Social Protection for Migrant Domestic Workers in Singapore: International Conventions,the Law,and Civil Society Action

Migrant Domestic Workers in Singapore

In an opinion piece making the case that “Foreign workers deserve protection” (2018), the national daily appealed to Singaporeans to accord foreign workers the “dignity of work” by looking into the mirror of their own ancestry:

Most Singaporeans can trace their ancestry to the arrival of workers whose industry and resilience turned an island into a nation—one that today’s foreign workers continue to build. Singaporeans would acknowledge that work as a cornerstone of their history.

Over the past few decades, the recruitment of migrant women from developing countries in South and Southeast Asia as live-in domestic workers continues this history of dependence on migrant labor to pave the way for Singapore’s rapid industrialization and accelerated growth. As Singapore faces demographic challenges typical of advanced economies—a plunging fertility rate, increasing life expectancy, and a growing proportion of elderly in its population—it continues to depend on migrant domestic workers as an economical solution to fill care deficits and avert social reproductive crises.

Approximately one in five families in Singapore employ a migrant domestic worker. In June 2018, there were 250,000 women, chiefly from Indonesia, Myanmar, and the Philippines registered in Singapore as domestic workers (Ministry of Manpower [MOM], 2018b). The growing numbers of migrant domestic workers have induced stringent control of their presence as a transient workforce, as policy makers construe these migrant women as potentially constituting social threats to Singapore society if not carefully managed (Yeoh & Annadhurai, 2008). Migrant domestic workers are thus governed under an illiberal immigration–citizenship regime of temporary migration. Regulatory measures that enact such transience include regular medical surveillance, 2-year employment contracts, and an employer-sponsored work permit regime.

Several features of live-in domestic work place migrant women in precarious situations: the blurred boundaries between work/home and between re/productive labor; their isolation within privatized households compounded in some cases by their restricted access to communication devices; the language barriers newcomers confront; curtailed access to social support systems (Baig & Chang, 2020); and their status as lowly paid foreigners that contribute to negative stereotyping (Lai & Fong, 2020). The persistent ideology around reproductive labor as unpaid women’s work sets paid domestic work up as a “site of multiple exploitations by gender, class, race/ethnicity, and nationality” (Yeoh et al., 2015, p. 298). “Home,” as experienced by migrant domestic workers, is paradoxically a “public space” out of their control and a “private space” within which curtailment of rights, and forms of abuse, are shielded from the public eye (Huang & Yeoh, 2007).

Considering the precariousness of live-in domestic work and the regulations in Singapore, it is pertinent to examine how migrant domestic workers access social support and protection (Fong & Yeoh, 2020). Social protection, in Devereux and Sabates-Wheeler’s (2004, p. 9) seminal definition, refers to:

all public and private initiatives that provide income or consumption transfers to the poor, protect the vulnerable against livelihood risks, and enhance the social status and rights of the marginalized; with the overall objective of reducing the economic and social vulnerability of poor, vulnerable and marginalized groups.

Their work expands the conceptualization of social protection beyond social assistance for the vulnerable poor to include social justice efforts that transform structural inequalities. In a similar spirit, Locke et al. (2013) go beyond formal social protection according to citizenship status to include informal social protection, such as practices found within kin and friendship networks, community relations, and the capability and agency of individual actors (see also Ye & Chen, 2020).

In tandem with the surge in cross-border mobility and the transnational migration of people across a wide spectrum of socioeconomic classes, scholars have advocated paying attention to how migrants are embedded in transnational social fields (Glick-Schiller & Faist, 2009). Levitt et al. (2017) write that the burgeoning number of migrants parallels the growing cross-border movements of religious, political, economic, and humanitarian organizations. In critique of the state–centric perspective of literature focusing on welfare state regimes as institutions of social protection, they advocate a “transnational social field approach which knits together allegedly separate sending and receiving country spaces into a single, sometimes seamless, and sometimes deeply fractured, social, political, and emotional imaginary” (Levitt et al., 2017, p. 4). Ideally, a transnational analysis of social protection for migrants should recognize the interplay of social protection at the scales of international conventions, origin country, and destination country; the collective provisions of non-state actors; the (limitations of) access to existing systems and migrant efforts to plug the gaps through social networks and informal ties; and the fundamental reorganization and transformation of social protection systems. This is particularly pertinent as migrants’ migration biographies can now often comprise of multiple moves, even at varying socioeconomic levels (Silvey & Parreñas, 2020). In theoretical terms, this approach advocates the concept of a “resource environment” constituted by clustering social protections emanating from the four potential sources of states, markets, third sector, and social networks (Levitt et al., 2017, p. 6). Such an approach provides an empirical basis to examine the reconfiguration of rights and relationships of mobile individuals vis-à-vis states and the transnational linkages that make social change for marginalized groups possible (Yeoh et al., 2004). Levitt et al. (2017) recognize that the cluster of protections accessible to migrants is dependent on the ecology of the third sector, the capacity of individual migrants, the services available on the private market, and the relationship between both origin and destination countries. This article uses Singapore as a case study to map the resource environment accessible to migrant domestic workers. We suggest that, as a consequence of limited regional regulation and the circumscribed reach of countries of origin, combined with a governance regime that places restrictions on the development of an advocacy-oriented third sector, legal recourse, as part of the state apparatus of the host nation, still constitutes the most powerful form of social protection for vulnerable migrants.

