Thinking Policy Through Migrant Domestic Workers’ Itineraries

Introduction

Many countries, including Canada, Singapore, the United Arab Emirates, and Hong Kong, have developed policy frameworks aimed at facilitating and managing the import of temporary foreign workers to fulfil their national care work demands. Simultaneously, labor export states, such as Indonesia and the Philippines, have developed policies aimed at training and governing an ordered flow of migrant workers from their home countries to their job contracts abroad and back again (Parreñas et al., 2018). A robust body of research has focused on the ways that these policies—and gaps within them—affect migrant workers (Koh et al., 2017; Oishi, 2005; Peng, 2017; Piper, 2017; Yeoh & Huang, 1998). Overall, research has found that existing policies not only fail to protect migrant workers’ rights but also, in many cases, contribute to the production and perpetuation of various aspects of vulnerability, marginalization, and exploitation among migrant workers. ¹

It is thus relatively straightforward to offer a critique of existing policy. First, states should stop enacting policies that have been found to create the conditions for abuse. This would involve a radical restructuring of domestic labor laws, as well as a reconfiguration of many of the existing laws and institutions ostensibly aimed at protecting migrant workers. ² Most fundamentally, it would require that governments pass and enforce laws that conform to classic labor demands: provide a higher minimum wage; legislate for a maximum number of working hours, require breaks, days off, and overtime pay, and enforce that legislation; see that the basic health care needs of workers are met and that the employer pay for these costs; and insist on a harassment-free workplace (J. Fish, 2014). Indeed, these basic labor rights, as well as additional elements of decent work, are outlined in the United Nation’s International Labor Organization’s (2019) Convention 189, the “Convention concerning decent work for domestic workers,” which was adopted by the United Nations in 2011, entered into force in 2013, and as of June 2019 has been ratified by 29 countries. The passage of Convention 189 has been important in terms of providing international recognition of domestic work as work, as well as developing some baseline norms for improving the basic rights and working conditions of domestic workers (J. N. Fish, 2017).

However, despite this shift in the international discourse toward support for domestic workers’ rights, our own research and that of many others notes the persistently uphill nature of these struggles (England, 2017; Silvey & Parreñas, 2019; Romero et al., 2014). We argue that to understand how policies can be more effective at protecting migrant workers in practice, we must situate labor conditions in their broader political economic context. Thus, in this article, we ask not what states or international organizations can do in terms of policy improvements but more generally how a series of systemic political-economic issues—with complex relationships to state policies and international conventions—affect the possibilities and constraints for migrant workers’ welfare over the life course.

The article is organized according to three phases of migrants’ journeys: (a) the premigration context, which in Indonesia and the Philippines is characterized by long histories of high rates of unemployment and underemployment, high levels of intergenerational poverty, and a growing dependence on migrants’ remittances as basic income for households, communities, and remittance-dependent nations (Chami et al., 2018; Wickramasekara, 2011); (b) the labor migration regimes of the destination countries, which vary across countries in terms of regulations affecting migrant domestic workers, but which as a whole are exclusionary vis-à-vis citizenship and geared toward extracting maximum labor while providing minimum benefits to workers (Parreñas, 2015); (c) the postmigration system, which—while beginning in some places to evidence some changes across generations at least in terms of potential migrants’ aspirations and education levels—continues, among the majority of our respondents in the United Arab Emirates, to be shaped by debt, dependency, and vulnerability (Parreñas et al., 2018). ³

The article draws on data collected in Dubai, the United Arab Emirates, in 2013, 2014, and 2015. We carried out in-depth, semistructured interviews with 85 Filipino and 79 Indonesian domestic workers. The respondents were identified via quota sampling, so that we could analyze workers employed by people from a range of ethnonational backgrounds (Emirati, other Arab, South Asian, Iranian, Filipino, and European), as well as workers with a range of different migration statuses, including documented and undocumented workers, workers who had absconded and were staying at migrant shelters, and workers who enjoyed a day off each week as well as some who had never had a day off. We managed to access a diverse group of domestic workers with the help of numerous Filipino and Indonesian frontstage workers in Dubai, as well as the support of both national embassies. ⁴

Although the interviews themselves were not focused on policy questions per se, they revealed the importance of several key policy issues embedded in this migration system. To illustrate the relationship between policies and their broader contexts, we focus on the histories and memories of two individual women whose experiences were similar to many of their peers. The two women we have chosen are not themselves average subjects, nor of course are their experiences statistically representative of those of the larger population. Nonetheless, their experiences converge with those of the majority of the domestic workers interviewed in Dubai as well as other locations, and the patterns that are reflected in their migration biographies cohere in large part with the findings of other scholars, as we detail in the following sections.

