‘Care drain’. Explaining bias in theorizing women's migration

94 The Romanian Journal of Society and Politics Speranța Dumitru “CARE DRAIN”. EXPLAINING BIAS IN THEORIZING WOMEN’S MIGRATION Speranta Dumitru Faculty of Law, University of Paris Descartes and CERLIS (CNRS), University of Sorbonne Paris Cité speranta.dumitru@parisdescartes.fr KEYWORDS ABSTRACT Migrant women are often stereotyped. Some scholars associate the feminization of migration with domestic work and criticize the “care drain” as a new form of imperialism that the First World imposes on the Third World. However, migrant women employed as domestic workers in Northern America and Europe represent only 2% of migrant women worldwide and cannot be seen as characterizing the “feminization of migration”. Why are migrant domestic workers overestimated? This paper explores two possible sources of bias. The irst is sampling: conclusions about “care drain” are often generalized from small samples of domestic workers. The second stems from the affect heuristic: imagining children left behind by migrant mothers provokes strong feelings of injustice which trump other considerations. The paper argues that neither source of bias is unavoidable and inds evidence of gender stereotypes in the “care drain” construal • care • • • • • • migrant women sexism feminization of migration domestic work global inequality globalization 1. Introduction This article explores the stereotyped construal of migrant women as a “care drain”. Coined by the sociologist Arlie Hochschild (2002), the phrase “care drain” associates feminization of migration with care workers and describes it as a cost for the Third World children[1]. Hochschild theorized women’s migration in the following terms. In the First World, women’s integration into the labor market releases them from domestic activities and leads to an increase of the labor demand in the care sector. The demand for care workers in the First World thus triggers women’s labor migration, especially from the Third World. According to Hochschild (2000), migrant women form “global care chains”: they work as caregivers for richer women in the First World, while sending money to poorer women in the Third World to care for their own children left behind. At the end of the “global 1 Throughout this paper, I use “First/Third World” to follow, without endorsing, Hochschild’s initial terminology. Speranta Dumitru 95 care chain”, there must be a “care drain”, a loss of care for the Third World children. The phrase “care drain” is coined as a female analogy of “brain drain”: “just as poor countries sufer a brain drain as trained personnel move from the South to the North, so too they sufer a care drain” (Hochschild: 2013, p. 147). This article investigates the source of bias in care drain theory. As a matter of fact, the 2.5 million migrant women employed as domestic workers in Northern America and Europe represent 2% of migrant women worldwide and cannot characterize the “feminization of migration”. Furthermore, the subset of domestic workers who are both mothers and who left their children in the Third World countries is even smaller. Therefore, the view characterizing First World’s “importation of care and love from poor countries” as “an important trend” is deeply biased (Hochschild: 2002, p. 17). How can the biased nature of “care drain” construal go unnoticed? The paper explores two possible responses to this question. One source of bias could come from sampling: research endorsing the “care drain” construal is generalized from small samples of migrant women who are domestic workers in developed countries and have been interviewed about their children left in developing countries. A second source of bias is emotion: imagining children left behind by their mothers provokes deep empathy and feelings of injustice that trump any other consideration and lead to overestimating the scope of “care drain”. This paper concludes that neither source of bias was unavoidable in the literature on “care drain”. The remaining article is divided into three parts. Part 2 presents the available data on migrant women and on domestic workers in order to assess the magnitude of bias involved in the construal of migrant women as “care drain”. Part 3 explores the sample of Filipina domestic workers interviewed by Hochschild (2000) and Parreñas (2001) and shows that the stereotyped construal was avoidable: more than half of them were college educated and almost half of them had left no children behind. Part 4 explores the emotional source of bias and shows that the “care drain” account fails to grasp the more severe global injustices which afect migrant domestic workers and poor children when they are considered separately. 2. Overestimating migrant domestic workers This section shows that the view which associates “feminization of migration” with domestic work is not supported by facts. Hochschild believes that “what is unprecedented is the scope and speed of women’s migration to [care] jobs” and that “many factors contribute to the growing feminization of migration…” (Hochschild: 2002, p.17). This view is widespread and even features in dictionaries and encyclopedias. For instance, it is asserted that “feminization of migration is characterized by (…) a concentration on female-speciic work such as domestic helpers, nurses, entertainers” (Yoshimura: 2007, p. 1515) and it is claimed that “demand for domestic work is recognized as a key factor behind the feminization of migration” (Oelz: 2014, p. 145). The view associating the “feminization of migration” with low-skilled jobs originates in the neo-Marxist theory of the “new international division of labor”. In 96 The Romanian Journal of Society and Politics the 1980s, this theory predicted that globalization would result in a new division of labor: outsourcing would industrialize Third World countries at the expense of Western countries which in turn deindustrialize and specialize in the service sector (Fröbel, Heinrichs, Kreye: 1980). The division is “new” by contrast to the “old” division of labor during the colonial times when European countries controlled the industrial transformation of raw materials extracted from the colonized regions. Another prediction of the neo-Marxist theory was the “feminization of migration”: the global expansion of capital “uproots” women irst from rural to industrial areas in the Third World countries and then from Third World to First World countries (Standing: 1989), (Sassen: 1988), (Mies: 1986). As the “feminization of migration” is understood as contributing to the new international division of labor, neo-Marxist scholarship tries to conirm the theory by focusing on migrant women from the Third World countries employed in low-skilled jobs, especially in the service sector. Migrant domestic workers thus become a textbook case for conirming the gendered international division of labor (Parrenas: 2000) and “care drain” is described as a “new imperialism” imposed by the First World which now “extracts love” and “emotional resources” from the Third World (Hochschild: 2002; Gündüz: 2013). Nevertheless, neither the “feminization of migration”, nor the association of women’s migration with domestic workers is supported by facts. Firstly, no evidence supports the phenomenon of “feminization of migration” as distinct from the “masculinization of migration” (Dumitru, Marfouk: 2015). The UN data reveals that over the last sixty years the proportion of women among international migrants has only slightly increased, from 47% to 48 % (see Table 1). While in absolute numbers, migrant women are nowadays three times more numerous than in the 1960s, the same is true of migrant men who still remain more numerous. However, there is no theorizing of a “masculinization of migration”.‬‬‬‬‬‬‬‬‬‬‬‬‬‬ Table 1: Evolution of the international migrants stock by sex (in thousands) 1960-2015 1960 1970 1980 1990 2000 2010 2015 77 115 84 460 101 983 152 563 172 703 221 714 243 700 Total Men 40 869 44 256 53 302 77 747 87 885 114 614 126 115 Women 36 246 40 204 48 681 74 816 84 818 107 100 117 585 % Women 47% 47,6% 47,7% 49% 49,1% 48,3% 48,2% Source: UN Population Division (2008 and 2015 revisions) and author’s calculation Secondly, asserting that the feminization of migration is linked to domestic work is a hasty generalization. According to the International Labor Organization, the total number of migrant women employed as domestic workers worldwide was 8.46 million in 2013 (ILO: 2015). In 2013, there were 111 million migrant women in total which means that the proportion of migrant women employed as domestic workers was 8% (see Figure 1). If the “First World” is equated with Northern America and Europe, where 2.45 million migrant women earned a living from domestic work in 2013, then domestic workers in those regions accounted Speranta Dumitru 97 for 2% of migrant women worldwide. It is noteworthy that “domestic workers” as deined by the ILO is a larger category than “nannies”. In any case, it is a mistake to generalize the profession of 8% to a characterization of the population of migrant women as a whole. Figure 1: Proportion of domestic workers among international migrant women Source: ILO (2015), UN Population Division (2013 revision) and author’s calculation Finally, to claim that “global care chains” characterize migrant women is to commit another logical error: the conjunction fallacy (Tversky, Kahneman: 1983). The conjunction fallacy is committed when the likelihood of a conjunction is assumed to be higher than the likelihood of its constituents taken separately. Hochschild deined “global care chains” as migrant women who work as caregivers in the First World and are mothers and have children left in the Third World countries. Therefore, migrant domestic workers involved in “care chains” are a subset of migrant domestic workers; and while migrant domestic workers represent 2% of migrant women, the subset is even less numerous and cannot characterize the “feminization of migration”. All in all, the literature about feminization of migration appears to be insensitive to information on base rates to calculate the prior probability of a random migrant woman being a domestic worker. In overestimating the likelihood of a migrant woman being domestic worker, scholars seem to use what has been called the “representativeness heuristic” (Kahneman, Tversky: 1973): that is, they evaluate the probability of an outcome (being domestic worker when one is a migrant woman) by its similarity with stereotyped characteristics of the parent population (women). Judgements made on the basis of representativeness ignore base rates and rely instead on stereotypes. 3. Small samples and sexist stereotypes This section considers sampling as a cause of overestimating the “care drain”. 