The debate over immigration enforcement has pitched through some striking twists and turns over recent years, and the roller coaster has certainly not completed its course. The mid-1990s were marked by political competition to show which party could be the toughest enforcer. In 1996 this contest gave birth to crackdown legislation that was massive, showy, tough-sounding and productive of real individual hardship, but as we now know, this legislation was ineffectual in controlling illegal migration.1 Well before its effectiveness could be tested, however—in fact, shortly after the reelection of President Bill Clinton to the White House—the political parties woke up to the fact that Congress’s 1996 approach seemed to drive away many Latino voters, one of the fastest-growing demographic groups among the electorate. Enforcement enthusiasm then took a backseat, and two measures that—strikingly—might properly be called amnesties, covering certain Central Americans and Haitians, found a quiet way to passage, with significant bipartisan support.2 President George W. Bush then came into office in 2001 hinting at a grand bargain with Mexico that might mean far more migration opportunities for Mexican citizens and possibly for low-wage workers of other nationalities.3 The September 11, 2001, attacks on New York and Washington brought enforcement back into vogue, but by January 2004 President Bush judged the time opportune once again to focus on new admissions in the form of a massive temporary worker program that would also give legal status (but only for six years) to the millions already illegally present in the United States.4
Work on this chapter, except for final copy editing, was completed before the author was appointed in November 2008 to President Obama’s transition team reviewing the Department of Homeland Security. The views expressed here are the personal views of the author and do not necessarily represent the position of the transition team or the Department of Homeland Security. This chapter is a revised and updated version of an article initially published in the New York University Journal of Legislation and Public Policy 10, no. 3 (2006–2007): 525–553.
By the time Congress took up the issue in earnest, in late 2005, however, politicians had discovered that the electorate still contained many people worried about enforcement. Moreover, it became apparent that legislators might face opponents happy to anchor their campaigns on denunciation of any measures that could be labeled amnesty for immigration lawbreakers. This made for more tumultuous politics, and it produced three different bills that received significant support and close attention over the subsequent two years—all of them unsuccessful.5 These legislative developments have been touched on in earlier chapters in this volume, but a brief summary here will set the stage for the closer scrutiny of the enforcement debate that I provide below.
The House was the first to act. In late 2005 it passed an enforcement-only package that toughened criminal penalties on unlawful migrants and those who aided them, mandated expanded employer screening of new hires (enhanced with mandatory electronic verification of the employee’s authorization to work), and emphatically declined to legalize the current undocumented population or adopt any significant new guest-worker provisions.6 Six months later the Senate passed a different kind of bill, called by its supporters “comprehensive immigration reform” because it included not only enforcement improvements (ambitious though less draconian than the House’s version) but also an expansive temporary worker program plus earned legalization for most of those aliens already present in the United States without documents.7 Unlike President Bush’s 2004 plan, this bill opened the way to full permanent resident status for such persons. The two houses were so far apart, however, that they could not even agree to meet in conference to reconcile the bills.
When the Democratic Party took control of Congress in 2007, hopes momentarily soared for a generous comprehensive approach because the Bush administration continued to profess support. But these hopes did not last long. Interbranch negotiations—apparently responding to the vehemence of the opposition to amnesty voiced during the 2006 campaigns—resulted in a toughened comprehensive reform package.8 It retained a large-scale guest-worker program and a “path to citizenship” for the undocumented population, but it incorporated more severe conditions for each. Guest workers could stay no more than two years at a time and would have far more limited rights to bring their families, as compared to the 2006 legislation. The path to citizenship for the undocumented already present was strewn with obstacles in the form of more complex procedures, lengthier waits, and higher fees. Signaling the political need to demonstrate commitment to enforcement, President Bush deployed the National Guard to the southwest frontier to fortify the Border Patrol’s efforts. Even so, the reform bill failed in the Senate and thus never reached the House. As of winter 2008, the issue has been left as one of the more daunting challenges bequeathed to the next president and Congress.
Throughout this turbulent ride, arguments for all sides have often rested on enduring and contradictory myths about the nature of enforcement and the problems and opportunities now before us. In this chapter, I identify and explore the eight most prominent of these myths. Probing the various factions’ assertions about enforcement and identifying the hollow spots can help to map the scope of the challenge if we are to build sound and comprehensive immigration reform—providing insights not only about enforcement but also about some of the changes proposed for our provisions governing legal migration. In particular, it reveals insufficiently appreciated flaws in the plans for a massive new temporary worker program, suggesting that legal migration reform should focus on changes to the permanent immigration provisions.
