Rethinking Care Through Social Reproduction: Articulating Circuits of Migration

Care has come to dominate much feminist research on globalized migrations and the transfer of labor from the South to the North, while the older concept of reproduction had been pushed into the background but is now becoming the subject of debates on the commodification of care in the household and changes in welfare state policies. This article argues that we could achieve a better understanding of the different modalities and trajectories of care in the reproduction of individuals, families, and communities, both of migrant and nonmigrant populations by articulating the diverse circuits of migration, in particular that of labor and the family. In doing this, I go back to the earlier North American writing on racialized minorities and migrants and stratified social reproduction. I also explore insights from current Asian studies of gendered circuits of migration connecting labor and marriage migrations as well as the notion of global householding that highlights the gender politics of social reproduction operating within and beyond households in institutional and welfare architectures. In contrast to Asia, there has relatively been little exploration in European studies of the articulation of labor and family migrations through Spring 2012 Pages 142–162 doi:10.1093/sp/jxr030 # The Author 2012. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com Advance Access publication January 18, 2012 Social Politics 2012 Volume 19 Number 1

the lens of social reproduction. However, connecting the different types of migration enables us to achieve a more complex understanding of care trajectories and their contribution to social reproduction.
As care has come to dominate much feminist research on globalized migrations and the transfer of labor from South to North, the older concept of social reproduction has declined in usage and been pushed into the background. Although receiving different levels of attention, the two concepts are nevertheless frequently linked. In particular, care work is seen as contributing to the globalization of social reproduction and thus a form of reproductive labor (Hill Maher 2004;Misra, Woodring, and Merz 2006;Perez Orozco 2010). UNRISD (2009, 3) notes that the social and political economy of care involves "the reconciling of the burden of reproduction with that of other social tasks as well as sharing the burden of reproduction between members of society." A useful definition of care refers to it as a range of activities and relationships that promote the physical and emotional well-being of people "who cannot or who are not inclined to perform these activities themselves" (Yeates 2004, 371). The notion of reproduction was widely debated in the 1970s when the emphasis was on the contribution of unpaid domestic labor to the reproduction of the labor force (Beneria 1979;Molyneux 1979;Vogel 1995). However, it incurred trenchant critiques (Molyneux 1979) in terms of its economic reductionism, abstraction from specific societies and histories, and narrow focus on labor undertaken in the household, the latter inadequately placed within a wider socio-economic context. 1 This is not to say that reproduction has fallen by the wayside. Unpaid labor and the domestic sector have been the mainstay of feminist economics (England and Folbre 2003;Elson 2000). Indeed, feminists, especially those working within a political economy perspective (Bakker 2007;Bakker and Gill 2003;Bakker and Silvey 2008;Bedford and Rai 2010;Bergeron 2011;Luxton and Bezanson 2006), have contributed to a resurgence of interest in reproduction, now generally more broadly conceptualized as social reproduction and extending beyond the reproduction of the labor force. Some feminist scholars have used the work of Bourdieu to critically engage with women's efforts in reproducing families and generations through different kinds of capital (Adkins and Skeggs 2005;Reay 2005); others have examined the reprivatization of reproduction through the privatization of state assets and forms and the varying degrees of commodified care work (Bakker 2007). Bakker and Gill (2003, 18) also include in their framework the institutions providing for the socialization of risk, health care, education, and other services.
Both physical 2 and socialization processes have been highlighted as crucial elements of social reproduction. For Petersen (2003, 79), writing from a feminist international political economy perspective, reproduction includes the material and symbolic processes required to reproduce human beings over time (daily and generationally) within the family and the private sphere. Peterson looks at how labor within the family sphere inscribes codes and norms of identities and divisions of labor through parenting practices. She also questions the heteronormative assumptions 3 underpinning the reproduction of families. Similarly, based on anthropological approaches, Bjeren (1997, 227) distinguishes two processes: the reproduction of people as physical beings and the reproduction of social identities within given social and cultural contexts. These two dimensions invoke the reshaping of familial and kin forms, such as marriage, over time and space, and reconfigure social relationships globally between localities and states. The reference to family and kinship highlights the fact that reproduction takes place not just through labor processes and clear-cut economic migration but also through other circuits of migration such as marriage, which create new families as well as the reunification of existing ones.