Premigration Stories: Unemployment, Debt, and the Dream House

Nursinih was a 35-year-old woman who grew up in a farming and fishing family in Indramayu, West Java. At the time of the interview in 2013, she had been working in Dubai for 4 years, having departed Java in 2009, leaving her 2-year-old daughter back home to be cared for by her husband and her mother. Before Dubai, she had worked in Saudi Arabia for 6 years, and prior to that, she had been unemployed. Her husband, whom she married in 2004 after her work in Saudi, earned only an unreliable and low income as a fisherman, and without her overseas income, they were only able to afford the most basic necessities. Before her first sojourn, she had lived in a house with a dirt floor, rattan walls, and a corrugated aluminum roof, and she dreamt of building a brick house someday with tile floors, glass windows, and a plumbing system. Before she was married, and with the income she earned on her first three contracts in Saudi, she began to build her dream home.

Nursinih had several neighbors who had a managed to finance what she considered beautiful home renovations with earnings from working as overseas domestics. When she scrolled through the photos of her home community, she pointed out the newly renovated houses that she especially liked. When she showed me an image of a neighbor’s freshly painted orange cement house with shiny white porch tiles, she said, “The owner of this house worked in Taiwan. . . .” She scrolled over to a photo of a three-story structure with bright green walls and pink trim, “This one worked in Saudi. . . .” Then, she showed me a newly constructed bungalow with large windows, saying, “Look! This one made a lot of money in Hong Kong.” From her perspective, it appeared that almost all of the renovations in this community had been funded by remittances (on the Philippines, see also Aguilar, 2009). Her view was confirmed by an interview later with a village head in Indramayu who explained that the remaining unimproved houses, like Nursinih’s premigration house, belonged to fishermen, farmers, and unemployed locals, all of whom lived with families without migrants. Visible poverty in this community was tied to nonmigration, whereas relative wealth via home improvements signified success gained through migrants’ remittances. In this context, migration was the most common way that local people could envision tiling their dirt floors and escaping from persistent unemployment and poverty. ⁵

At 18 years, 2 years after Nursinih had dropped out of high school [“because we could not afford the school fees”], she began looking for work. Like many of her neighbors, she put herself on the waiting list for factory work. There was substantial competition for the existing factory positions, and she had to borrow money from a local moneylender in order to cover the cost of reserving a spot on the waiting list. There was no certainty about how long she would have to wait for an interview, nor any guarantee that she would ever be offered a job. The factory jobs themselves, for those who were fortunate enough to secure them, were similar to those in other parts of Java, promising only meagre wages, few benefits, and no job security. In order to add her name to each waitlist, she had to borrow additional money. She began to worry that she would be unable to repay the debts she had incurred simply looking for work (see also Sim, 2009).

It was in this context that Nursinih decided to pursue work abroad. Her anxiety about her indebtedness and her poor prospects for employment led her to respond to the recruitment agent making overtures in her neighborhood (Killias, 2010; Lindquist, 2010). At this point she was 20 years old, indebted, unmarried, and feeling, in her words, “ . . . bored and without options.” The agency’s offer of a 2-year contract in Saudi Arabia, paying 600 riyals (or US$200) per month, seemed more than acceptable, given that the salary on offer was almost five times more than she would have earned in the factories.

Nursinih’s decision to go abroad, and the ways she considered the possibility, were similar to those of many of the young women in her neighborhood. As she put it,

It wasn’t really a decision; it was the only choice. Even my parents felt I should go. Other friends were also suffering under debt loads. None of my friends wanted to work as farmers, and while we were all willing to work in factories, there weren’t enough job openings. So, actually, when it turned out that I could work in Saudi, I was grateful to God. A lot of us growing up in Indramayu have this fate.

