‘Also by Stephen Castles Immigrant Workers and Class Structure in Western Europe (with Godula Kosack) The Eaucation ofthe Future (with Wiebke Wistenberg) Here for Good: Western Burope's New Ethnic Minorities CChizenship and Migration: Globalization and the Politics of Belonging (with Alastair Davidson) Etinicty and Globalization: From Migrant Worker to Transnational Citizen Migration, Citizenship andthe European Welfare State: A European Dilemma (with (Car-Uihik Schierup and Peo Hansen) Migration and Development: Perspectives from the South (edited with Raul Delgado Wise) ‘Also by Mark J. Miler Foreign Workers in Europe: An Emerging Political Force ‘Administering Foreign Worker Program: Lesson from Europe (with Philip L. Martin) ‘The Unavoidabe tesue: Unie G. Papademeriou) The War on Terror in Comparative Perspective (with Boyka Steanova) | States Immigration Policy inthe 1980s (with Demetios The Age of Migration Intemational Population Movements in the Modern World 4th edition Stephen Castles Mark J. MillerChapter 2 Theories of Migration International migration is hardly ever a simple individual action in which ‘person decides to move in search of better life-chanees, pulls up his or her roots in the place of origin and quickly becomes assimilated in the new country. Much more often migration and settlement are a Jong-drawn-out process that wil be played out for the rest of the migrant’ life, and affect subsequent generations too. Migration can even transcend death: members of some migrant groups have been known to arrange for their bodies to be taken back for burial in their native soil (see Tribalat, 1995; 109-111). Migration is a collective action, arising out of social change and affecting the whole society in both sending and receiving areas, Moreover, the experience of migration and of living in another country often leads to modification of the original plans, so that migrants’ intentions a the time of departure are poor predictors of actual behaviour. ‘Similarly, no government has ever set out to build an ethnically diverse society through immigration, yet labour recruitment policies often lead tothe formation of ethnic minorities, with far-reaching consequences for social relations, public policies, national identity and international relations, ‘The study of international migration has often fallen into two rather separate bodies of social sciemttic investigation: frst, esearch on the determinants, processes and patterns of migration, and, second, research ‘on he ways in which migrants become incorporated into receiving societies (compare Massey etal, 1998: 3). We argue that this distinction is artificial, and detrimental to a full understanding of the migratory process. We use the term “migration studies’ inthe widest sense, to embrace both bodies of investigation. In addition, we believe the second area should be understood more broadly as the ways in which migration brings about change in both sending and receiving societies. ‘This chapter provides a theoretical framework for understanding the ‘more descriptive accounts of migration, setlement and minority formation in later chapters. However, the reader may prefer to read those fist and come back tothe theory later. In various places, our account draws attention to the links between internal and international migration, In many countries — especially those with very big populations like China, India, Brazil or Nigeria ~ internal migration is far larger than international, The ‘wo are often closely linked, and internal rural-urban migration may be a prelude (0 cross-border movement (Skeldon, 1997). However, this book does not deal systematically with internal migration, 20 Theories of Migration 21 Explaining the migratory process “The concept of the migratory process sums up the complex sets of factors and interactions which lead to international migration and influence its course. Migration is a process which affects every dimension of social existence, and which develops its own complex dynamics. The great ‘majority of people in the world (97 per cent in 2000) (UNDESA, 2005) fare not international migrants, yet their communities and way of life are changed by migration, The changes are generally much bigger for the ‘migrants themselves, and they can be seen at every stage of the migratory process, whether in countries of origin, transit or destination Research on migration is therefore intrinsically interdisciplinary: sociology, political science, history, economics, geography, demography, psychology, cultural studies and law are all relevant (Brettell and Hollifield, 007). These disciplines look at different aspects of population mobil- ity, and a full understanding requires contributions from all of them. Within cach social scientific discipline there isa variety of approaches, based on differences in theory and methods. For instance, researchers who base their work on quantitative analysis of large data-sets (such as censuses of representative surveys) will ask different questions and get different results from those who do qualitative studies of small groups. Those who examine the role of migrant labour within the world economy using historical and institutional approaches will again get different findings. ‘Each of these methods has its place, as long as it lays no claim to be the only correct one. As interest in migration research has grown in recent years, theoretical approaches have proliferated and interacted, leading to ‘more complex understanding of migration and its links with broader processes of change. Of special importance has been the application of theories of globalization and transnationalism to migration. A detailed survey of migration theory is not possible here (see Massey ct al., 1993, 1994, 1998; Portes and DeWind, 2004; Brettell and Hollifield, 2007), but ‘we will outline some main issues and provide pointers for futher reading. Economic theories of migration Neoclassical theory remains the dominant paradigm in economics, and hhas had an important role in migration studies. However, some of its key assumptions and findings have been questioned through alternative approaches, The neoclassical perspective has its antecedents inthe earliest systematic theory on migration: that