Despite the revival of interest in social reproduction in feminist political economy and disciplines such as geography and sociology, there has relatively been little inquiry among scholars of gendered migration (but see Bakker and Silvey (2008) and Beneria (2007) who include it as one of the aspects in the globalization of social reproduction) who have been more disposed toward the concept of care as the primary object of inquiry, leaving the complexities of social reproduction in the background. I would, however, suggest that engaging with the concept of social reproduction enables us to place care work within a wider landscape of activities and sites and to connect supposedly disparate circuits of migration, in particular labor, family, and education, which are usually analyzed separately but which are in fact interconnected. These diverse circuits all contribute to different configurations of social reproduction which may be initiated by different migratory flows. On the one hand, migration due to marriage may lead to the formation of new families with implications for the care of the able bodied spouse and children. In seeking to work, spouses, especially females, may call upon siblings and parents, who are effectively moving in order to take care of their children. On the other hand, a different configuration results from economic migrants undertaking the care of non-migrants at the same time as their own family, both in the sending and in the receiving countries. However, in leaving the different circuits (family and labor) disconnected, we seem to be hanging onto traditional gendered knowledge in terms of categories of migration (Schwenken 2008). 4 Thus my main objective in this article is to suggest how we might articulate the diverse circuits of migration and achieve a better understanding of the different modalities and trajectories of care in the reproduction of individuals, families, and communities, both of the migrant and of the non-migrant populations. Unfortunately, it is not possible within the scope of this article to explore the different articulations and configurations in great depth but merely to indicate a few ways in which it may be achieved, using in particular the insights derived both from earlier North American research on racialized minorities and the emergence of globalized social reproduction and current Asian studies of gendered circuits of migration.
In the first section, I briefly outline the early debate on waged domestic labor and the initial incorporation of migrant women into the analysis of social reproduction in the 1990s. The second section explores how migrant and minority women have been incorporated into theorizations of social reproduction, not just through labor flows, but also in the articulation with other flows, especially family ones, which contribute to the social reproduction of individuals, households, and communities. The significance of migrant and minority labor in commodifying and geographically extending social reproduction within this approach can be traced back to North American analyses in the 1990s (Colen 1995;Nakano Glenn 1992) and more recently in Asian studies of the articulation of diverse regional and global migratory flows (labor and marriage) (Constable 2009;Lan 2008;Palriwala and Uberoi 2005). Although these approaches have not received the same attention in European studies, I highlight in the third section some developments that are beginning to make connections between different types of migration relevant to a more complex understanding of care and social reproduction.

Social Reproduction and the Incorporation of Migrant and Minority Labor
As previously noted, debates about the relationship between production and reproduction go back to the 1970s and the domestic labor debate which revolved around the non-recognition of the added value of labor in the household and the reproduction of the worker (Dalla Costa 1973;Mackintosh 1979;Smith 1979). However, the debates among Northern socialist feminists did not take into account waged domestic labor provided by non-family members, such as rural migrants (Kofman and Raghuram 2006). Waged domestic labor represented a pre-modern form of labor that had supposedly largely disappeared in the modern world (Andall 2003;Coser 1973). Hence, the study of servants in Europe was left to historians (Sarti 2008).
In the 1970s and 1980s, the growing visibility of migrant women in the economy in Northern European countries meant that academic studies focused more on their presence in manufacturing, sweatshops, less skilled work in public services (Condon and Ogden 1991;Phizacklea 1983; Knocke and de los Reyes cited in Cederberg 2010) and, with some exceptions in France, on domestic work (Leonetti and Lévi 1979). In terms of sociological analyses, domestic reproductive labor and its ethnicized composition was the object of analysis in South Africa (Cock 1980), the USA (Colen 1995;Nakano Glenn 1992) and Latin America (Radcliffe 1990). In the UK, Gregson and Lowe (1994, 123) concluded that "no close association existed between ethnicity, female migration and waged domestic labour" in relation to the resurgence in demand for waged domestic labor, especially nannies and cleaners, in servicing the middle classes in the 1980s. 5 Instead, they argued that the demand from dual career middle-class households led to a reworking of labor along class lines with different strata of women (lower middle class as nannies and older working class as cleaners) once again taking up employment in the domestic sphere. And what concerned them was not so much the issue of care but how waged domestic labor had become in the past decade "a necessary aspect of social reproduction" (p. 75). So, when migrant and minority ethnic women began to be incorporated into discussions about domestic labor in Europe in the 1990s, it was often in terms of reproductive labor rather than care (Anderson 2001).