The misunderstandings about enforcement, however, will claim center stage here for an essential reason. The one indispensable component of viable immigration reform, whatever its precise content, must be the patient construction of a stable, enduring, and functional enforcement system. Only by developing the capacity to enforce the deliberate choices that the nation makes about immigration (including periodic revisions) can we reduce the bitter polarization these issues have produced and thereby calm the wild and disabling swings of policy that we have witnessed for two decades. Furthermore, though they rarely recognize this counterintuitive reality, it is the proponents of liberal and generous immigration policies who have the greatest stake in achieving effective federal enforcement. Demagoguery on immigration feeds on many dark impulses, but it also taps into a more admirable one, respect for the rule of law. It can continue to claim wider support, to the detriment of generous admissions policy, as long as the citizenry lacks confidence that the nation really can enforce the legislative choices it makes about immigration. Restoring confidence need not mean, of course, that the system becomes completely bulletproof against lawbreaking. Rather, it requires that lawbreaking becomes marginal instead of a widely practiced and profitable enterprise.
The first cluster of myths I examine below enjoyed a heyday from about 2001 through early 2005, a period when most of those expressing views on immigration reform focused on opening up new benefits either for those illegally here already or for employers allegedly in need of new and legal labor flows. Once proposals of that type started to pick up momentum, however, opponents became better organized and began pressing for legislation that would reject amnesty or extensive new admissions and would instead concentrate on enforcement. Their arguments then propounded a different set of myths, which appear later in the list.
Myth 1: A large guest-worker program will eliminate most need for enforcement.
Following 9/11, some claimed that a guest-worker program would bring all the well-intentioned and hardworking unauthorized migrants out of the shadows to identify themselves, permitting law enforcement to focus on a far smaller and more easily separated class of terrorists and criminals.9 For some, this prospect even suggested that we could significantly reduce the resources spent on preventing the migration of unauthorized foreign workers.10
A moment of reflection should reveal the flaws in this assertion. For this idea to have any chance of succeeding in reducing the enforcement task, whatever legalization is offered must be attractive enough to ensure that nearly all of the currently undocumented (estimated at 11.5–12 million11) come forth to register. That seems dubious if the plan offers them only a time-limited work period, such as the one contained in the president’s 2004 plan. Even if we lay that doubt aside, however, the new guest-worker scheme would also have to offer enough future spaces to satisfy nearly all those foreigners who might wish to come, especially the low-skilled workers who now constitute the primary customer base for smugglers. Daniel Griswold of the Cato Institute writes, “If a wide enough channel were opened so that the supply of workers from Mexico could be legally matched with the demand for their labor in the United States, the rationale for the current illegal flow would vanish.”12 (For Griswold’s argument in full, see chapter 12.) The “if” clause masks a truly heroic assumption. The net inflow of illegal migration this decade has been estimated at an average of 500,000 annually,13 and there is no guarantee that the current supply figure will remain stable. In fact, whenever the economy is relatively healthy this number seems likely to increase, perhaps dramatically, once a legal avenue toward working in the United States becomes available, so that prospective migrants no longer need to think of paying a smuggler’s fees or risking a hazardous trip through the desert.
Hence the only channel wide enough to reduce reliance on immigration enforcement would be a guest-worker program free of fixed numerical ceilings. Under such a policy, admissions would be limited only by the quantity of petitioning employers who successfully demonstrate that they posted the job opportunity and found an insufficient response from American workers. Some early versions of what became the Senate’s Comprehensive Immigration Reform Act of 2006 envisioned such a ceiling-free system, but as reported from committee, the legislation bowed to political realities and imposed a ceiling of 325,000 annually. A floor amendment then reduced that limit to 200,000. Even if the next Congress could enact a far higher ceiling, fierce pressure to reduce the quota would likely result from the first serious upturn in unemployment levels, particularly if the program by then is regularly bringing in hundreds of thousands of workers each year. In short, if guest-worker admission spaces do not keep up with worker interest in sending countries—and they almost surely would not—enforcement officers would have to remain engaged in policing against old-fashioned garden-variety immigration violations, lest the system again lose all credibility. Immigration officials would not be able to divert much, if any, of their enforcement attention to criminals and terrorists.