98 The Romanian Journal of Society and Politics One possible reason why domestic work is taken as characterizing the “feminization of migration” is that scholars generalize from small samples of migrant domestic workers. However, a quick inspection of the initial sample used by Hochschild (2002) and Parreñas (2001) when theorizing the “care drain” and the “globalization of mothering” shows that hasty generalization was avoidable. First of all, the research assumed that “care drain” – that is, a loss of care for the children left behind in the Third World – can be proven not by investigating in the Third World the variation of care the children received before and after their mother’s departure, but on interviews conducted in the First World with domestic workers (Parreñas: 2001; Hochschild: 2002). Research on the children left behind was conducted by Parreñas (2005) only at a later stage. The initial sample was constituted exclusively of women domestic workers: 46 domestic workers in Rome and 26 in Los Angeles, all from the Philippines. Based on this sample, conclusions were drawn about global trends: Parreñas theorized the “globalization of mothering” and Hochschild analyzed the global injustice of “care drain”. However, the sample size is not the only problem as the demographic characteristics of Parreñas’ sample could have avoided hasty generalization. First, as Parreñas (2001, p.19) indicated: “women with children living in the Philippines constitute a greater portion of my sample in both Rome and Los Angeles: twentyive of forty-six in Rome and fourteen of twenty-six in Los Angeles”. That means that almost half of the interviewed women had no children at all living in the Philippines and the sample did not allow for generalizations about “globalization of mothering”. Second, the sample of domestic workers included for more than half college-educated women (Parreñas: 2002, p.262). What draws attention is the skill mismatch – the gap between women’s education and the profession they exercise. Nonetheless, faced with college-educated women, Hochschild chose to coin the term “care drain” as a feminine analogy for “brain drain”. Stereotyped thinking provides a better explanation of the overestimation of care drain. To be sure, both metaphors – “care drain” and “brain drain” – are ways to stereotype people by reducing them to a single characteristic. While educated emigrants employed in jobs for which they are overqualiied are usually termed “brain waste”, Hochschild chose to create a new category for women: the “care drain”. “Care drain” is deined as it follows: “Rowena’s life relects an important and growing trend: the importation of care and love from poor countries to rich ones. For some time now, promising and highly trained professional have been moving from ill-equipped hospitals, impoverished schools, antiquated banks, and other beleaguered workplaces of the Third World to better opportunities and higher pay in the irst world (…). But in addition to this brain drain there is now a parallel but more hidden and wrenching trend, as women who normally care for the young, the old, and the sick in their own poor countries move to care for the young, the old, and the sick in rich countries, whether as maid and nannies or as day-care and nursing-home aides. It’s a care drain” (Hochschild: 2002, p.17, my emphasis). Actually, Rowena, whose “life” is claimed to illustrate a “care drain”, could have illustrated a brain drain or a brain waste. Rowena, a Filipina migrant woman, Speranta Dumitru 99 certainly works as a nanny in the US, but according to the description given by Hochschild, she had “worked three years toward an engineering degree” in the Philippines (Hochschild: 2002, p.16). We are not used to describing the migration of domestic workers as “brain drain”, but having two years of tertiary education commonly classiies a migrant as being among the highly-skilled and thus as characterizing the so-called “brain drain”. As the International Organization for Migration observes, “the most basic deinition of highly skilled migrants tends to be restricted to persons with tertiary education, typically adults who have completed a formal two-year college education or more” (IOM: 2008, p.52). Rowena completed three, not two years of college education. Besides, the probability that other Filipina maids are highly-skilled migrants is indeed sizeable since no less than 64% of migrant women from the Philippines were highly-skilled in 2000 (Brücker, Capuano, Marfouk: 2013). Hochschild is aware that the pool of domestic workers she studies “includes college-educated teachers, businesswomen, secretaries (…) and more than half of the nannies [Parreñas] interviewed had college degrees” (Hochschild: 2000). Still, Hochschild does not qualify any migrant maid as “brain drain”. The kind of stereotypes on which the metaphor of “care drain” is built is sexist. Sexism is usually deined as unequal treatment of men and women based on a traditional ideology about sex roles. Traditional gender ideology especially separates sex roles in the family: men are supposed to fulill their family roles through instrumental, breadwinning activities, while women are supposed to do so through nurturing, homemaking, and parenting activities (Kroska: 2007; Lind: 2007). In migration studies, the representation of women as attached to the household and/or to the family could explain the assumption that they immigrate either less than men or mainly as wives; the “discovery” of women’s labor migration in the 1980s gave rise to the literature on the “feminization of migration”. In previous work, I have deined “methodological sexism” as a research bias which involves three criteria: women are studied only as caregivers, only women are studied as caregivers (men are excluded) and women’s failure to fulill their traditional family roles is judged regrettable (Dumitru: 2014). The three conditions are met by Hochschild’s description of the “care drain”. Firstly, women are overconstrued as caregivers. Though the deinition cited above suggests that any woman produces “care drain” since “women normally care for the young, the old and the sick” (Hochschild: 2002, p. 17), Hochschild ampliies the scope of a particular subset of migrant women. In order to support her thesis about “the global care chains” through which the First World “extracts love” from the Third World, she describes the migrant women as conjointly meeting two criteria: having children (in a poor country) and being employed in care work (in a rich country). “Care drain” is supposed to be produced by hiring as caregivers women who would otherwise have cared for their children in the poor country. But there are obviously one too many caregiving functions in this picture. When one aims to study the “care drain”, the variation of care a child is supposed to experience in the absence of her mother does not logically depend on the kind of job that the mother has at destination: once absent, whether she is working in engineering or in the care sector is prima 100 The Romanian Journal of Society and Politics facie irrelevant for what happens at home. Conversely, if one aims to study migrant women employed as caregivers, one should not discriminate according to their family status: whether they have children or not and, if they have, whether they reside with them or not, they are all migrant women employed as domestic workers. Yet, migrant women aren’t studied as workers. Not only is their condition of caregivers unnecessarily ampliied, but they are studied as “natural” caregivers. Their professional ambitions and diiculties, the fact that women may choose domestic work as the only legal way for them to cross international borders (Momsen: 1999), their work relationships, and the way they respond to eventual exploitation or to tasks for which they are overqualiied are not of Hochschild’s concern. More than that, caregiving paid work is analyzed not as work but as an extension of motherhood, an extraction of love from one’s own children to the beneit of First World children (Hochschild: 2002, p.26). Hochschild looks to Freud for support in maintaining that “immigrant nannies and au pairs often divert feelings originally directed toward their own children” and to the work of Sau-ling C. Wong to argue that “time and energy (…) are diverted from those who, by kinship or communal ties, are their more rightful recipients” (Wong: 1994, p. 69). While some women might experience their work in this way, it is not the aim of social analysis to assume or endorse a conception of domestic work as an extension of motherhood. As a matter of fact, viewing domestic workers, not as workers, but as family members is consistent with depriving them of their rights. Secondly, only the migration of women, not of men, is studied as a cause of “care drain”. Fathers’ migration doesn’t give rise to any comparative analysis. Rowena’s case is again a good example. As Hochschild reveals, “the father of her children went to Korea in search of work and, over time, he faded from his children’s lives” (Hochschild: 2002, p.16). By contrast, Rowena migrated without having faded from her children’s lives: she phones them, writes to them and sends $450 of her $750 monthly earnings “for her children’s food, clothes, and schooling” (Hochschild: 2003, p.18). Yet, the mother’s migration, rather than the father’s, is analyzed as depriving the children of care. If Hochschild parenthetically remarks that “it is men who have for the most part stepped aside from caring work, and it is with them that the ‘care drain’ truly begins” (Hochschild: 2003, p.29), she actually refers to the men in rich countries whose involvement in care could reduce the incentives for women’s labor migration. However, there is no reason not to interview migrant men, whether in domestic work or not, to study the way they feel and cope with the family roles and with the emotional labor at their workplace. Some scholars work this way (Näre: 2010), while others warned that “an exclusive focus on women has promoted an unfortunate attachment to sex-role theory” (Hondagneu-Sotelo: 1999, p. 566) and “reiies stereotypical gendered conceptions of domesticity and afect” (Manalansan: 2006, p.238). Thirdly, women’s failure to fulill traditional family roles is judged not only as regrettable but as an injustice which is global in scope. Commenting on one migrant woman, “Vicky Diaz, a college-educated schoolteacher”, Hochschild maintains that “she has taken part in a global heart transplant” (Hochschild: 2002, p.22). The “heart transplant” doesn’t refer to the pupils’ loss of a good teacher, or to Speranta Dumitru 101 an empirically documented loss of care afecting her biological children. Although the global injustice at stake is initially described as a “care drain”, the “new gold” drained from the poor countries, is not exactly “care”, but “love”. This is not surprising, since Hochschild’s other sociological writings focus on the concept of “emotional labor”; and she thus casts the relevant global inequality in terms of “emotional resources” and compares “the emotional deprivation of these [Third World] children with the surfeit of afection their First World counterparts enjoy” (Hochschild: 2002, p.22). Unfortunately, the asserted inequality in emotional resources is not based on any comparative study of how the emotional environment changes for the children living in nuclear families in the North and the children who live in extended families in certain regions of the South. While Hochschild never claims that migrant women would love their own children less, she maintains that love is extracted from those children. Now, what the children are certainly deprived of is not their mothers’ love, but their bodily physical presence. And inferring a loss of care from the mothers’ bodily physical absence alone is a way of downgrading care and reducing it to the women’s bodily closeness (Dumitru: 2014). By contrast, there is a programmatic neglect of the “care gain” in terms of “children’s food, clothes and schooling” (Hochschild: 2002, p.18), and this is consistent with Hochschild’s belief that a mother’s “love” is non-fungible. This suggests that however good the mothers are as breadwinners, their failure to fulill the traditional role of housekeeper is viewed as a “global injustice”. To sum up, while some scholars warned against the trap of reifying stereotypical gendered conceptions (see e.g. (Hondagneu-Sotelo: 1999), (Manalansan: 2006), (Kilkey: 2010), (Dumitru: 2011), (Akpinar-Elci, Elci, Civaner: 2014), (Dumitru: 2014), many others adopt the construal of “care drain” (Bettio: 2006), (Lutz, Palenga-Mollenbeck: 2012), (Gündüz: 2013). Why despite those warnings, is the metaphor of “care drain” still so popular? 