Theorizations of the role of racialized minorities, established and recent, within a national context (Colen 1995;Nakano Glenn 1992) and in the globalization of social reproduction (Truong 1996), emerged in the 1990s. Nakano Glenn's writing has been very influential. She defined (1992, 1) reproduction as "the creation and recreation of people as cultural and social as well as physical human beings who engage in an array of activities and relationships involved in maintaining people both on a daily basis and intergenerationally". This was applied historically to the interlocking and interactive gendered and racialized dimensions of reproduction through the study of three groups of racialized and ethnicized women-African American, Mexican American, and Japanese American in different regions of the USA. In the course of the twentieth century, there was a shift for such women from working in households as commodification moved to low-level institutional service work, e.g., as nursing aides.
On the basis of fieldwork conducted between 1984 and 1986 with West Indian women who had migrated to New York since the changes in immigration legislation in 1965, Colen (1995) analyzed their economic and familial positions in terms of stratified reproduction. This concept described the power relations through which some categories of people gain greater sustenance and support (material and normative), while others face greater obstacles and hardships in sustaining their own reproduction and care. Care, which Colen distinguished from other activities in the household, is only one of the activities contributing to social reproduction. Undertaking care as a separate activity in the household also places this employment on a higher level in the job market than those positions where care is combined with other domestic tasks, such as cleaning. Her sample of interviewees had often occupied all the different positions within this labor niche, with many remaining in it even after obtaining the Green Card because they could not find better qualified work.
Such physical and social reproduction of bearing, raising, and socializing children, and of creating and maintaining households from infancy to old age is achieved, experienced, valued, and rewarded differently according to inequalities of access to material and social resources in particular historical and cultural contexts (Colen 1995, 78). The patterns of these inequalities of reproduction in turn reproduce stratification by reflecting, reinforcing, and intensifying the inequalities on which this stratification is based. The resultant transnational system of inequality plays itself out in intimate, daily, and local events in sending and receiving states in which a number of different institutions (states, labor markets, families, local communities, media) contribute to particular living and familial arrangements and their representations as being legitimate or marginalized. Thus, the provision of reproductive labor for wealthier professional women in turn shapes the ability and modalities of different categories of migrants to reproduce their own families. How rapidly migrant women could reunite their families depended to a great extent on their position within the domestic labor market (live in work was often occupied by the undocumented seeking to get their employer to sponsor them) and their immigration status. A close relationship with a citizen can also be a means of facilitating the entry of other members of the migrant's family.
Colen and Nakano Glenn focused their analysis on the nationstate; it was Truong (1996) who subsequently extended the analysis to a more general thesis of the relationship of production and reproduction through the globalized transfer of labor. She argued "[n]o production system operates without a reproduction system and it should not be surprising that the globalization of production is accompanied by its intimate 'Other' i.e. reproduction". The transfer of reproductive labor from one class, ethnic group, nation, or region, was beginning to attract some attention but she recommended that this insight required a global perspective which could explain three elements. These were first the structural gaps of labor in reproduction which affected different classes and economies; secondly, the processes by which the transfer occurred and thirdly the implications this has for states, capital, communities, and the reproductive workers themselves. The current gaps in reproductive labor stem from three main areas: the withdrawal of the state from supporting children and the elderly; the increased participation of women in paid work and the inflexibility of the sexual division of labor within the household and the expansion of the hospitality industry and associated sex services, on the one hand, and increased mobility of the female work force (Truong 1996, 34).
However, in the mid-1990s, the implications of the shifts in the organization of different forms of reproduction and the impact of the transfer of labor from the South to the North were not fully understood. A perceptive insight into the segmentation and stratification of services was offered by Wolkowitz's (2002) analysis of occupations related to body services underpinning the simultaneous incorporation of migrant labor and the globalization of social reproduction. She argued that body services had become "a vital replacement for the production of objects as a source of profit and employment in affluent societies" while the items of food and clothing required for human reproduction have been exported to thirdworld countries. She lists a wide range of occupations falling within the remit of paid body work-beauticians, hairdressers, care assistants, fitness instructors, doctors, nurses, and maids. Higher status groups are much more likely to be drawn into occupations which are seen as control, supervision, and management, rather than those constructed in terms of caring for or servicing other bodies. Lower status occupations deal with what is rejected, spills out, or pollutes. These distinctions also operate starkly in particular occupations, for example, between staff nurses and nursing aides.
