Genuinely new ideas are rare in international political life, particularly new ideas that happen also to be good ideas. One such prospective new idea has been slowly rising and has recently taken on protean aspirations. That idea is to expand international labor migration as a means not only to boost economies in the developed world, but also in the developing world, especially through the expansion of remittances (money sent by migrant workers to relatives in their home countries). The hope, too, is that by expanding legal migration, illegal migration, with its host of public health and security liabilities, can be sharply reduced. This new “big idea” has come to be characterized as a “win-win-win” proposition: Migrants win by earning higher wages; recipient countries win with additional workers to expand their own productivity and affluence; and sending countries win through remittances and the return of workers who have gained skills abroad.
If this big new idea sounds almost too good to be true, it’s because it is. If migrant labor is to be controlled and temporary, then we are talking about guest worker programs. Guest worker programs have a history, and this history shows that both the economics and the sociology of migration are not so easily controlled. Some good could come from this big new idea, but only if the lessons of the past are understood. In particular, the hope that expanded legal migration will depress illegal migration will be cruelly dashed unless governments work cooperatively to create clear incentives and disincentives, respectively, for legal and illegal migration. This is not easy to do. It would also be a huge mistake to see migration policies as a cure-all form of development policy.
Guest worker programs can play a role in better managing labor migration, but they should be thought of as a last or lesser, not a first, resort. The best way to deal with illegal migration issues is to pass good laws that don’t flout economic realities—and then to enforce those laws. The best way to deal with development is to change economic realities—that is, to generate development inside poor countries, rather than to export their human capital to rich ones.
What’s the Big Idea?
The idea that labor migration could be a good thing, notably to finance economic development in poorer countries, is not strictly speaking all that new. UN High Commissioner for Refugees Rudd Lubbers, for example, urged European nations in the late 1990s to open more doors to economically motivated migrants so that they would not clog Europe’s asylum channels with spurious cases, thus undermining the EU’s evolving refugee and asylum system altogether. Other European leaders at that time pointed to their aging societies and shrinking labor forces to argue for the inevitability of immigration. If immigration was inevitable, all the better that Europeans actively select their immigrants rather than allow immigrants to “select” them.
These concerns led to the Berne Initiative in June 2001. The Berne Initiative was a creation of the Swiss government that brought together countries grappling with migration issues and seeking best practices to improve migration management. Governments shared their concerns and policy priorities, but the initiative did not get broad traction, partly because the U.S. government was not in the least “seized by the idea”—as the saying goes—but also because some developing countries attacked the concept as a one-way effort to reduce illegal migration. While the initiative did little to slow rising illegal migration, it did draw more attention to the problem and stimulated thinking on what could be done about it—all the more so after the terror attacks of September 11, 2001, as three separate strands of policy concern came together.
First, economic development was widely seen as a means of “draining the swamp” or “getting to the root cause” of terrorism. (Liberals were more inclined to see poverty as a key “root cause” of terrorism; others saw it as a lesser contributor to the problem, but worth attention nonetheless.)
Second, the problem of illegal migration in the United States finally spilled over a new political threshold. Migration had always been a “third-rail” issue in American politics, but now it was clear that someone, somehow, would have to touch that rail. On January 7, 2004, President Bush introduced a series of proposals to Congress for dealing with illegal immigration. He characterized his ideas in terms of matching “willing foreign workers with willing U.S. employers when no American can be found to fill those jobs.”
That initiative, in turn, enabled the State Department to begin “thinking out loud” about how to cope with the international aspects of the migration problem. Thus Secretary of State Colin Powell declared before the Bretton Woods Association on September 30, 2004, that we
. . . want to get at the problem of restricted labor migration and remittances. The international community needs to do better at matching labor that wants to work with markets that need that labor. President Bush has taken the initiative here, specifically with regard to the U.S.-Mexican relationship, but it is a worldwide problem that takes a toll on all of us. Because there’s too little legal labor mobility, there’s too much illegal migration—with all the security, public health and humanitarian liabilities that go with that illegal migration. The global economy also pays a cost in the reduced flow of remittances, which contributes more to developing countries each year than all the official foreign assistance combined.