4. Emotions and World’s Injustices This section considers the afect heuristic as a possible cause of overestimating “care drain”. The afect heuristic, that is a way to make judgements based on one’s emotions, is recognized as a source of bias (Slovic, Finucane, Peters: 2007). Indeed, imagining the children one’s nanny has left in the Third World may provoke deep unease and feelings of global injustice that trump considerations about their actual numerical importance. The story of “care drain”, although built on small amounts of information, looks plausible. The available information seems to tell an obvious story linking three kinds of actors: (1) women in the First World (2) hiring migrant nannies (3) whose children are sometimes left in the Third World. Psychologists call the ability to create a coherent story from limited information WYSIATI: what you see is what there is (Kahneman: 2011). As the story unfolds before our eyes, there is no pressure to check the likelihood for a migrant woman to be a domestic worker who satisies all the “care chain” criteria, and generalization is easy. After all, the combination of gender, class, and colonial domination in the story makes the injustice of “care drain” global in scope. The appeal to emotion is irresistible in the literature about “care drain”. 102 The Romanian Journal of Society and Politics Emotions are at the center of Hochschild’s account: women are viewed as “emotional resources” that the “new imperialism” extracts from the Third World; migrant caregivers perform “emotional labor”; they and their children pay the “emotional costs” of globalization; migration subtracts something from the “emotional commons”; “emotional well-being” of children left behind is impaired and inequality between children from the First and Third World increases (Hochschild: 2013). Some philosophers ind this approach convincing and argue that emotional needs are basic needs and their frustration might seriously harm the children’s development, as well as their mothers’ wellbeing (Gheaus: 2013a; Gheaus: 2013b). Being able to experience emotions, “to love, to grieve, to experience longing, gratitude, and justiied anger” are sometimes described as central human capabilities (Nussbaum: 2003, p. 41). As mental health is part of the human right to health, emotional wellbeing or, at least, “not having one’s emotional development blighted by fear and anxiety” is an important right, too (Nussbaum: 2003, p. 41) While the appeal to emotion appears irresistible, feeling touched about what you see in the First World leads you to ignore what there is at the global level. At a global level, the injustices afecting migrant domestic workers and children from the Third World, considered independently, are much more severe than those afecting migrant domestic workers and the children considered together as involved in a “global chain”. The temptation to tell a coherent story about “global care chains” is a way to hide all the “unchained” injustices, those which do not link women in the First World, migrant nannies, and children in the Third World. When you think about it, the coherent story of “global care chains” is more Eurocentric than global: the story is ultimately about the First World which extracts love and care from the Third World; and the end of a care chain is assumed to be in the First World. That means that the story is of low level of generality and ignores everything unrelated to the First World. However, women domestic workers in some non-First World countries and children unrelated to nannies in the First World sufer far more severe injustices, both in emotional and in economic terms. Consider the emotional injustice: the emphasis on the First World leads to ignoring higher emotional costs in the rest of the world. On Hochschild’s terms, emotional costs are incurred when mothers are separated from their children. While Hochschild never mentions the existence of immigration policies and their impact on family life, a brief comparative analysis can be useful. In most countries in the First World, family reuniication is not always respected but it is legally recognized as a right. On the contrary, in some Asian countries, such as Singapore and Hong Kong, as well as in the Gulf countries, migrant domestic workers are mostly hired as live-in and are legally forbidden to bring dependents. It follows that in such cases the separation of domestic workers from their family is institutionalized and the likelihood of “emotional costs” is greater. The intensity and duration of emotional loss are also higher, since migrant domestic workers are not allowed to travel until the end of their work contract. In the Gulf countries, the sponsorship system (kafala) implies that domestic workers can enter, work and leave the country only with the assistance or explicit permission of their sponsor (kafeel) (Chammartin, Esim, Smith: 2004). The legal system allows employers to withhold passports in Speranta Dumitru 103 order to prevent their domestic worker from leaving the country and breaking the contract. Besides, part of the employers’ abuses consists in keeping the passport beyond the duration of the two or three years work contract. In all these countries, the emotional costs sufered when there are children left behind are higher since the duration of the separation is longer and mothers are legally deprived of the means to visit their children. While in the First World employers’ abuses or the illegal status of migrants might de facto prevent domestic workers from traveling, abusive employers are not legally protected. On the contrary, in the United States for instance, the application for domestic employee visas requires a statement by the employer that he or she will not withhold the passport, employment contract, or other personal property of the employee. Therefore, being hired as a domestic worker in the First World is likely to produce less emotional costs to the possible children left behind, then being hired in other, non-First World, countries. Consider now injustices more severe than the emotional costs of separation. The injustice sufered by migrant domestic workers can be deined as more severe when they are not only separated from their children, but sufer some other harm in addition. For instance, domestic workers’ freedom of movement may be restricted at the international, as well as at the local level. Thus, in Lebanon, a survey showed that 71% of employers believed that they have the right to restrict a domestic worker’s movement outside the house and 40% believed that they have the right to lock the door on the domestic worker (Jureidini: 2002; Abdulrahim: 2010). Forced coninement, excessive work demands, employer abuse, and inancial pressures are factors explaining why, for instance, in a single year, of 95 cases of deaths of migrant domestic workers reported in Lebanon, 40 were suicides (Human Rights Watch: 2008). Besides, in the Arab League countries, domestic workers are not covered by, but are explicitly excluded from, the labor law (Chammartin, Esim, Smith: 2004). The reason sometimes invoked is that “house workers are to be treated as part of the family”, as a spokesman from Bahrain’s Ministry of Labor explained (Chammartin Esim, Smith: 2004, p.17). The consequence is that many domestic helpers work more than 100 hours a week in the GCC countries, the average days of are at most two days per month, one in ive domestic workers reported unpaid wages, and half reported physical, psychological, or sexual abuse by their employers. In the First World, the work conditions are diferent. While no rich country signed the 1990 UN Convention on the Protection of Rights of all Migrant Workers and only seven countries ratiied the 2011 Convention concerning Decent Work for Domestic Workers, nowhere in the First World is domestic work wholly excluded from the labor law. Labor law respects freedom of movement during the time of, and guarantees at least one day of per week, a considerably reduced number of working hours, and a legal basis for complaints against abuses. The Eurocentric story of “global care chains” conveys an inaccurate view of migration. Scholars in the neo-Marxist tradition expect to ind evidence that the First World “extracts” love from the Third World as it once extracted raw materials from colonized regions. They tend to neglect that immigration policies in the First World are in no way seeking to “extract” people from developing countries, on the contrary, they are very restrictive, especially with lower-skilled women whose 104 The Romanian Journal of Society and Politics emigration rates are dramatically decreasing (Dumitru, Marfouk: 2015). As far as Filipino migrant workers are concerned, their main destinations remain the Asian and the Middle East countries and not the First World (POEA: 2015). Instead, data from the US census shows that Filipina migrants in the United States in 2010 are most often educated (52% have at least a Bachelor’s degree and 90% are at least High School graduates); that they most often work in “management, business, science, and arts occupations” (42%); and that their mean annual earnings stand at $52,020 (US Census Bureau: 2011-2013). ‬‬‬‬ Finally, consider the second and more severe injustice that is neglected by the “care drain” story: child poverty. Global inequality of emotional resources may be important, but children in the Third World are still ighting premature death, undernourishment, poverty, illiteracy, and dangerous working conditions. What is more emotionally unbearable is that 16,000 children under ive died every single day in 2015; that of these almost 6 million annual deaths, half were caused by nutrition related causes; that in South-Eastern Asia, about one in seven children under the age of ive is underweight. What is also emotionally disturbing is that about 58 million children and 63 million adolescents are not in school. While most of them are in Africa and in Southern Asia, there were 5,9 million children and 8,4 million adolescents out of school in Eastern Asia in 2012 (UNESCO, UNICEF: 2015). What is emotionally worrisome is that about 215 million children are trapped in child labor and that half of them are in hazardous child labor, in mines, agriculture or industry. In the Philippines, there were 5.5 million working children in 2011, of which 2 million were in hazardous child labor (ILO: 2011). Hazardous labor implies exposure to chemical (e.g. silica dust, oil, gasoline, mercury etc.), physical (e.g. noise, temperature, pressure etc.) and biological (viral, bacterial, infected water) risks. Given these igures, defending children’s interests in not having migrant mothers appears misplaced. Hochschild invokes the UN Convention on the Rights of the Child which maintains that children “should grow up in a family environment, in an atmosphere of happiness, love and understanding”[2] and should “not to be separated from his or her parents against their will”[3]. Hochschild seems to interpret these rights as children’s rights to grow up with their mothers who should provide them with “happiness, love and understanding”. The afect heuristic is appealing. However, for children dying before their ifth birthday or living malnourished, unschooled or working in hazardous labor, the right to live in an “atmosphere of happiness, love and understanding” does not seem the most urgent priority. Fighting poverty is. An increasing body of literature has shown that migration is among the most efective means to ight poverty (for a review of literature, see Dumitru (2013)). Migration decreases global inequalities (Hatton, Williamson: 1998) and increases global GDP (Clemens, 2011). Since 1996, the inlow of migrant remittances to developing countries has overcome Oicial Development Aid (ODA); in 2016, the 2 Cf. 6th Recital of the 1989 UN Convention on the Rights of the Child (my emphasis). 