This study is based on archival research on court cases of abuse in Singapore and media coverage in the national broadsheet The Straits Times as well as documentary research drawing from government websites, the Singapore Parliamentary Reports, and research reports published by nongovernmental organizations. This is supplemented by the authors’ research focusing on the conditions of migrant domestic work in Singapore drawn from two separate projects. The first is a quantitative survey of migrant domestic workers in Singapore in 2013, comprising 201 surveys administered to Indonesian domestic workers to understand women’s migration experiences and remittance-sending behaviors (Platt et al., 2013). We also drew from another project that traced the development of the “day off” policy for domestic workers in Singapore, where we conducted 14 qualitative semistructured interviews with key informants, such as employment agents, government officers, journalists, embassy officers, domestic worker groups, and representatives from nongovernmental organizations from 2015 to 2016 (Koh et al., 2017a). Supplementing our archival and documentary research, these projects have helped enrich our understanding of how the scope of social protection for migrant domestic workers in Singapore has shifted in the past decade.

We begin by adopting a transnational social protection research agenda to examine the international conventions and agreements pertaining to migrant domestic work and the efforts of origin countries to protect their female nationals working as domestic workers abroad. These efforts, however, have limited reach as destination countries often do not extend the full range of formal social protection to migrants as a means of “regulat[ing] migration by blocking access to services so that less desired migrants are encouraged to return home” (Levitt et al., 2017, p. 5). Following this premise that the social support and protection of migrant domestic workers are structurally constrained by host society, we devote the remainder of the article to two interrelated developments in Singapore. First, we examine the reach of immigration, labor, and criminal law in recent “maid abuse” cases. Second, we consider the changing landscape and existing reach of civil society through their campaign for a “mandatory” rest day.

International Conventions and Agreements

The vulnerabilities of migrant domestic work have long presented a subject of concern for international organizations such as the United Nations and the International Labor Organization (ILO). These organizations lay out international labor standards—conventions, protocols, and recommendations—that articulate fundamental principles, rights and obligations, and procedural recommendations so as to facilitate decent work conditions (Tayah, 2016). Selectively reviewing the raft of international conventions pertaining to migrant domestic workers, this section provides an overview of the key agreements most relevant to Singapore and sets out the limitations of multilateral governance.

The Universal Declaration of Human Rights (UDHR) established in 1948 posits equality of treatment and nondiscrimination, as well as the right to rest, access to regular holidays, and reasonable limitations on working hours for everyone. As a member state of the United Nations, Singapore underwent its most recent Universal Periodic Review of its human rights record in 2016, during which many countries proposed recommendations to improve migrant worker protections in Singapore. Adopted in 1990 by the United Nations, the International Convention for the Protection of the Rights of all Migrant Workers and Members of their Families (MWC) not only calls for equal treatment between migrants and nationals in the workplace, it also notes the increased risk that migrant domestic workers face of exploitation, abuse, and gender-based violence. Most recently, the Domestic Workers Convention No. 189 was passed in 2011, with a view to addressing the invisibility and undervaluation of domestic work, the vulnerability of marginalized women and girls who carry it out, and the state of developing countries with limited formal employment opportunities from which domestic workers hail. Singapore has not ratified the above conventions relating to migrant workers or domestic workers, or ILO Convention No. 97 (Migration for Employment, Revised) and Convention No. 143 (Migrations in Abusive Conditions and the Promotion of Equality of Opportunity and Treatment of Migrant Workers). At the regional level, while the Association of South East Asian Nations (ASEAN) adopted in 2007 the Declaration on the Protection and Promotion of the Rights of Migrant Workers that undertook to develop a binding and enforceable instrument on the protection and promotion of the rights of migrant workers, this has not materialized after more than a decade (Marti, 2019). The 2017 ASEAN Consensus on the Protection and Promotion of the Rights of Migrant Workers apparently represents renewed urgency to improve social protection provisioning for migrant workers, but given the continued dependence on nonbinding principles, any vision of developing a set of minimum labor standards with enforceable penalties appears difficult to achieve, if not illusory. The slow rate of progress is particularly worrying in view of the countervailing tendencies for a quickened race to the bottom as a result of the “confusion, conflicts, and abuses” stemming from “the lack of uniform, regional rules or standards governing entry, deployment, and national treatment of migrant workers within the ASEAN region” (Olivier, 2018, p. 20).