Such labor markets are segmented by class, sex, and "race" where the lower ranks of occupations such as nursing, cleaning, and caring in public institutions and private homes are racialized and occupied by "subordinate-race" women. However, the boundaries between occupations are not necessarily fixed, for example, the emergence of core highly paid knowledge nurses who may push lower paid care workers further down the occupational ladder and into what is seen as dirty and symbolically polluting work.

Articulation of Labor and Family Circuits
Although most analyses of transnational and global reproduction focused on labor flows, and to a lesser extent, the stratifying implications of labor inequalities for migrant family life, we can also examine more directly reproduction through marriage and family reunification of migrant families, whose own reproduction has been hidden and privatized (Arat-Koç 2006, 88). In this way, a theoretical framework constructed around social reproduction allows us to connect different forms of reproductive labor with the reproduction of individuals, families, generations, and communities and to link a diversity of intersecting migratory circuits (Lan 2008). Palriwala and Uberoi (2005, 31) have noted that marriage migration tends to operate in parallel with flows of labor migrants. Despite the complexity and diversity in cross-border marriages, it is usually women rather than men who emigrate to marry. As Constable (2005, 15) nicely puts it, this pattern illustrates "gendered geographies of power" in which "women, by virtue of their social positioning, can take advantage of opportunities for mobility that are sometimes unavailable to men." They often marry a man in a higher socio-economic location or what she (2005) terms spatial or global hypergamy. This is not to say that women inevitably marry men more educated or wealthier than themselves. In some instances, marriage may involve geographical hypergamy without higher socioeconomic status as is the case of educated women marrying less-educated rural men in South Korea and Japan, or among diasporic populations, for example Vietnamese women marrying US citizens of Vietnamese origin (Thai 2008).
The articulation of circuits is strongly shaped by immigration and settlement policies. In Asia, wealthier countries such as South Korea, Japan, Malaysia, and Taiwan have attracted large numbers of migrants but do not permit family reunification or make it very difficult, especially for less-skilled migrants (Seol and Skrentny 2009). On the other hand, transnational marriages have increased sharply (Douglass 2006;Hugo 2010). Immigration policies make a clear distinction between bringing in labor (paid) as opposed to spouses (unpaid labor for the husband and often parents) into the household. Both categories provide reproductive labor and care in different ways, but the distinction has profound consequences for the rights and citizenship of migrants and their children (Turner 2008).
In Asian countries, accessing settlement and citizenship for those doing low-skilled work is almost impossible through a labor as opposed to marriage route.
Drawing on the close connection of labor and marriage migrations, studies of Asian migration have traced the paths both from paid reproductive to unpaid reproductive labor and vice versa in the case of female labor migrants. Female migrants may start off either as entertainers and sex workers or as carers who marry their employers. Women may seek marriage as a strategy for exiting domestic work and deskilling, as Filipinas entering through the Live-In Caregiver Program in Canada have done (McKay 2003). In other instances, the opposite movement from unpaid to paid occurs where a woman enters for marriage purposes and at the same time or subsequently enters the labor market, often in a reproductive occupation. Switching from one type of caring work to another may be even more complicated as in the case of Filipina entertainers to Japan who have married Japanese men (they may also be looking after his parents) and, in some cases, doing paid care work for others as local foreign carers (Siampukdee 2011;Suzuki 2007, 373). Accepting a socially respectable job is a way of overcoming negative representations. Thus, over time care and social reproduction is undertaken for strangers and family members and as paid and unpaid work. Lan (2008) also envisages different strategies as class-specific solutions to the shortage of reproductive labor. While upper-class and middle-class households hire migrant domestics to outsource housework and care work, working-class households seek foreign wives to provide unpaid domestic labor (Wang 2007).