3 Cf. art. 9 of the 1989 UN Convention on the Rights of the Child (my emphasis), which protects the children against a forcible separation, non-consented by the parents. Speranta Dumitru 105 volume of remittances is estimated at 442 billion US$, almost four times higher than ODA. Remittances signiicantly decrease poverty, understood as the number of persons living with less than $1.00 per day (Adams, Page: 2005). The Philippines is the third country in the world, after India and China, by the volume of remittances received from migrants. The most signiicant source of remittances for the country is the United States: for instance in 2014, more than one third, that is, 10 out of 28 billion US$ were sent to the Philippines by migrants living in the United States alone. Remittances improve children’s health and school attendance, especially in families with poor educational attainment (Yang, Choi: 2007; Acosta: 2011; Alcaraz, Chiquiar, Salcedo: 2012). Still, Hochschild and her colleagues insist on the non-fungible nature of love as they fear that the “children, as well as their care givers, can come to experience money as a substitute for shared experiences and love” (Isaksen, Devi, Hochschild: 2008). To sum up, using afect heuristic to support children’s right to not having migrant mothers is a way to bias judgements about the world’s injustice. While both domestic migrants and children from the poor countries may experience emotional costs, both have much more urgent needs. Construing a coherent story about the “new emotional imperialism” thus diverts our attention from their genuine needs. 5. Conclusion There is an important gap between real and perceived proportion of domestic workers among migrant women. Migrant women employed as domestic workers in the European and Northern American countries represent 2% of the migrant women worldwide. In total, migrant women employed as domestic workers all over the world represent 8% of migrant women. However, researchers often commit hasty generalization and associate the “feminization of migration” with domestic work. This paper is the irst attempt to understand the source of bias. Here, the analysis is limited to the theory of “care drain” elaborated by Arlie Hochschild and adopted by numerous scholars. In this case, the biased perception is more prominent as “care drain” refers to a subset of domestic workers: those who are both mothers and who left their children in Third World countries. As this subset is logically smaller than the group of domestic workers, it is even less likely that it characterize the “feminization of migration”. How can the bias go unnoticed? The paper explored two possible mechanisms of bias – reasoning from small samples and using afect heuristic – and found that none of them fully explain the bias. In the samples, more than half of Filipina migrant were college educated and almost half of them had left no children behind. A better explanation has been found in a stereotyped thinking which systematically associates women with care. The literature on “care drain” unnecessarily ampliies this association, by representing migrant women as both care-workers and mothers. Such a stereotyped thinking can go unnoticed for anyone reasoning from afect heuristic. Indeed, imagining children left behind by migrant mothers provokes strong feelings which trump any other consideration. However, emotions are more rightfully experienced when informed by data. 106 The Romanian Journal of Society and Politics Acknowledgements I am grateful to Alex Sager, David Hirst, two anonymous referees and the editors, for their helpful comments and editorial work. I would also like to thank Ruth Achenbach, Yonson Ahn, Brunella Casalini, Solange Chavel, Ryoa Chung, Anca Gheaus, Lisa Eckenwiler, Milena Jaksic, Lena Näre, Patrick Pharo, Camille Schmoll, and Tiina Vaittinen for stimulating discussions References Abdulrahim, S. (2010), Servant, daughter, or employee? A pilot study on the attitudes of Lebanese employers towards migrant domestic workers, KAFA, Beirut. Acosta, P. (2011), “School attendance, child labour, and remittances from international migration in El Salvador” Journal of Development Studies, 47(6), pp. 913-936. Adams, R. H., Page, J. (2005), “Do international migration and remittances reduce poverty in developing countries?” World development, 33(10), pp. 1645-1669. Akpinar-Elci, M., Elci, O. C., Civaner, M. (2015), “Care Drain”, in H. ten Have, (ed.), Encyclopedia of Global Bioethics,Springer International Publishing, Dordrecht, pp. 1-7. Alcaraz, C., Chiquiar, D., Salcedo, A. (2012), “Remittances, schooling, and child labor in Mexico” Journal of Development Economics, 97(1), pp. 156-165. Bettio, F., Simonazzi, A., Villa, P. (2006), “Change in care regimes and female migration: the ‘care drain’ in the Mediterranean” Journal of European Social Policy, 16(3), pp. 271-285. Brücker, H., Capuano, S., Marfouk, A. (2013), “Education, gender and international migration: insights from a panel-dataset 1980-2010”. Chammartin, G. M. F., Esim, S., Smith, M. (2004). Gender and Migration in Arab States: The Case of Domestic Workers. International Labour Organization, Regional Ofice for Arab States, Geneva. Clemens, M. A. (2011), “Economics and emigration: Trillion-dollar bills on the sidewalk?” The Journal of Economic Perspectives, 25(3), pp. 83-106. Dumitru, S. (2011), “Care drain: le piège sexiste du nationalisme”, Tiziana Caponio, F. Giordano, and L. Ricaldone, (eds.), World Wide Women. Globalizzazione, Speranta Dumitru 107 generi, linguaggio. University of Turin: CIRSDe, pp. 51-60. Dumitru, S. (2013), “Des visas, pas de l’aide! De la migration comme substitut à l’aide au développement” Éthique publique. Revue internationale d’éthique sociétale et gouvernementale, 15(2), pp. 77-98. Dumitru, S. (2014), “From “brain drain” to “care drain” : Women’s labor migration and methodological sexism” Women’s Studies International Forum, 47, Part B, pp. 203-212. Dumitru, S., Marfouk, A. (2015), “฀ Existe-t-il une féminisation de la migration internationale?฀ Féminisation de la migration qualiiée et invisibilité des diplômes” Hommes & Migrations, 1311(3), pp. 31-41.฀฀฀฀฀฀฀฀ Fröbel, F., Heinrichs, J., Kreye, O. (1980), The new international division of labour : structural unemployment in industrialised countries and industrialisation in developing countries, Cambridge University Press, London. Gheaus, A. (2013a), “Care Drain as an Issue of Global Gender Justice” Ethical perspectives, 20(1), pp. 61-80. Gheaus, A. (2013b), “Care drain: who should provide for the children left behind?” Critical Review of International Social and Political Philosophy, 16(1), pp. 1-23. Gündüz, Z. Y. (2013), “The Feminization of Migration: Care and the New Emotional Imperialism” Monthly Review, 65(7), pp. 32-43. Hatton, T. J., Williamson, J. G. (1998), The age of mass migration: Causes and economic impact, Oxford University Press, Oxford. Hochschild, A. R. (2000), “The nanny chain” The American Prospect, 11(4), pp. 32-36. Hochschild, A. R. (2002), “Love and gold”, Global women: Nannies, maids, and sex workers in the new economy,Owl Books, New York, pp. 15-30. Hochschild, A. R. (2013), So How’s the Family? : And Other Essays, University of California Press, Berkeley, CA, USA. Hondagneu-Sotelo, P. (1999), “Introduction: Gender and Contemporary U.S. Immigration” American Behavioral Scientist, 42(4), pp. 565-576. 108 The Romanian Journal of Society and Politics Human Rights Watch. (2008). Deaths of Migrant Domestic Workers in Lebanon. Human Rights Watch. ILO. (2011). Survey on children. ILO, Geneva. ILO. (2015). Global estimates on migrant workers. ILO, Geneva. Isaksen, L., Devi, U., Hochschild, A. (2008), “Global care crisis: mother and child’s-eye view” Sociologia(56). Jureidini, R. (2002), Women migrant domestic workers in Lebanon, International Labour Organisation, Kahneman, D. (2011), Thinking, fast and slow, Macmillan, New York. Kahneman, D., Tversky, A. (1973), “On the psychology of prediction” Psychological review, 80(4), pp. 237. Kilkey, M. (2010), “Men and Domestic Labor: A Missing Link in the Global Care Chain” Men and Masculinities, 13(1), pp. 126-149. Kroska, A. (2007), “Gender ideology and gender role ideology”Blackwell encyclopedia of sociology [electronic resources]. Blackwell Publishing: Malden. Lind, A. (2007), “Sexism” Blackwell encyclopedia of sociology. Blackwell Publishing. Lutz, H., Palenga-Mollenbeck, E. (2012), “Care Workers, Care Drain, and Care Chains: Relections on Care, Migration, and Citizenship” Social Politics: International Studies in Gender, State & Society, 19(1), pp. 15-37. Manalansan, M. F. (2006), “Queer Intersections: Sexuality and Gender in Migration Studies” International Migration Review, 40(1), pp. 224-249. Mies, M. (1986), Patriarchy and accumulation on a world scale : women in the international division of labour, Zed, London. Momsen, J. H. (1999), “Maids on the move: victim or victor”, in J. H. Momsen, (ed.), Gender, Migration and Domestic Work,Routledge, London & New York. Näre, L. (2010), “Sri Lankan men working as cleaners and carers: Negotiating masculinity in Naples” Men and Masculinities, 13(1), pp. 65-86. Nussbaum, M. (2003), “Capabilities as fundamental entitlements: Sen and social justice” Feminist economics, 9(2-3), pp. 33-59. Speranta Dumitru 109 Oelz, M. (2014), “The ILO’s Domestic Workers Convention and Recommendation: A window of opportunity for social justice” International Labour Review, 153(1), pp. 143-172. Parrenas, R. S. (2000), “Migrant Filipina Domestic Workers and the International Division of Reproductive Labor” Gender and Society, 14(4), pp. 560-580. Parreñas, R. S. (2001), Servants of Globalization: Women, Migration, and Domestic Work, Stanford University Press, Stanford, CA. Parreñas, R. S. (2005), Children of global migration: Transnational families and gendered woes, Stanford University Press, Stanford. POEA. (2015), “Statistics OFW”. Sassen, S. (1988), The Mobility of Labor and Capital: A Study in International Investment and Labor Flow, Cambridge University Press, Cambridge Slovic, P., Finucane, M. L., Peters, E., MacGregor, D. G. (2007), “The affect heuristic” European journal of operational research, 177(3), pp. 1333-1352. Standing, G. (1989), “Global Feminization through Flexible Labor” World Development, 17(7), pp. 1077-1095. Tversky, A., Kahneman, D. (1983), “Extensional versus intuitive reasoning: The conjunction fallacy in probability judgment” Psychological Review, 90(4), pp. 293315. UNESCO, UNICEF. (2015), “Fixing the Broken Promise of Education for All: Findings from the Global Initiative on Out-of-School Children”. UNESCO Institute for Statistics, Montreal. US Census Bureau. (2011-2013), “3-Year American Community Survey”. Wong, S.-L. (1994), “Diverted mothering: Representations of caregivers of color in the age of multiculturalism”, in E. Nakano Glenn, G. Chang, and L. Forcey, (eds.), Mothering: Ideology, experience, and agency,Routledge, London, pp. 67-91. Yang, D., Choi, H. (2007), “Are remittances insurance? Evidence from rainfall shocks in the Philippines” The World Bank Economic Review, 21(2), pp. 219-248. Yoshimura, M. (2007), “Feminization of Labor Migration”, in G. Ritzer, (ed.), Blackwell Encyclopedia of Sociology,Blackwell Publishing. 