Moreover, the employer petitioning process set forth in the previous Senate bills was far too vague to afford reliable protection for American workers, particularly because it lacked a serious enforcement mechanism to monitor employer job searches or the wages later paid to guest workers. Employers hungry to bring in workers willing to work for lower wages will make maximum use of any enforcement loopholes or soft spots. Serious enforcement would be required to ensure the integrity of the petitioning process and the maintenance of adequate wages and working conditions. Enforcement resources would be saved only if the government really did not care about eroding the latter standards—and there has been reason to wonder whether such an attitude informed some of the arguments that latched onto this myth.
Perhaps a more glaring shortcoming of new guest-worker proposals has been that President Bush and other proponents consistently omitted any discussion of the need to enforce departures at the end of the temporary admission period. This is a telling omission because the temptations for employer and employee to want to continue the employment relationship will be substantial. Perhaps this silence about enforced departures merely reflected the usual short time horizon of political officeholders. But such blindness is at least misleading, if not manipulative. This convenient obscuring of an inescapable component of any guest-worker plan is closely linked to myth 2.
Myth 2: This time we can make temporary worker programs truly temporary.
The history of temporary worker programs, here and in Europe, gives little confidence that guest workers’ tenure would truly be temporary, at least in the absence of serious, well-designed, and adequately resourced enforcement.14
The bracero program, which allowed over 4 million Mexican workers to enter the United States for ostensibly temporary employment from 1942 to 1964, helped to establish employment relationships that proved far more durable than policymakers expected when they officially ended the program. In the absence of effective enforcement thereafter, employers were glad to continue using workers who by then were well-trained and familiar with the farm or business. Workers likewise found it convenient to continue the cycle by sneaking across the border.15 Moreover, they often brought friends or family members along to replenish the worker pool. Migration is a phenomenon built on social networks, and the contacts established even through temporary work can become the basis for a wider circle of future migration.16 Similarly, when most European countries ended their postwar guest-worker initiatives after the oil shock of 1974, millions of supposedly temporary workers remained in those countries.17
Different results for the current U.S. temporary worker plans seem unlikely. The jobs at issue are mostly permanent jobs, and there would be every reason for employers to want to keep experienced employees when the temporary period expires. The immigrant worker would also have strong incentives to stay after several years of growing linguistic, cultural, and technical familiarity, both with the particular job and with life in America.
President Bush’s January 2004 proposal addressed this question with only carrots, not sticks. As an incentive to return, he suggested methods of crediting work time in the United States toward the national retirement system in the home country through social security totalization treaties.18 That this would be a sufficient incentive seems highly dubious. Pension benefits—scheduled to begin only after retirement—based on perhaps six years of U.S. work would rarely outweigh the economic gain the worker could expect if weak enforcement permitted continuing employment at full U.S. wages for additional decades.
Other guest-worker plans, to be sure, have spoken of imposing harsher measures in an effort to ensure that the work period truly remains temporary. After the 2006 elections, for example, the Republican leadership of Congress pressed for much tighter restrictions on guest workers than what the Republican-controlled Senate had accepted in 2006.19 (That earlier bill, like the president’s 2004 proposal, would have allowed workers in for a three-year stay, readily renewable up to a maximum of six years. And workers were generally allowed to bring their families.) The Bush White House backed this tougher approach, working most of the proposed changes into the 2007 bill before it was introduced. Under that bill, most guest workers would be permitted to come to the United States for no more than a two-year stay. Two renewals would be possible, but only after the worker spent one year outside the country between stays. The bill made it difficult for family members to accompany the worker. They had to have full medical insurance and had to show family income that exceeded 150% of the poverty level. Even then, the family could come for a maximum of only two years. Other guest-worker plans have suggested that a portion of wages be withheld and made available to the worker only on return to the home country.20
Proposals in this vein recognize that, in general, wide-scale temporary worker programs have the best chance of remaining temporary only if accompanied by severe limitations and resolute enforcement. But the limited examples history affords of successful measures along these lines are almost surely measures the United States would have difficulty imposing—and indeed that we should not want to impose.21 Occasional mass roundups of overstaying guest workers, such as those Nigeria conducted in 1983,22 would be inhumane, riddled with error and injustice, and quite likely unconstitutional. Preventing the workers’ families from living together in effect keeps hostages in the home country. Although this would probably be constitutional, it is bad public policy. Human beings should be allowed to live with their families, as the Senate’s 2006 and the president’s 2004 bill wisely recognized. Yet a humane concession of that sort obviously makes it considerably more difficult to insist on removal at the end of six years of lawful, even if nominally temporary, residence. The Swiss novelist Max Frisch captured this dilemma in his incisive comment on the European guest-worker situation some forty years ago: “We asked for workers, but human beings came.”23
If we need people for their labor, we should recognize that they will come with a full set of human yearnings and needs, including a natural inclination to sink roots and seek stability. Therefore, before changing our provisions for employment-based migration, we should first ask hard questions about which persons or classes of workers we truly need for their labor and not just because businesses prefer to pay wages lower than the level they now find necessary to attract a sufficient pool of American workers. Then, if more work-related migration is truly needed, we should admit persons in that narrower group, along with their families, as lawful permanent residents. The migrants—rather than Congress or the Department of Homeland Security—would then be able to decide whether or not to return to the home country after a few years of U.S. earnings. Some doubtless would, though almost surely a minority. But if such workers want to remain, that should be their option.24
Myth 3: Enforcement is doomed without a large-scale guest-worker program because undocumented workers take jobs Americans do not want.