In 1998, she was far from alone in her sense of urgency and limited options; the Asian financial crisis had tested most Indonesian people’s ability to “accept one’s fate” (terima nasib). Jobs had become harder than ever to find, layoffs were widespread, and inflation rates, especially of basic food prices, made everyday survival even more challenging than usual (Raharto et al., 2013). Even middle-income professionals, when faced with the desperation and unemployment wrought by the 1998 financial crisis, had become newly willing to consider migrating abroad to work as domestics in jobs they would have generally considered well “below their station in life.”

When Nursinih accepted her first overseas contract, she was responding to economic pressures that were felt across the country and the region. However, in Indramayu, working abroad had already become commonplace well before the crisis. Here, as well as in other parts of Java (Raharto et al., 2013; Spaan, 1999), the labor export industry had already become relatively well established by the early 1980s (Lindquist, 2010). Since the Indonesian government began collecting data and formally managing out-migration in 1979, the province of West Java (including the regency of Indramayu) has registered more overseas migrant workers than any other region (BNP2TKI, 2016), and more of these have worked in Saudi Arabia over time than in other destinations. ⁶

Over the course of her multiple journeys, the economy in Indramayu grew significantly, but the sources of income remained largely the same, particularly for women with limited levels of education. More factories were built in the regency, so more positions became available on the assembly lines of garment and electronics producers, and Indramayu’s proximity to Jakarta meant that as the capital city’s growth accelerated, people were drawn to migrate internally from the coastal city’s poor prospects to the city where employment (formal and informal) options were growing. In the 2000s, Indonesia’s economic growth has skyrocketed, increasing by an average of 5% to 6% per year between 2005 and 2015. For Indramayu, much of this growth has arrived in the form of overseas migrants’ remittances, which totaled 9.42 billion rupiah (or over US$66 million) in 2015 for the nation as a whole (Listiani, 2017). The dream homes continue to be built. Satellite dishes adorn the rooves of the renovated houses. Many migrants return home at the end of a contract only to turn around and depart for another labor stint.

In terms of policy, how could Nursinih’s predeparture experiences have been improved? Most obviously, if state policies had been effectively directed toward poverty alleviation, public education, and basic income, she may not have felt the need to look for work abroad. If state policies had managed to make more jobs available, and jobs that offered living wages, the migration industry would have had to compete with other employers to find workers. Education levels had risen, and more jobs were available by the mid-2000s than had been in the 1980s. However, most of the job growth that took place in Indramayu was in factories, and these jobs remained scarce and paid little (Badan Pusat Statistik, 2010). So, like the majority of migrants interviewed, Nursinih felt that she had little choice but to work overseas.

Some of the other policies that affected her origin experience had to do with the process of recruitment, and especially the ways that the government distanced itself from governing labor recruitment, training, and export (see also Yeoh et al., 2020). The sponsor who escorted her from the village to the agency in Jakarta, and who assisted her with the extensive paperwork required for processing a visa, was not a formal government employee. Nor was the labor agency itself run by the state. Rather, the government credentialed the agency, and the sponsors who brought prospective migrants to the agency were self-employed. ⁷ By being “hands-off” in this way, the government left a lot to chance and to the discretion of the agencies. The agency required Nursinih to sign over the first 6 months of her wages to them, and out of these withheld wages, she paid her sponsor’s fee (approximately US$50, plus additional if he had assisted her with paperwork prior to arriving), her training, her visa processing, her round-trip airfare and ground transport to her job placement. In effect, this meant that all migrant domestic workers who agreed to such terms, and these were the norm in the late 1990s, worked as indentured servants for the first 2 to 8 months of their contracts. ⁸

Visa regulations were also a way that the premigration policy context shaped Nursinih’s experience and those of people in similar positions. The immigration bureaucracy at the time required extremely complicated, time-consuming paperwork to be delivered to and approved by officials at multiple government offices, such that prospective migrants had little choice but to depend on the services of agencies to shepherd them through the process. Others (Tirtosudarmo, 2009) have detailed the various stages involved in the process; its Byzantine nature emerged at least partially as the result of multiple offices hoping for a slice of the fees (or additional bribes, as the case often was). The opacity of the system meant that migrants generally accepted the agencies’ offers to take care of the application process for them, and the agencies passed the cost (plus additional fees for profit for themselves) on to the migrants (Jones, 2000).

