Towards a theoretical model for the study of productive and reproductive strategies in transnational families: Latin American migration and social mobility in Spain

ABSTRACT The aim of this article is to study the transnational social mobility strategies, focusing on the interaction between reproductive and productive strategies, of Latin American migrant households in Spain. We have created an analysis model based on the assessment of the differentiated strategic investments of transnational families in terms of four types of resources: physical and financial; educational; social and emotional. Our intention is to trace variations in strategic behaviour in accordance with the migrants’ family generation (grandparents, parents, children,) their migratory position (pioneering or regrouped migrants, family members remaining in the country of origin, etc.) and their family situation. A longitudinal approach has been adopted, in order to analyse the changes in social and spatial mobility strategies throughout the course of the migratory process, using biographical narratives and multi-situated fieldwork carried out in Spain with Latin American immigrants (Brazilians and Ecuadorians), as well as in certain countries of origin (Ecuador).

1. Gender, transnational families and social reproduction: An approach from the perspective of social mobility Traditional theoretical approaches have centred on providing explanations for population movements from an economistic perspective (the sum of individual decisions resulting from the rational weighing up of costs and profit) with the focus on the productive sphere. However, from the late 1980s onwards, a number of authors began to question this approach, considering migratory flows not merely in terms of individual decisions, but also within the framework of family and community strategies (Stark, 1984). This marked the start of a growing interest in gender and migration, particularly from the 1990s onwards. Furthermore, scientific production also began to address the reproductive sphere in migration studies (Catarino & Oso, 2014). 1 Literature started to highlight the growing demand in northern households for immigrant female workers from southern countries, as part of a global transfer of reproductive services associated with traditional female roles (care, domestic work and sex work) which in turn is attributable for the globalization of the labour force (Truong, 1996). Changes in fertility rates, life expectancy and the rise in the female workforce are at the roots of the 'care crisis' and the worldwide commodification of care, generating a demand for migrant women to work in paid care work, such as domestic service, childcare and nursing (Benería, Deere, & Kabeer, 2015). Particular mention must be made of the studies addressing global care chains, which first appeared at the start of the twenty-first century with the pioneering work carried out by Ehrenreich and Hochschild (2002). 2 Within the framework of this approach, transnational maternity has been one of the most frequent objects of research, including the study of the impact of female migration on the children left behind in the country of origin (Parreñas, 2003, among others). Kofman and Raghuram (2015) also posit the importance of analysing subject gender and migration from the perspective of social reproduction, stressing that migration is both a cause and an outcome of the depletion in social reproduction. Of particular interest is the analysis model devised by Baldassar and Merla, namely the 'care circulation approach', which they defined as 'the reciprocal, multi-directional and asymmetrical exchange of care that fluctuates over the life course within transnational family networks subject to the political, economic, cultural and social context of both sending and receiving societies' (Baldassar & Merla, 2014, p. 25).
However, and as Catarino and Morokvasic (2005) point out, the focus on reproductive roles has meant that migrant women's contribution to the productive sphere has been largely overlooked. Indeed, it entails the maintenance of the separation between the productive/male sphere and the reproductive/female sphere, perpetuating this classic dichotomy in literature on migration (Catarino & Oso, 2014).
In an attempt to move beyond this dichotomy and articulate both spheres, Oso and Ribas-Mateos (2015) analysed global production and reproduction chains, highlighting the new models that have developed around the emerging trends played out by women in contemporary mobility flows. They consider the importance of the link between transnational migration, remittances, the circulation of care and the process of change and development in the places of origin and transit of migrants. In turn, Catarino and Oso (2014) also posit this articulative approach, providing an analysis framework for assessing the migration/gender and development nexus through a prism of interconnected production and reproduction strategies, deployed by transnational households, taking into consideration the relationship between production and reproduction holistically.
It is therefore clear that recent literature on 'global care chains', gender and remittances has shed considerable light on the articulation of social reproduction strategies based on their economic and financial dimensions. However, with the exception of studies such as those of Singh and Cabraal (2014), or Reynolds and Zontini (2014), 3 very little research has been carried out to date linking the study of the migrant population's social reproduction strategies and practices with the analysis of one of the core dimensions of literature addressing international migration, namely social mobility trajectories.