Third, as interest in development rose, disappointment with the traditional methods of generating it was rising, too. In the United States this disappointment led to the Millennium Challenge Account, a program whose underlying assumptions challenged the paradigm of development economics on which the U.S. Agency for International Development had been founded. And when it dawned not only that remittances exceeded the entire amount of official development aid, but also that remittances more readily escaped the webs of corruption that afflict so many governments in developing economies, the makings of the big idea were in place.
Several major new reports reflect this epiphany. An April 2006 report of the International Organization on Migration noted that there is now a growing “consensus that international migration can have important impacts on development, and that it is an important opportunity to help develop policy interventions that will help realize the full potential of international migration.”1 The UN’s Global Commission on International Migration (GCIM), the World Trade Organization’s GATS Mode 4 negotiations and the United Nations’ High-Level Dialogue on Migration and Development have all endorsed more guest workers to ease migration problems.
The World Bank, too, has endorsed more guest workers for economic reasons. Its Global Economic Prospects 2006 report estimates that moving an additional 14 million migrant workers from developing to high-income countries would generate a global income gain of more than $350 billion—that’s more than the $300 billion gain the completion of the Doha Round of trade negotiations would have produced. The World Bank report argues, according to a press release accompanying its presentation, that more “managed migration programs, including temporary work visas for low-skilled migrants in industrial countries . . . would contribute to significant reductions in poverty in migrant sending countries, among the migrants themselves, their families and, as remittances increase, in the broader community.” The GCIM, launched in December 2003, similarly recommended “carefully designed temporary migration programs as a means of addressing the economic needs of both countries of origin and destination.”
Sex, Money & Moving Around
The aspirations of what is often called the “migration-development nexus” rest on some basic facts concerning demography and economics—or what may be expressed a little more tartly as sex, money and moving around. The basic picture starts with some large numbers.
The world’s population in 2004, the last year for which we have verified data from the Population Reference Bureau, was 6.4 billion. Roughly half of the world’s people and half of those in most countries are in the labor force, so the global labor force is 3.2 billion, and the labor force of the United States, which has 300 million residents, is 150 million.
That’s fairly simple; things begin to get more complicated, however, when we note that the distributions of the world’s workers and the world’s economic output are strikingly different. About 85 percent of the world’s workers are in low- and middle-income countries, while 80 percent of the world’s $40 trillion Gross National Income (GNI) accrues to high-income countries with 15 percent of the world’s workers.2 This is a demographic fact that is not going to change soon, except perhaps to become more pronounced. That is because virtually all of the annual 75 million increase in global population occurs in developing countries. As these children enter the labor market, and as the populations of most wealthy countries age, an even higher percentage of the global labor force will be found in low- and middle-income countries.
The existing gap in per capita GNI between high-income and low- and middle-income countries is already very large. In 2004, global per capita GNI was $6,300; it was $32,100 in high-income countries and $1,500 in low- and middle-income countries—a ratio of more than twenty to one. (Even adjusting for the lower cost of living in the developing world, the ratio is greater than six to one.) This wage gap, which has not changed much in the past quarter century, clearly encourages (especially young) people to cross national borders in search of opportunity.
Of course, migration from place to place is an age-old response to differences in economic opportunity, security and other factors—and we all know that in much of the 19th century international migration constituted a major social and political fact in the first age of globalization. But most labor migration before the 20th century took place within political entities, often within large multinational empires. After all, there were only 43 generally recognized countries in 1900, versus almost 200 today. Statistically speaking, therefore, large-scale international migration over regulated national borders is a relatively recent phenomenon. Indeed, the number of international migrants—almost 200 million, about 3.1 percent of the global population—has doubled in just the past two decades. About half of these 200 million migrants are in the labor force of the destination area, and almost two-thirds of the world’s migrant workers—about sixty million of them—reside in high-income countries.3 These sixty million migrant workers make up, on average, about 12 percent of the entire labor force in high-income countries.
Migrant workers who move from developing to high-income countries—as from Mexico to the United States, for example—are different in significant ways both from the workers they join abroad and from those they leave behind. Globally, about 40 percent of the world’s workers are employed in agriculture, 20 percent in industry and construction and 40 percent in services. Developing country migrants are drawn from societies with this 40-20-40 distribution. The industrial countries to which migrants move, however, have only about 3 percent of their workers employed in agriculture, 25 percent in industry and 72 percent in services.