The age restrictions on workers also affected the premigration experience. Specifically, the minimum age for overseas work in 1998 was 25 years, whereas Nursinih was only 18 years. However, her sponsor assured her that he could arrange for her documents to falsely vouch that she had reached the required age. He arranged for several of her required documents (her birth certificate, medical test, and household registration) to be falsified in this way, but he assured her that it was “normal” for migrants to carry “fixed” documents. She never met an official face-to-face and remained dependent on her sponsor for information. This overall context thus left migrants in the dark about existing regulations and kept them at arm’s length from officialdom. The agencies and the sponsors, as well as migrants themselves, justified these arrangements with reference to the low education levels of the migrants and the complexity of the forms and the process. Until many years later, the government made no effort to simplify the process or to create systems that would enable migrants to have more autonomous, direct access to official visa processing systems. ⁹

Finally, it is worth noting that prior to departure, candidates for migration were required to pass health tests, psychological tests, and skills training. They were forced to undergo pregnancy tests, tests to prove that they were free from sexually transmitted diseases, general physical exams, and psychological tests to prove their mental fitness (“fit mental”). These requirements reflect the overall thrust of the overseas labor program, which is aimed at identifying and exporting individual workers—separated from their families—who are “fit” for labor under stressful conditions. Once accepted into the program, the agency trains them to prepare them for their employment; the training periods run from 1 to 4 months, depending on the agency. For Nursinih, this involved staying at a training facility in Jakarta for 1 month during which time she was taught some basic Arabic. The agency’s goal in the training period was to improve the workers’ intercultural and linguistic fluency to the point that their employers would not lodge complaints. The overall strategy on the part of the agency was geared toward minimizing complaints from either employers or employees, and the training was oriented toward cultivating docility, self-sacrifice, a strong work ethic, and fear among the migrants (see Rodriguez, 2010, on the Philippines; Chang, 2018). The training led Nursinih to pray that she would end up with a good employer, as she had learned that neither the agency nor her government would help her if she found herself in an abusive situation (see also Killias, 2018).

Migrant Work Experiences: Vulnerability and Luck

When Nursinih initially left home on her first overseas contract, she was placed with a family with two small children. The work was more or less what she had expected, long days of cleaning, cooking, and caring for the children, rising at dawn for early-morning prayers, going to bed after she had cleaned the kitchen and put the children to sleep. The money she earned was eventually sufficient to invest in beginning to build her house, though she had been required to endure many months of indentured servitude prior to receiving her first wage payment. She felt “fortunate” to have been placed with an employer who was “nice,” in her words, who offered her plenty to eat, time to rest during the day, and a workload that “wasn’t too heavy.” The most difficult part of adjusting to her new job was learning to speak Arabic, since she had only received one month of language instruction as part of her job training prior to leaving. Although her Arabic was good enough to serve the family and understand their requests (“Thank you”; “I’m sorry”; “Do you need anything?” “May I wash that for you?”), she was unable to communicate about anything beyond her duties. Over the 2-year contract period, however, she learned to express her thoughts with relatively ease. The family had grown fond of her, and they asked her to return to work for them again after her planned short break between contracts.

She had heard many stories and read news reports about women whose experiences had been much worse than her own (see Fernandez & de Regt, 2014). She laughed nervously when she reported that she was “glad to never have had any sexual harassment; always received pay on time; and never been beaten up.” When I asked her to elaborate on these thoughts, she told me that she had repeatedly been warned prior to departure that she could be one of the “unlucky ones,” and she had worried about whether or not she would “get placed with a good family.” She said she was only now able to laugh because she had avoided being one of the victims of the widely reported worst case scenarios, but she also admitted that discussing the topic made her nervous. She was cognizant that a “high risk” existed of falling victim to various abuses, and she knew that she would face these dangers anew if she were to start a new contract with a different employer (cf. Jureidini, 2010). Thus, when the family invited her to return for a second contract, she welcomed the invitation, feeling that her “fate had been good,” and that she would like to continue to earn what she termed her “high salary” (of approximately $200/month with a salary deduction taken—in the form of six months of withheld wages—by the agency). Her interest in taking a second contract was driven in large part by the fact that she had only earned enough to money to build the first bit of her house; the foundation had been laid, and the cement had been poured for the walls, but with another contract, she hoped she could complete the roof, the tiles, and the porch.