Speranța DUMITRU 7 Speranța Dumitru Speranța Dumitru “CARE DRAIN”. EXPLAINING BIAS IN THEORIZING WOMEN’S MIGRATION Faculty of Law, University of Paris Descartes and CERLIS (CNRS), University of Sorbonne Paris Cité speranta.dumitru@parisdescartes.fr KEYWORDS ABSTRACT Migrant women are often stereotyped. Some scholars associate the feminization of migration with domestic work and criticize the “care drain” as a new form of imperialism that the First World imposes on the Third World. However, migrant women employed as domestic workers in Northern America and Europe represent only 2% of migrant women worldwide and cannot be seen as characterizing the “feminization of migration”. Why are migrant domestic workers overestimated? This paper explores two possible sources of bias. The irst is sampling: conclusions about “care drain” are often generalized from small samples of domestic workers. The second stems from the affect heuristic: imagining children left behind by migrant mothers provokes strong feelings of injustice which trump other considerations. The paper argues that neither source of bias is unavoidable and inds evidence of gender stereotypes in the “care drain” construal • care • • • • • • migrant women sexism feminization of migration domestic work global inequality globalization 1. Introduction This article explores the stereotyped construal of migrant women as a “care drain”. Coined by the sociologist Arlie Hochschild (2002), the phrase “care drain” associates feminization of migration with care workers and describes it as a cost for the Third World children[1]. Hochschild theorized women’s migration in the following terms. In the First World, women’s integration into the labor market releases them from domestic activities and leads to an increase of the labor demand in the care sector. The demand for care workers in the First World thus triggers women’s labor migration, especially from the Third World. According to Hochschild (2000), migrant women form “global care chains”: they work as caregivers for richer women in the First World, while sending money to poorer women in the Third World to care for their own children left behind. At the end of the “global care chain”, there must be 1 Throughout this paper, I use “First/Third World” to follow, without endorsing, Hochschild’s initial terminology. 8 The Romanian Journal of Society and Politics a “care drain”, a loss of care for the Third World children. The phrase “care drain” is coined as a female analogy of “brain drain”: “just as poor countries suffer a brain drain as trained personnel move from the South to the North, so too they suffer a care drain” (Hochschild: 2013, p. 147). This article investigates the source of bias in care drain theory. As a matter of fact, the 2.5 million migrant women employed as domestic workers in Northern America and Europe represent 2% of migrant women worldwide and cannot characterize the “feminization of migration”. Furthermore, the subset of domestic workers who are both mothers and who left their children in the Third World countries is even smaller. Therefore, the view characterizing First World’s “importation of care and love from poor countries” as “an important trend” is deeply biased (Hochschild: 2002, p. 17). How can the biased nature of “care drain” construal go unnoticed? The paper explores two possible responses to this question. One source of bias could come from sampling: research endorsing the “care drain” construal is generalized from small samples of migrant women who are domestic workers in developed countries and have been interviewed about their children left in developing countries. A second source of bias is emotion: imagining children left behind by their mothers provokes deep empathy and feelings of injustice that trump any other consideration and lead to overestimating the scope of “care drain”. This paper concludes that neither source of bias was unavoidable in the literature on “care drain”. The remaining article is divided into three parts. Part 2 presents the available data on migrant women and on domestic workers in order to assess the magnitude of bias involved in the construal of migrant women as “care drain”. Part 3 explores the sample of Filipina domestic workers interviewed by Hochschild (2000) and Parreñas (2001) and shows that the stereotyped construal was avoidable: more than half of them were college educated and almost half of them had left no children behind. Part 4 explores the emotional source of bias and shows that the “care drain” account fails to grasp the more severe global injustices which affect migrant domestic workers and poor children when they are considered separately. 2. Overestimating migrant domestic workers This section shows that the view which associates “feminization of migration” with domestic work is not supported by facts. Hochschild believes that “what is unprecedented is the scope and speed of women’s migration to [care] jobs” and that “many factors contribute to the growing feminization of migration…” (Hochschild: 2002, p.17). This view is widespread and even features in dictionaries and encyclopedias. For instance, it is asserted that “feminization of migration is characterized by (…) a concentration on female-speciic work such as domestic helpers, nurses, entertainers” (Yoshimura: 2007, p. 1515) and it is claimed that “demand for domestic work is recognized as a key factor behind the feminization of migration” (Oelz: 2014, p. 145). The view associating the “feminization of migration” with low-skilled jobs originates in the neo-Marxist theory of the “new international division of labor”. In the 1980s, this theory predicted that globalization would result in a new division of labor: outsourcing would industrialize Third World countries at the expense of Western Speranța DUMITRU 9 countries which in turn deindustrialize and specialize in the service sector (Fröbel, Heinrichs, Kreye: 1980). The division is “new” by contrast to the “old” division of labor during the colonial times when European countries controlled the industrial transformation of raw materials extracted from the colonized regions. Another prediction of the neo-Marxist theory was the “feminization of migration”: the global expansion of capital “uproots” women irst from rural to industrial areas in the Third World countries and then from Third World to First World countries (Standing: 1989), (Sassen: 1988), (Mies: 1986). As the “feminization of migration” is understood as contributing to the new international division of labor, neo-Marxist scholarship tries to conirm the theory by focusing on migrant women from the Third World countries employed in low-skilled jobs, especially in the service sector. Migrant domestic workers thus become a textbook case for conirming the gendered international division of labor (Parrenas: 2000) and “care drain” is described as a “new imperialism” imposed by the First World which now “extracts love” and “emotional resources” from the Third World (Hochschild: 2002; Gündüz: 2013). Nevertheless, neither the “feminization of migration”, nor the association of women’s migration with domestic workers is supported by facts. Firstly, no evidence supports the phenomenon of “feminization of migration” as distinct from the “masculinization of migration” (Dumitru, Marfouk: 2015). The UN data reveals that over the last sixty years the proportion of women among international migrants has only slightly increased, from 47% to 48 % (see Table 1). While in absolute numbers, migrant women are nowadays three times more numerous than in the 1960s, the same is true of migrant men who still remain more numerous. However, there is no theorizing of a “masculinization of migration” Table 1: Evolution of the international migrants stock by sex (in thousands) 1960-2015 1960 1970 1980 1990 2000 2010 2015 77 115 84 460 101 983 152 563 172 703 221 714 243 700 Total Men 40 869 44 256 53 302 77 747 87 885 114 614 126 115 Women 36 246 40 204 48 681 74 816 84 818 107 100 117 585 % Women 47% 47,6% 47,7% 49% 49,1% 48,3% 48,2% Source: UN Population Division (2008 and 2015 revisions) and author’s calculation Secondly, asserting that the feminization of migration is linked to domestic work is a hasty generalization. According to the International Labor Organization, the total number of migrant women employed as domestic workers worldwide was 8.46 million in 2013 (ILO: 2015). In 2013, there were 111 million migrant women in total which means that the proportion of migrant women employed as domestic workers was 8% (see Figure 1). If the “First World” is equated with Northern America and Europe, where 2.45 million migrant women earned a living from domestic work in 2013, then domestic workers in those regions accounted for 2% of migrant women worldwide. It is noteworthy that “domestic workers” as deined by the ILO is a larger 10 The Romanian Journal of Society and Politics category than “nannies”. In any case, it is a mistake to generalize the profession of 8% to a characterization of the population of migrant women as a whole. Figure 1: Proportion of domestic workers among international migrant women Source: ILO (2015), UN Population Division (2013 revision) and author’s calculation Finally, to claim that “global care chains” characterize migrant women is to commit another logical error: the conjunction fallacy (Tversky, Kahneman: 1983). The conjunction fallacy is committed when the likelihood of a conjunction is assumed to be higher than the likelihood of its constituents taken separately. Hochschild deined “global care chains” as migrant women who work as caregivers in the First World and are mothers and have children left in the Third World countries. Therefore, migrant domestic workers involved in “care chains” are a subset of migrant domestic workers; and while migrant domestic workers represent 2% of migrant women, the subset is even less numerous and cannot characterize the “feminization of migration”. All in all, the literature about feminization of migration appears to be insensitive to information on base rates to calculate the prior probability of a random migrant woman being a domestic worker. In overestimating the likelihood of a migrant woman being domestic worker, scholars seem to use what has been called the “representativeness heuristic” (Kahneman, Tversky: 1973): that is, they evaluate the probability of an outcome (being domestic worker when one is a migrant woman) by its similarity with stereotyped characteristics of the parent population (women). Judgements made on the basis of representativeness ignore base rates and rely instead on stereotypes. 