Migrant workers from developing countries who reside in industrial countries have a labor force distribution unlike that in sending or receiving countries. About 10 percent are employed in agriculture, 40 percent in industry and construction, and 50 percent in services, reflecting a tendency of three types of industrial-country employers to request migrants: those in sunset industries, such as agriculture, and light manufacturing, such as sewing; those in industries in which international trade is difficult, such as construction; and those in various service sectors, from janitorial to health care.
Developing-world migrant workers in industrial countries also have personal characteristics that are different from the averages of sending and receiving countries. The best single determinant of individual income in industrial countries is education. In most developing countries, the distribution of adults by education has a “pyramid” shape, with a few well-educated persons on the top of the pyramid and most adults at the bottom, with less than a secondary-school certificate or high-school diploma. Native-born adults in high-income countries have a “diamond” shaped profile when arrayed by years of education. About 25 percent have a college degree, 60 percent a secondary school certificate, and 15 percent less than a secondary or high-school diploma. Migrants from developing countries in industrial countries are different from adults at home and abroad, forming more of an hourglass or barbell shape when arrayed by years of education. About 40 percent have a college degree, 25 percent a secondary school certificate, and 35 percent less than a high-school diploma. In short, international migration from developing to industrial countries takes persons from the top and bottom of a pyramid-shaped education-distribution and adds them to the top and bottom of a diamond-shaped distribution.
Rising demographic differences juxtaposed against mostly stable relative economic conditions suggest that the incentives for labor to want to cross national borders will continue to rise. Meanwhile, revolutions in communications and transportation are making it easier to learn about opportunities abroad, and about how to travel across national borders to take advantage of them. Over the past few decades these factors have combined to produce a sense among many policymakers that migration could quickly spin out of control—or is doing so already.
One reason that the new big idea of the migration-development nexus is so popular is that older responses have not been effective. Like the man with a hammer who sees all problems as nails, the usual policy reaction to growing migration pressures has been to restrict the rights of migrants. European nations, having opened their borders to guest workers in the 1960s and 1970s (to fill factory jobs during the Wirtschaftswunder), restricted family unification and access to asylum in the 1980s and 1990s. For its part, the United States limited access to means-tested welfare benefits in the mid-1990s.
The main result of these efforts to manage legal migration has been to spur illegal migration. In Europe, the number of asylum seekers has fallen over the years, but the share of migrants smuggled over national borders has risen. These migrants now generally destroy their documents and assert they are from countries to which return is difficult, such as Iraq or Somalia. This reinforces the smuggler’s promise that, once inside an industrial country, migrants can stay. In the United States, illegal migration has been high and rising; today almost a third of the 37 million foreign-born U.S. residents are unauthorized. This, in turn, launched the immigration issue into the forefront of U.S. domestic politics. So it’s not as though U.S. and European policymakers have done nothing over the past decade or two. But rather than solve demographic and economic imbalances, all they’ve managed to do is shift around their effects—making some migrants legal, and encouraging more to arrive illegally.
Now, wherever the marketplace of problems is large, the supply of proposed solution will rise to meet it. So there is no shortage of suggestions for how to deal with rising illegal migration—especially in the United States. President Bush’s calls for a new guest worker program echo the earlier European proposals. But the history of European guest worker programs has not been happy. Still earlier U.S. experiences with such programs were not happy either. Before the United States and other countries forge ahead with new guest worker initiatives, the least their leaders should do is reflect on that history.
If they were actually to do that, the first fact that would arrest their attention is this: There are already more guest worker programs and more legal migrant workers in the world than ever before, but there are more smugglers and unauthorized migrants than ever before, too. The United States today issues more than 15 types of visas to non-immigrants that permit employment, from “H”-visas for foreign workers to “TN” visas for Canadian and Mexican professionals with at least college degrees. Despite it all, there are nearly 11.5 million unauthorized migrants here. A similar point can be made about France, Italy and other European countries. Why is this?
Distortion and Dependence
Guest worker programs tend to get larger and to last longer than anticipated (and to attract more, not fewer, illegal migrants) thanks to two interwoven phenomena: distortion and dependence. These two phenomena can stake a claim to universality: They have operated in the U.S.-Latin American context, the European-Mideast and African contexts, the Arab Gulf-South Asia context, and even in the smaller-scale context of Romanian and Thai workers in Israel.