Traditionally, studies addressing the connection between migration and social mobility have adopted a quantitative perspective (Borjas, 2006;Chiswick, 2000). An exception to this is the work of Bertaux and Thompson, who address the issue from a qualitative approach, based on the analysis of family histories (1997). Scientific production has mainly attempted to highlight the labour insertion trajectories of the migrant population (Papademetriou, Somerville, & Sumption, 2009), revealing a tendency to exclude other dimensions of social mobility that are not linked to occupational indicators (Aysa-Lastra & Cachón, 2013). 4 This includes studies into 'second generations' or immigrant children, which fall directly within the debate surrounding social integration (Portes & Zhou, 1993;Thomson & Crul, 2007), and which also analyse the intergenerational educational and occupational trajectories of migrant origin population in the host societies (Meurs, Pailhé, & Simon, 2006, among others).
Nevertheless, scientific production has so far failed to consider other dimensions of social mobility (such as social and emotional resources), focusing instead on the study of social mobility trajectories from a productive perspective, rather than addressing their articulation with household social reproduction strategies. Furthermore, the studies carried out to date have not taken into account the transnational aspects of social mobility within the context of geographical mobility, reflecting the epistemological and methodological nationalism associated with traditional migratory studies (Wimmer & Glick Schiller, 2002).
Widespread globalization has challenged the theoretical and epistemological model of mobility, based on the notion of the nation-state and a territorial approach (Favell & Recchi, 2011), with emerging analysis frameworks that attempt to adapt the study of individual movements throughout the social hierarchy to the context of the new logics of migration, such as the transnational approach (Glick Schiller, Basch, & Blanc-Szanton, 1992;Levitt, 2001) or the 'mobility turn' (Sheller & Urry, 2006). This latter analytical paradigm considers the study of new forms of mobility, moving beyond mere geographical movements that cross the borders of national territories. 5 Faced with the need for research that considers the analysis of the articulation between productive and reproductive strategies from a transnational approach, 6 this article describes a theoretical model based on the multidimensional investments of transnational families in terms of physical, social, financial and human capital for the analysis of the social mobility strategies and trajectories of migrant families. We will also take into account emotional aspects, considering the impact of migratory and social mobility strategies on migrants' affective resources.
Our intention is to trace variations in strategic transnational behaviour in accordance with three key variables that form the bases of our analysis proposal: (1) gender; (2) migrants' family generation, understood as the situation of each individual in terms of their genealogical tree (grandparents, parents, children, etc.) and their role in the transnational household's migratory project (pioneering migrants, regrouped migrants, family remembers remaining in the country of origin, etc.); (3) family circumstances (marital status, the existence of dependents in the household, etc.). The following questions are posed: What is the impact of migrants' reproductive and productive strategies on the social mobility trajectories of transnational households? And on those of their children? How do they differ in terms of gender and generation? Do certain transnational strategies represent social stagnation for some household members, yet revert on others in the form of upward social mobility trajectories? 7 In order to answer these questions, the article begins with a theoretical framework that includes the study of productive and reproductive social mobility strategies in transnational households. This section describes both the proposed analysis model and the methodological approach employed, based on the use of biographical narratives and empirical multi-situated fieldwork carried out in Spain with Latin American immigrants (Brazilians and Ecuadorians), as well as in certain countries of origin (Ecuador). This is followed by a discussion of the main results obtained from the analysis. Finally, a series of concluding arguments is provided regarding the articulation of productive and reproductive practices employed by Latin American families in Spain, within the framework of their transnational social mobility projects, strategies and trajectories.
2. The articulation of productive and reproductive social mobility strategies as tangible and intangible resources The theoretical model we propose is based on a multidimensional approach to the interaction between the productive and reproductive strategies used by the migrant population. The aim is to go beyond the analysis of the traditional dichotomy the majority of scientific literature has drawn between these two spheres. It seeks to understand the way the relation between them defines and shapes the social mobility strategies and trajectories of transnational households.

Production and reproduction strategies: an interacting system of tangible and intangible resources
Our analysis proposal conceptualizes the reproductive dimension associated with care, emotions and also affection as an intangible asset in which Latin American migrants invest and transfer throughout the transnational space during the various phases of the migratory process. 8 We therefore consider reproductive strategies not only as relational aspects, but also as an asset included in the ongoing interaction with the other resources in which the migrant population invests.
