5/9/12 - Typepad



To appear in 28 Georgetown Immigration Law Journal #1 (2014)April 2013 draft.The Law and Economics of Family UnificationBy Alan HydeAbstractThe US is internationally unique in the percentage of immigrant visas reserved for family members. The practice has been criticized on the grounds that visas for parents, siblings, and adult children might better be reallocated to skilled workers, and Congress has come close to adopting such reforms. This Article argues by contrast that family unification is a very good economic deal for the US. First, the limited data that exist, contrasting immigrant earnings by type of visa, show no sharp distinction in earnings by type of immigrant visa. This is a surprise to one operating entirely within a human capital framework in which all earnings in labor markets are necessarily a function of skills and education. The paradox is resolved by incorporating three economic models of migration or labor markets that are not usually applied to this debate: (1) New Economics of Migration or family investment models, in which families, not individuals, are economic units; (2) network models of labor markets, in which productivity reflects social ties rather than skills as such; and (3) family economics in which a woman’s economic contribution is often reflected in the earnings of her husband, children, or a family firm. Incorporating these models makes sense of three frequently-observed patterns of immigrant economic activity that illustrate why family unification visas are a good economic deal. Immigrant women frequently perform paid or unpaid intrafamily child care. Immigrants frequently work in family businesses. Immigrant wives frequently work while husbands build small businesses or attend graduate school. In each of these common scenarios, the immigrant admitted on a family unification visa is economically productive and almost certain not to be taking a job from a US worker. However, the immigrant’s economic contribution is importantly realized by the immigrant’s children, spouse, or family business. While skilled migration is crucial to US economic success, there is no reason that it should come at the expense of family unification. The US stands alone internationally in the importance of family unification to its system of immigration law. Family unification is the biggest category of immigrant visas and takes up, through derivative visas, much of other visa categories as well. While long established, this preference for family unification has been controversial. Academics, study commissions, and Congressional Republicans have frequently advocated trimming the number of visas devoted to family unification in order to increase the number of visas available for skilled workers, and some Democrats have shown themselves willing to do this deal. There has been little academic or policy defense of family unification.This Article is the first such defense from an economic perspective. I argue that family unification visas are an excellent economic deal for the US. I need no convincing of the importance of skilled migrants to the US economy. US methods for recruiting skilled workers can be improved in many ways that lie outside the subject of this Article. This Article argues rather that any such increase or improvement in recruitment of skilled migrants should not come at the expense of current visas for family members.The defense of family unification proceeds in three steps. Part III of this Article reviews the limited existing literature that attempts to examine the economic contributions of immigrants depending on the type of visa under which they were admitted. Most show surprisingly little difference, with any such difference at admission dissipating over time. At the same time, other countries’ attempts to direct visas to skilled workers—Canada’s has been influential in US policy debates—have often had disappointing unintended consequences.Part IV summarizes the case against visas for family members, which takes place inside a narrow human capital framework, within the assumptions of which it is tautologically true. In a human capital framework: (A) Decisions to migrate are made by individual maximizers, who migrate when the anticipated benefits to migration exceed the anticipated costs. (B) Migrants enter labor markets in which compensation is entirely or predominately a function of “human capital” or “skills,” typically measured by degrees held or work experience. (C) Unpaid labor is ignored, as difficult to measure and probably reflecting “low human capital.” Under these three simplifying assumptions, common to labor economics within a human capital framework, it is true tautologically that an immigrant with a lot of skills will outearn a migrant who lacks them, and it is just irrational to issue a family member a visa that might have gone to a skilled worker.However, substantial economic analysis departs from these three simplifying assumptions of the human capital approach. This Article is among the first to apply these in the legal literature and certainly the first to apply them to the analysis of visas for family members. Part V will broaden our perspectives to take in three bodies of economic literature that helpfully correct the human capital approach. This enriched economic theory better explain the empirical results in Part III. It shows why, contrary to the tautologies of human capital, immigrants sponsored by family members do well even in existing economic studies, while suggesting simultaneously that those studies almost certainly underestimate their economic contribution. Part V will introduce:A The New Economics of Migration, or family investment model, under which the unit making the decision to migrate is the family, not an isolated individual; B Network models of labor markets, in which individual earning importantly reflects social ties and location in networks, rather than just human capital; and C Family economics, in which unpaid intrafamily care is valued. Similarly, low paid work that grows a family business is given its full value even though reflected in the income of the business rather than the individual. With these powerful lenses before our eyes, we can make sense of the otherwise surprising findings in Part III that immigrants admitted with family ties are just about as successful as immigrants admitted for skills. Part VI assembles ethnographic illustrations. These illustrate three common patterns. (A) Immigrants admitted on family unification visas are particularly likely to perform unpaid intrafamily care, particularly in Latina and Asian families. Their economic contribution shows up, not in their individual earnings (which may be zero), but in the earnings of their daughters and sisters who are able to work full-time, and in the health and well-being of the children in their care. (B) Immigrants admitted on family unification visas often work in family-owned businesses. Their economic contribution shows up in the earnings of the business. (C) Immigrant wives are often in the labor force while their husbands build small businesses or attend graduate or other school. In all these patterns, the immigrant will be making crucial contributions to economic growth that will be reflected, not in her individual earnings, but in the earnings of other family members or their business.Part VII concludes. The Argument, as may be seen, will have ranged through economic data, economic theory, and ethnography. It will involve inference and assumption and cannot be definitive. But it will strongly suggest that the US benefits from visas for family members, and that these contribute powerfully to economic growth and immigrant integration. We should think very carefully before changing a system that has brought such benefits to the US.Family Unification in Current Immigration LawAbout two-thirds of lawful immigration to the United States is authorized because of the immigrant’s relation to a US citizen or lawful permanent resident. The Immigration and Nationality Act sets aside 480,000 visas annually for family-sponsored immigrants. It also creates large categories not subject to this world-wide level, most of which are also for family members. By contrast, the statute caps employment-based immigrant visas at 140,000 annually, a number that has not changed