I am very much honored to be the keynote speaker here at this conference on immigration. I have been asked to talk about immigration in the context of the current economic crisis. But I find it very difficult to speak to this topic here in Spain: My own understanding of both immigration and of the current crisis is based primarily on the experience of the United States. Moreover, the crisis has taken different forms in different countries, and even withinEuropethere is substantial variation. So I am going to try to lay out a general framework for thinking about this problem without pretending to capture the specific experience of each country.
It is conventional to understand migration in terms of an income differential between the sending and receiving countries; this differential constitutes the reward or return to migration. That return must be balanced against cost of movement. Short-run policy typically looks to control the migration process by increasing the cost of policing the borders. And within this framework, the crisis is typically thought of as lowering the return by increasing the difficulty of finding work or reducing the wages of jobs which might still be available. Since wages and employment opportunities for migrants are difficult to control in general, but certainly without affecting national workers and aggravating the crisis, policy tends to focus on border control and on undocumented workers. Certainly this has been the case in the United States.
But I would submit that international migration has to be understood as structural, embedded in the technological trajectory of industrial society and the sociology of the labor force. Thus, historically, mass international migration has grown out of the interaction between the economies of the advanced developed countries, which generate a demand for labor with certain characteristics that are difficult to find within their own territory, and foreign labor that has a very different perspective on the labor market from that of national workers (Piore 1979).
For a long time the trajectory of the industrial world was fairly clear and well defined, and the problems which that trajectory posed for continuous economic growth and development evident. As a result, the potential role of international migration and its advantage and disadvantage as a solution to these problems was clear as well. The trajectory of development is no longer so clear, and that I think is the challenge for migration research and for immigration policy. The challenge I should make clear is really twofold: On the one hand it is the confusion about what exactly the trajectory looks like; and on the other, the challenge that the shift in the trajectory poses for public policy.
Under the old industrial trajectory, the problem was that the economy had a tendency to generate a series of jobs that national workers found unattractive, often so unattractive that they felt the work degrading and humiliating. Migrants, particularly from rural, agricultural areas on the fringe of the industrial world, were on the other hand willing and often eager to perform the work. Their willingness to do so was predicated on the fact that they thought of their sojourn abroad as temporary. The temporary nature of their commitment to the foreign jobs meant that they were less affected by any social stigma attached to the work or by the instability, uncertainty and lack of career advancement, all of which were viewed very negatively by national workers who thought of themselves as committed to these jobs permanently. Any social stigma attached to the work was further eased by the fact that migrants performed it in a foreign country out of the eyes of their home communities in which their identity was rooted. Immigration thus seemed an almost ideal solution to this particular labor market bottleneck. It also seemed not only to solve the labor market problems of industrial work, but to contribute to economic development at the place of origin as the temporary migrants return with developmental resources in the form of capital, skills, and knowledge of the broader market for the goods they were able to produce. Here too, in fact, the accumulation of funds for investment in productive projects at home -in agricultural areas, in land, farm equipment and livestock- often motivated the migration process.
Thus, when this process worked well, the migrants took the jobs which national workers did not want, used the jobs to accumulate capital to invest at home, left after a relatively short period of time. Because one of the characteristics of the jobs which rendered them unattractive to national workers was their instability, they often disappeared with the migrants who held them, but if the jobs remained, the initial wave of migrants was replaced by newcomers with similar motivations. Meanwhile the migrants invested their savings at home in projects which promoted economic development and expansion there, strengthening the incentives for their compatriots to return.
Whether or not the process ever worked smoothly in this way, it virtually never did so for a very long period of time. Out of the stream of initially temporary migrants, a permanent settlement invariably seemed to emerge sooner or later. Initially temporary workers stayed because it took longer than anticipated to accumulate the funds which had originally motivated their migration and/or because they formed families and other attachments at the destination. The emergence of such a settlement has three distinct effects. First, the motivation and aspirations of the migrants began to shift toward those of national workers. As their time horizon in the destination country expanded, they become increasingly interested in the job security and upward mobility offered by the jobs they held and concerned that any social stigma associated with the work they were performing would come to be attached to them as persons and undermine their stature in their community and their self-conception as an individual. As their stay became more permanent, they thus become competitors for the opportunities to which national workers aspired but also less easily managed on the job and less attractive to their employers as workers. Second, as they settled, the first generation migrants began to build families around themselves and raise children at the destination. Indeed, the family formation was often responsible for the changing motivation of initially temporary migrants, and the children attached to these families, wherever they were actually born, and whatever the aspirations of their parents, were essentially native to the communities where their parents settled, with the attitudes and aspirations of native workers. Third, as permanent settlements emerged, the relationship between the original home community and the new one began to change. People began to move back and forth between the two regions for different reasons: Family problems which were once invariably resolved by the migrants returning home were now often addressed by bringing distressed relatives to live abroad; income which was once reserved for investment projects at home was now retained abroad. Or, when it continued to be sent home, it was increasingly used to support family members who could not migrate, i.e., for consumption that undermined work incentives at home rather than for investment in development projects.