She worked with the same family for two additional contract periods, totaling 6 years in all with short breaks (1-2 months) in between to visit family and renew her visa. During this time, her salary remained unchanged. As her fluency in Arabic improved, and her familiarity with her employers and the context deepened, she began to feel “at home” in Saudi Arabia. She continued to invest in the building of her house in Indramayu, but it remained unfinished even after she completed her third contract. At this point, she was in her mid-20s, and her family and friends were urging her to get married; she herself was feeling “ready to stop working and start a family.” So, when her longtime boyfriend, Asep, proposed to her, she readily accepted. They used her remaining savings to fund the wedding festivities, which were, according to her, “pretty fancy by our village standards,” and they moved into the unfinished house to begin their lives together. Both she and Asep hoped her days working overseas were over, and they were especially keen to stay in Indramayu together once their first child was born. However, by the time their daughter was almost 2 years old, they felt they had little choice but to consider another sojourn for her (see also Babar & Gardner, 2016). They wanted to finish building the house, and they wanted to be able to afford to send the girl to school; most important, according to Nursinih, was her feeling that she had become accustomed to working and earning an income, and it was difficult to just “be bored in the house.”

By 2009, there were many more sponsors offering positions in a wider variety of destinations than when she had first travelled, so this time she decided to try her luck in another country. By the time I met her, she had been employed in the same household for 4 years (with one visit home after the first 2 years). The work was similar to what it had been in her first job, as she had been “lucky” again with another placement in a “nice” family. She preferred Dubai to Saudi because she felt “free” [bebas]. The family employed two full-time, live-in domestic workers (one in addition to her), as well as a driver, all of whom were from Indonesia. She liked her coworkers, and she appreciated this “bustling/lively” [ramai] family, especially because caring for the six children, two of whom were still younger than 5 years, made her feel “at home” and consumed most of her energy. ¹⁰

Even with two domestic workers, however, the workload was heavy; cleaning the four-bedroom, four-bathroom house, caring for all the children, and cooking for the large family meant that she and her coworker were busy from dawn until after dark every day with no days off. What constituted a break in her eyes were the family outings she was invited to join. Although her duties continued during these outings, including caring for the children, carrying shopping bags, and fetching snacks or cleaning up spills (while being seated just outside of the restaurants to wait with her coworker when the family ate inside), she viewed these outings as a reprieve from her regular routine. More than anything, though, she appreciated the salary, which started at 800 dirham (approximately US$400) per month and had increased to 1,000 dirham (approximately US$500) per month. At this rate, she told me, she would finally be able to finish building her house, and this was her plan, to continue working in Dubai “with patience and fortitude,” so that she could eventually return home to her family.

The contracts for domestic workers in Saudi Arabia and Dubai are not regular labor contracts subject to the usual labor laws; the domestic sphere is considered private and beyond the purview of the law. When Nursinih was working there, no legislation existed to regulate rest days, hours of work, holidays, sleeping space, or minimum wages. Employers (then as now) have the power to fire and deport their domestics at will. The kafala system binds workers in a relationship of patron–client servitude with their employers, meaning workers are subjected to the complete authority of their employers. ¹¹ This means that it is impossible for workers to change employers, a situation that exacerbates their vulnerability. The lack of labor protections intensifies the precarity of all migrant workers in the United Arab Emirates (Ali, 2010; Kanna, 2011; Kathiravelu, 2016).