for decades. About 45% of these employment-based visas go to family members of the principal visa beneficiary.The largest single group of migrants comprises the “immediate relatives” of US citizens: spouses, unmarried minor children, and parents of a citizen of the US. In fiscal 2011, 453,158 such immediate relatives became lawful permanent residents of the US. There are no numerical quotas or waiting times for the admission of such “immediate relatives.” A smaller group-- 234, 931--became lawful permanent residents under the family preference categories, which are subject to numerical limitations. Combining the immediate relatives, and those admitted in the family preference categories, yields about 65 percent of lawful migration to the US.Family members also take a substantial percentage of the immigrant visas issued under other provisions of the statute, as derivative beneficiaries of the principal visa recipient. As mentioned, 45 percent, of the paltry 140,000 visas set aside for employer sponsorship or special skills or abilities, actually go to such family members “accompanying or following to join” their spouse or parent. Similarly, 40 percent of the visas issued under the diversity lottery go to spouses and children of the lottery winner. Refugees, too, may bring in their spouses and unmarried minor children.Finally, family ties are important in other aspects of US immigration law. For example, ties to spouses or children who are US citizens or lawful residents may be crucial in considering whether to waive grounds that would make a noncitizen inadmissible or cancel a noncitizen’s removal from the US. Critique of family unificationThis importance given to family unification in US law and practice is long-standing. Throughout all the immigration reforms of the twentieth century, from the era of national origin quotas to the present, there have always been preferences for family members, and their numbers have always dwarfed the numbers available for skilled workers. Such a preference for family unification is, so far as I know, unique to the US. Immigration policy often compares the US to Canada and Australia, two culturally-similar countries whose populations include a higher proportion of immigrants than the US. Both devote a higher percentage of immigrant visas to “skilled” immigrants. International law instruments relating to migration do not protect family unification as a value. Neither the Refugees Convention nor the Convention on Migrant Labor refers to any right to be accompanied by family.Beginning in the early 1980s, the preference for family unification became controversial politically among free-market economists associated with the Council of Economic Advisers and American Enterprise Institute, who argued that immigration in general was or could be positive for the US, but had to be rebalanced away from family unification and toward skilled workers. (Again, this Article does not take on the question of skilled workers, but is focused on the critique of family migration). The rationale for limiting family unification visas varied. Barry Chiswick, writing in a human capital framework, assumed that visas awarded either for employer sponsorship, or for degrees, would naturally attract immigrants with higher human capital than those awarded visas because of their relation to current US residents. George Borjas, writing more than a decade later, understood that the two general types of visas attract generally similar individuals from any given country. He frankly favors reallocating visas from family preferences, to skilled workers, as a way of limiting immigration from Latin America in favor of immigration from the rest of the world.Beginning in the 1980s, Republicans have repeatedly attempted to reduce the number of visas for family unification. Democrats have shown themselves willing to agree as part of larger deals in which they got other things they valued more. Such proposals have come close to adoption.In the run-up to the 1986 legislation, Sen. Alan Simpson, a leading Republican voice on immigration, tried and failed to eliminate visas for siblings. As noted, this did not make it into the 1986 legislation. Simpson returned to the fight in 1988, introducing legislation, this time co-sponsored by Sen. Edward Kennedy, to limit family visas. Sibling visas would have been restricted to those who had never married, and the numbers gained then redirected into a new category of so-called independent migrants who would qualify on combination of education, skills, English fluency and diversity of source country. This category of independent migrants had been advocated by the Select Commission. Simpson-Kennedy passed the Senate but not the House, where business interests, supported by the administration of the elder President Bush, opposed its firm cap on admissions and joined with ethnic and religious organizations supporting family unification. When reintroduced in 1989, Simpson-Kennedy no longer touched family unification. Legislation in 1990 created a new Commission on Immigration Reform which did not really get underway until President Clinton named former Representative Barbara Jordan to its head. The Commission issued a series of reports in the 1990s that represented no clear vision for immigration. One advocated eliminating visas for siblings and adult children for redistribution to other categories. Like almost everything in the Jordan Commission, this sent a highly-confused message about the economic value of all immigration. The eventual 1996 legislation retained visas for siblings and adult children, but instead sought to limit family unification by requiring economic sponsorship and restricting access to public benefits. In its final September1997 Report, following the1996 legislation, USCIR reiterated its earlier recommendations to narrow family reunion. These failed to find Congressional sponsors at that time. However, a decade later, when negotiations around comprehensive immigration reform became intense in 2007, Congressional Republicans once again sought to restrict family unification. One version would have eliminated entirely visas for siblings, parents, and adult sons and daughters. In another version, parents could come in but only if they were over 65, had health insurance, and majority of their children lived in US. Republicans also unsuccessfully proposed increasing the financial liability of the visa sponsor from the current 125% of poverty level to 200%, a level at which 45% of US families would be unable to sponsor a family member. Sen. Kennedy, the leading Democratic negotiator, proposed to reduce backlogs in current categories, then eliminate visas for siblings and adult children (except for 5000 hardship visas for “medical or other hardships that require their relatives’ presence”) and over eight years shift 140,000 slots to a new points system for skilled migrants. It is reported that Sens. Kennedy and McCain signed off on this proposal, at which point Sen. Robert Menendez stormed out of the room. However, negotiations eventually collapsed and no comprehensive immigration reform was enacted in 2007. Republicans have continued to advocate limiting family unification, while Democrats have proposed expanding it. The issue may surface in forthcoming discussions of immigration reform. Neither side, however, has ever put forward much in the way of economic support for its proposals..Economic data on Family UnificationThe thesis of this Article is that, on this issue, Sen. Menendez was right and Sen. Kennedy was wrong. Family-based visas are a very good economic deal for the US. They do not detract from recruitment of skilled individuals and in fact explain why US modes of recruiting skilled workers (employer and family sponsorship) are so much more economically successful than the Canadian points system. Making this case, however, requires three steps and will inevitably involve inference and suggestion. This is because neither the government, nor any other source, collects systematic economic data on immigrants, let alone by type of visa. The 2000 Census was the last to ask respondents where they were born; since that Census, researchers must use the much smaller samples of the American Community Survey or other special surveys. My argument thus proceeds in three steps. Part III