The long term viability of migration as a solution to the labor shortage in these basically unattractive short-term jobs thus depends on the ability of the economy to generate enough stable, secure, and socially attractive jobs to accommodate the aspirations of the members of the migrant communities as they settle, and, often more to the point, to make those jobs accessible to them. Thus, while the migration process in its early stages assures a rough equivalency between the aspirations of the migrants and the needs of industrial society, there is nothing in the process which assures that equivalency in the long run once the second generation has emerged.
On the whole, the notion that migration would lead to economic development in the underdeveloped region has proven to be disappointing as well. Returning migrants often invested in the home region in the early stages of the migration process, but these regions were seldom the natural locus of development. The regions were agricultural and development was centered in urban, industrial areas. The skills, as opposed to capital, which the migrants brought home were trivial because their work abroad was largely confined to unskilled jobs, and where they did return with industrial skills useful in the developing country, the industries which employed them were not located in their home communities but in other parts of the country, often as distant socially, if not physically, as the destinations abroad. When economic development failed to ensue, more and more of the labor force of productive age was drawn into the migration process. Over time, therefore, the income sent home was increasingly used for consumption rather than developmental investment, and the sending community atrophied economically. Many of you have, I am sure, seen the villages in the late stages of the evolution of the migration process that are dominated by enormous houses in the style of the places to which the migrants had gone. These villages are empty of all but the aged and young children, except in certain holiday periods when the migrants came home to visit and boast of their success abroad.
In this late stage, the migration process has become completely entrenched. The migrant community at “home” is more and more dependent on the income of the migrants, and at the same time the migrants at the destination are deeply embedded in the economy there with well established channels for finding jobs for their brethren and circumventing physical and legal barriers to residence and employment.
The institutional structures that created the spaces in the labor market structure for which the circular migrants from backward agricultural areas were particularly adapted to fill were greatly strengthened in two great waves of labor reforms: First in the 1930’s and then in the late 1960’s and early 1970’s, in the United States and also in much of Western Europe. Political pressures in these periods and organized economic actions such as strikes and street demonstrations, as well as outright civil disobedience, led to enhanced worker protections and restrictions on managerial authority in the workplace. The pressures were in no small part, at least in the U.S., generated by second generation migrants reacting against the conditions of work in the jobs they inherited from their parents. The results were considerable limits on what has come to be known as labor market flexibility, and especially on the ability of management to layoff and discharge workers in response to fluctuations in demand. By and large these restrictions were imposed without any corresponding reduction in the flex and uncertainty of the economy in which the businesses operated. For firms seeking ways of managing the costs of variability, which they could no longer transfer to their core labor force, a fringe of pliant migrant workers became increasingly attractive, not only to the firms themselves but to the core labor force whose working conditions the fringe of migrants helped to sustain. The history in Spainis a little different: Here the restrictions on layoff and discharge were typically a heritage of the Franco regime, but they were not challenged in the democratic transition.
2. Changes in the structure of the labor market and some implications for migration
Over the course of the last several decades, these processes, which I think capture an important part of the immigration phenomenon for most of the 20th century, and a good deal of the 19th, has begun to change. The changes pose significant challenges to how we think about immigration and the policies which we evoke in an attempt to control it. I will not try to cover all of those changes here -indeed I will not pretend to fully understand them. But I will focus on two of these, again particularly salient in the U.S., but I think of some relevance here in Europe as well.