Some advocates of comprehensive immigration reform assert that enforcement is doomed without a large guest-worker provision because market forces ensure that this volume of migration will occur anyway. Only by providing a legal channel to satisfy an inevitable demand can we reduce the magnitude of the enforcement task to an achievable level.25 Underlying this assertion is the claim that Americans do not want the jobs that undocumented workers occupy, leaving a vacuum that employers and unauthorized workers rush to fill. In this account, employers such as those in the hotel and restaurant field or in the construction industry, and particularly in farmwork, have a demand for labor that can be met only by foreigners. It is better for them to use a legal and temporary channel than to be forced to reach out to unauthorized workers. President Bush’s version of this tale, which he has often voiced, is that our system cannot realistically stand in the way of “willing employers and willing employees.” He has often spoken of filling with temporary workers jobs that “Americans won’t do.”26
But in fact, Americans people every one of the employment fields cited in these discussions, even farmwork. According to U.S. census data, 55% of those employed in farming (including forestry and fishing) are native-born, as are 65% of those involved in building cleaning and maintenance, 74% in construction and extraction, and 76% in food preparation.27 Because these percentages do not include the foreign-born who are naturalized citizens or who have a legal immigration status, the numbers actually understate the involvement or availability of Americans and of authorized noncitizen workers in each of these fields.
When employers encounter a shortage of available workers, the normal market response would be an increase in pay so as to draw in a greater supply of interested workers from whatever pool is available. That market process comes into play whatever the constraints on supply. If some hypothetical hemispheric catastrophe virtually shut down the supply of new foreign workers, American farming would not end, hotels and restaurants would continue to operate, and we would still build new buildings. Prices would be affected, but the impact would be far lower than these arguments typically assert. For example, Professor Philip Martin, an agricultural economist who has done extensive work on migration, calculated in 1998 that if farm wages rose by 50 percent, the average American family’s cost for fresh fruits and vegetables would increase only $10 a year because farm labor accounts for only a small percentage of the cost of our food.28
President Bush’s mantra about “willing employees” thus masks the key question: Willing on what terms? Many more Americans would be willing to take up these employment opportunities if the pay were higher or the working conditions better. The effect of a massive guest-worker program—and surely the intent of at least some of its proponents, though never admitted—will be to keep wages low and working conditions marginal.29
This is not to say that it will be easy to meet the employment needs in some of these fields by tweaking wages upward. In fact, if significant wage increases are required, some of the business will simply move overseas, with complex effects on American business and labor. Yet, because many of the enterprises at the center of this debate—especially hotels, restaurants, and construction—cannot relocate overseas owing to the nature of their business, they would have to find another way to adjust. If there were effective enforcement limiting these businesses to authorized workers, the firms might well face a difficult transition. But we would not wake up to find all such businesses shuttered.
Studies show that a few of the fastest-growing employment fields in the United States, such as personal and home-care aides or medical assistants, do not require extensive education and training.30 But these trends do not prove that we must make room for vast numbers of imported temporary workers. If these labor needs truly cannot be met by wage increases or other changes in U.S. practices, then those admitted from abroad to meet these needs should come with permanent residence rights, either immediately or on terms that leave the choice to obtain permanent residence largely in the hands of workers, not employers.
Another problem with temporary worker programs is the temptation they pose for exploitative behavior by employers. Workers who risk deportation on losing their jobs are unlikely to complain about unfair treatment or substandard working conditions. Even if one might theoretically make up for this vulnerability through increased centralized enforcement by the Department of Labor, resource increases of the magnitude that might be needed seem highly unlikely. Giving temporary workers more scope and freedom to change employers could ameliorate this particular problem, but only at the cost of making administration far more complicated. Could the temporary worker go only to other employers who had previously pursued the application process to show that they could not find other American workers? If not, the rationale that we are importing workers only to fill jobs unwanted by Americans would be defeated. But if so, an unfairly treated worker will have only limited real-world options for leaving the first employer, given the inevitable delays and bureaucratic hoop-jumping such a system would entail. This is another reason to impose on ourselves the discipline of taking as workers only those we are prepared to admit for permanent residence. Permanent residents are fully able to leave an exploitative employer and move anywhere else in the labor market.