ofthe nineteenth-century geographer Ravenstein, who formulated statistical Jaws of migration (Ravenstei 1885, 1889). These were general statements unconnected with any actual ‘migratory movement (Cohen, 1987: 34-35; Zolberg, Suhrke and Aguso, 1989: 403-405), Such “general theories’ emphasize tendencies of peo- ple to move from densely to sparsely populated areas or from low- to22. The Age of Migration Theories of Migration 23 high-income areas, or link migrations to fluctuations inthe business cycle, These approaches are often known as ‘push-pull’ theories, because they perceive the causes of migration to lie in a combination of ‘push factors” impelling people to leave the areas of origin, and ‘pull factors’, attracting them to certain receiving couttries. ‘Push factors" include demographic srowth, low living standards, lack of economic opportunities and political repression, while ‘pull factors’ include demand for labour, availability of land, good economic opportunities and political freedoms, ‘Today this model is mainly found in economies, but is also used by some sociologists, demographers and geographers. It is individualistic «and ahistorical. It emphasizes the individual decision to migrate, based on rational comparison of the relative costs and benefits of remaining at ‘home or moving. Neoclassical theory assumes that potential migeants have perfect knowledge of wage levels and employment opportunities in destination regions, and that their migration decisions are overwhelm. ingly based on these economic factors. Constraining factors, such as government restrictions, are mainly dealt with as distortions of the rational market. The central concept is ‘human capital’: people decide to invest in ‘migration, in the same way as they might invest in education or vocational training, and will migrate ifthe expected rate of return from higher wages in the destination country is greater than the costs incurred through ‘migrating (Chiswick, 2000). Borjas puts forward the model of an “immigration marke Neo-classical theory assumes that individuals maximize utility individuals ‘search’ for the country of residence that maximizes their well-being ... The search is constrained by the individual's financial resources, by the immigration regulations imposed by competing host ‘countries and by the emigration regulations ofthe source country. In the immigration market the various pieces of information are exchanged and the varinns options are compared. Ina sense, competing, lust counties make ‘migration offers’ from which individuals compare and choose. ‘The information gathered in this marketplace leads many individuals to conclude that itis ‘profitable” to remain in their birthplace ... Con. versely, other individuals conclude that they are better off in some other country. The immigration market nonrandomly sorts these individuals across host countries, (Borjas, 1989: 461) Borjas claims that ‘this approach leads toa clear ~ and empirically testable — categorization of the types of immigrant flows that arise in a world where individuals search for the “best” country” Borjas, 1989: 461). The mere «existence of economic disparities between various areas should be sufficient to generate migrant flows. In the long run, such flows should help to equalize wages and conditions in underdeveloped and developed regions, leading towards economic equilibrium, Borjas has argued that this may lead to negative effects for immigration countries, notably the decline of levels and lower wages for lowersilled local workers sxe oy Bejan, 2001), However tis hing is contested within ores val research: Chiswick clams that migrants are positively sel- neces ore Highly skilled are more likely to move because they cin a higher return on their human capital javestment in mobility. This hasnt elects for countries of ongin, by easing "rat dra (Chien Mics of specific migntion experiences cast doubt on eet chory. ts rarely the poorest peopl from te lent developed asthe move lo he ices counters more requ the migrants a ple of intermediate social status from areas which are undergoing ae sop tnd social. change. Similar, the pashepll model, predicts sor from densely populated areas o mere sparsely peopled regions, = countries of immigration like the Netherlands and Germany are among Je cris more deecly populated. Finally the pusipull model cannot saan why certain group of migrants goes to one county rae than Str Fria yvonne mgt cea hile the Opposite aplics to Turks? ONzocasical migration theories have therefore been criticized as inetpble of explaizing actual movements or predicting future ons (soe exer Toss; Boyd, 1989, Portes and Rumba, 2006. 16-17) It seems Ssocrd't eat migrants a individual market players who have ful internation on their options an eed o make raina cies, tsead vipanisbave limited and ctenconsaditoy information, anlar subject qavtange of consrains (especialy lack of power inthe face of employers snippet Any compeni yon cil an a iat (ee below), Moreover, historians, anthropologist, sociologists tedgengraphershaveshownthmgran bshavioistong ences by Htrel experiences as well by famaly and commanity dynamics (Portes and Béricz, ee 7 a : Tsessetal oinuoducca wider range of fat ny exouumic ea (ne atempt odo this dual or segmented) labour marke theory, which Shows te iponanee of iatutondl factors a wel as race and ger in beingng show labour market segmentation ior (1979) argues tha inte baton migration is caused by sructral demand within advanced econ0- Ines foc both highly skilled workers and lowerskiled manval workers t0 any Oo produion aks (eg assembly line werk or garment manufac ture) and to staff service meeps, (catering, cleaning. aged ome, oe. ‘\aison into primary and secondy labour mae emerges ow, 1979), while the most dynamic ‘global cities’ are marked by economic polaizaton a growing gulf between the highly paid core workers in France, managenent and esearch, and the por paid workers who sr ice their needs (Sassen, 1991). The workers in the primary labour market are positively selected on the basis of human capital, but also often through membership ofthe majority etnic group, male gender an, i the eae of rmigeants, regular legal status, Conversely, those in the secondary labour