First, there is a change in the institutional environment, basically a move away from the so called rigidities introduced in the 1930’s and the 1960’s/early 1970’s, giving employers greater latitude to deploy labor as they see fit and to layoff and discharge workers in response to variations in product demand. To this degree, the “need” for the kind of malleable labor force which immigrant workers provided, at least in the early stages of the migration process, has declined. At the same time, the forces of technological change and globalization appear to be “hollowing out” the job structure. Mid-level jobs, especially in manufacturing, appear to be gradually disappearing, leaving a bi-modal job distribution, increasingly bifurcated between low wage unskilled jobs at the bottom and relatively skilled jobs, requiring high levels of education, at the top. We have yet to see what the net effect of these changes is on the willingness of native workers to take the jobs which immigrants have traditionally filled. On one hand, there is less need for the kinds of flexibility which immigrants have supplied and less of the more stable job opportunities which enabled national workers to escape “immigrant” work. And, of course, the crisis has narrowed the job opportunities of natives still further -although it has narrowed the opportunities of immigrants as well, in some places more even than those of natives. On the other hand, many of the constraints which appear to be institutional are actually social: Or rather, the institutional restraints imposed by law and collective bargaining are an expression of social attitudes, and those attitudes may continue to affect the way workers behave on the job in ways that impose costs upon employers and constrain their behavior even when the laws and collective bargaining contracts through which they were once expressed are no long in force. Thus, national workers may accept the jobs where there is no alternative but perform poorly in them or express resentment toward management. And managers may as a result continue to recruit immigrants, whom they view as less resentful and more malleable, even when national workers are available.
A recent participant observation study of the displacement of black workers in a South Carolina plant by newly arrived Latino immigrants is particularly to the point here. The study by Laura Lopez-Sanders (2012) documents how management policy deliberately and self-consciously pushed out the black labor force in order to replace them with Latinos, whom, according to the comments of the managers quoted in the study, were perceived as harder working and more malleable employees. The study does not really discuss why management sought to make this change at this time, but one can speculate that the black labor force was first recruited in an earlier period when most blacks in the South were coming into industry directly from agriculture, had little education, and were perceived as particularly pliant and industrious laborers. The rural labor reserve in the South has now been essentially exhausted and the labor force clearly has other ambitions. The patterns here are very similar to those we observed in the North in the 1970’s in low wage urban manufacturing, when Latino immigrants were replacing blacks workers, who had been emboldened by the black political movement, in urban manufacturing (Piore 1973).
Quite aside from the way in which attitudes toward unstable and/or socially stigmatized work remain even when the institutions through which they were once expressed are eliminated, countervailing changes in the native labor force may be working at the same time to increase the demand for pliant immigrant labor. Young workers who were not yet committed to full-time work, many still in school, once had attitudes similar to rural migrants: They too were basically looking for income to support leisure time activity or finance their education, and, like the migrants, were untouched by the social status of the work or its instability because they did not expect to stay in those jobs after they grew up and/or graduated from school. They thus played a similar role to that of immigrants in the labor market. But the labor force participation rates among young workers are declining. Women who in a similar way were willing to take jobs without stability or chances of advancement because they conceived of themselves as housewives temporarily in the workforce but committed in the long run to their roles at home, now aspire and have access to permanent jobs with some social standing and status.
3. Highly educated migration
The second major shift in the patterns of immigration which had prevailed in the 19th and 20th centuries has been the emergence of significant streams of highly educated, skilled migrants moving into positions at the top of the job structure. This new pattern has attracted considerable attention among policymakers in both the sending and receiving countries. In the latter, it is often viewed as a net contributor to economic growth and prosperity in a way in which unskilled migrants who are (generally wrongly) perceived as a net drag on public resources are not. And, in the United States at least, there has been considerable pressure to redesign the requirements for residency and work permits to favor this type of immigration. In the developing world, the emigration of highly educated workers was initially viewed as a brain drain, but more recently the role of the Diaspora in creating new industries linked to markets abroad utilizing skills and knowledge acquired there has altered the views of what it might contribute to economic development (although the balance of these negative and positive features is seldom drawn).
How does the emergence of this new high wage migration change the perspective I have just articulated? An initial point is that while the specific patterns of these new migrations may be very different, the analytical framework through which they must be understood is essentially the same. They cannot be understood in terms of income differentials alone, but rather require an understanding of the structure of jobs and employment opportunities in the sending and receiving countries, of the sociology and self-conception of the migrants themselves, particularly relative to that of native workers, and the dynamic through which the job structure and the self-conception of the different elements of the labor force change over time.