Myth 4: Enforcement has had no effect, and we really cannot enforce the immigration laws anyway.
Some depictions of current border policing go beyond calling it inadequate and assert that it “stops almost no one.”31 In a variant of the myths described above, this one claims that market forces are too strong, inevitably overwhelming the state’s efforts at control. The chief evidence put forth for this assertion is the fact of continuing high-level migration, indeed an increase in the flow, through the late 1990s, even as the Border Patrol grew enormously and adopted new and seemingly better-designed patrolling strategies, such as Operation Gatekeeper in the San Diego sector.32
This evidence is sobering testimony to the difficulties and disappointments of immigration enforcement, and it must be taken quite seriously in planning future immigration control initiatives. In my view, it primarily shows the inefficacy of heavy border policing without serious internal enforcement, but the problems are multifaceted. Nonetheless, the idea that border policing has no effect has hardly been proved. The relevant indicator is not how illegal migration levels compare before and after these deployments, but rather what migration would have looked like during those same years without the new deployments. We cannot empirically test this counterfactual, but it is not credible that migration would have been no higher without the Border Patrol changes of the mid-1990s. Seasoned enforcement officers do not conceive of their jobs as designed to hold illegal migration to zero. Instead they understand their objective as reducing it to a level well below what it would be in the absence of their efforts.
At least one piece of reasonably hard evidence refutes the claim that border enforcement is futile. By nearly all accounts, smugglers’ fees increased enormously during the time the Border Patrol was implementing its new strategy. A common estimate is that the prices rose from a few hundred dollars in the early 1990s to over two thousand dollars in recent years.33 Surely some people contemplating migration found the new smuggling prices too high. Other factors that simultaneously shifted the demand curve to the right likely caused overall illegal migration to stay constant, or even to increase in spite of the rise in prices. The superheated U.S. economy of the late 1990s, with record low unemployment, surely played a role in producing that result. Many who lean toward the “enforcement is futile” perspective base their claims on a view that the market simply overwhelms state restraints. It is at least ironic for such proponents to overlook solid market evidence, in the form of this crucial price rise, that appears to negate the futility hypothesis. At the very least, they owe some explanation of why this example of supposedly market-driven behavior—the choice to migrate—utterly escapes the usual market effects of price increases.
Myth 5: Improved enforcement means tighter border enforcement.
Even many who spoke of immigration reform in 2004 primarily as the vehicle for new guest-worker admissions now offer at least rhetorical support to the need for better enforcement as part of a complete reform package. Former president Bush provides the paradigmatic example. His January 2004 speech launching the latest reform round said little about enforcement. By mid-2006, however, he was deploying the National Guard to the southwest border in the hope that a show of enforcement seriousness would improve the chances of passage for the legislation he favored.
Welcome as these late conversions may be, far too often they imply that enforcement means only more Border Patrol officers, more fencing, and basically a greater emphasis on policing the frontiers.34 Border measures are undoubtedly important as part of a sound overall strategy, and the last decade’s increases in the Border Patrol combined with the new forward deployment approach have given us a solid border enforcement component. But the border has received disproportionate attention. Enforcement that addresses only the border is bound to fail.
As long as potential migrants believe that a single successful foray past the border defenses (even after several failed attempts) opens wide prospects in the U.S. workplace, smugglers will have a ready market for their skills, and hundreds of thousands of immigrants will still come. Even if we build fences across the entire southwest border, smugglers will find other ways to transport their clients. This nation has over 88,000 miles of tidal shoreline, and some migrants already arrive clandestinely on boats. Entry along the 5,500-mile Canadian border is also a real possibility. Furthermore, an estimated 25 to 40% of our current unauthorized population never entered clandestinely, instead overstaying a legal admission period.35 Border enforcement contributes almost nothing to the deterrence or apprehension of overstayers.
What we really need is to complement border enforcement with major new efforts at interior enforcement, especially enforcement through worksite verification of new hires. Jobs draw migrants to America. The key to winning the enforcement battle is reducing the job magnet.