The analytical model considers the strategic investment choices of transnational households in terms of four types of resources or assets: (1) physical and financial; (2) educational; (3) social; (4) care and emotional resources. Our definition of the first four types of resources is based on the contribution by Moser (2009), 9 whilst we have opted to define care and emotional resources as those affective assets related to the emotional sphere and individuals' reproductive dimension.
Our multidimensional analysis framework also considers the interaction between the various resource investments. In this sense, the general hypothesis of our research (H G1 ) is that investment in some types of assets may impact directly on the accumulation of the remaining resources and therefore transform the social mobility strategies and trajectories of transnational households. This process, whereby assets feed on one other, does not establish a lineal connection between them, nor does it adopt a necessarily positive nature. As a result, investment in certain types of resources, such as those related to emotions and affection, may impact negatively on the accumulation of other types of resources, leading to the loss of financial and monetary assets.
We also posit that tangible and intangible resources can be defined and considered both from individual and collective spheres (H G2 ). As a result, they can be possessed, accumulated and transferred by both the migrants themselves or jointly by the transnational household. This implies that migrants' investments in assets may lead to varying consequences at both levels of analysis, and therefore may not affect the various members of the domestic group included in the social mobility project in equal measure. This could produce differentiated socially ascending or descending trajectories between specific members of the household, as well as the transnational family as a whole. This often results in the build-up of serious transnational tension between the migrants and their families, particularly when conflicts arise between investments aimed at obtaining an improved position in the productive sphere or the accumulation of assets destined for enhancing care, affective and emotional resources. 10

The transnational dimension of social mobility
The idea of production and reproduction strategies as part of an interacting system of tangible and intangible resources throughout the transnational space requires the inclusion of a spatial dimension, including at least three social spaces: (a) the community of origin; (b) the receptor social space and; (c) the transnational social space, made up of the community created within the framework of transnational social relations. All three spaces will be governed by differing social hierarchies and the position of the migrants and their families may vary in each. We believe that investment in a specific asset in the country of origin may have direct positive or negative consequences on the accumulation and transfer of other resources within the receptor context, leading to differentiated mobility expectations and trajectories in both spaces and to varying effects at both ends of the transnational space.

Gender, generation and family situation as analytical variables
Finally, it must be stressed that our analysis proposal considers that the process, whereby productive and reproductive resources feed mutually on one another, is determined by the convergence of three key analytical variables that should be considered in the study of transnational social mobility strategies and trajectories: (1) gender; (2) generation; (3) family circumstances.
Firstly, and regarding the question of gender, we consider that the various itineraries are traced by both men and women in the transnational social space. Our specific research hypotheses include the idea that this variable plays a crucial role in explaining changes to the strategic behaviour of the migrant population, through the definition of differentiated transnational social mobility trajectories resulting from the divergence in the investments made by men and women in productive and reproductive resources (H E1 ). We have therefore included the study of all those agents participating in or forming part of the social mobility strategies of the transnational household, rather than focusing our analysis exclusively on the link between migrant women and reproductive roles which has traditionally centred the attention of scientific literature.
Secondly, our analysis model is based on an intergenerational approach, which takes into account all members of the transnational households, including those that remain in the country of origin, and considers the various generations that take part in the construction of social mobility strategies. These generations can be defined not just as the demographic cohort each individual belongs to, but rather as the role played by each actor within the transnational household and their position in terms of the family migratory project. We posit that the articulation between productive and reproductive social mobility strategies, set up within the transnational space, will vary depending on which member of the family initiates the migratory chain and their role within the family structure (H E2 ). Furthermore, and in addition to these analysis categories, we consider that the structure of the transnational household, and more specifically migrants' family circumstances, are crucial in understanding the reasons underlying the choice of productive and reproductive social mobility strategies. For the purpose of our model, 'family circumstances' are defined as the migrants' marital status, and particularly the existence or lack of family responsibilities (children) when embarking on the migratory process. This is due to the fact that scientific production has revealed that the existence of children in the household heightens feelings of responsibility and obligation (Instraw, 2006;Parreñas, 2003), bringing to the fore the importance of articulating the analysis of collective projects and expectations through the study of individual decisions and trajectories. In the light of this, we believe that households with transnational family responsibilities deploy strategies that differ from those of migrants embarking on their migratory project without this type of responsibility, investing more heavily in reproductive and emotional resources stemming from caring for and maintaining the transnational domestic group (H E3 ).