A conceptual framework connecting the different processes and trajectories is that of global householding, an entity broader than the family and allowing for the formation of units not necessarily composed of relations through marriage or direct lineage, has received theoretical and empirical attention in the past few years (Douglass 2006;Petersen 2010;Safri and Graham 2010). For Douglass (2006, 424), it builds on the different stages of the life cycle enabling the continuity of social reproduction. 6 Such stages are exemplified by marriage, adoption of children, sending children abroad for education and settling them in other countries, demand for domestic helpers, retirement of couples from higher to lower income countries to save resources, and the sending of remittances. Studies drawing upon this framework, unlike many of those based on the familial domestic sphere, have connected the diverse activities in the household to highlight how the gender politics of social reproduction operate beyond households to perpetuate economic and other inequalities that are materially constituted and culturally normalized (Petersen 2010). Global householding also incorporates other dimensions of migration such as education. Sending children abroad for education, for example from Korea and Taiwan, may require their mothers moving to look after them (Chee 2003). Furthermore, the increasing migration of students in higher education (Hugo 2010) may result in international marriages. In all of these instances, care is an element of social reproduction.
In contrast to Asia, there has been relatively little exploration in European studies of the articulation of labor and family migrations, either generally, or more specifically through the lens of reproductive labor and social reproduction. Theoretically, and in policy terms, the rationalities of each form of migration are deemed to be distinctive and separate. Family migration is located in the social world and in effect disconnected from the economic realm (Kofman 2004). It is therefore often thought not necessary to probe into the employment or training of family migrants. Yet, as Escrivà and Skinner (2006) highlight in their Spanish study, family migrants are above all concerned with the economic and the support of transnational and reconstituted families, either through paid work or unpaid care (i.e., reproductive labor).
In making connections between the different circuits of migration, several major differences between Asian and European states should be noted. The first relates to the intensification of intra-European mobility following EU eastward enlargement since 2004 7 and which has given rise to transient and longer term circulations where labor and familial movements and strategies closely mesh (for Polish migrations see Ryan et al. 2009;White 2010). The opening up of migratory spaces has meant that whole families have been able to settle, unlike the earlier migrations where only one parent circulated (Morokvasic 2004). In addition, the parents of European migrants can also be reunited with them, make regular visits to provide practical and emotional support and come at short notice in case of emergencies, such as parents or children falling ill. In this way, the familial disembedding and disruption so commonly forced upon third-country nationals have been avoided to some extent by recent Eastern European migrants who have access to mobility and citizenship rights.
A second difference is that, although subject to more restrictive criteria in recent years, European states allow for the settlement of designated family members from third-country nationals (Kraler 2010;SOPEMI 2000). 8 Family migration remains the major source of permanent migration (Grillo 2008;Kraler 2010) and binational and transnational marriages with a person of immigrant background have become much more common (Beck-Gernsheim 2007;Bryceson and Vuorela 2002;Charsley 2011;Eggebø 2010). The relationship between family migrants and paid work has, however, not been investigated, although Gonzá lez-Ferrer (2007 has shown for both Germany in the 1960s and Spain in the past decade that family reunification took place very rapidly after the migration of the initial migrant and that women were joining the labor market and at times initiating the migration (Erdem and Mattes 2003;Gonzá lez-Ferrer 2011), indicating that migration for labor purposes was probably a family strategy from the outset. Of course at a time of increasingly restrictive immigration policies, it can be problematic to highlight the economic contribution of family migrants, who are constructed in policy terms as dependants (Eggebø 2010), and also often accused of being a burden on the welfare state and a threat to the modernity of European societies as a result of their low levels of participation in the labor market (Roggeband and Verloo 2007). In other words, they denote an excess of alterity or otherness in their reproduction of traditional practices (Grillo 2007).