Distortion is fairly easy to explain. Most employers in host countries do not hire guest workers. Even at the peak of the Mexico-U.S. Bracero program in the mid-1950s—when 450,000 Mexicans a year were admitted to work on U.S. farms—fewer than 5 percent of U.S. farmers hired Braceros, and Braceros made up only an eighth of all U.S. farm workers in that period. Nonetheless, the relatively few employers who do hire migrants tend to make significant investments based on the assumption that these “temporary” guest workers will always be here. Farmers may plant fruit trees in areas with few people and then assert that they will go out of business without migrants to pick their crops, hence resisting efforts to reduce the number of guest workers. If governments nonetheless reduce or end the programs, labor costs will likely rise, which may lead to mechanization—as indeed occurred at the end of the Bracero program in the mid-1960s. These distortive changes in the economy would never have occurred if the guest worker program hadn’t existed in the first place. On the other side, illegal migration tends to rise if the government fails to enforce its migration and employment laws—including the prescribed limits to guest worker programs.
A Mexican Bracero, 1959 [credit: Bettmann/CORBIS]
Dependence is in many respects the flipside of distortion. It reflects the fact that some migrants, families and regions (or entire countries) of origin assume that foreign jobs, earnings and remittances will continue indefinitely. During the Bracero program, the Mexican government neglected rural areas in its own west central states, from which most migrants came, and focused its development efforts elsewhere. It did so assuming that rural Mexicans in Bracero-heavy states could always work in the United States.
When the U.S. government unilaterally ended the Bracero program in 1964, the first policy response was for Mexico and the United States to make exceptions to their trade laws and develop tariff-free maquiladora assembly operations along Mexico’s northern border. This did not help ex-Braceros, most of whom were men, since maquiladoras preferred to hire young women. The Mexican men in areas that had not developed alternatives to small-scale farming thus had an incentive to migrate illegally, or face reductions in their incomes. In that way, the Bracero program sowed the seeds of subsequent unauthorized Mexican immigration into the United States. Subsequent immigration magnified the distortion effects in rural America, and also magnified the dependence effects in rural Mexico.
A similar story played out in Germany, where the number of job vacancies exceeded the number of unemployed workers during most of the 1960s. The response was to recruit guest workers from southern European countries and from Morocco to Turkey. The expectation was that Gastarbeiter would work for a year or two on German assembly lines and construction sites, and then return to their countries of origin with savings and skills that would accelerate development.
The employment of guest workers rose as Germany and other European countries kept their currencies undervalued and U.S. consumers bought VW Beetles and other European goods. Japanese auto makers in the early 1970s were rebuffed when they asked for guest workers, encouraging them to take the lead in the use of robots in manufacturing.
When oil price hikes and flexible exchange rates led to structural adjustments in European economies in the 1970s, most guest workers decided to stay in Germany rather than return to their countries of origin, where economic conditions were worse and there was a thinner safety net. Migrants united their families abroad, often settling in ethnic enclaves in which a significant share of residents were unemployed. Children raised in such enclaves are some of the disaffected youth that produce worries about an Islamic underclass. The distortion in the European case was to hire migrants rather than develop labor-saving technologies, and dependence was heightened in some Muslim and African countries as governments neglected fundamental reforms because workers could go abroad.4
As these examples show, distortion and dependence should make governments very cautious about launching new guest worker programs to “get control” of illegal migration. This does not mean that guest worker programs never make sense, but it does mean that governments launching new programs must plan to use economic mechanisms such as taxes and subsidies to minimize distortion and dependence. Taxes can encourage employers to look for alternatives to migrants while subsidies can give foreign workers incentives to return to their countries of origin as program rules require. Without them, we’ll find that there is nothing more permanent than a temporary worker.
Current guest worker programs are rule based, setting out what employers must do to hire migrants and what migrants must do before they are hired and after their work permits expire. Such programs frequently do not work as anticipated because obeying the rules often means employers and migrants must act against their own economic interests. Most programs call for worker rotation, for example, meaning that the migrant must return home at the end of a year or two. Rotation may work in seasonal industries such as agriculture, but factories, hotels and restaurants are unlikely to want to see a trained and experienced worker depart simply because a work permit is expiring. Meanwhile, migrants accustomed to higher wages certainly will not want to return to unemployment or lower wages.