Nonetheless, a notable convention (specific to women rather than migrants) to which Singapore has acceded since 1995 is the International Convention on the Elimination of All Forms of Discrimination against Women (CEDAW) adopted in 1979. It obliges states to eradicate employment discrimination against women and recommends that destination countries safeguard the same rights for female migrant workers as the female citizen workforce. While this has not borne direct fruit for migrant domestic workers’ social protection, the convention provides a mechanism for migrant rights NGOs to advocate for migrant domestic workers to be included in the Employment Act and entitled to sick leave benefits, maximum working hours, periods of rest, overtime payment, and maternity rights (National NGO Singapore, 2017).

There are formidable challenges to the implementation and ratification of these international legal instruments. Parreñas (2017) argues that the Domestic Worker Convention only stipulates obscure recommendations for member states, which stymies the force of international law and provides latitude for members to avoid explicitly legislating and enforcing decent work conditions. Additionally, the signatories of migrant worker conventions are primarily countries of origin aiming to protect their citizens abroad. Without the cooperation of destination countries, however, extraterritorial regulations remain ineffectual and migrant women fall outside the protections of these international labor standards.

It would not be difficult to agree with Marti (2019, p. 1350) that international human rights and labor standards codified in international conventions “have not diffused into the domestic level, and their impact on the internal policies of Singapore with regard to migrant workers is limited.” Nevertheless, even without states acceding to them, these international agreements can contribute to advancing social protections for migrant domestic workers in roundabout ways, whether by moving the needle toward recognizing domestic work as “real work” or in crystallizing international norms of decent work for domestic workers (Koh et al., 2017b). Taking our cue from Yee’s (2009) argument that human rights in Singapore are neither universal nor substantive but tactically contingent, we consider in a later section how civil society activists promoted a day off for migrant domestic workers through framing migrant rights claims strategically.

The Social Protections Extended by Origin Countries

Apart from ratifying international conventions, governments in countries of origin have resorted to several legislative and institutional arrangements to protect their citizens working abroad. Typical protections enacted by origin country governments include dispatching labor attaches to be stationed at embassies of major destinations, administering predeparture training, processing job contracts that regulate particular aspects of employment, and licensing (and blacklisting) private recruitment and employment agencies. Origin states’ protective measures may also resort to restrictions on the mobilities of migrant women, reinscribing patriarchal relations of power and subsuming women’s agency to the collective will of the household (Yeoh et al., 2015). Despite the explicit discouragement in international conventions such as the UDHR, CEDAW, and MWC against restrictions on women’s migration, origin countries within ASEAN have resorted to banning female nationals from moving to particular countries for domestic work ostensibly as a means of protection. These bans often occur in the wake of a publicized case of abuse involving a migrant domestic worker or as “political leverage” to negotiate employment conditions with destination country governments (Napier-Moore, 2017). However, bans are often ineffective in reducing the number of emigrants or gaining the ears of destination country governments, who pay little heed and continue facilitating the entrance of migrant women. Furthermore, bans are not just ineffective but are often counterproductive in protecting women because prospective migrants bear the costs of bribing immigration officials at exit as well as the risks of emigrating through irregular channels (Napier-Moore, 2017).

Overall, the weaker bargaining power of origin governments in relation to destination governments circumscribes the ability of the former to extend social protections to migrant domestic workers across borders (Yeoh et al., 2004). When destination countries like Singapore are unwilling to enter into binding bilateral labor agreements for regulating migration and recruitment with origin governments, the scope of formal social protection emanating unilaterally from origin states is rather limited. Consequently, the resource environment that constitutes transnational social protection for migrant domestic workers shrinks, and the receiving state weighs heavily in determining domestic workers’ access to social protection. Under these circumstances, the provisions and regulations at destination become most consequential for migrant domestic workers.