Why then have we spent so much energy and money on border measures? One study found that between 1986 and 2002, about 60% of all appropriated enforcement resources went to border work, leaving only 10% for interior investigations and related enforcement.36 Interior enforcement, through sending investigators out to track down and remove persons already illegally present, is costly and inefficient.37 It is also unglamorous and unpopular work in the eyes of most of Homeland Security’s enforcement personnel; those officers who are successful at it regularly gain less administrative credit and prestige than others who are involved in criminal alien apprehension or participation in antiterrorist initiatives.
A different specialty within internal enforcement focuses on employer compliance with the law’s workplace screening requirements.38 Enhancing this type of enforcement should be far more efficient than one-by-one apprehensions at deterring illegal migration, but ramping up worksite enforcement carries other medium-term political disadvantages. Primarily, it imposes visible burdens on business. As a result, significant interest-group pressure quietly helps push Congress toward underfunding these enforcement endeavors, and there has been no equivalently organized constituency pushing back.39 Border measures, in contrast, step on almost no influential toes. Border crackdowns are therefore used to demonstrate enforcement seriousness, alienating few and placating many. But focusing only on the border is an ineffective way to master our enforcement problems. The key fulcrum for effectiveness is the workplace.
Myth 6: We have tried employer enforcement, and it does not work.
It is quite true that the current version of employer enforcement does not work, but flaws in that system are correctable—not easily, but with the right combination of resources, smart design, and patience. Under the system established by the 1986 Immigration Reform and Control Act (IRCA), all U.S. employers must require each new employee to present documents showing identity and demonstrating work authorization.40 The company must then record on Form I-9, signed by both employer and employee, what documents were produced. By law, however, the employer is required to accept any document “that reasonably appears on its face to be genuine.”41 False documents are widely available, but a company runs afoul of the antidiscrimination provisions of the immigration laws if it demands more of an employee whose documents have satisfied the facial validity test.42 Further, as these respective enforcement regimes have evolved, employers have usually had more to fear from antidiscrimination enforcement than from enforcement related to the immigration screening. No wonder this system has been utterly ineffective at closing off the attraction of the U.S. job market.
A few crucial and helpful procedures have nonetheless been established by IRCA’s mechanisms. It is now commonplace for every legitimate business in the United States to go through the I-9 verification procedure, empty ceremony though it may often be. This routine is a valuable behavioral foundation on which to build a modified worksite immigration screening system that will actually work. Employers need not be asked to adopt a whole new set of actions; they need only be led to adjust modestly an already familiar practice. The key is to give employers access to a reliable and minimally burdensome system that can verify the documents presented—and also to provide ready means to check that the person presenting them is really the one to whom the documents relate.43 This task is large but technically within reach. In fact, the immigration agencies already have eleven years of experience with a mechanism, known as the Basic Pilot, which has led to the development of an efficient computerized system to verify the documents that a new hire presents.44
Both of the immigration reform bills passed by the 109th Congress, one in the House and one in the Senate, contained significant provisions meant to implement this kind of computerized verification nationwide.45 And for all the striking differences in the two bills (one proudly touted as “enforcement-only,” and the other as “comprehensive reform”), they were not far apart in what they envisioned for this particular, crucial component. That harmony is important, but it does not mean that we should simply enact the verification measures appearing in those bills, for they lacked the essential patience we need to practice in calling a nationwide verification system into being. The bills commanded that the new process be up and running for all the nation’s employers within eighteen months from full funding in the Senate’s version or twenty-four months from enactment in that of the House. That task would have entailed ramping up the system from the ten thousand employers registered for the Basic Pilot in 2005 to the full range of 8 million U.S. employers. (Since the 109th Congress, the verification system, rechristened E-Verify, has in fact already expanded considerably, though it is still not mandated by federal law. It had enrolled 100,000 employers as of January 2009 and fields inquiries at a rate of 5 million annually.46 This growth was largely stimulated by recent state laws requiring that employers in the state enroll in E-Verify as a condition of licensing or of obtaining state and local contracts.)