The cost and complication of the process of securing her visa increased Nursinih’s dependence on her sponsor and her agent, and the debt she shouldered from the agency—as well as the lack of laws against such predatory lending—had pushed her into indentured servitude. Moreover, her formal training and her low-income background had socialized her to believe that she was “fortunate” to have been placed in the job “with the nice employer” and a “big salary.” The generalized fear and disempowerment that Nursinih felt were deepened by the broad context of political repression. Prior to 1998 in Indonesia, and throughout her time in Saudi and Dubai, civil society organizations were repressed by authoritarian governments (under President Suharto in Indonesia until 1998 and the monarchies of the King of Saudi Arabia and United Arab Emirates), and she, like most other migrants, was aware that the forces were stacked against her and anyone who might want to help her. She spoke in idioms of fortune and fate; she did not speak in the language of rights, entitlements, or freedom.

Return: Unfinished Dreams, Persistent Precarity

The Indonesian government’s policies for overseas migrants, and the broader policy climate in the country changed noticeably between 1998 when Nursinih first migrated and 2013 when I met her. However, her migration and labor experiences stayed remarkably similar over time. In 1998, Indonesian President Suharto’s reign ended after 32 years of authoritarian rule, and although his ouster was a momentous national-scale transition, it had little immediate impact on overseas workers.

As the reformasi era began, however, various policy changes were beginning to affect public expectations of the Indonesian government with respect to its role vis-à-vis migrant workers. Prompted by public pressure, in the first 2 years of reformasi, two ministerial decrees were issued aimed at increasing the government’s role in ensuring migrants’ welfare. The first, passed in 1999, required high-ranking officials at national and regional levels to produce regular progress reports regarding the placement and recruitment of migrants. ¹² In addition, it outlined the scope of the government’s obligations, including responsibility for information management, interagency coordination, and law enforcement/enactment (International Organization of Migration, 2010). It also mandated the simplification and improvement of the placement system and services provided to migrants, as well as “the empowerment of Indonesian labour migrants and improving the quality of protection for migrants and their families; [and] improved performance of recruitment agencies” (International Organization of Migration, 2010, p. 12).

Between the time Nursinih had first left home and the beginning of her third contract, another ministerial decree was issued, this one specifically geared toward offloading some responsibility for protection from the state onto the private sector and individual migrants. ¹³ It authorized private companies to recruit and place domestic workers, thereby relieving the state of responsibility should anything go awry, and it simultaneously increased the sanctions directed at individual migrant workers if they “‘resigned’, ‘violated the terms of the work agreement’, [or] ‘took actions that were threatened with criminal sanctions’” (as cited in International Organization of Migration, 2010, p. 17). In practice, at least in Nursinih’s experience, these decrees changed little, but they represented efforts on the part of the newly democratic state to remove itself from the line of fire being driven by civil society actors. And, although this change in the political climate did not affect policies directly, and thus did not improve Nursinih’s salary or working conditions, the spirit of reformasi was in the air as she finished her third contract, and it was likely part of what encouraged her to start expecting a higher wage and better working conditions.

Indonesia as a whole, and Java and Bali in particular, have been framed by states as “labor surplus” regions, and as such they have long been targets of state interventions aimed at population redistribution (e.g., transmigration policies which began in the Dutch colonial period and have been continued by postindependence and reform-era leaders; Elmhirst, 1999). States have rationalized these interventions—whether transmigration policy or recent overseas labor policies—with reference to poverty alleviation and employment creation. The policies have supported and promoted migration for the sake of national development, and they have done so in the name of assisting individual migrants, their families, and their communities. For Nursinih at least, the policies directed at her migration—and the limited social supports provided for basic income generation—left her in something of a policy vacuum, one where she was able to access little in the way of state supports, struggling with poverty and unemployment, and therefore vulnerable to agents and brokers looking to earn some money off of delivering her for export.

For another woman we interviewed, Mishra, who was from Zaboanga, Basilan, the story was similar. Despite the Philippines Overseas Employment Agency and the Philippines embassy in Dubai having provided her with some better support both before and during migration than the Indonesian government provided, her overall experience was equally negative. She had left the Philippines for many of the same reasons Nursinih had left Indonesia: few job prospects, family needs, limited local options, and the desire to earn a lot of money. But she was less “fortunate” than Nursinih in terms of her employer, eventually being treated so cruelly that she ran away. Her employer had confiscated her passport, ¹⁴ so she worked in the informal sector, taking on live-out domestic work and earning an unreliable and inadequate income. Her undocumented status left her structurally vulnerable to abuse by her employers, but in her experience, she was “luckier” without her papers in the informal sector than she had been with her formal employer.