3. Small samples and sexist stereotypes This section considers sampling as a cause of overestimating the “care drain”. One possible reason why domestic work is taken as characterizing the “feminization of migration” is that scholars generalize from small samples of migrant domestic Speranța DUMITRU 11 workers. However, a quick inspection of the initial sample used by Hochschild (2002) and Parreñas (2001) when theorizing the “care drain” and the “globalization of mothering” shows that hasty generalization was avoidable. First of all, the research assumed that “care drain” – that is, a loss of care for the children left behind in the Third World – can be proven not by investigating in the Third World the variation of care the children received before and after their mother’s departure, but on interviews conducted in the First World with domestic workers (Parreñas: 2001; Hochschild: 2002). Research on the children left behind was conducted by Parreñas (2005) only at a later stage. The initial sample was constituted exclusively of women domestic workers: 46 domestic workers in Rome and 26 in Los Angeles, all from the Philippines. Based on this sample, conclusions were drawn about global trends: Parreñas theorized the “globalization of mothering” and Hochschild analyzed the global injustice of “care drain”. However, the sample size is not the only problem as the demographic characteristics of Parreñas’ sample could have avoided hasty generalization. First, as Parreñas (2001, p.19) indicated: “women with children living in the Philippines constitute a greater portion of my sample in both Rome and Los Angeles: twenty-ive of forty-six in Rome and fourteen of twenty-six in Los Angeles”. That means that almost half of the interviewed women had no children at all living in the Philippines and the sample did not allow for generalizations about “globalization of mothering”. Second, the sample of domestic workers included for more than half college-educated women (Parreñas: 2002, p.262). What draws attention is the skill mismatch – the gap between women’s education and the profession they exercise. Nonetheless, faced with college-educated women, Hochschild chose to coin the term “care drain” as a feminine analogy for “brain drain”. Stereotyped thinking provides a better explanation of the overestimation of care drain. To be sure, both metaphors – “care drain” and “brain drain” – are ways to stereotype people by reducing them to a single characteristic. While educated emigrants employed in jobs for which they are overqualiied are usually termed “brain waste”, Hochschild chose to create a new category for women: the “care drain”. “Care drain” is deined as it follows: “Rowena’s life relects an important and growing trend: the importation of care and love from poor countries to rich ones. For some time now, promising and highly trained professional have been moving from ill-equipped hospitals, impoverished schools, antiquated banks, and other beleaguered workplaces of the Third World to better opportunities and higher pay in the irst world (…). But in addition to this brain drain there is now a parallel but more hidden and wrenching trend, as women who normally care for the young, the old, and the sick in their own poor countries move to care for the young, the old, and the sick in rich countries, whether as maid and nannies or as day-care and nursing-home aides. It’s a care drain” (Hochschild: 2002, p.17, my emphasis). Actually, Rowena, whose “life” is claimed to illustrate a “care drain”, could have illustrated a brain drain or a brain waste. Rowena, a Filipina migrant woman, certainly works as a nanny in the US, but according to the description given by Hochschild, she had “worked three years toward an engineering degree” in the Philippines (Hochschild: 2002, p.16). We are not used to describing the migration of domestic workers as “brain drain”, but having two years of tertiary education commonly classiies a migrant as 12 The Romanian Journal of Society and Politics being among the highly-skilled and thus as characterizing the so-called “brain drain”. As the International Organization for Migration observes, “the most basic deinition of highly skilled migrants tends to be restricted to persons with tertiary education, typically adults who have completed a formal two-year college education or more” (IOM: 2008, p.52). Rowena completed three, not two years of college education. Besides, the probability that other Filipina maids are highly-skilled migrants is indeed sizeable since no less than 64% of migrant women from the Philippines were highlyskilled in 2000 (Brücker, Capuano, Marfouk: 2013). Hochschild is aware that the pool of domestic workers she studies “includes college-educated teachers, businesswomen, secretaries (…) and more than half of the nannies [Parreñas] interviewed had college degrees” (Hochschild: 2000). Still, Hochschild does not qualify any migrant maid as “brain drain”. The kind of stereotypes on which the metaphor of “care drain” is built is sexist. Sexism is usually deined as unequal treatment of men and women based on a traditional ideology about sex roles. Traditional gender ideology especially separates sex roles in the family: men are supposed to fulill their family roles through instrumental, breadwinning activities, while women are supposed to do so through nurturing, homemaking, and parenting activities (Kroska: 2007; Lind: 2007). In migration studies, the representation of women as attached to the household and/or to the family could explain the assumption that they immigrate either less than men or mainly as wives; the “discovery” of women’s labor migration in the 1980s gave rise to the literature on the “feminization of migration”. In previous work, I have deined “methodological sexism” as a research bias which involves three criteria: women are studied only as caregivers, only women are studied as caregivers (men are excluded) and women’s failure to fulill their traditional family roles is judged regrettable (Dumitru: 2014). The three conditions are met by Hochschild’s description of the “care drain”. Firstly, women are over-construed as caregivers. Though the deinition cited above suggests that any woman produces “care drain” since “women normally care for the young, the old and the sick” (Hochschild: 2002, p. 17), Hochschild ampliies the scope of a particular subset of migrant women. In order to support her thesis about “the global care chains” through which the First World “extracts love” from the Third World, she describes the migrant women as conjointly meeting two criteria: having children (in a poor country) and being employed in care work (in a rich country). “Care drain” is supposed to be produced by hiring as caregivers women who would otherwise have cared for their children in the poor country. But there are obviously one too many caregiving functions in this picture. When one aims to study the “care drain”, the variation of care a child is supposed to experience in the absence of her mother does not logically depend on the kind of job that the mother has at destination: once absent, whether she is working in engineering or in the care sector is prima facie irrelevant for what happens at home. Conversely, if one aims to study migrant women employed as caregivers, one should not discriminate according to their family status: whether they have children or not and, if they have, whether they reside with them or not, they are all migrant women employed as domestic workers. Yet, migrant women aren’t studied as workers. Not only is their condition of Speranța DUMITRU 13 caregivers unnecessarily ampliied, but they are studied as “natural” caregivers. Their professional ambitions and dificulties, the fact that women may choose domestic work as the only legal way for them to cross international borders (Momsen: 1999), their work relationships, and the way they respond to eventual exploitation or to tasks for which they are overqualiied are not of Hochschild’s concern. More than that, caregiving paid work is analyzed not as work but as an extension of motherhood, an extraction of love from one’s own children to the beneit of First World children (Hochschild: 2002, p.26). Hochschild looks to Freud for support in maintaining that “immigrant nannies and au pairs often divert feelings originally directed toward their own children” and to the work of Sau-ling C. Wong to argue that “time and energy (…) are diverted from those who, by kinship or communal ties, are their more rightful recipients” (Wong: 1994, p. 69). While some women might experience their work in this way, it is not the aim of social analysis to assume or endorse a conception of domestic work as an extension of motherhood. As a matter of fact, viewing domestic workers, not as workers, but as family members is consistent with depriving them of their rights. Secondly, only the migration of women, not of men, is studied as a cause of “care drain”. Fathers’ migration doesn’t give rise to any comparative analysis. Rowena’s case is again a good example. As Hochschild reveals, “the father of her children went to Korea in search of work and, over time, he faded from his children’s lives” (Hochschild: 2002, p.16). By contrast, Rowena migrated without having faded from her children’s lives: she phones them, writes to them and sends $450 of her $750 monthly earnings “for her children’s food, clothes, and schooling” (Hochschild: 2002, p.18). Yet, the mother’s migration, rather than the father’s, is analyzed as depriving the children of care. If Hochschild parenthetically remarks that “it is men who have for the most part stepped aside from caring work, and it is with them that the ‘care drain’ truly begins” (Hochschild: 2002, p.29), she actually refers to the men in rich countries whose involvement in care could reduce the incentives for women’s labor migration. However, there is no reason not to interview migrant men, whether in domestic work or not, to study the way they feel and cope with the family roles and with the emotional labor at their workplace. Some scholars work this way (Näre: 2010), while others warned that “an exclusive focus on women has promoted an unfortunate attachment to sexrole theory” (Hondagneu-Sotelo: 1999, p. 566) and “reiies stereotypical gendered conceptions of domesticity and affect” (Manalansan: 2006, p.238). Thirdly, women’s failure to fulill traditional family roles is judged not only as regrettable but as an injustice which is global in scope. Commenting on one migrant woman, “Vicky Diaz, a college-educated schoolteacher”, Hochschild maintains that “she has taken part in a global heart transplant” (Hochschild: 2002, p.22). The “heart transplant” doesn’t refer to the pupils’ loss of a good teacher, or to an empirically documented loss of care affecting her biological children. Although the global injustice at stake is initially described as a “care drain”, the “new gold” drained from the poor countries, is not exactly “care”, but “love”. This is not surprising, since Hochschild’s other sociological writings focus on the concept of “emotional labor”; and she thus casts the relevant global inequality in terms of “emotional resources” and compares “the emotional deprivation of these [Third World] children with the surfeit of affection their First World counterparts enjoy” (Hochschild: 2002, p.22). Unfortunately, the 14 The Romanian Journal of Society and Politics asserted inequality in emotional resources is not based on any comparative study of how the emotional environment changes for the children living in nuclear families in the North and the children who live in extended families in certain regions of the South. While Hochschild never claims that migrant women would love their own children less, she maintains that love is extracted from those children. Now, what the children are certainly deprived of is not their mothers’ love, but their bodily physical presence. And inferring a loss of care from the mothers’ bodily physical absence alone is a way of downgrading care and reducing it to the women’s bodily closeness (Dumitru: 2014). By contrast, there is a programmatic neglect of the “care gain” in terms of “children’s food, clothes and schooling” (Hochschild: 2002, p.18), and this is consistent with Hochschild’s belief that a mother’s “love” is non-fungible. This suggests that however good the mothers are as breadwinners, their failure to fulill the traditional role of housekeeper is viewed as a “global injustice”. To sum up, while some scholars warned against the trap of reifying stereotypical gendered conceptions (see e.g. (Hondagneu-Sotelo: 1999), (Manalansan: 2006), (Kilkey: 2010), (Dumitru: 2011), (Akpinar-Elci, Elci, Civaner: 2014), (Dumitru: 2014), many others adopt the construal of “care drain” (Bettio: 2006), (Lutz, Palenga-Mollenbeck: 2012), (Gündüz: 2013). Why despite those warnings, is the metaphor of “care drain” still so popular? 