We need a longer lead time because the vital contribution of such a system is not what it can do two years after enactment. The objective instead is the mastering of a difficult problem for decades to come. It is far more important that the new system work well from its opening days, thereby cementing confidence in the verification endeavor (a confidence badly eroded by the inefficacy of the 1986 system), than that it start up swiftly.47 Working well requires more than reliably screening out the persons not authorized to work. Equally important is that the system provide prompt positive verification of valid workers. The biggest complaints about the Basic Pilot centered on the frustrations of those persons wrongly sent to secondary or tertiary verification steps, even though they were entitled to keep working while that process proceeded.48
One other deficiency of the earlier congressional bills was that they failed to provide a dedicated funding source that would consistently ensure adequate resources to the verification system in the future. To be sure, annual appropriations are meant to provide such resources, but the history of appropriations in this field is not encouraging. As noted, Congress has quietly placated various constituencies over the last two decades by skewing resources in the direction of the border (60%) and away from employer-focused enforcement and other interior investigative measures (10%).49
A new verification system may attract abundant resources in the first flush of enthusiasm for a comprehensive immigration reform bill. But once the legislative debate results in new legislation, much of America, assuming that the problem has been solved, will turn its attention elsewhere, making it easy for later Congresses to reduce funding. The computer system has to be sustained with enough resources to keep it at the technological cutting edge. But, more important, a new and truly effective system will generate more temptations for employers either to hire off the books or otherwise evade the system. Therefore, a well-funded audit and follow-up enforcement capacity has to be a critical and perpetual part of the new system. All employers must know that they run a real and substantial risk of detection if they fail to verify an employee or if they enter false information or otherwise misuse the system. Sending such a message requires reliable funding, sustained year after year. The best measure may be to base funding automatically on the intake of specified fees, such as fees on employers participating in any temporary worker system, fees imposed on immigrants who enter in the occupational preferences, or fees for E-Verify inquiries.50
Finally, reform of worksite verification must pay close attention to another challenge that has only recently gained close scrutiny from key officials: identity fraud. If the new system really does reliably reject bogus Social Security numbers and false immigration documents, smugglers will work to steal the identities or documents of persons who have legitimate status. It serves no purpose to create a whole new system for verifying documents only to have it fail its real mission because of such evasions.
Reliable identity documents, bearing biometric identifiers, must be an essential part of the procedure. The 2005 REAL ID Act launched identification improvements by mandating changes and harmonization in state systems for issuing drivers’ licenses.51 Congress evidently focused on the state ID systems to avoid the sensitive charge that it was building a “national identity card”52—but the difference may be more cosmetic than genuine. In any event, more and more states have expressed resistance to the REAL ID mandates, particularly because they have so far been forced to carry the primary expenses for developing reliable identifiers. As a key deadline for interim implementation steps approached in May 2008, the Department of Homeland Security was able to finesse some of that resistance through carefully designed waivers, but this has only postponed a full reckoning on the issue.53 Congress may ease the problem through enhanced federal funding, but the states’ role will remain a sticking point.
The crucial point is that a new employment verification system, if it is to be successful, must incorporate effective measures to ensure that each new hire reliably establish his or her identity. DHS is now in the process of implementing a new system that can provide employers, at the time of verification, with a computer-screen copy of the photo that is supposed to appear on the certain types of documents the individual presents, to guard against impostors who have manipulated a photograph.54 This is only an early and partial version of a fully effective safeguard against imposters, and ongoing refinements and experiments are needed. Should it take a somewhat longer implementation schedule to develop mechanisms that will enhance assurance of a worker’s identity, the added time would be well spent.
If such an employment verification system can be built and sustained, we can move away from the current wrong-headed paradigm of immigration enforcement: catching people and incarcerating them until they are removed one by one. No law enforcement system is healthy if ensuring compliance must rely primarily on direct enforcement through a police officer’s personal attention. Healthy systems are based on widespread voluntary compliance, leaving police to target misbehavior around the margins.
In immigration enforcement, we are miles away from such a system, but the crucial leverage for getting there lies in the workplace. If, after full implementation of a verification system, new unauthorized arrivals cannot readily find work in the American workplace, far fewer will attempt the journey. That audience—people who have not yet migrated—must remain the crucial focus of the endeavor. Deterring such new migration through strong workplace screening is an achievable goal—daunting but achievable. Moreover, the transition to an effective workplace screening system will go far more smoothly and win wider support from the employers who must implement it, if we do not at the same time ask employers to sever preexisting relationships. This means, at a minimum, that the new verification system should not be applied to existing employees but should be focused only on new hires. And it suggests that realism in immigration reform counsels incorporating some mechanism for legalizing most of those currently present, so as to minimize business opposition to this necessary but ambitious transition to a new employment screening regime.
Myth 7: Serious enforcement is at odds with humane immigration policy.