Mishra’s experience was not unique. According to Soraya, who had also been working informally in the United Arab Emirates for 10 years:

A migrant worker isn’t protected by the law even when she is “legal”; she isn’t allowed to leave the house or the country unless her employer lets her. So, when we’re working without visas [informally] here, we don’t feel more scared than we do when we’re stuck [formally] in a cruel employer’s house. In fact, once we run away we can actually walk around somewhat freely, and the only thing we have to worry about is a police raid [on undocumented workers].

Surely the awareness that a police raid could lead to their imprisonment and deportation contributed to the experience of precarity among migrants without papers. Indeed, the depth of their precarity with or without papers is simply underlined by these reports by some of feeling less fearful and less confined as undocumented workers than they had felt with papers in the confines of their employers’ homes in the United Arab Emirates (see also Johnson & Lindquist, 2019). ¹⁵

Mishra, like Nursinih, was a serial migrant (Parreñas et al., 2018). She had been unable to meet her aspirations for migration by the end of her first contract, and soon after returning home she was arranging for her next round of overseas work. She, like Nursinih, had dreamt of building a house with her income, and she too had been unable to fulfil her dream. ¹⁶ She had two sons who were still in school, and she felt it necessary to continue to work until they graduated. She had never finished high school, and she aspired to provide her sons with the education she had never completed. During the almost two decades she had worked abroad, she had accumulated no savings, having sent all of her earnings to support her sons and their educations (on return, see Xiang et al., 2013). So, she began the process of signing up to go abroad yet again.

Both Mishra and Nursinih were embedded in serial migration prompted by persistent poverty and limited employment prospects at home. Elsewhere we have noted that serial labor migrants similar to them are involved in “precarity chains,” a concept that refers to the relational, intra-familial reinforcement and transfer of precarity across space and generations. It is a cycle that is more pronounced among migrants with lower education levels whose labor migration itineraries keep them in cycles of repeated work contracts located in transit states (Khalaf et al., 2015). Transit states are the destinations that offer no options for permanent settlement and that cost significantly less in recruitment fees compared to countries with pathways to permanent residency or citizenship (e.g., $8-$12,000 for Canada or Italy; Parreñas, 2015).

The financial insecurities of domestic workers affect them in the immediate term, as well as over the longer term. For many women we interviewed, one of their primary goals in working abroad in addition to building a house was to see that their children would never have to take on the same kind of work they had done. Recent research (Bittiandra et al., 2019) indicates that some children of former migrant domestic workers from Indonesia are choosing not to migrate in the footsteps of their mothers. Some have developed a sense of having “enough” without going abroad; they feel an obligation to their mothers who have sacrificed through migration for them to make use of their education to find work in Indonesia. This appears to be changing the aspirations of the younger generation among Bittiandra et al.’s (2019) respondents, as fewer of the second generation in their study are migrating to work abroad than we see through the lens of “precarity chains.” ¹⁷ However, among our respondents in Dubai, it appeared that many daughters of domestics were still following in their mothers’ footsteps, including Mishra for instance whose mother had been a domestic worker for over 20 years in Saudi Arabia by the time Mishra accepted her first contract.

After almost two decades of migrant labor, Mishra still had no savings. If there had been any social welfare schemes in place for migrants, Mishra could have been protected from persistent poverty. But the state had already offloaded precarity and poverty onto individual migrants and their families, so the risks and the costs of the journeys fell entirely on them. ¹⁸ In a glaring contradiction, while the costs and risks of migration were shouldered by migrants and migrants’ family members, the benefits accrued disproportionately to employers and nation-states, the well-being and economic development of which have been made possible by the labor of migrant workers.