A further difference compared with Asian countries is that members other than spouses may also settle. Whilst parents constitute a small percentage of family members, and must demonstrate in many states that they are dependants of their children (Kraler 2010), they may play a very significant role in the social reproduction of the family by caring (usually unpaid) for grandchildren whilst their children, and especially mothers, go to work (Escrivà and Skinner 2006;Escrivà 2005). Siblings too may provide paid care or may take in children after school whilst the mother works (Herrera 2008). We have more evidence of this in Southern European countries where family members previously entered initially as tourists in the absence of family reunification policies (Gonzá lez-Ferrer 2011). Canadian research (Creese, Dyck, and McClaren 2011) has also highlighted the advantages of extended families, commonly depicted as traditional, for successful integration due to the greater amount of material and affective resources available to the family. Its worth noting that connecting family and labor migrations, be that of European or Third-country nationals, renders more complex the chains of care and reproductive labor than envisaged in the literature on chains of care. Established and more recent migrant families in receiving countries too require care and those with greater resources may employ other migrant workers (Kofman and Raghuram 2010;Krøger 2003;Wall and Sã o José 2004; see also Escrivà 2005).
As we have seen in the Asian literature, there has been some discussion on policy proclamations about the extensive abuse of marriage or marriages of convenience ( de Hart 2006), but we have little evidence of how frequently this change in status from employment to spouse occurs. Nevertheless, as in Asia, the marriage route does confer permanent residence and citizenship after a stipulated period of duration of the marriage, and increasingly, the successful passing of integration tests for the spouse. Marriage with a citizen may also yield advantages such as greater ease in bringing in family members. Drawing on Colen (1995), a study of Latin American female migrants in Milan noted that migrants with Italian family members were more easily able to regularize their status and gain citizenship, and hence more easily bring in other family members and reconfigure their family (Bonizzoni 2011). It may ease the pressures on the social reproduction of their own families, although children are not in all cases happy in being reunited with their mother or parents in a new country after a period of absence and often living in difficult conditions (see also Herrera 2008).
To investigate the different forms, orientations, and directions of care, one would need as a first step to adopt an approach that follows longitudinally and spatially the migrant so as to capture care giving and receiving. Given the recent increase in the use of domestic and care labor, many studies of such work in the household are concerned with the demand for labor and the relationship between employer and employee at a particular time and place. The problems encountered in reconstituting the families of domestic workers or other care workers are dealt with as a separate issue. Bringing together labor and family life would enable us to build up a more complete picture of the migrant caring subject within a broader perspective of the social reproduction of their own and other families. To this end, the use of the global householding framework, as the nexus of the life cycle and its diverse activities, including care, would be a concrete way of articulating the different circuits in the overall social reproduction of the household. Just as importantly, it is a way of acknowledging that migrant and non-migrant households in destination states are both part of the narrative of care and of social reproduction. It is also important that the global household is not just represented as an inward looking nexus but analyzed in relation to external institutional and welfare architectures (Kofman 2010;Kilkey, Lutz, and Palenga-Mö llenbeck 2010;Williams 2010), as the earlier writers on globalized social reproduction emphasized, and which is increasingly evident in writing on the contradictory and complex manifestations of care trajectories.

Conclusions
In recent years, social reproduction has once again become the subject of debate among a number of feminist scholars. Recent writing has considered the part played by the commodification of care in the household and the effect of changes in welfare state policies. So too have studies analyzed the globalization of social reproduction and the role played by migrant women in the uneven deployment of economic and social resources in providing care for different categories of individuals and types of families and for ensuring social reproduction, including their own, in different sites and transnationally. Although boundaries may be blurred between care and other activities, one of the distinctive aspects of much of the literature derived from a social reproduction perspective (Colen, Duffy, Nakano Glenn, Truong) is its delineation between the different activities that contribute to social reproduction, whether it be within the household (cleaning, caring, food preparation, socialization of children, etc.) or in other sites. Concern with these differentially valued activities, materially and symbolically, has also been matched by an interest in the stratification, including the ability and right to reconstitute the family, to which these activities give rise.
The concept of global householding, too, draws on the notion of social reproduction, and is based on different activities undertaken throughout the life cycle to sustain and maintain households, some of which are multi-sited and multi-national. Global householding highlights the way in which economic and social resources are accumulated and deployed and how different types of migration enable such resources to be embedded and disembedded across space and time. The act of caring for oneself, other members of the household and the distant and unrelated other, traverses the different stages of the life cycle experienced by members of the household. Migration may lead to the reconstitution of households, the continuation of dispersed forms, or the formation of new households. The extent to which one is able to physically and socially reproduce families and households depends in turn on rights of mobility, immigration, residence, and citizenship status. The growing and rich literature on global gendered migrations and social reproduction have also provocatively drawn connections and trajectories between the reproductive aspects of labor and family migrations and hopefully resulted in a questioning of knowledge paradigms pertaining to these supposedly distinct circuits.