of this Article surveys the limited economic literature that attempts, by inference, to compare the economic performance of immigrants according to the type of visa under which they were admitted. While the studies employ different methodologies to estimate this, and do not all agree, on balance they suggest very small differences in economic performance in the year following admission, followed by convergence. Part IV examines the difficulties that a human capital approach encounters in explaining the data in Part III. Part V employs broader economic theory than the customary human capital framework in order to explain the results in Part III, and to suggest further that these may systematically undervalue the economic contributions of family unification migrants. Specifically, we will employ the New Economics of Immigration, in which economic actors are precisely families, not individuals; network models of labor markets, in which compensation partly reflects ties to others, not just individual human capital; and family economics, valuing economic contributions that are realized, as is disproportionately true for immigrants, by other family members or family businesses. Part VI illustrates the economic theories with ethnographies and other social science on immigrant families. It will show that immigrants admitted on family unification visas are particularly likely to perform unpaid intrafamily labor that enables other family members to be more productive; to work in family businesses (so that their contributions show up as returns to the business, not the individual); and to work to support other family members’ education or building small businesses. The result will be a suggestive, but alas not definitive, portrait of the economics of family unification.Of course, one need not base one’s views on family unification on economic data. I am not one of those legal scholars who thinks arguments are illegitimate because they do not seek to maximize output. As noted, neither the case for or against family unification has much invoked economics. The case for family unification normally speaks exclusively in the language of affection and sentiment. I have no objection to such argument, but find it indeterminate, to put it mildly. How close are you to your brothers and sisters? parents? Responses to these questions vary sharply and provide little basis for drawing the preference lines in one place rather than another.So let us turn to the data attempting to link behavior to type of visa at admission. How different are the characteristics of those on each visa? How many family members will the median immigrant sponsor for future immigration? How does the economic performance of different immigrants compare by type of visa?The first thing we learn from the data are that the skills and education of immigrants to the US do not differ sharply by type of visa. Since more visas are available for family members, this has at times been the most efficient way of securing a visa for a trained professional who might have qualified under another legal provision. Such visa strategies are endogenous to changing policies and practices of visa administration. For example, when national origin quotas were eliminated in 1964, relatively few Asians had a US citizen parent, child, or sibling to sponsor them. The first generation typically arrived under employer sponsorship, but this ratio did not last long, as that first generation soon sponsored the migration of their parents, children, and siblings. By the early 1970s, the majority of migration from Asia was on family unification visas, including thousands of professionals who, in education and profession, were similar to the relatives who sponsored them.I think it important to confront directly the power of images. “Immigrant on family unification visa” is a good example of an abstract category that, the cognitive scientists tell us, can be grasped in the brain only through the image of a prototype. I think that our mental pictures of visa categories are likely to reflect racial and ethnic stereotypes. I have not done any research. I am speaking precisely for myself here, identifying a habit of thought that is both necessary, and necessary to struggle against. When I have to picture an immigrant sponsored by an employer, I form a mental image of an engineer from Asia, often the father portrayed by Irrfan Khan in the movie The Namesake. This is probably common and in my case surely reflects my earlier research and interviews in Silicon Valley. And when I say “family unification immigrant,” my mind typically flashes to the family checking in ahead of me the last time I flew Aeromexico, returning to be with family in Mexico, checking enormous, strangely-wrapped parcels while a great many children raced around JFK airport. Even though I obviously have no knowledge of this family’s immigration status, the experience of waiting in line after their protracted check-in made the image indelible. I expect that most readers will have similar visual prototypes, irrespective of the reader’s politics or ethnicity. But, obviously, Asian engineers have parents and siblings, and Latinos have skills sought by employers, and we cannot assume from their position in the US economy that we know what sort of visa brought them to the US, particularly, as noted, given the numerical domination of family unification visas even for Asian immigrants.Second, the data clearly dispel a common trope of the political forces opposed to family unification: fear of so-called “chain migration.” The fear is that each immigrant admitted can potentially sponsor dozens if not hundreds of siblings and children, who in turn sponsor their spouses, and the spouses’ siblings and children, and so on. While theoretically possible, this bears no resemblance to the actual behavior of immigrants to the US, each of whom over a lifetime will sponsor, on average, 1.2 family members.Third, how well do immigrants on different visas perform? A handful of studies use different estimation techniques to attempt to compare immigrant economic performance by type of visa. Jasso and Rosenzweig found little difference in skills (meaning education levels) at the time of entry. Entrants on family unification visas resemble the family members who sponsored them. Recent papers by these authors use the New Immigrant Survey to assemble a clearer picture of sponsorship by recent immigrants. Sponsorship is interestingly related to sending remittances. It appears that recent immigrants sponsor for immigration their more educated children, which the authors interpret as maximizing return to the family, since those more educated children will do the best in the US labor market, with its rewards for education. Less-educated children remain in the country of origin and receive remittances from their US-based parents and siblings. “Thus we find that family reunification is positively selective on skill…” Duleep and Regets compared economic performance, by country of origin, of cohorts who entered in years with relatively greater or lesser proportion of family unification visas. Cohorts with high percentages of family unification immigrants made less money on arrival but converged with the normal over time, that is, had relatively high earnings growth. They also observed a high positive correlation in the percentage admitted as siblings, and both years of education (for immigrants from Asian and Latin America), and propensity to start a business. The authors interpret the data on business startups as showing the importance of siblings’ financial and social capital in immigrant business startups. (We will return to this finding). Scholars analyzing data from Canada and Australia, where family unification migrants comprise a smaller proportion of the total, have similarly found little difference in skills at entry between immigrants admitted for skills and immigrants admitted for family unification. Finally, other studies exploit the difference between the American preference for family unification, and the Canadian and Australian preference for education, and compare results. Perhaps surprisingly, a points system like Australia’s or Canada’s does not result in immigrants that are higher-earning, educated, or more proficient in English, than US system, once country of origin