In short, our multidimensional analysis framework posits that the articulation between the production and reproduction strategies deployed by migrant households can be defined in accordance with a model of interaction between varying types of resources (financial and educational resources, social assets, emotional and care resources) that families invest in as part of their migratory process. This interaction is determined and shaped by the convergence between the model's three principal variables (gender, generation and family situation), leading to the emergence of differentiated social mobility trajectories that are formed, modified and transformed in three spatial contexts: the host country, the country of origin and the transnational social space. This analytical approach can be summed up in the following graph: (Figure 1).
This theoretical framework is based on the analysis of the empirical data obtained from qualitative fieldwork. The methodological strategy was carried out using multi-situated research in various geographic contexts and conducted with migrant families from Ecuador and Brazil. Interviews were conducted with Ecuadorians living in Madrid and Ecuador, specifically in the district of Turubamba Bajo in the south of Quito, where relatives of migrants settled in Madrid were interviewed. The fieldwork was carried out at two different time points, during 2007-2009 (immediately prior to and during the onset of the financial crisis) and in 2013-2015. 11 With regard to the Brazilian participants, interviews were conducted in A Coruña and Barcelona between 2012 and 2015; no fieldwork was carried out with this group in the country of origin. 12 A total of 69 people were contacted for the purpose of our study, including both migrant groups. We have attempted to apply a longitudinal approach, monitoring the biographical narratives of six migrant families, recorded from various sessions held with different members of each family, who were contacted on various occasions during the course of the fieldwork. In addition to tracking these families, we conducted additional interviews with other transnational households that were interviewed only once. 13 The technical model was based on the analysis of the full set of interviews conducted during our research. Nevertheless, and for the purpose of illustrating the model, the following section provides a detailed analysis of a selection of four transnational families. 14 The intention is not to draw a series of generalized conclusions, but rather to illustrate through empirical material how the model can be put into practice.
The families were selected for their highly illustrative profiles, in terms of their typological representation 15 : (1) Unmarried children with no family responsibilities that lead the migratory process; (2) Married mothers acting as heads of transnational households; (3) Married fathers acting as heads of transnational households; (4) Grandparents moving within the transnational space.
3. Reproductive and productive social mobility strategies of Latin American transnational households 3.1. From individual expectations to family social mobility projects: unmarried children with no family responsibilities that lead the migratory process Estela was single when she left Ecuador in 2000 with the idea of travelling to Spain in order to take up her studies at a later date. On arrival in Madrid, she first worked in domestic service before moving into the catering industry. However, Estela's migratory project eventually fell within the framework of a family social mobility strategy. In the initial stages of the migratory process, our informant would send half her salary back home, which was managed by her mother who invested it in extending the home and the family business (telephone booths and computers were installed in the ironmonger's located on the ground floor of the family home). Nevertheless, when Estela decided to marry an Ecuadorian in Madrid (2005), this family social mobility strategy came to a halt, as did the sending of remittances. Estela and her husband bought a flat in Madrid, investing their earnings in paying off a mortgage. The couple decided to separate (2007), and Estela once again started sending remittances to her family in Ecuador to buy business premises in her name, this time in response to her own individual strategy (aimed at forging a professional future for herself, should she decide to return). Nevertheless, she continued to support her family, as she decided to grant her brother and sister-in-law, who had a baby, rent-free use of the premises so that they could set up a business that would provide them with a livelihood. In time, Spain would be affected by the economic crisis and Estela's mother would fall ill. She looked after her mother until her death (2010).
Estela decided to go back to Spain to earn more money for two more years and returns to Ecuador in 2013, taking some savings with her that she deposited in a bank account, this time for her personal expenses. However, she was unable to settle in her country of origin and decided to migrate again, this time to Buenos Aires in 2014, where she gave birth, as a single mother, to her daughter.