It is thus timely that the literature on care chains and trajectories in Europe, which as the papers in this volume demonstrate, are highlighting the complexity and diversity of care arrangements, begins to engage with the very rich and insightful literature on social reproduction. Reproduction always involves a complex interplay of a variety of actors in different sites so that it "is secured through a shifting constellation of sources encompassed within the broad categories of the state, the household, capital and civil society" (Katz 2001, 131). However, the literature stemming from the global chains of care has tended to dichotomize care giving and receiving. On the one hand, the transfer of labor to the global North enables households and other institutions of care to benefit from a care gain, on the other hand, the households of the South must resolve the problems of a care deficit and the redistribution of care giving between members of the family and other nonhouseholds carers. Indeed, assigning reproductive work to particular groups of people across the sites is very much a question to be addressed in contemporary processes of global restructuring (Bakker and Silvey 2008).
As highlighted in this article, care giving and receiving transcends labor migration and calls for an understanding of the articulation of different circuits of migration generating different configurations and dynamic trajectories. It should also not be forgotten that care, as an aspect of the social reproduction of families, both migrant and nonmigrant, in the different sites (households, communities, state, NGOs), is performed both by less-skilled and skilled workers, the latter often subjected to severe deskilling. Care labor is stratified by class, immigration, and citizenship status which in turn determines access to resources for familial reproduction from the state. I have not been able in the context of this article to address more fully these very complex issues and their contribution to social reproduction in different European welfare regimes but the care provided by migrants also ensures the reproduction of welfare states more generally, whether it be in the household, the private residential home, or a state institution.

NOTES
Eleonore Kofman is a Professor of Gender, Migration and Citizenship and Co-Director of the Social Policy Research Centre, Middlesex University, The Burroughs, London NW4 4BT, UK. Email: e.kofman@ mdx.ac.uk. She has written extensively on theoretical and empirical aspects of gendered migrations in Europe and has co-edited Gender, Generations and Family in International Migration, University of Amsterdam Press, 2011.
1. Molyneux interestingly notes that the emphasis was on the labor benefitting the male breadwinner rather than child rearing and care. It also failed to acknowledge the reproductive labor carried out by extra-familial agencies.
2. Bledsoe and Sow (2008) comment on the surprising lack of research on (biological) reproduction in anthropology and sociology which has received more attention from demographers.
3. Heteronormativity can also be critiqued in relation to families in the chains of care literature (see Kofman 2010;Manalansan 2006). 4. Schwenken divides knowledge into three forms: tacit and unreflected everyday knowledge of experience; knowledge and meaning generated by institutions and academia or law; and popularized knowledge in media, guidebooks, and social movements. This division draws upon the conceptualization of gender knowledge as the explicit and implicit negotiations about gender meanings and relations in society by Dö lling and Andresen (2005). 5. They probably reached this conclusion because the Irish were not considered as migrants and because of the sites selected for their fieldwork, namely Reading in Berkshire and Newcastle upon Tyne, which at the time had very few migrants or established minorities from the former colonies. A few years later Cox (1999) highlighted the ethnicization of domestic employment in London where a somewhat varied group of migrants with diverse immigration statuses (Australians, Irish, Filipinas, Europeans) filled the various jobs in this sector.
6. Safri and Graham (2010) prefer the term 'production' as the moment of economic activity within the cycle of reproduction and list a series of activities which occur within the household as a site of production. These include care of children and older people, food preparation, shopping, cleaning, sewing, household maintenance, etc. 7. Ireland, Sweden, and the UK opened their borders in 2004 with a number of other states in Southern Europe following in 2006 -07. Germany remained closed for the full seven-year transition period which came to an end on 1 May 2011.
8. Different rights in relation to entry and family formation in the EU need to be considered. Third-country national (non-EU) spouses of EU citizens exercising their mobility rights often have far greater advantages than those seeking to join citizens who do not enjoy freedom of movement rights. In contrast, the application of the visa regime makes it very difficult for African migrants compared with other non-EU migrants who do not require visas, e.g., from selected Latin American nationalities to reproduce their families biologically in Europe (Bledsoe and Sow 2008).