is controlled for.Some studies observe bigger differences among immigrants by type of visa. While as noted, Jennifer Hunt found some groups of immigrants admitted on student visas to surpass similar native-born residents in economic performance, the same was not true of immigrants admitted on family unification visas, who have labor market experiences similar to native-born citizens of similar education. In 1980, the last year for which noncitizens were required to register annually, the Immigration and Naturalization Service asked a random sample to list their class of admission. These were matched with Social Security data. Immigrants who had been admitted during the 1970s on employment-sponsored visas earned more, and were more likely to be employed as professionals, than immigrants sponsored by family members. This study was not published until 1992, however, by which time it was clear to its authors that it represented a snapshot of the 1980 immigrant population, not a timeless verity. Employment-sponsored visas, half of which in 1976 went to professionals, came to include a higher percentage of less-skilled workers in the 1980s until this category was itself eliminated in 1990. I would also add that a 1980 snapshot would have just begun to pick up the numerically-significant category of professionals from India, Korea, and the Philippines who nevertheless entered, increasingly in the 1980s, under family sponsorship. Despite these ways in which a 1980 snapshot might have exaggerated the economic advantages to employer sponsorship, the differences were still not great. Family-sponsored migrants were just as likely to work as employer-sponsored migrants, and were less likely to be taking jobs from U.S. workers or exerting negative effect on their compensation.So the picture is cloudy. But isn’t there something perverse about these studies that find little or no difference in the economic performance between migrants specifically selected for their skills, typically meaning education, and migrants selected for family ties that have no necessary relation to skills? How can it be that letting employers select for skills admits immigrants who differ only slightly from those admitted by relatives? We noted the suggestion of Jasso and Rosenzweig that US-based parents are more likely to sponsor their more educated children, but surely this is only a small part of the story.A bigger part of the story is that immigrants who arrive sponsored by family arrive with networks and connections. This is why, irrespective of visa (or even whether their migration is authorized at all), migrants with family ties do better than similar migrants who lack them. Immigrants who arrive sponsored by employers have a connection with that employer, but not necessarily anywhere else in the economy. And immigrants who arrive in Canada or Australia under those countries’ points system have no necessary connection with anyone in the economy—and their dismal economic performance reflects this. Recall that the McCain-Kennedy agreement in 2007 would have reallocated visas from family unification to a new points system on the Canadian model. Adoption of such a Canadian system, long a favorite of George Borjas, was a major Republican demand in the 2007 debates. Examining the failures of that system—as Congressional Republicans refused to do in 2007--will help set the stage for our search in Part V for a richer economic theory of immigrants’ economic performance.The Canadian points system in place in 2007 was a disaster, precisely because it admits migrants on the basis of degrees and work experience, without any assurance that there is any need for these “skills” in the Canadian economy.The system was heavily backlogged. In 2007, 800,000 applications were pending, with wait times of 4 years or more. Without any need for employer sponsorship, thousands apply. Most importantly, the program admits engineers, physicians, and other highly-educated migrants without any concern for whether these skills are needed in the Canadian economy or these degrees valued by Canadian employers. The result is that shockingly few of these professionals ever practice their professions in Canada. I had heard the Vancouver joke that, if you need to see a doctor in a hurry, go the curb and yell “Taxi!” This turns out to be no joke at all. A physician from India who migrates to Canada has a 19% chance of working as doctor during five years after admission; from Eastern Europe 8%, China 4%, Philippines 3%. Only 65% of Canadian economic immigrants are working at all, fewer than 30% as professionals. While the Canadian points system at that time thus admitted thousands of engineers and physicians to Canadian unemployment, it was neither flexible nor efficient enough to obtain oil field workers for Alberta, who must be skilled, but do not typically have advanced degrees. As a result, such oil field workers, and other skilled workers without graduate degrees, are admitted to Canada as temporary workers, even though their jobs are in no sense temporary and even though this opens up possibilities for abuse of the worker. Republican obduracy in insisting on the adoption of a points system in the US is, obviously, the triumph of ideology over evidence. This Article will anatomize that ideology: it is envious of paper skills and blind to the more important indicia of economic success: ties, networks, family support, child care. The Economic Critique of Family UnificationSo the data present us with a puzzle. Immigrants admitted because employers want them, or because they are well-educated, should, it would seem, logically do better economically than immigrants who gain admission simply because they are related to US residents. This would seem to be one of those assumptions that should just naturally be true, and it certainly sounds plausible to many people of good will who want nothing but what is best for the US economy. Yet the data instead suggest that the differences are not great between the economic performance of US migrants admitted under family sponsorship and US migrants admitted under employer sponsorship, while attempts to admit strictly for degrees omit an important dimension.The critique of family unification visas is true tautologically under a human capital framework. Since the purpose of Parts V and VI will be to broaden or supplant such models, I feel I owe human capital models a decent introduction.Legal readers are familiar with the basic outlines of human capital economics, though the concept-metaphor is normally introduced without any indication of how controversial it is within the labor economics profession. Human capital models originated to explain why individuals or societies invest in education. Instead of treating this as consumption, a human capital model treats it as an investment. If it is a rational investment, it must pay returns. In human capital models, individual earnings in labor markets are entirely explained as returns on each individual’s investments. “[T]he essence of human capital theory is that the job is unimportant. Wages and wealth are determined by the individuals’ skills. Occupation and industry variables are secondary and almost an embarrassment to the theory.” Thus, the proposition that the migrant with advanced degrees should be expected to earn more than the migrant who lacks them is not an empirical question in a human capital model. It is tautologically true under the assumptions of the model. Since the data suggest otherwise, we need to broaden the assumption that economic success is entirely a function of degrees held.Human capital models cannot explain migration. If we must model human capital as an asset, it is a peculiar one. Normally assets are worth more in markets in which supply is scarce. Human capital, by contrast, is worth more in markets in which supply is plentiful, and thus migrates from labor markets in which it is scarce, to labor markets in which it is plentiful, as this author has pointed out.V Economic Theory of Family Migration: Interpreting the Data Through Three Economic TheoriesThe reader undoubtedly has intuited that the economic case for family unification will involve treating the family (rather than the individual) as the unit of analysis; understanding the role of networks and personal ties in labor markets; and