The distortion that arises from investors who assume migrants will continue to be available requires policymakers to recognize that there are always alternative ways to get work done and ensure that products are available. The minority of employers who hire migrant workers usually find them before asking government agencies to certify their need for guest workers. Certification usually involves a sham period of supervised recruitment that rarely yields local workers, which is not surprising because the employer has already identified the foreign workers he wants to hire. Government agencies are ill-suited to second-guess employer recruitment choices—one reason why the labor certification process is often contentious.
Once employers learn how to game the system to get guest workers certified, most do not invest in alternatives to migrants. Migrant-dependent sectors thus become isolated from national labor markets, allowing distortions to increase over time. This is what occurred, for example, in U.S. Western agriculture during the 1950s, when American citizens of Mexican descent moved in substantial numbers from agricultural areas where wages were held down by Braceros to Los Angeles and San Jose, where there were no Braceros. Networks evolve to allow current migrants to bring their friends and relatives to fill vacant jobs. Information soon flows far more freely from a migrant workplace to the migrant’s country of origin than to nearby pockets of unemployment. This is demonstrably true not only of Mexicans in the United States but of Turks in Germany, Kurds in Sweden, and Algerians in France; each group came to dominate jobs in particular sectors and work places.
International norms and most national laws call for migrant workers to be treated equally and be paid the same wages and benefits as local workers. One way to minimize distortion is to realize that Social Security and unemployment benefits add 20 to 40 percent to wages. Payroll taxes should therefore be collected on migrant wages to level the playing field between migrant and local workers; if there are no payroll taxes on migrant worker wages, employers have an incentive to hire them because they would be cheaper.
Berlin’s Kreuzberg district, also known as “little Istanbul” [credit: Reuters/CORBIS]
However, in most cases, guest workers are not eligible for the benefits financed by these taxes, such as Social Security and unemployment benefits. To deal with the problem, guest workers could be given unique tax-reporting numbers so that the employer’s share of payroll taxes on their earnings could be used specifically to address distortion problems. For example, in an industry such as agriculture, it is often hard for one farmer to finance or implement mechanization, since peach packers and processors want either hand or mechanically picked fruit, but not both. Similarly, it may be hard for individual homebuilders to cooperate to develop labor-saving innovations such as off-site wall construction, but such cooperation could be fostered if organizations with representatives from industry, labor and government had the authority to invest payroll taxes collected on migrant earnings to reduce dependence on migrants over time. The important point is that an economic incentive is necessary to have employers continue to seek alternatives to migrants rather than to fall into patterns of reliance on them.5 Otherwise, the economic conditions conducive to guest workers leaving will never arise.
Indeed, this is why, despite sincere plans to return to their country of origin, many young migrants stay longer than planned and form or unite families in the countries in which they work. Few connive from the beginning to overstay their legal welcome; many get trapped by circumstances. To reinforce rules requiring returns after a year or two abroad, the worker’s share of payroll taxes could be refunded when the migrant surrenders his or her work visa in the country of origin. Development agencies interested in using remittances to hasten development could match payroll tax refunds to support projects that create jobs in the migrants’ home area. Taxes paid by migrants who did not return, on the other hand, could be used for emergency health care for migrants and for additional enforcement capacity.
Of course, guest worker programs that minimize distortion and dependence with taxes and subsidies will not work if unauthorized workers are readily available because labor laws are not enforced. Some employers hire unauthorized workers to pay lower wages and save payroll taxes, and they may continue to do so if unauthorized workers are available. Logic might suggest that, if legal guest workers are available, employers will hire them rather than illegal workers in order to spare themselves the risks of being caught violating the law. But during the 1942–64 Bracero program, there were more apprehensions of illegal Mexicans in the United States (4.9 million) than legal Bracero admissions (4.6 million).6 At the time, there were no sanctions on U.S. employers who knowingly hired unauthorized foreigners, and Mexicans who had to pay bribes to get on guest worker lists had added incentives to go to the United States illegally. U.S. employers, too, had incentives to hire unauthorized Mexicans so that they did not have to pay for workers’ transportation or provide them with inspected housing.