Estela's family history reveals how her individual social mobility strategy, articulated through migration, is embedded in a wider family strategy, which extends to other members (parents, brothers and sisters), even though she migrated as an unmarried daughter, without an initial migratory project aimed at supporting the transnational household. Estela's decision to invest in financial and physical resources led to a better economic situation for the family in the social space of origin (Ecuador), although this had a negative impact on the human assets (our informant was unable to study in Spain) and her professional trajectory (saving was only possible through work in mainly unskilled areas such as domestic service and later the catering sector). Furthermore, our case study reveals how marriage in Spain can curb the family's transnational social mobility strategy, as migrants tend to channel their investment efforts towards a new social mobility project with their partners. Lastly, her mother's illness led Estela to return to her country of origin, in driven partly by the financial crisis, but also by the responsibility of caring for her mother at the end of her life (a reproductive strategy). Estela's trajectory also includes a re-emigration project, which, following her arrival in Buenos Aires and the birth of her child, results in the further restructuring of her social mobility strategies, which this time are focused on the reproduction of her new singleparent family. Her emotional and affective concerns are once again centred outside Ecuador. The following graph shows the sequence of Estela's resource investments and how her productive and reproductive strategies are articulated (Figure 2). 3.2. Production, reproduction and delegated social mobility projects: married migrants with family responsibilities

Married mothers acting as heads of transnational households
Aparición travelled to Spain alone as a pioneer migrant in 2009 in order to accumulate the necessary financial capital to invest in education resources for the daughter she had left behind in Brazil under her husband's charge, with the intention of eventually returning to her country of origin with the financial capital necessary to guarantee improved opportunities for her domestic group. However, several years after her arrival in Spain, she suddenly found herself out of work, as a result of the economic crisis.
She considered the possibility of returning home, a bitter pill to swallow for the mobility expectations of the transnational household. The family strategy behind Aparición's decision to remain in Spain in order to guarantee investment in financial resources even led her to prostitution for a while. Although this episode implied major gains in financial assets and a significant rise in the family's social status in Brazil, it resulted in a complete loss of social status and empowerment for our informant. Aparición therefore stood back completely from the decision-making process carried out in her transnational household as a result of the feelings of shame and humiliation derived from working in the sex industry.
Finally, Aparición found work in a geriatric home in 2014 looking after pensioners, which started an upward individual mobility trajectory enabling her to acquire sufficient resources to bring her daughter to Spain. From that point on she would focus her efforts on a strategy aimed at improving their situation in the host country, putting aside transnational family expectations and strategies which had previously determined her social status. This reduction of remittances being sent home resulted in the loss of affective and emotional resources in Brazil, but implied a certain degree of empowerment.
The history of this informant reveals how the migration of women with family responsibilities in transnational households often requires significant economic investment for the provision of domestic and care tasks, which even if they are assumed by the husband, are in fact often carried out by another woman who takes responsibility for them. Aparición is therefore forced to make major investments in reproductive resources in order to cover the domestic and care tasks of a household she is unable to cope with, delegating in her mother and husband. This produces feelings of obligation and responsibility not only in terms of attention and care for her daughter, but also in her relationship with her husband.
Her testimony equally reflects the stigma that has built up over time towards the negative effects of transnational motherhood on children. As a result, pioneer migrant women earn a certain reputation as 'bad mothers', a fact that has already been highlighted by other researchers (Pedone, 2008). Therefore, although the migrant gains in financial capital by leaving the household, it may lead a loss of emotional resources (dual affective distancing from both the husband and children) along with social resources (regarding the community in the country of origin), resulting in reduced individual social mobility trajectories.
It is therefore clear that emotional considerations and care resources are of major importance in the definition of this migrant's upward social mobility strategies, as well as in the analysis of her social mobility trajectory. Aparición's investment and asset accumulation decisions are largely determined by the role played by her reproductive strategies and investment in transnational affective assets. At various points throughout her migratory process, they are responsible for the loss of other types of resources, including financial assets, or are even directly responsible for her downward occupational trajectory. Indeed, our informant sacrifices her social position and opportunities for gender empowerment within her transnational household in favour of the welfare of the domestic group, shaping a 'delegated' social mobility project aimed at promoting the upward social mobility of the household in the country of origin (Figure 3).