properly valuing unpaid labor. At the risk of belaboring what to some readers will be obvious, let me sketch some of the substantial economic literature on just these three points. These are well-established moves in economic analysis. They just haven’t been applied to understand the economic performance of migrants.To make sense of the above data and explain why family migration is usually economically successful, we must add:A. Methodological individualism (neoclassical) should be supplemented or replaced by New Economics of Migration (discussed Part VA).B. Models in which immigrants necessarily substitute for domestic labor must largely be replaced by models of complementary labor (discussed Part VB).C. Family economics: Understanding the economics of unpaid intrafamily care and unpaid contributions to family businesses (discussed Part VC). A. New Economics of Migration: Families are Economic UnitsIf one starts from the assumption that the only economic actors who matter, in migration or anything else, are individuals (methodological individualism), then it would seem uncontroversial that such isolates should be as skilled as possible. It would just be irrational to prefer an unskilled migrant over a skilled migrant. Such methodological individualism is normally a part of a human capital analysis. On this view, decisions whether or not to migrate are made by individual maximizers familiar from economic theory. They weigh the costs and benefits of migrating and migrate when the benefits outweigh the costs. Such models face an immediate embarrassment. They would seem to overpredict the amount of migration that we observe. Almost everyone in the world could substantially increase his or her earnings by moving to the US. Why don’t all these individual maximizers, maximize their way to the US or a similarly-developed economy?Much economic analysis of immigration by professional economists solves these problems within a framework styled family investment models or the “New Economics of Migration.” In this framework, the principal economic actors are families, sometimes communities, not individuals. Individuals normally live in the world with ties to others and it is only an economist’s coup de main that severs these. It turns out that this coup de main does not improve the analysis.In the New Economics of Migration, households diversify investment, and protect against risk, by having family members work in different labor markets. The source of income matters lest the family be at risk when a shock to one labor market results in lost income for all family members.While neither framework is without value, the New Economics certainly does a better job explaining the immigration that we observe. In a neoclassical model, individual maximizers perform cost-benefit analyses of staying or going. Since nearly everyone in the world would make more money in the US than at home, why don’t all these maximizers move? The model requires, as we have seen, an assumption, that individuals place enormous subjective costs on moving (or benefits to staying), just to get off the ground. The New Economics better explains observed migration, which so frequently involves young people traveling long distances and sending remittances home. The New Economics also better explains migration patterns from Mexico. Migrants do not come from the poorest regions of Mexico, individuals for whom migration would represent the greatest ratio of benefit to cost. They come from midsize towns and cities, where diversification is particularly rational. But if the New Economics understands migration as family diversification, how does it understand the family decision to reunite in a foreign land—seemingly the opposite of diversification? We will postpone this analysis until we get a few more economic tools onto the table, but, as we have already hinted, family unification, like emigration, is a rational family investment strategy, for example when it represents intrafamily division of complementary labor (one relative watches the others’ children), or similar division of labor in ethnic networks or communities. We will see that most economic behavior by immigrants, such as business formation and women’s labor force participation, is best modeled as the result of family investment decisions rather than decisions by individuals.B. Immigrants Complement Rather Than Substitute for US Labor. Network models of labor markets vs. human capital models. The analyst who demands that each individual migrant be individually skilled (possess high human capital) is necessarily assuming that the migrant is likely to replace a US worker doing the same job. Under this assumption, but only under this assumption, immigration policy must steadfastly ensure that the replacement have higher skills, not merely lower wages.The assumption that immigrants typically replace (substitute for) native labor is another one of those assumptions that has been substantially exploded in the economic literature. It cannot explain the immigration that we observe, which so often exploits the gains from networks of complementary labor. The point is well established in the economic literature. In a world of globalized communication technology, human capital theory cannot explain migration at all. We have noted that a software developer who wins the lottery for an H-1B visa—let’s call him Rajiv-- will make six times as much in California as his identically-educated coworker who works on the same team in India. Rajiv has the same human capital in San Jose as in Bangalore. The only point in bringing him from Bangalore to San Jose is that he will be more productive in San Jose, and that can only be because the proximity to other individuals on his team, in his firm, or across firm lines will make him more productive. Without those networks, his human capital, which we assume to be high, is worth no more in San Jose than in Bangalore. Inside those networks, his earnings will sextuple in the US. Obviously the migrant is not substituting for a native developer. (We might be more concerned that the poorly-paid developer in India, like other foreign labor to whom work is outsourced, substitutes for US labor). Rajiv is more valuable in San Jose because of his position in networks.As noted, most professional labor economists understand that compensation in labor markets partly reflects institutional characteristics, not merely individuals’ human capital. I believe that I am the first to describe labor markets as networks, in what might not be a felicitous metaphor. The idea is to analogize labor markets to markets that economists have analyzed, such as the markets for communication technology, in which the value of an asset increases as the supply increases. In the classic aphorism, one telephone is an expensive paperweight. With two one can communicate, and the telephone is worth more with each user. Markets for information exhibit similar network effects: increasing returns to scale under which information becomes more valuable if shared. Similarly, as we have seen, high human capital workers maximize gains from that human capital by migrating from regions where it is scarce to regions where it is plentiful, at which point they compliment, not substitute for local labor. The use of the term “network” is only a metaphor, since in a labor market, unlike a market for information, the marginal cost of reproducing a unit never becomes zero.However, it is a powerful metaphor. Understanding high-tech labor markets makes it clear that this industry, which heavily employs immigrant labor, hires for networks, not for human capital; for complementary, not replacement, labor. If Rajiv continues to work in information technology in California, he will, if he is like the median IT professional, work for his employer for six months. However, Rajiv will continue to be of value to his old employer even if he leaves after six months. His old employer’s patent applications will be much likelier to cite patents from the firms to which Rajiv moves. His old employer will call Rajiv for advice about problems they both face. His old employer may outsource work to Rajiv’s new employer or perhaps to Rajiv himself, perhaps if he returns to India and starts a firm to handle jobs outsourced from outside India. The economic importance of these, and