The Reach of Labor, Immigration, and Criminal Law in Singapore

Imbued with the power to confer rights and protections on subjects, to dictate the organization and activity of collectives, and to define their experiences and welfare, the law is a powerful institution that undergirds the subject positions of migrant women (Yeoh et al., 2004). Given that Singapore has not signed most of the international conventions relevant to migrant domestic work or acceded to bilateral agreements with origin governments, the country’s domestic legislation becomes most consequential for the social protection of migrant women. We provide a summary mapping of the immigration, labor, and criminal law at whose intersection the fates and fortunes of migrant domestic workers substantively depend before turning our attention to recent legal developments and landmark cases to consider the reach of the law.

A migrant domestic worker enters Singapore on a work permit, which is bound to a 2-year employment contract, and tied to a specific employer and address. Additionally, a work permit may be cancelled at the will of employers and can only be renewed by the employer or agent. This renders migrant bodies deportable (De Genova 2002), bolsters the power that employers can exercise over a migrant’s sociolegal status (Rosewarne, 2010), and constitutes a migrant workforce that is flexible, controllable, and replaceable.

Migrant domestic workers are excluded from Singapore’s Employment Act—which stipulates limits on working hours, number of rest days, entitlements to annual and sick leave, and reasonable notice periods (Humanitarian Organization for Migration Economics [HOME] and Transient Workers Count Too [TWC2], 2017)—because the government finds it “[im]practical to regulate specific aspects of domestic work” (MOM, 2018a). Instead, work permit conditions as well as the punishments for violating them are legislated under the Employment of Foreign Manpower Act (EFMA), last updated in 2012, which offers a less robust set of welfare provisions compared with the Employment Act. The EFMA delineates the responsibilities of employers in relation to migrant women, stipulates prohibitions against ill-treatment, and codifies the safety procedures around the cleaning of windows in high-rise apartments. The language of the act, which requires employers to provide “adequate food and medical treatment,” “adequate rest daily,” “acceptable accommodation,” and “reasonable notice of repatriation,” tends to be ambiguous and limits its efficacy in safeguarding the welfare of migrant women (Wee et al., 2018).

Due to the “nature” of live-in domestic work, the Singapore government has adopted a hands-off approach to governing the employment conditions of migrant women, preferring to leave them to market forces: that is, the private negotiations between employer and migrant worker as facilitated by the recruitment and placement agent. Rest days, placement loans, notice periods, and compensation-in-lieu are stipulated in an employment contract between employer and migrant worker, as well as a service agreement between employer and agent. Although the Association of Employment Agencies (Singapore) and the consumer protection body CaseTrust jointly issued model contracts in 2006, employers are not required by law to sign an employment contract with migrant domestic workers.

Scholars have observed that these labor migration regulations may compound the vulnerabilities migrants face when performing live-in domestic work (Devasahayam, 2010). For instance, the employment contract that prescribes “mutual agreement” between employer and migrant domestic worker ignores the asymmetry in power relations, the latter’s unfamiliarity with local legislations, and the brutal reality that migrants are often expected to be subservient to their employers (Yeoh et al., 2004). Although the Employment Agencies Act caps the amount that migrant domestic workers pay their Singapore agents in fees at 1 month of their salary per year of their employment contract, the Singapore government has abstained from regulating the recruitment fees that migrant women pay to the recruiters at origin by insisting that these are private loans (Kaur, 2007). In the 2013 survey of Indonesian migrant domestic workers earlier referred to (Platt et al., 2013), 90% of respondents surveyed did not pay any recruitment fees to a migration intermediary prior to their arrival in Singapore, but paid them via monthly salary deductions. The loans they undertook were substantial: 64% took between 7 and 9 months to complete their salary deduction period. During this time, approximately two thirds of the respondents received S$10 or less per month, while 8% of the respondents received no salary at all. Coupled with the additional fees of seeking a transfer of employer, the obligation to pay off onerous loans and their lack of access to capital often pressures migrant women to remain in employment relationships even if these have become difficult or turned abusive.