A common vignette in media coverage of immigration issues is the detention and deportation of a person, perhaps based on a years-old deportation order and in spite of relationships to citizen or lawful resident family members (some of these relationships may have come into existence only after the person’s illegal migration, for example, through the birth of a child on U.S. soil). The coverage is usually sympathetic to the individual, for understandable reasons, and it often conveys a message that enforcement—that is, ensuring the person’s deportation instead of allowing him or her to remain—is the antithesis of a humane immigration policy.55
We cannot realistically build American immigration policy on the notion that long residence, even if illegal, must always generate an equitable claim to remain. As the recent backlash against amnesty demonstrates, such a notion flies against a strong popular headwind derived from a widely held (and publicly valuable) aversion to lawbreaking. It will be a sufficient achievement if this round of immigration reform can incorporate a one-time amnesty for a finite population already present. In my view, that can happen only if a wide enough segment of the public is convinced that such an amnesty would truly be a single event—primarily because of the simultaneous deployment of the resources and systems necessary to sustain resolute enforcement thereafter. Such promises accompanied IRCA’s amnesties when they were enacted in 1986, but the enforcement promise was not kept. Hence pledges of enforcement seriousness carry a far heavier burden this time around.
Viewed in wider compass, the only politically durable foundation for generous legal immigration policy in the future is the assurance that immigration is under control. Without reliable enforcement, the political field is open to those who blow the negative effects of immigration out of all proportion and who seek to ride fears of widespread lawbreaking to political success. At times, such efforts even threaten to cut back on legal migration—the easiest part of our overall immigration patterns for a frustrated Congress to affect. After all, we only narrowly avoided a serious reduction in legal immigration ceilings in 1996 as part of that year’s illegal migration control legislation.56
Furthermore, without reliable federal enforcement that greatly reduces the job magnet, frustration with visible lawbreaking also leads state and local governments to experiment with harsh measures meant to discourage illegal migration. Since federal legislation failed in 2007, such measures have enjoyed phenomenal success and have proliferated throughout the country57—but they carry significant disadvantages. Many wind up enhancing hostility toward all foreigners, or encouraging racial profiling, or simply forcing problems on neighboring jurisdictions. Even when carefully administered, they still provide at best a patchwork answer to a national problem. But without credible and determined federal enforcement, we face a future of continued growth in these initiatives. Such a prospect provides another significant reason for supporters of generous overall immigration policy to help develop truly effective federal-level enforcement.
Frustration also sometimes leads Congress to lash out, as it did in 1996, to impose ever harsher measures on the unfortunate few who do fall within the toils of the enforcement system—visiting on them new mandates for detention, exaggerated reactions to minor infractions, or new restrictions on forms of relief that once allowed immigration judges to take account of humanitarian reality and forgive deportation on a case-by-case basis in recognition of rehabilitation or serious family hardship.58 Reliable enforcement would arm the opponents of such harsh measures with better arguments and bring into the fold wider constituencies for resisting such harshness—perhaps someday even for rolling back some of the severity of the 1996 amendments. Reliable enforcement, in short, empowers generous legal immigration policy in the long run. This insight suggests one final myth deserving examination here.
Myth 8: Immigration reform is really only about enforcement.
I have focused here on enforcement, and I definitely want to see the United States develop an effective, balanced, and well-designed enforcement system, over a five-to ten-year time frame, centered around workplace verification and follow-up, though bolstered by effective border policing. Enforcement, for the reasons I have indicated, is a crucial component of reform. No combination of guest-worker provisions or legal migration expansion, short of virtual open borders, will obviate or even reduce significantly the enforcement requirement. Without effective enforcement, demagoguery and polarization on immigration will persist and eat away at generous legal immigration provisions. But enforcement is not really the most important focus for our current immigration reform debate. Instead it is merely a means—a crucial means that must be well crafted and well implemented—toward a greater objective.
The greater objective is to reform our laws so that we as a nation can best take advantage of the benefits of immigration in the twenty-first century. After all, we have at our disposal a huge and globally rare asset because, by tradition and self-image, we are a nation reasonably comfortable with high levels of legal immigration. We readily admit 1 million persons a year for legal permanent migration, and in my opinion, the citizenry would tolerate a higher level if it really felt assured that illegal migration has come under control. Our nation struggles with ongoing problems of discrimination and excess, to be sure, but viewed on the world scene, the United States is one of only a handful of countries that has been reasonably successful in welcoming immigrants, providing real opportunity, and making good use of the restless talent of those who come to take up a new life. Immigration is important to this nation’s economy, its scientific advancement, and its artistic and cultural richness.59 We need to get enforcement right so that we can preserve, nurture, and expand these national advantages.