Married fathers acting as heads of transnational households
Natividad had five children when her husband, Mariano, decided to migrate to Spain in 1999, with the idea of returning after two or three years. At the time of the first interviews (held in 2008 in Quito with Natividad and two of her sons and in Madrid with Mariano), the regular remittances ($700 a month) that Mariano sent were the principal source of income for Natividad and the children that remained behind in Ecuador. Natividad managed the money she received from her husband, which provided her with a certain independence and power through the possession and management of financial capital. However, during the interview she repeatedly stressed the negative consequences of having to take full reproductive responsibility for raising her children alone.
Initially the family invested practically all the remittances in improvements to the home (physical capital) and, above all, in human capital, taking children out of state education and placing them in private schools, which is one of the main social mobility strategies migrants families apply (Egüez & Acosta, 2009). Natividad was also able to open a small grocery business in her own home that gave her a supplementary income without having to abandon her family-related reproductive tasks.
Natividad and Mariano were a very close couple, but the husband's extended stay in Spain eventually led to their separation. Mariano met an Ecuadorian woman in Spain (2002) and had another child, who was born in Madrid, with his new partner (2005). Remittances were maintained, but the financial management of this transnational household became increasingly complex following the appearance of Mariano's 'second family' in Spain. Mariano decided to buy a flat in Madrid with the help of his new partner, thereby investing in physical capital in Spain, thinking about the future of his new family in Madrid.
We held a second interview with Mariano in Madrid in 2014. Time and the onset of the economic crisis in Spain had altered the family's transnational geographical and social mobility strategies. Mariano's purchasing power was lower following a cut in his salary and he was forced to reduce the remittances he sent to Ecuador, which dropped to 500 dollars a month. When we interviewed Natividad for a second time in Quito in 2015, she informed us that she had lost the home the couple had owned before her husband migrated. Poor financial management had led to debts and the property was eventually seized. The family in Ecuador still depends on the remittances Mariano sends, as the small food business run by the mother only provides a small boost for the household's income.
In addition, the economic crisis had a negative impact on Mariano's physical resources in Madrid. Our informant acted as a guarantor for a relative who purchased a flat but lost his job during the recession and left Spain without repaying his mortgage, so Mariano's own property (apartment) now is at risk, as it can be seized by the bank.
As a result, the 'dual' transnational household has lost its physical capital (both in Ecuador and in Spain), and Mariano's economic resources are now used almost exclusively for the basic subsistence costs of the two family units (Natividad and her children in Ecuador, and his family in Spain). Investment is now channelled into improving his children's human resources (education) both in Quito and Madrid.
It is therefore clear that this dual transnational strategy has led to an increase in investment in reproductive resources, allowing for the maintenance of two family units in Ecuador and Spain, although it has impacted negatively on the possibility of accumulating other types of resources (physical, financial, etc.) (Figure 4). 3.3. Beyond transnational care: grandparents moving within the transnational space Nelly embarked on a migratory project in order to help her two sons back in Brazil in 2008. On arrival in Spain she found work in domestic service, enabling her to acquire financial capital within a relatively short space of time, and sending large remittances, that enabled her eldest son to set up a business with his wife and younger brother. Once her sons had achieved a degree of financial stability, Nelly decided to adjust her strategy, turning her attention to saving up for her return to Brazil.
Nelly decided that she had saved up enough to guarantee her a decent standard of living in Brazil, so she gave up her job in Spain and returned to help her son with the family business in 2011. However, a year and a half later, the business went under, coinciding with the birth of Nelly's first grandchild. Faced with this new situation, Nelly returned to Spain (2013), but this time as a grandmother. Indeed, on this occasion, her mobility strategy was centred on sending back as much money as possible to Brazil to provide for her grandchild. Nelly's current expectations are therefore centred on guaranteeing upward social mobility for the transnational household, albeit at the expense of her own mobility strategy.
Scientific production has repeatedly revealed the major role grandparents play as reproductive anchors for the transnational household in the country of origin, bringing provision of care and affection in the absence of the children's mothers (Herrera, 2013). However, Nelly's history reveals a completely different scenario in which grandmothers play an active part in the migratory strategy, abandoning their traditional role as carers and becoming the heads of transnational households. Nelly first embarked on the migratory process as a mother and later as a grandmother. She projects her emotional and affective resources on her domestic group in Brazil, sending large remittances in order to boost the upward mobility of her transnational household.