Conclusions: Thinking Policy Through Migrants’ Itineraries

Following Mishra and Nursinih over chapters of their journeys has provided some context and texture for understanding how policy gaps and failures reproduce the precarity of migrant domestic workers. Their cases illustrate the patterns that we identified in the larger samples of migrants (79 Indonesian and 85 from the Philippines) we interviewed in Dubai, specifically the patterns of (a) limited state supports for migrants prior to migration (i.e., policies that promote migration as a “solution” to poverty but do not simplify or facilitate migration training or work visas); (b) limited rights for workers during their sojourns (i.e., major gaps in labor and human rights protections for domestic workers); and (c) finally persistent financial insecurity of workers after migration (i.e., policies and an economic development context that leaves them with few livelihood options other than returning to repeat their migration). While Indonesia and the Philippines offer distinct programs for overseas workers, the two countries exhibit broad policy similarities that contribute to reinforcing migrants’ persistent poverty and serial migration.

The patterns of our findings—the ways in which sending country policies contribute to persistent poverty and precarity—are particularly relevant for the context of Dubai, a destination that is less costly to access and as such attracts lower income migrants than other destinations such as Hong Kong, Israel, Singapore, or Canada. Indeed, research in other locations finds that migrants may migrate out of poverty (from Indonesia to Singapore, as in Goh et al., 2016) or engage in stepwise migration (from the Philippines and Indonesia to Hong Kong and then Canada, as in Paul, 2012). Among migrants in Dubai, we did not observe stepwise migration patterns, and we argue that their socioeconomic immobility results in part from their relative entrapment in the specific migration trajectories that reproduce lower incomes. Further research is required to determine precisely how these observed differences in outcomes of migration are shaped by relational geographies, specifically how socioeconomic contexts of sending communities prior to migration articulate with specific political economies of destination contexts for migrants.

Both the Philippines and Indonesian governments have actively promoted the out-migration of low-wage laborers as a policy “solution” for underemployment and poverty. Global migration policy makers have trumpeted a “win-win-win” narrative about temporary labor migration, promoting the idea that sending countries, receiving countries, and migrants themselves can all benefit from temporary labor migration systems (Sinatti, 2015). However, our research joins that of critical work (e.g., King & Collyer, 2016) that highlights the hierarchies of origin and destination to which migrants travel and from which it can be difficult to escape, such that migrants from lower-income countries and lower income households tend to concentrate their mobility within lower income circuits. ¹⁹ This finding suggests that the emphasis on “win-win-win” (framing temporary labor migration as a win for sending countries; win for receiving countries; and win for migrants) in much recent policy work on migrant remittances is unduly optimistic, particularly for the migrants and the origin communities that are more impoverished. ²⁰

Our findings indicate that policy research would do well to move beyond conceptualizations of migration in terms of bi-national or even transnational trajectories, and strive to more fully engage the global context of today’s migrations. This is especially important for the study of labor migrants who are increasingly understood to participate in “serial migration” (also sometimes called multinational migration, stepwise migration, or repeated migration), such that over the life course they migrate on repeated contracts to multiple destinations. ²¹ For policy makers concerned with migrant welfare, this finding indicates that rather than considering binary pairs of migrant “sending” and “receiving” countries, or even to transnational ties, attention would be more fruitfully directed toward the more complex itineraries that migrant workers actually follow in practice. The cracks between the jurisdictional scopes of individual nation-states exacerbate the vulnerability and exploitation of migrant workers, and they reinforce the conditions that perpetuate inequalities between migrant workers and their employers.

Finally, we find that indebtedness and precarity travel with the migrant domestic workers in our case studies. The policy lesson based on this finding is again at odds with the optimism of “win-win-win” approaches to migration and development, and encourages a reframing of development priorities away from migration-led solutions, toward spatial equity policies and basic income grants. Only by intervening in the pervasive sources of poverty and precarity for domestic workers can their welfare be structurally improved. The welfare of migrant workers is contingent on their having a greater universe of choice in terms of livelihoods, greater protections and benefits when they take on employment, and greater financial and social security for themselves and their families over the longer term. Policies that place migrants’ human needs, complex migration itineraries, and life course aspirations at the center of dialogue will be more likely to provide meaningful benefits to migrants and their communities than have policies aimed at the promotion and protection of temporary labor flows.