other, channels of diffusing information across firm and national boundaries through employee mobility, explains the significant advantage in economic and technological growth that is achieved when jurisdictions forbid enforcement of covenants not to compete. Most observed immigration to the United States follows this pattern: the migrant’s skills complement, rather than substitute for, labor already being performed in the US. (We can be confident of this because, when US employers actually want to replace US workers with cheaper foreign labor, they do not embark on the expensive uncertainty of obtaining visas, often including the bizarre and cumbersome certification process of the US Department of Labor. Employers that want to replace US workers with cheaper foreign labor outsource work to foreign facilities or employers. This is done openly, notoriously, and completely without shame; it does not require certification from the US Department of Labor or the approval of any other governmental agency or labor union; and normally is rewarded in financial markets’ valuation of company shares).In short, it is easy to misunderstand the economic contribution even of immigrants sponsored by employers. Their contribution often reflects their position in networks and ability to complement existing labor. So it will not be surprising to find exactly these overlooked mechanisms at work when visas are issued to family members, who arrive networked in the economy and almost never displace U.S. labor. In other words, were the U.S. to adopt some variant of the Canadian points system (something I do not recommend), issuing visas to applicants who amass points for degrees and work experience, it –in order to avoid having physicians drive cabs—would have to find a way of awarding points for ties and networks to the U.S economy. Instead, we achieve this result by awarding visas instead to family members.C. Family economicsBy “family economics” in this Article I mean only that women should be counted, and that unpaid labor should be valued. In particular, women’s economic contributions often show up, not in her earnings, but in the earnings of another: her husband, children, a family firm. Most of the attention in feminist economics has gone to “unpaid domestic labor.” But women also perform “unpaid market labor”, for example in family businesses, which is conceptually distinct from “unpaid domestic labor.” Immigrants, as we shall see, are particularly likely to do both.Applying the theories: three common economic activities by immigrantsThe relatively similar economic performance of immigrants sponsored by relatives to those admitted for skills may now be explained. In almost every context, family migrants arrive networked; complement rather than substitute for domestic labor; take jobs that advance family investment rather than individual advancement; and may make economic contributions that are not paid as such but show up in increased earnings for family members or family businesses.We will discuss three particularly common contexts for these observed economic results: immigrants who (A) care for children and other family members; (B) work in family businesses; or (C) support other family members who are in school or building family businesses. These are all “family investment” models. Immigrants who care for children, or build family businesses, function economically as part of a family, and seek to maximize family income or growth. They may be paid little or nothing, and thus their economic contribution will be systematically underestimated in snapshots limited to immigrants’ earned income in their first year after migration.Caring for ChildrenShould Congress ever eliminate, as it nearly did in 2007, visas for parents, the immediate impact would be a radical decrease in the availability of child care for immigrant and native families. Working families in the US are heavily dependent on relatives for child care; Latina families are even more dependent; and immigrant families seem to be particularly heavy users of relatives for child care. Moreover, immigrants also disproportionately perform paid child care and thus contribute to increased earnings for working mothers.The Government collects good data on who is caring for America’s children through several instruments: the Census; a special Early Childhood Program Participation Survey; and the Survey of Income and Program Participation. In addition, private groups, such as the Urban Institute, conduct their own research. All their findings are broadly similar and I will not discuss technical differences among the studies that might account for the relatively slight disagreements.Working families rely heavily on relatives For many working families, child care is grandma and nobody else. The US, unlike many European countries, does not routinely provide free public preschool education. Among all US children of pre-school age, 26.3% receive care exclusively from parents; 57.2% are in some kind of organized program; 22.6% receive care from other relatives; and only 11.6% receive any care from a nonrelative. When mother works and the child is pre-school aged, grandparents and organized programs are just about equal, each reaching about 30% of such children. When mothers work, fathers care for 25% (fewer than grandmothers), siblings for 3%, and other relatives 8%. Nationally, close to half of all grandparents provide some child care assistance, averaging 23 hours per week.The Urban Institute recently looked in depth at child care arrangements among working families in Providence and Seattle. Sixty percent of the families in the survey contained an immigrant parent, but these families were broadly similar to the others in the survey. Two-thirds of these working families depended on family members for child care. Other countries of heavy immigration have similar experiences.Latina families depend on family members for child care even more than other working familiesThe Government’s surveys on child care arrangements do not specifically distinguish immigrant families. They do, however, break out Latina, African-American, and Asian families for separate analysis. Since Latinos in recent years have been the largest group of migrants to the US, we can use these surveys as a rather noisy look at immigrant families, particularly since, as we shall see momentarily, the surveys directed at immigrant families confirm the picture.Latina families, for reasons on which there is no consensus, are particularly low users of formal child care programs, and particularly high users of family members for child care. The survey of all preschool children reveals that Hispanic families are low users of formal programs and nonrelative care: All preschool childrenHispanic preschool childrenParents exclusively26.3%38.0%Other relatives22.622.7Organized programs57.243.4Nonrelatives11.6 8.1Similarly, the survey limited to preschool children of working mothers shows heavy Hispanic reliance on family members, less on formal programs:All preschool childrenHispanic preschool childrenParents exclusively24.8%25.4%Grandparents20.928.5Siblings and Other Relatives 6.313.1Day Care Centers19.212.8. Only among Hispanic families is child care by aunts and older sisters statistically significant; for white and Afro-American families such arrangements are rare. It is precisely such adult siblings who are the most frequent targets of proposals to trim family visas.Immigrant families, particularly Latina and Asian, are heavy users of family child careThe last US Census to ask respondents their country of birth was 2000. While the Census web site does not make it easy to search only among families with a foreign-born member, a team of demographers at the Center for Social and Demographic Analysis SUNY Albany has helpfully done so for some data categories. They broke out, as “immigrant families,” families with at least one foreign-born parent. Children in immigrant families are significantly more likely to live with two parents: 84% of children in immigrant families vs. 76% of children in all-native families. Children in immigrant families are two to four times more likely than children in native European-American families to have a grandparent in the home. And, among some immigrant groups, aunts and uncles living in the home is common: 