The principal difference between the two phases of the migratory process (as a mother and as a grandmother) lies in the fact that during her initial stay in Spain, Nelly was able to invest part of the financial assets acquired in her own individual mobility strategy, which consisted of securing a certain standard of living and welfare on her return to Brazil. When she migrated to Spain for a second period, this time as a grandmother, all her assets were channelled directly into the social mobility project of her domestic group in Brazil, trying to achieve the best possible position and opportunities for her grandson.
In this sense, and although Nelly abandoned the traditional role associated with these women in her country of origin, namely that of carers, it is also true that her migratory project falls within a clearly reproductive strategy, designed to ensure the upkeep of her children and grandchild. Her decision to migrate to Spain for a second time is attributable to a strategy for the accumulation of reproductive and emotional assets for the family members that remain behind in Brazil. As she is responsible for leading the migratory process, she allows her son and daughter-in-law to care for their child without forcing them to migrate, as a delegated social mobility project, targeting the domestic group left behind in Brazil.
This family history also reveals the importance of including a longitudinal perspective in the empirical application of the theoretical model. The changes to this informant's family situation, in terms of her position in the generational structure of the household (namely the switch from mother to grandmother), shows how the productive and reproductive practices of the migrant population shift and are reconfigured over time, part of a dynamic process defined through the changing positions of individuals within their transnational domestic group. This process is articulated jointly with the other spheres of the migratory experience, giving rise to differentiated mobility trajectories ( Figure 5).

Conclusions
The main aim of this article was to construct a theoretical model capable of articulating the study of productive and reproductive social mobility strategies of transnational households, through a multidimensional framework that moves beyond the traditional dichotomy of both spheres conceptualized by international literature (Catarino & Oso, 2014).
The article has shown how the various types of tangible and intangible resources in which transnational Latin American families invest interact and feed upon one another, allowing for the definition of various types of productive and reproductive strategies, which in turn, impact directly on transnational households social mobility trajectories. Productive and reproductive resources are transferred through the various members of transnational domestic groups, in an attempt to improve individual and family social positions. This leads to social mobility strategies, projects and itineraries that are continuously being renegotiated and readjusted within the transnational space. 16 These investments do not always have a positive accumulative effect on other assets, and may have a variable impact when transferred between origin and destination, leading to the breakdown of relationships and the loss of emotional and other types of resources, attributable to transnational tension resulting from the clash between individual and collective productive and reproductive strategies. In this sense, the central role of the transnational family as a core analysis unit is clear: it forms a framework for negotiations and the decision-making processes that determine production and reproduction strategies as well as social mobility trajectories. Based on the empirical application of the proposed analysis framework, it can be also seen how the decision to adopt a certain social mobility strategy is determined largely by gender; the type of migrant household studied, which in turn is defined by the generation to which each individual belongs to; their role within the family, geographical mobility, understood in this case from the perspective of pioneer migrants; and family circumstances, understood as the existence or absence of family responsibilities in the household.
As the analysis of our informants' family histories reveals, those migrants who are not bound by major responsibilities, due to their family's circumstances (unmarried migrants), can focus their investment efforts on individual social mobility projects, although they are also directly linked to the upward social expectations of the transnational domestic group. These projects tend to be associated with the accumulation of physical and financial assets, capable of securing an upward social mobility in the country of origin that is compatible with satisfying the material demands of the family left behind.
In contrast, in the case of married migrants with family responsibilities playing out the role of parent or grandparent, the transnational demands of the domestic group tend to exert greater pressure on the family social mobility plans, which results in greater investment in reproductive, emotional and affective resources. This is particularly true in those cases where caring for children and the home has to be assumed in the country of origin. In these cases, reproductive strategies acquire a greater significance, often to the detriment of other types of assets such as financial and physical resources. Consequently, individual social mobility projects are often delegated for the benefit of second and third generations.