The poster child for the new immigration patterns and their contribution to economic development in both the sending and the receiving countries is the Indian offshore information technology (IT) industry. My own understanding of this phenomenon derives from a small but instructive study that I did several years ago in an attempt to understand whyIndiahad come to be the major offshore software developer for the U.S. but not Mexico (Piore 2004). IT development requires close personal interaction between the client and the developer. The need for face-to-face contact is underscored by the fact that a rotating pool of nearly one-third of the Indian personnel were in the United States on temporary business visas living in hotels, at any point; anecdotal evidence suggests that number is still very high today. Despite this extensive direct interaction, constant real time communication by telephone and on chat lines was also required. These contacts should have been much easier with Mexico which is in close physical proximity to the U.S. and in the same time zones than with India, which is nearly a full day’s travel away and working on a clock that is almost the inverse of that in the U.S. Language, which we initially thought was a major barrier, did not seem to be a factor.
The answer to the puzzle of why India and not Mexico is that the groundwork for the Indian industry was laid by the Indian Institutes of Technology (ITT), elite engineering schools financed by the Indian government. Many of the undergraduates of these schools came to graduate school in the U.S. in the 1960’s and 1970’s, and then stayed to make their careers here, moving over time to high positions in the management and engineering ranks of major U.S. corporations, but maintaining close contact with their colleagues who remained in (or returned to) India. These India educated, but U.S. based engineers created the bridge which made off-shoring possible. Highly educated Mexicans who studied in the U.S. have until quite recently tended to return home to Mexico upon graduation, without moving into the kinds of positions in the U.S. that the Indian intermediaries held. In this sense, it is the Indian professional network and the way it spread out internationally that is the key element of the story.
There is a parallel here to the networks through which unskilled migrants navigate the labor market. The Indian engineers were long removed from the rural agricultural villages that are so central to the low wage immigrant story (if indeed their families were ever rooted in rural India), but their allegiance and social identities were rooted in a way similar to the agricultural migrants in their home villages, in the professional communities that grew out of the ITT, and they moved back and forth between these communities in much the same way that the low wage immigrants moved back and forth in their transnational villages. A distinguishing feature is that these professional communities were in turn embedded in an institutional structure that spanned the two countries and linked them together: Many of the ITT professors had themselves studied in the U.S., maintained continuing relationships with their erstwhile teachers and scholars there, and modeled their curriculums on those of the U.S. universities, updating the Indian curriculums as those in the U.S. evolved. Thus, the Indian students were able to move seamlessly into graduate education in the U.S.and then back into the Indian institutions if and when they returned. I will return to the question of whether there are analogues to these institutional links for low wage communities below.
Less clear in this story is precisely what role these networks play in the actual operation of the IT industry and its development. The current literature in economics and sociology would emphasize the role of information and/or trust. But the literature on software development would emphasize the role of tacit knowledge and face-to-face communication in its development and communication (Brooks 1995, McBreen 2002). By tacit knowledge, I mean understanding that cannot be communicated through formal instructions, written or verbal alone, but depends on a shared framework through which verbal instruments are interpreted, and the way in which interpretation emerges (or perhaps evolves) in face-to-face interaction and communication (Polanyi 1966). In this view, the framework for that communication is laid through education, although not necessarily through formal classroom instruction. But it is notable that despite the shared background which the transnational professional community created, so many of the Indian engineers working on a specific engineering project had to be in the U.S. talking directly to their customer’s engineers at any moment of time.
The IT industry is only one example of high-skilled migration. AnnaLee Saxenian’s work in Silicon Valley (2002, 2006) and the linkages it established with India, Taiwan and China makes clear that it is not unique. Still, in other industries, immigrants, despite their professional status, are used in very much the same way as unskilled migrants to play roles which nationals are unwilling to assume. In the medical care industry, for example, foreign doctors and nurses are used to fill temporary jobs, or jobs that do not or cannot promise job security (although they may not actually prove to be temporary), or in isolated rural areas, or to staff hospitals in very poor urban neighborhoods. Moreover, all high-skilled immigrants do not move in well-established professional channels. In the U.S., they are attractive to employers in part because the visa system puts enormous power in the hands of employers; and, it is easier to obtain these visas, and the power over the workers who hold them, for professionals who arguably have unique skills not otherwise available on the labor market.
