their "prosecutorial discretion," the administration made clear that it views illegal presence in the United States (and all its ancillary crimes, such as document fraud and identity theft) as a secondary offense, like not buckling your seat belt while driving. In other words, they seek to pursue immigration offenses only in conjunction with other crimes. That means they will deport a rapist after he has completed his U.S. prison sentence, but they are uninterested in ordinary illegal aliens.
While prioritizing limited resources is part of any enforcement regime, this wholesale downgrade of an entire body of law is unprecedented. It's as though the Internal Revenue Service were to announce that ordinary citizens who are not terrorists or money launderers don't need to comply with the tax law because, as a matter of policy, no one would pursue them. The unprecedented and even lawless nature of the administration's approach is perhaps why so many immigration agents in the field have simply refused to obey what they see as an illegal order—one even their labor union said they should resist.
Do these things matter if the supply of illegal immigrants is drying up? Unfortunately, lack of sufficient enforcement remains a problem. Lower fertility and declining population are no guarantee that emigration from Mexico will come to an end. Emigration is not analogous to an overflowing cup that stops spilling liquid when the level falls. Migration is based on networks of family, clan and village that can continue to operate long after the conditions that may have sparked the original emigration have disappeared. For example, the states of western-central Mexico—far from the border—that sent farmworkers in the 1940s and 1950s through the so-called bracero program are still disproportionately important sending areas nearly a lifetime after the program began.
Looking at fertility rates in other countries confirms that low birthrates and low emigration are not necessarily connected. Mexico's current TFR is almost identical to that of other countries of emigration—Burma and Indonesia—but also the same as countries of immigration—Saudi Arabia and Argentina. Likewise, South Korea and Russia have some of the lowest TFRs in the world—1.23 and 1.42, respectively. In fact, Russia's population is already declining and is expected to fall about 11 percent by midcentury. And yet both South Korea and Russia are major source countries for immigration to the United States. Though correlation doesn't necessarily imply causation, in both cases the U.S. immigrant populations from those countries have risen just as fertility rates have fallen.
And there's always the rest of the world. Mexicans account for more than half of the current illegal population, but close to another 20 percent comes from Central America, which is even poorer and less developed. And TFRs are still very high in much of Africa and the Middle East. The importance of networks means we get little immigration from, say, Chad, but the U.S. legal-immigration system, especially refugee resettlement and the visa lottery, actually creates new networks for future illegal immigration.
As important as it is to have a functioning immigration-control program, legal immigration is ultimately more consequential. Of the forty million foreign-born people living in the United States, nearly three-quarters of them are legal. Even in 2008–2009, during the two years of the worst recession in living memory, 2.5 million people moved here from abroad, most of them legally. And the problems associated with illegal immigration—burdens on schools, pressure on public services, even wage suppression—have nothing to do with legal status and everything to do with numbers. For example, a fourth of all people in the United States living in poverty are immigrants (legal and illegal) and their young children. Immigrant families account for a third of the uninsured, while 36 percent of immigrant households use at least one federal welfare program.
Thus, there is no reason to conclude this big national crisis, or the intense political emotions it generates, will fade from the scene anytime soon. Those who think otherwise are engaging in wishful thinking, likely born of their own favorable view toward the immigration wave of recent decades. The problem is ongoing, as is the civic and political imperative that it be confronted.