In conclusion, the theoretical model presented in this article provides a multidimensional approach to the productive and reproductive social mobility strategies of Latin American transnational families. Although the study of aspects such as care or emotional and affective resources does not represent an innovative scientific approach to international migration phenomena, the analysis of their interrelation with the productive sphere highlights the need for new theoretical challenges. Considering care, affection and emotions as resources, and analysing their articulation with other type of investments (physical, financial, educational and social assets), sheds considerable light on the analytical complexity associated with the study of the interaction between productive and reproductive social mobility strategies, highlighting the need to include analytical dimensions beyond mere economic considerations when analysing transnational migratory processes.
Finally, it is particularly important to conceptualize these strategies as a highly complex and dynamic process that involves all members of the domestic group, as well as adopting a transnational perspective that allows for their consideration beyond the host community. Notes 1. See Catarino and Oso (2014) a greater insight into the shift from literature's productivecentred approach to a growing interest in social reproduction. 2. On this issue see the review of the literature by Oso and Parella (2012). 3. These authors analyze the interaction between family ties and care with other resources and forms of capital (namely social and cultural) circulated by individuals in order to boost their social status. 4. Although certain researches have considered the role played by variables such as social capital (Massey & Aysa-Lastra, 2011), or gender, when analysing the shifting positions of individuals within the social structure. 5. A more detailed bibliographical review of the connection between migration and social mobility can be found in Oso and Suárez Grimalt (forthcoming). 6. Some studies have addressed the analysis of Latin American migration to Spain from a multidimensional, gender and transnational approach (Herrera, 2013;Sanz Abad, 2015;Vega, 2016). Nevertheless, they have not assessed the issue from an intergenerational social mobility perspective. 7. Our theoretical approach distinguishes between the concepts of social mobility strategies, projects and trajectories. Indeed, we understand social mobility strategies to be the actioninducing decisions individuals make in order to climb the social ladder, differentiated from the social mobility projects, which are considered as future dreams or expectations. In turn, we define the social mobility trajectory as the group of personal or collective itineraries that individuals develop throughout the social mobility hierarchy in relation to one or various social spaces (origin/destination), or within the transnational space (Oso & Suárez-Grimalt, forthcoming). 8. By assets in which migrants invest, we are referring to all those resources capable of generating social impacts that may be acquired, developed, improved and transferred through the generations (Ford, 2002;cited in Moser, 2009). Tangible assets refer to all those physical resources such as money, housing, education and productive assets, whilst intangible assets are understood to be relational, affective, psycho-social, civic and political resources, etc. 9. Physical capital refers to the stock of plant, equipment, infrastructure, and other productive resources owned by individuals, the business sector, on the country itself. Financial capital is defined as the financial resources available to people (such as savings and supplies of credit). Human capital refers to investments in education, health, and the nutrition of individuals. Labor is linked to investments in human capital, health status determines people's capacity to work and skills and education determine the returns from their labor. In terms of social capital, it is considered an intangible asset, defined as the rules, norms, obligations, reciprocity, and trust embedded in social relations, social structures, and society's institutional arrangements. It is embedded at the micro-institutional level (communities and households) as well as in the rules and regulations governing normalized institutions in the marketplace, political system, and civil society. (Moser, 2009, p. 18) 10. This hypothesis has been developed in the framework of the following research project Oso, dir, (2011Oso, dir, ( -2014. See Oso, Sáiz-López, and Cortés (2017). 11. A total of 44 persons were contacted during the course of the two periods of fieldwork with Ecuadorian population. This fieldwork was conducted by Laura Oso. 12. The qualitative research in the case of the Brazilian participants included 25 semi-structured interviews and was conducted by Laura Suárez-Grimalt. 13. For a more detailed description of this methodological approach see Oso and Suárez-Grimalt (forthcoming). 14. A detailed insight into family dynamics is necessary in order to shed light on the articulation between productive and reproductive strategies within the framework of interaction among the various types of resources. In this article, space restrictions prevent us from including more than four cases. 15. Typological representation is based on identifying population trends and uniformity through the identification and analysis of representative cases. It does not refer to statistical representation. 16. Concerning the evolutions over the life-course of the roles undertaken by migrants in their family network and the evolutions also of the expectations they face from their relatives, see Wall and Bolzman (2014).

Disclosure statement
No potential conflict of interest was reported by the authors.