International migration is playing out a long run process, creating durable links between the advanced developed and the less developed world. What initially starts out as a temporary migration facilitating labor market adjustment that is advantageous to both countries leads over time to permanent settlement. This crisis is a moment when policymakers would like to take advantage of the temporary nature of the migration to help absorb the impact of unemployment. And it is certainly a moment when one finds out just how far the process of settlement has proceeded and how temporary the migration actually is. But it is typically too late to adjust the process to the needs of the moment. And policy measures such as tighter border enforcement often act perversely by inducing people who might otherwise have gone home, at least temporarily until the crisis has passed, to stay and settle permanently.
The structural elements which do affect the nature of the migration process over the long run are, however, in the process of changing. I have tried to identify some of these changes in the body of this text. Some are indeed the subject of public policy, such as the policies surrounding labor market flexibility, student financing, female labor force participation and job access. But again, the crisis is typically a moment when one sees the impact of these policies, which have previously been obscured by prosperity. But their impact is also long run and not amendable to short-term crisis management.
The most salient development in international migration patterns in the last several decades, however, has been the emergence of significant streams of high-skilled, highly educated migrants between developed and developing countries. These flows may be similar analytically to those of low-skilled migrants but they are responsive to different structural factors. The one is not a substitute for the other. There is no reason to believe that the existence of high-skilled migration will affect the forces drawing immigrants into the bottom of the labor market. And, if we adjust our formal institutions to accommodate high-skilled at the expensive of low-skilled workers, we will simply drive the low-skilled migration further underground.
High-skilled migration is likely, on the other hand, to complicate the problems generated by low-skilled migration because it will limit the chances for upward mobility for the second generation. In the United Sates, foreign doctors have definitely served as a way of circumventing the pressure for the expansion of U.S. medical schools. And it would appear that off-shoring of software development toIndiawas an alternative to the training of software engineers in theUnited Stateswho might have been drawn from the ranks of U.S. natives and second generation immigrants.
There are, however, lessons to be drawn from the Indian software industry, and those, I think, have to do with the role of professional networks in the development and transmission (and conservation) of tacit knowledge. Natasha Iskander (2010, 2011) has emphasized the role of tacit knowledge and the way in which it emerges and evolves over time in lower wage migration between the U.S. and Mexico and between France and Morocco. The processes she identifies are very similar to those in software development, but since the channels are not institutionalized and the professional communities which emerge in the process not generally recognized, the transfer of knowledge between countries does not occur and the potential contribution for the development process is often lost. If we could develop comparable formal institutions, they would serve as an instrument for increasing the backflow from the migration process to the developing countries and strengthen the forces promoting circular (and temporary) migration. And this might indeed help to address economic crises, not of course the one we are presently living through, but, one might hope, the next.
Brooks, Frederick P.
1995 The Mythical Man-Month: Essays on Software Engineering. Boston, Addison Wesley.
Iskander, Natasha N.
2010 Creative State: Forty Years of Migration and Development Policy in Morocco and Mexico. Ithaca, ILR Press.
Iskander, Natasha (and Nichola Lowe)
2011 “The Transformers: Immigration and Tacit Knowledge Development”, Wagner Research Paper (NYU), nº 1: 1-66
2012 “Racialized and Gendered Labor Queues in New Immigrant Destination” (mimeo). Berkeley, University of California.
2002 Software Craftsmanship, the New Imperative. Boston, Addison-Wesley.
Piore, Michael J.
1973 “The Role of Immigration in Industrial Growth: A Case Study of the Origins and Character of Puerto Rican Migration to Boston”, MIT Department of Economics Working Paper (Massachusetts), nº 112A.
1979 Birds of passage: migrant labor and industrial societies. Cambridge, New York, CambridgeUniversity Press.
2004 “The Limits of the Division of Labor in Design
and the Prospects for Off-Shore Software Development in Mexico” (mimeo), Department of Economics, MIT.
1966 The Tacit Dimension. N. Y., Garden City.
2006 The New Argonauts: Regional Advantage in a Global Economy. Cambridge, Mass., Harvard University Press.
Saxenian, AnnaLee, Yasuyuki Motoyama, and Xiaohong Quan
2002 Local and Global Networks of Immigrant Professionals in Silicon Valley. San Francisco, CA, Public Policy Institute of California.