25-37% of children in families with at least one parent from Mexico, Central America, Caribbean, South America, Philippines, Indochina, Pakistan, Bangladesh, Afghanistan, Iraq, or Africa. Again, it is precisely these elderly parents, and siblings (that is, aunts and uncles), who were targeted for elimination of visa eligibility in earlier legislative proposals. Immigrants are also low users of formal programs and concomitantly highly reliant on family members for child care. One working immigrant (ethnicity alas undisclosed), “Brianna, whose sister takes care of her daughter, assumed that her sister would be available purely on account of their relationship. When asked how they came to an agreement where her sister would watch her daughter, Brianna said, ‘We don’t agree. I just drop her off.’”Studies of the Asian immigrant population reveal the crucial role played by grandparents, especially grandmothers, in child care. “[M]any Asian-American elders moved to the United States to perform childcare tasks for daughters or daughters-in-law who work in a family business.” While by this author’s reckoning, 11% of grandparents nationwide care for grandchildren, the figure is three times that high for Asians. A survey of these grandparent caregivers revealed that the majority lived with one or both of the grandchild’s parents. Nearly all supplemented care by those parents but were not primary caregivers themselves. Just over half cared for two grandchildren. The most common pattern was for both parents to work in a family-owned small business and for grandparents to provide child care during the day when both parents work. So when a U.S. citizen family succeeds in getting a visa for the wife’s mother to come care for the grandchildren—typically to enable the wife to move to full-time employment—all three of our alternative economic models are illustrated. The immigration decision represents a family investment. Grandma performs labor that complements the labor of US workers, and does not take a paying job from a US worker. And Grandma’s own labor is not compensated.But how does this show up in the immigration statistics? Suppose for ease of illustration that the visa sponsors are naturalized immigrants from Colombia and that Abuela is a Colombian national. Abuela may never have finished high school. She arrives in the US on a much-maligned “family preference” visa. She is a “low-skilled” or “unskilled” migrant. She may not have been in the paid labor force in Colombia. She is definitely not in the paid labor force in the US. Her tax returns will show earned income of zero in the year she arrives in the US, and every year thereafter until she enrolls for Medicare. At this point, we may imagine that George Borjas sits up and says, see, I told you that today’s immigrants are lower quality than the immigrants of a generation ago.But nothing could be further from the truth. Abuela is “unskilled” only because society does not value the skill of taking care of small children. She has earnings of zero because nobody pays her for this important skill. But, because Abuela is here, her daughter can earn more for the family. And Abuela is in no sense “taking a job from an American.” She performs classic complementary labor, except that hers is uncompensated. Is child care by grandparents and other family members good for children and society?We have tried to put a human face on three categories of family unification that have frequently been targeted for elimination: parents, siblings, and adult children of U.S. citizens. There can be no doubt that elimination of such visas would force hard-pressed working families to scramble for child care, and lead parents to reduce their working time, or devote more resources to child care. But would it be a good thing, on balance, to deny US families a visa for Abuela, if it increased immigrant participation in formal child care programs? Perhaps it is enough to say that our society has made no such value choice. Nor am I aware of any data showing any harm to children from being raised by relatives. All over the globe, involvement by grandmothers and other female relatives is correlated with higher survival and better outcomes for children. Examination of two studies of immigrant child rearing suggests that care by grandmothers, aunts, and big sisters may instead be highly functional, in ways scholars are only beginning to understand. Consider, for example, the urban working poor households studied by Katherine S. Newman, each with members working in the fast food industry. She well describes the somewhat chaotic family life of an extended immigrant family from the Dominican Republic. At the beginning of the book, a teenager from this family had dropped out of school, left work, is married to a man who makes little, and in a difficult pregnancy. “Fortunately, Carmen was literally surrounded by her father’s kin, organized into four or five related households, all living on adjacent floors of the same building in Washington Heights. She did not have to worry about starving, and when her phone was cut off…she could use her aunt’s phone (as did all the other members of the family.)” And, at the end of the book, this is the only family to be poised on the edge of economic security. One aunt in Michigan calls to say that the factory where her husband works is hiring; some of her sisters join her; in their absence grandparents and older siblings care for the younger grandchildren. Consider also the so-called “Latino paradox”: Latin immigrants are much healthier than their incomes would suggest. An example is the potentially puzzling finding that rates of asthma among Latino immigrants in Chicago are sharply lower if they live in a community numerically dominated by Latino immigrants. The puzzle disappears when one remembers the crucial role of unpaid intrafamily or intracommunity care. Care of asthma requires family members to take the child outside; support the patient; share health information. Such unpaid intrafamily care may be the key to the entire “Latino paradox.”. B. Working in family businesses: network economics; family investment model; complementary laborImmigrants sponsored by family members are particularly likely to work in family businesses. They may or may not have reported income (unlike the unpaid child care worker). However, their reported individual earnings will undervalue their economic contribution, which will show up instead in the income of the business.Immigrants are particularly likely to found businesses, mostly small—twice as likely as similar native-born. This ability to found businesses depends on a supply of relatives. Family members support business formation by providing labor; pooling financial resources; pooling living arrangements; intrafamily loans; highly-productive labor despite low wages; “family labor can be trusted to handle sensitive transactions in which the risk of opportunism and malfeasance is high.” While English-language proficiency is normally positive for immigrant business formation, its advantage disappears for immigrants whose compatriots are geographically concentrated: that is, immigrants who are not proficient in English may have the option of founding a business to serve the community whose language they speak. While a comprehensive picture of family business is not possible, ethnographies illustrate some of the ways that immigrant businesses benefit from visas for parents, siblings, and adult puter distribution in Los Angeles is completely dominated by family businesses of Chinese from Taiwan. New immigrants are integrated into the family business despite the fact that none had any contacts with computer distribution in Taiwan. Koreans are similar in their ability to integrate adult siblings into business networks. Pyong Gap Min describes a Korean who immigrated in 1970 under employer sponsorship, worked for a jewelry company for three years, then started his own beauty supply store. After he naturalized, he and his wife sponsored all their siblings, seven in total, who arrived between 1978 and 1987. All seven newcomers were trained by their siblings to run beauty supply stores, extended loans by the first to arrive, and all seven established beauty supply stores within six months of arrival. Stories of this kind could be multiplied. What do they prove? The