4. Emotions and World’s Injustices This section considers the affect heuristic as a possible cause of overestimating “care drain”. The affect heuristic, that is, the way to make judgements based on one’s emotions, is recognized as a source of bias (Slovic, Finucane, Peters: 2007). Indeed, imagining the children one’s nanny has left in the Third World may provoke deep unease and feelings of global injustice that trump considerations about their actual numerical importance. The story of “care drain”, although built on small amounts of information, looks plausible. The available information seems to tell an obvious story linking three kinds of actors: (1) women in the First World (2) hiring migrant nannies (3) whose children are sometimes left in the Third World. Psychologists call the ability to create a coherent story from limited information WYSIATI: what you see is what there is (Kahneman: 2011). As the story unfolds before our eyes, there is no pressure to check the likelihood for a migrant woman to be a domestic worker who satisies all the “care chain” criteria, and generalization is easy. After all, the combination of gender, class, and colonial domination in the story makes the injustice of “care drain” global in scope. The appeal to emotion is irresistible in the literature about “care drain”. Emotions are at the center of Hochschild’s account: women are viewed as “emotional resources” that the “new imperialism” extracts from the Third World; migrant caregivers perform “emotional labor”; they and their children pay the “emotional costs” of globalization; migration subtracts something from the “emotional commons”; “emotional wellbeing” of children left behind is impaired and inequality between children from the First and Third World increases (Hochschild: 2013). Speranța DUMITRU 15 Some philosophers ind this approach convincing and argue that emotional needs are basic needs and their frustration might seriously harm the children’s development, as well as their mothers’ wellbeing (Gheaus: 2013a; Gheaus: 2013b). Being able to experience emotions, “to love, to grieve, to experience longing, gratitude, and justiied anger” are sometimes described as central human capabilities (Nussbaum: 2003, p. 41). As mental health is part of the human right to health, emotional wellbeing or, at least, “not having one’s emotional development blighted by fear and anxiety” is an important right, too (Nussbaum: 2003, p. 41) While the appeal to emotion appears irresistible, feeling touched about what you see in the First World leads you to ignore what there is at the global level. At a global level, the injustices affecting migrant domestic workers and children from the Third World, considered independently, are much more severe than those affecting migrant domestic workers and the children considered together as involved in a “global chain”. The temptation to tell a coherent story about “global care chains” is a way to hide all the “unchained” injustices, those which do not link women in the First World, migrant nannies, and children in the Third World. When you think about it, the coherent story of “global care chains” is more Eurocentric than global: the story is ultimately about the First World which extracts love and care from the Third World; and the end of a care chain is assumed to be in the First World. That means that the story is of low level of generality and ignores everything unrelated to the First World. However, women domestic workers in some non-First World countries and children unrelated to nannies in the First World suffer far more severe injustices, both in emotional and in economic terms. Consider the emotional injustice: the emphasis on the First World leads to ignoring higher emotional costs in the rest of the world. On Hochschild’s terms, emotional costs are incurred when mothers are separated from their children. While Hochschild never mentions the existence of immigration policies and their impact on family life, a brief comparative analysis can be useful. In most countries in the First World, family reuniication is not always respected but it is legally recognized as a right. On the contrary, in some Asian countries, such as Singapore and Hong Kong, as well as in the Gulf countries, migrant domestic workers are mostly hired as live-in and are legally forbidden to bring dependents. It follows that in such cases the separation of domestic workers from their family is institutionalized and the likelihood of “emotional costs” is greater. The intensity and duration of emotional loss are also higher, since migrant domestic workers are not allowed to travel until the end of their work contract. In the Gulf countries, the sponsorship system (kafala) implies that domestic workers can enter, work and leave the country only with the assistance or explicit permission of their sponsor (kafeel) (Chammartin, Esim, Smith: 2004). The legal system allows employers to withhold passports in order to prevent their domestic worker from leaving the country and breaking the contract. Besides, part of the employers’ abuses consists in keeping the passport beyond the duration of the two or three years work contract. In all these countries, the emotional costs suffered when there are children left behind are higher since the duration of the separation is longer and mothers are legally deprived of the means to visit their children. While in the First World employers’ abuses or the illegal status of migrants might de facto prevent domestic workers from traveling, 16 The Romanian Journal of Society and Politics abusive employers are not legally protected. On the contrary, in the United States for instance, the application for domestic employee visas requires a statement by the employer that he or she will not withhold the passport, employment contract, or other personal property of the employee. Therefore, being hired as a domestic worker in the First World is likely to produce less emotional costs to the possible children left behind, then being hired in other, non-First World, countries. Consider now injustices more severe than the emotional costs of separation. The injustice suffered by migrant domestic workers can be deined as more severe when they are not only separated from their children, but suffer some other harm in addition. For instance, domestic workers’ freedom of movement may be restricted at the international, as well as at the local level. Thus, in Lebanon, a survey showed that 71% of employers believed that they have the right to restrict a domestic worker’s movement outside the house and 40% believed that they have the right to lock the door on the domestic worker (Jureidini: 2002; Abdulrahim: 2010). Forced coninement, excessive work demands, employer abuse, and inancial pressures are factors explaining why, for instance, in a single year, of 95 cases of deaths of migrant domestic workers reported in Lebanon, 40 were suicides (Human Rights Watch: 2008). Besides, in the Arab League countries, domestic workers are not covered by, but are explicitly excluded from, the labor law (Chammartin, Esim, Smith: 2004). The reason sometimes invoked is that “house workers are to be treated as part of the family”, as a spokesman from Bahrain’s Ministry of Labor explained (Chammartin Esim, Smith: 2004, p.17). The consequence is that many domestic helpers work more than 100 hours a week in the GCC countries, the average days off are at most two days per month, one in ive domestic workers reported unpaid wages, and half reported physical, psychological, or sexual abuse by their employers. In the First World, the work conditions are different. While no rich country signed the 1990 UN Convention on the Protection of Rights of all Migrant Workers and only seven countries ratiied the 2011 Convention concerning Decent Work for Domestic Workers, nowhere in the First World is domestic work wholly excluded from the labor law. Labor law respects freedom of movement during the time off, and guarantees at least one day off per week, a considerably reduced number of working hours, and a legal basis for complaints against abuses. The Eurocentric story of “global care chains” conveys an inaccurate view of migration. Scholars in the neo-Marxist tradition expect to ind evidence that the First World “extracts” love from the Third World as it once extracted raw materials from colonized regions. They tend to neglect that immigration policies in the First World are in no way seeking to “extract” people from developing countries, on the contrary, they are very restrictive, especially with lower-skilled women whose emigration rates are dramatically decreasing (Dumitru, Marfouk: 2015). As far as Filipino migrant workers are concerned, their main destinations remain the Asian and the Middle East countries and not the First World (POEA: 2015). Instead, data from the US census shows that Filipina migrants in the United States in 2010 are most often educated (52% have at least a Bachelor’s degree and 90% are at least High School graduates); that they most often work in “management, business, science, and arts occupations” (42%); and that their mean annual earnings stand at $52,020 (US Census Bureau: 2011-2013). Finally, consider the second and more severe injustice that is neglected by Speranța DUMITRU 17 the “care drain” story: child poverty. Global inequality of emotional resources may be important, but children in the Third World are still ighting premature death, undernourishment, poverty, illiteracy, and dangerous working conditions. What is more emotionally unbearable is that 16,000 children under ive died every single day in 2015; that of these almost 6 million annual deaths, half were caused by nutrition related causes; that in South-Eastern Asia, about one in seven children under the age of ive is underweight. What is also emotionally disturbing is that about 58 million children and 63 million adolescents are not in school. While most of them are in Africa and in Southern Asia, there were 5,9 million children and 8,4 million adolescents out of school in Eastern Asia in 2012 (UNESCO, UNICEF: 2015). What is emotionally worrisome is that about 215 million children are trapped in child labor and that half of them are in hazardous child labor, in mines, agriculture or industry. In the Philippines, there were 5.5 million working children in 2011, of which 2 million were in hazardous child labor (ILO: 2011). Hazardous labor implies exposure to chemical (e.g. silica dust, oil, gasoline, mercury etc.), physical (e.g. noise, temperature, pressure etc.) and biological (viral, bacterial, infected water) risks. Given these igures, defending children’s interests in not having migrant mothers appears misplaced. Hochschild invokes the UN Convention on the Rights of the Child which maintains that children “should grow up in a family environment, in an atmosphere of happiness, love and understanding”[2] and should “not to be separated from his or her parents against their will”[3]. Hochschild seems to interpret these rights as children’s rights to grow up with their mothers who should provide them with “happiness, love and understanding”. The affect heuristic is appealing. However, for children dying before their ifth birthday or living malnourished, unschooled or working in hazardous labor, the right to live in an “atmosphere of happiness, love and understanding” does not seem the most urgent priority. Fighting poverty is. An increasing body of literature has shown that migration is among the most effective means to ight poverty (for a review of literature, see Dumitru (2013)). Migration decreases global inequalities (Hatton, Williamson: 1998) and increases global GDP (Clemens, 2011). Since 1996, the inlow of migrant remittances to developing countries has overcome Oficial Development Aid (ODA); in 2016, the volume of remittances is estimated at 442 billion US$, almost four times higher than ODA. Remittances signiicantly decrease poverty, understood as the number of persons living with less than $1.00 per day (Adams, Page: 2005). The Philippines is the third country in the world, after India and China, by the volume of remittances received from migrants. The most signiicant source of remittances for the country is the United States: for instance in 2014, more than one third, that is, 10 out of 28 billion US$ were sent to the Philippines by migrants living in the United States alone. Remittances improve children’s health and school attendance, especially in families with poor educational attainment (Yang, Choi: 2007; Acosta: 2011; Alcaraz, Chiquiar, Salcedo: 2012). Still, Hochschild and her colleagues insist on the non-fungible nature of love as they fear that the “children, as well as their care 2 Cf. 6th Recital of the 1989 UN Convention on the Rights of the Child (my emphasis). 