The point here is twofold: Laws concerning guest worker programs that spite market realities will not produce their intended effects, and laws that are not enforced will help to create market realities that, more often than not, are perverse from the point of view of the programs’ intent. That was certainly the case with the Bracero program, the ultimate consequence of which is that today there are roughly 11.5 million unauthorized foreigners (of which seven to eight million are workers), over half of whom were born in Mexico. To date, no one has come up with a reliable way to legalize their status without simultaneously making future migration problems worse. The fact that the U.S. Congress has struggled for more than three years over this issue without yet coming close to an agreement is testimony to how difficult the problem is. Alas, if it were easy to design a program to match “willing foreign workers with willing U.S. employers when no American can be found to fill those jobs”, to repeat President Bush’s words, we would already have done so. The conclusion is inescapable: Guest worker programs don’t simply satisfy the needs of labor markets, they also create artificial labor markets that, left unperturbed, produce their own economic and social momentum.
Coming and Going
With illegal migration high and rising despite vast increases in expenditures on border controls, and with the frequent tragedies associated with smugglers guiding migrants via dangerous routes, there is a sense that something new must be done about international migration flows. The allure of guest worker programs as a solution to these problems seems compelling. In receiving countries, the benefits of having migrant workers arrive to fill jobs are immediate and measurable in economic terms. These benefits are concentrated among the employers who hire migrants, but complementary local workers (supervisors or managerial support staff, for example) also benefit, as do the customers of migrant-produced goods and services who pay lower prices. The costs of migrant workers, on the other hand, tend to be delayed, difficult to measure and diffuse. These costs arise mostly from the facts of permanent settlement, the dynamics of family unification, and the difficulties many have in adapting to diversity. With the timing and nature of the benefits being immediate and clear, while the costs are delayed and diffuse, proponents of guest worker programs have a natural debating advantage. And when the programs eventually fail to live up to their promises, it’s already usually too late to find effective and painless ways to deal with them.
The new twist in this old story is the development angle, the notion that sending workers abroad can speed economic development at home. The past half-century’s efforts to accelerate development have produced very uneven results. Some East Asian tigers have grown rapidly and become destinations for migrants from neighboring countries, but most poorer countries have grown only slowly or have stagnated and even fallen backward. The recent upsurge in migration and remittances raises the question of whether more migration from developing to industrial countries will lead to stay-at-home development or will instead produce a set of countries whose permanent comparative advantage seems to be producing workers to send abroad. The data is mixed.
There are cases where remittances have led to significant poverty alleviation, according to studies done by the World Bank. But studies also show that remittances go almost entirely—about 85 percent, according to World Bank estimates—to basic survival needs, leaving very little money saved, spent on education or invested.7 Nor do many guest workers return to their countries of origin with either capital or plans to work; they tend to return home to retire, if they return at all. Thus, the nurseries and nursing homes that dot rural Mexico, or Filipino villages more attuned to events in the Middle East than in Manila, highlight the fact that some migration is more likely to beget more migration than to aid development efforts at home. Moreover, remittances can even have negative effects on development, as for example when the infusion of cash from abroad inflates the value of local currencies and thus harms exporters.
Indian temporary workers in Dubai, United Arab Emirates [credit: Bettmann/CORBIS]
The demographic differences between countries that ensure most labor force growth is in developing countries, coupled with economic differences that put most high-wage jobs in developed countries, suggest that moving more workers over national borders is natural and inevitable. If so, the only question is how to manage international labor migration to make it legal and orderly. Finding the right or optimal amount of international labor migration is the difficult challenge, and managing this migration in a way that does not lead to distortion, dependence and more unauthorized migration is especially daunting. Guest worker programs may be part of the answer, but simply re-launching past programs is likely to produce the same problems, only on a larger scale.
Managing international labor migration can sometimes appear to be an insoluble problem. It reminds some of us of the (admittedly less critical) problem of finding parking for professors on campus. After an especially contentious parking meeting, the chair concluded that those offering solutions did not understand the problem, while those who understood the problem did not have a solution. The solution to the problem of unwanted migration is economic development in both migrant-sending and receiving countries. Well designed and implemented guest worker programs may play a role in hastening such economic development, but only under carefully controlled circumstances. Experience should make us skeptical that such programs can be a quick fix, or any kind of real fix, for anything. ?