firmest point, but the narrowest, is, again, the critique of conventional accounting of whether immigrants are succeeding or failing. The conventional measure, earned income in the year after arrival, is plainly too narrow for a sector statistically prone to found new businesses that, like all new businesses, may struggle in the early years. Immigrants probably take out little in the way of salary—for example, wives may not be paid at all—in order to build the business. So our accounting for immigrant success may need adjusting.For the larger question of immigration policy, however, the story may be ambiguous. There may not seem to be all that much policy difference between the Canadian professional immigrant who arrives without family or networks and thus drives a taxi in Vancouver, and the US counterpart who, by dint of family connections, opens a beauty supply store in the Bronx. Small businesses loom large in the national imaginary but one might question how important they should be in a development or immigration strategy, outside of high technology. While the many technology companies founded by immigrants from Asia have captured the imagination, I have never heard of one that was founded by an immigrant who arrived in the US as an adult on a family unification visa. Occasionally a technology company or its founder thereafter sponsors the founder’s family members for immigration, but this is rare. One cannot, in other words, argue that family unification visas contribute much to the founding or maintenance of technology companies. So is it a good deal for the US to admit parents and siblings of US citizens on the assumption that many will work in family businesses?Comprehensive examination of immigrant small business as a justification for family unification visas lies well beyond available data and certainly beyond the scope of this paper. We are just beginning to understand the role of the immigrant-founded small business beyond the first generation. My Rutgers colleague Jamie Lew has studied Asian young people who drop out of high school in New York. The US economy is generally cruel to high-school dropouts. Nevertheless, “Most of the Korean American high school dropouts worked in menial service jobs in the ethnic economy—working mostly for other Korean entrepreneurs as manicurists, cashiers, valet parking attendants, nightclub bouncers, and the like. With their ties to ethnic enclaves and immigrant networks, these jobs were relatively easy to secure.” I do not think we understand all the ways the economy benefits by giving visas to siblings and adult children who start such nail salons and restaurants, and we should probably think very hard before altering that system.C. Women working dead-end jobs while men build their careers or businesses: family investment modelA final family investment model was first described by Baker and Benjamin. Using Canadian data, they observed the following pattern. Immigrant women participated in the paid labor force at higher rates than native women, often in jobs offering little future growth, such as child care. Immigrant men, as in the U.S., start businesses at higher rates than similar native men. Baker and Benjamin described this as a kind of family investment model in which both wives and husbands seek to maximize household income. Wives support men while men invest in the future, perhaps by graduate education, perhaps by building a small business, in either case commanding low initial returns but larger potential. The authors show that this family investment model (akin to the New Economics of Immigration) better explains immigrant wage patterns than other economic models, for example neoclassical. A similar pattern has been noted for Australia. Like any family investment model, the economic contribution of the Baker-Benjamin model is underestimated when immigrant economic contribution (or “immigrant quality”) in evaluated by examining only earned income in the year of arrival.Francine Blau and associates have shown that the Baker-Benjamin family investment model is not consistent with the overall picture of immigration to the US. It is true that immigrant women have high labor force participation, and, as we have seen, that immigrants disproportionately found businesses. However, for several reasons, it does not seem the best interepretation of the data generally, to understand women’s labor force participation as a family investment model. Women’s labor force participation seems to respond to opportunities. It is not very different from men’s labor force participation. Women and men immigrating to the US both invest in their own human capital. Women’s labor force participation increases with assimilation to the US; it is not a temporary measure while husbands invest in education or businesses. But if the Baker-Benjamin family investment model is not the best overall model of labor force participation by women immigrants to the US, the pattern that it describes is certainly part of the mix. It is not difficult to find examples of women in non-career jobs while men are in graduate school or building a business. Further research might disclose specific subgroups of the US immigrant population that display the Baker-Benjamin family investment model. CONCLUSIONOf course the US benefits economically when its doors are open to those who promise unusual creativity or entrepreneurial success. As mentioned, the first priority in increasing such skilled migration should award permanent residence, not just nonimmigrant work visas, to those who come to the US for postdoctoral work. The US could also experiment, if it likes, with some version of the Canadian points system for those who have accumulated degrees and experience in their own countries, although for reasons suggested in this Article, such a points system should include points for US experience, employer sponsorship, US family, and other indicia of network connections in the US economy.But neither these, nor other programs that one might imagine to encourage migration of skilled workers, should come at the expense of existing visas for family unification. In particular, the ability of US citizens to petition for their parents, siblings, and adult children, is probably a very good deal for the US. Conclusions must be tentative, given the limited data that actually compares immigrant economic performance by visa type at admission. We have attempted to draw plausible inferences from other data sets. The picture cannot be definitive, and more study has been suggested. Some things are clear. If visas for parents of US citizens were eliminated or trimmed, many working families would scramble for child care. Mothers would trim or eliminate their work hours, whether inside or outside a family business. The bodega or greengrocer would not be able to stay open as late, eliminating its chief comparative advantage over the supermarket.It is harder to form a complete economic picture of immigrant siblings of US citizens. However, we know that these aunts are a statistically significant source of child care in Latina families. Indeed, under a family investment model, it may be rational for the most recently-arrived aunt to watch all the children while her sisters, better networked and more proficient in English, are in the paid labor force. We also know that siblings are particularly likely to work in family businesses.Overall, then, it appears that the US is well-served by family unification. Our search for the skilled and talented should not blind us to the crucial kind of economic growth that comes when children are well cared-for and educated; their health needs observed and attended; when family businesses employ family members and obviate their demand for social services; when family members pool earned income while one builds a small business. This kind of economic growth can be observed, with our eyes of course, but in diffuse statistics on the new economics of migration; complementary labor; employment in networks; and the economics of unpaid work and care. Economic statistics that don’t observe this economy will underestimate the economic value to the US of its policies on family unification. We should hesitate before changing policies that have served us so well. ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download