3 Cf. art. 9 of the 1989 UN Convention on the Rights of the Child (my emphasis), which protects the children against a forcible separation, non-consented by the parents. 18 The Romanian Journal of Society and Politics givers, can come to experience money as a substitute for shared experiences and love” (Isaksen, Devi, Hochschild: 2008). To sum up, using affect heuristic to support children’s right to not having migrant mothers is a way to bias judgements about the world’s injustice. While both domestic migrants and children from the poor countries may experience emotional costs, both have much more urgent needs. Construing a coherent story about the “new emotional imperialism” thus diverts our attention from their genuine needs. 5. Conclusion There is an important gap between real and perceived proportion of domestic workers among migrant women. Migrant women employed as domestic workers in the European and Northern American countries represent 2% of the migrant women worldwide. In total, migrant women employed as domestic workers all over the world represent 8% of migrant women. However, researchers often commit hasty generalization and associate the “feminization of migration” with domestic work. This paper is the irst attempt to understand the source of bias. Here, the analysis is limited to the theory of “care drain” elaborated by Arlie Hochschild and adopted by numerous scholars. In this case, the biased perception is more prominent as “care drain” refers to a subset of domestic workers: those who are both mothers and who left their children in Third World countries. As this subset is logically smaller than the group of domestic workers, it is even less likely that it characterize the “feminization of migration”. How can the bias go unnoticed? The paper explored two possible mechanisms of bias – reasoning from small samples and using affect heuristic – and found that none of them fully explain the bias. In the samples, more than half of Filipina migrant were college educated and almost half of them had left no children behind. A better explanation has been found in a stereotyped thinking which systematically associates women with care. The literature on “care drain” unnecessarily ampliies this association, by representing migrant women as both care-workers and mothers. Such a stereotyped thinking can go unnoticed for anyone reasoning from affect heuristic. Indeed, imagining children left behind by migrant mothers provokes strong feelings which trump any other consideration. However, emotions are more rightfully experienced when informed by data. Acknowledgements I am grateful to Alex Sager, David Hirst, two anonymous referees and the editors, for their helpful comments and editorial work. I would also like to thank Ruth Achenbach, Yonson Ahn, Brunella Casalini, Solange Chavel, Ryoa Chung, Anca Gheaus, Lisa Eckenwiler, Milena Jaksic, Lena Näre, Patrick Pharo, Camille Schmoll, and Tiina Vaittinen for stimulating discussions Speranța DUMITRU 19 References Abdulrahim, S. (2010), Servant, daughter, or employee? A pilot study on the attitudes of Lebanese employers towards migrant domestic workers, KAFA, Beirut. Acosta, P. (2011), “School attendance, child labour, and remittances from international migration in El Salvador” Journal of Development Studies, 47(6), pp. 913-936. Adams, R. H., Page, J. (2005), “Do international migration and remittances reduce poverty in developing countries?” World development, 33(10), pp. 1645-1669. Akpinar-Elci, M., Elci, O. C., Civaner, M. (2015), “Care Drain”, in H. ten Have, (ed.), Encyclopedia of Global Bioethics,Springer International Publishing, Dordrecht, pp. 1-7. Alcaraz, C., Chiquiar, D., Salcedo, A. (2012), “Remittances, schooling, and child labor in Mexico” Journal of Development Economics, 97(1), pp. 156-165. Bettio, F., Simonazzi, A., Villa, P. (2006), “Change in care regimes and female migration: the ‘care drain’ in the Mediterranean” Journal of European Social Policy, 16(3), pp. 271-285. Brücker, H., Capuano, S., Marfouk, A. (2013), “Education, gender and international migration: insights from a panel-dataset 1980-2010”. Chammartin, G. M. F., Esim, S., Smith, M. (2004). Gender and Migration in Arab States: The Case of Domestic Workers. International Labour Organization, Regional Ofice for Arab States, Geneva. Clemens, M. A. (2011), “Economics and emigration: Trillion-dollar bills on the sidewalk?” The Journal of Economic Perspectives, 25(3), pp. 83-106. 20 The Romanian Journal of Society and Politics Dumitru, S. (2011), “Care drain: le piège sexiste du nationalisme”, Tiziana Caponio, F. Giordano, and L. Ricaldone, (eds.), World Wide Women. Globalizzazione, generi, linguaggio. University of Turin: CIRSDe, pp. 51-60. Dumitru, S. (2013), “Des visas, pas de l’aide! De la migration comme substitut à l’aide au développement” Éthique publique. Revue internationale d’éthique sociétale et gouvernementale, 15(2), pp. 77-98. Dumitru, S. (2014), “From “brain drain” to “care drain” : Women’s labor migration and methodological sexism” Women’s Studies International Forum, 47, Part B, pp. 203-212. Dumitru, S., Marfouk, A. (2015), “ Existe-t-il une féminisation de la migration internationale? Féminisation de la migration qualiiée et invisibilité des diplômes” Hommes & Migrations, 1311(3), pp. 31-41. Fröbel, F., Heinrichs, J., Kreye, O. (1980), The new international division of labour : structural unemployment in industrialised countries and industrialisation in developing countries, Cambridge University Press, London. Gheaus, A. (2013a), “Care Drain as an Issue of Global Gender Justice” Ethical perspectives, 20(1), pp. 61-80. Gheaus, A. (2013b), “Care drain: who should provide for the children left behind?” Critical Review of International Social and Political Philosophy, 16(1), pp. 1-23. Gündüz, Z. Y. (2013), “The Feminization of Migration: Care and the New Emotional Imperialism” Monthly Review, 65(7), pp. 32-43. Hatton, T. J., Williamson, J. G. (1998), The age of mass migration: Causes and Speranța DUMITRU 21 economic impact, Oxford University Press, Oxford. Hochschild, A. R. (2000), “The nanny chain” The American Prospect, 11(4), pp. 32-36. Hochschild, A. R. (2002), “Love and gold”, Global women: Nannies, maids, and sex workers in the new economy,Owl Books, New York, pp. 15-30. Hochschild, A. R. (2013), So How’s the Family? : And Other Essays, University of California Press, Berkeley, CA, USA. Hondagneu-Sotelo, P. (1999), “Introduction: Gender and Contemporary U.S. Immigration” American Behavioral Scientist, 42(4), pp. 565-576. Human Rights Watch. (2008). Deaths of Migrant Domestic Workers in Lebanon. Human Rights Watch. ILO. (2011). Survey on children. ILO, Geneva. ILO. (2015). Global estimates on migrant workers. ILO, Geneva. Isaksen, L., Devi, U., Hochschild, A. (2008), “Global care crisis: mother and child’s-eye view” Sociologia(56). Jureidini, R. (2002), Women migrant domestic workers in Lebanon, International Labour Organisation, Kahneman, D. (2011), Thinking, fast and slow, Macmillan, New York. Kahneman, D., Tversky, A. (1973), “On the psychology of prediction” 22 The Romanian Journal of Society and Politics Psychological review, 80(4), pp. 237. Kilkey, M. (2010), “Men and Domestic Labor: A Missing Link in the Global Care Chain” Men and Masculinities, 13(1), pp. 126-149. Kroska, A. (2007), “Gender ideology and gender role ideology”Blackwell encyclopedia of sociology [electronic resources]. Blackwell Publishing: Malden. Lind, A. (2007), “Sexism” Blackwell encyclopedia of sociology. Blackwell Publishing. Lutz, H., Palenga-Mollenbeck, E. (2012), “Care Workers, Care Drain, and Care Chains: Relections on Care, Migration, and Citizenship” Social Politics: International Studies in Gender, State & Society, 19(1), pp. 15-37. Manalansan, M. F. (2006), “Queer Intersections: Sexuality and Gender in Migration Studies” International Migration Review, 40(1), pp. 224-249. Mies, M. (1986), Patriarchy and accumulation on a world scale : women in the international division of labour, Zed, London. Momsen, J. H. (1999), “Maids on the move: victim or victor”, in J. H. Momsen, (ed.), Gender, Migration and Domestic Work,Routledge, London & New York. Näre, L. (2010), “Sri Lankan men working as cleaners and carers: Negotiating masculinity in Naples” Men and Masculinities, 13(1), pp. 65-86. Nussbaum, M. (2003), “Capabilities as fundamental entitlements: Sen and social justice” Feminist economics, 9(2-3), pp. 33-59. Oelz, M. (2014), “The ILO’s Domestic Workers Convention and Speranța DUMITRU 23 Recommendation: A window of opportunity for social justice” International Labour Review, 153(1), pp. 143-172. Parrenas, R. S. (2000), “Migrant Filipina Domestic Workers and the International Division of Reproductive Labor” Gender and Society, 14(4), pp. 560-580. Parreñas, R. S. (2001), Servants of Globalization: Women, Migration, and Domestic Work, Stanford University Press, Stanford, CA. Parreñas, R. S. (2005), Children of global migration: Transnational families and gendered woes, Stanford University Press, Stanford. POEA. (2015), “Statistics OFW”. Sassen, S. (1988), The Mobility of Labor and Capital: A Study in International Investment and Labor Flow, Cambridge University Press, Cambridge Slovic, P., Finucane, M. L., Peters, E., MacGregor, D. G. (2007), “The affect heuristic” European journal of operational research, 177(3), pp. 1333-1352. Standing, G. (1989), “Global Feminization through Flexible Labor” World Development, 17(7), pp. 1077-1095. Tversky, A., Kahneman, D. (1983), “Extensional versus intuitive reasoning: The conjunction fallacy in probability judgment” Psychological Review, 90(4), pp. 293-315. UNESCO, UNICEF. (2015), “Fixing the Broken Promise of Education for All: Findings from the Global Initiative on Out-of-School Children”. UNESCO Institute for Statistics, Montreal. US Census Bureau. (2011-2013), “3-Year American Community Survey”. 24 The Romanian Journal of Society and Politics Wong, S.-L. (1994), “Diverted mothering: Representations of caregivers of color in the age of multiculturalism”, in E. Nakano Glenn, G. Chang, and L. Forcey, (eds.), Mothering: Ideology, experience, and agency,Routledge, London, pp. 67-91. Yang, D., Choi, H. (2007), “Are remittances insurance? Evidence from rainfall shocks in the Philippines” The World Bank Economic Review, 21(2), pp. 219-248. Yoshimura, M. (2007), “Feminization of Labor Migration”, in G. Ritzer, (ed.), Blackwell Encyclopedia of Sociology,Blackwell Publishing.