While Singapore’s labor and immigration laws have not accorded full protection to migrant domestic workers, its criminal law has been enhanced to punish and deter the abusive behavior of employers. The Penal Code was amended in 1998 to include a section for increasing the penalties for errant employers who abused their migrant domestic workers. Under Section 73, employers, or a member of the household, found guilty could be penalized up to 1.5 times the maximum limit on fines, imprisonment, and caning of a particular charge. The then-Minister for Home Affairs, explained that the heavier penalties are “intended to send a strong signal to those employers who have a tendency to abuse their domestic maids that we [the government] take a very stern view against such abuse” (Yeoh et al., 2004, p. 14). In 2007, newer amendments to the Penal Code have further enhanced penalties for employers who voluntarily cause hurt (Section 323). Chief Justice Sundaresh Menon explained that the special provisions arise from the recognition that live-in migrant domestic workers form a class of victims particularly vulnerable to abuse. The latest round of public consultations by the Penal Code Review Committee in August 2018 proposed increasing the penalty of an offending employer or agent up to two times the maximum charge (Ministry of Law, 2018).

Although the statistics on the abuse of migrant domestic workers in Singapore remain patchy, the national newspapers have tracked the number of abuse cases filed over the years, noting that the number of cases filed with MOM fell from 157 in 1997 to 53 in 2008 (Pereira, 2005; TWC2, 2011). However, the number of cases filed with State Courts doubled from 14 in 2012 to 26 in 2015, bringing the total number of cases between 2011 and 2015 to 90 (Seow, 2015). The Judge of Appeal V. K. Rajah noted that at the level of the courts, “the number of substantiated domestic maid abuse cases, in which employers inflicted bodily hurt on the maid, has not meaningfully decreased over the years [ . . . ] despite the introduction of enhanced penalties.” While there were eight such cases dealt with in 2001, the number of cases rose to 26 in 2004 and 19 in 2008. It should be noted that the cases filed and processed by the police and courts comprise the most severe abuses and only a subset of the number of calls handled by embassies and civil society organizations. The Counsellor for the Protection of Indonesian Citizens in Singapore reported that the embassy handled 1,579 cases of abuse, “including salaries not being paid for months and even years, sexual and/or physical abuse, migrant domestic workers working up to 19 to 20 hours a day and the restriction of religious prayers” in 2017, and received 571 domestic workers seeking help from January to May 2018 (Parlambang, 2018). HOME reported a total of 872 cases and an average of 17 per week from April 1, 2017, to March 31, 2018 (HOME, 2018).

Most recently, in March 2018, the High Court established a new sentencing framework to account for the psychological abuse of migrant domestic workers. Although the High Courts had long acknowledged the significance of psychological abuse (Huang & Yeoh, 2007), the revised sentencing framework adjusted punishments upward according to the degree of physical as well as psychological harm in order to rectify the underemphasis of the latter in prosecuting cases of abuse. The High Court has thus extended its sentencing powers to take into account that the psychological distress of abuse can be as acute and permanent as physical harm (Tay Wee Kiat v. PP, 2018).

The landmark case during which this sentencing framework was introduced was the latest in a series of cases manifesting “blatant displays of violence, cruelty, and humiliating and demeaning (even dehumanizing) behavior” (Tay Wee Kiat v. PP, 2018). A Singapore employer, Tay Wee Kiat, was convicted on 10 charges, among others, of voluntarily causing hurt to Fitriyah, a 33-year-old Indonesian woman who worked in his household as a domestic worker from December 2010 to December 2012. Criminal investigations only began when another migrant domestic worker, Moe Moe Than, reported she had been abused when she was leaving Singapore. The abusive incidents that occurred over the course of 2 years included “slapping her, hitting her head with a bundle of three canes tied together, [ . . . ] pushing her head against a cabinet.” Tay also “made the victim stand on a plastic stool on one leg and hold up another plastic stool with one hand, while he pushed an empty plastic bottle into her mouth (‘the Stool Incident’),” “made the victim and Moe Moe Than kneel and get up before a Buddhist altar in the home 100 times and then slap each other 10 times (‘the Prayer Incident’)” although the victim is Muslim and Moe Moe Than Christian, and “made the victim and Moe Moe Than assume a push-up position and then kicked the victim on her left wrist (“the Push-Up Incident”).” Applying the new sentencing framework to Tay, the judges denounced the humiliation, degradation, bullying, cruelty, and sadism of his behavior before increasing his sentence from 28 to 43 months of imprisonment (Tay Wee Kiat v. PP, 2018).

The prosecution of employers also calls into question the responsibility of “mute bystanders,” particularly family members who were involved in, if not aware of, the ongoing abuse (Yeoh et al., 2004). Chia Yun Ling, the wife of Tay, was convicted on two charges for two physically abusive incidents, slapping Fitriyah in the face and punching her forehead. Since Chia’s offences did not form a prolonged pattern of abuse, the judges ruled that she was not as culpable as Tay and kept her sentence the same before trial at 2 months of imprisonment.

Depending on the relevant charge at hand, however, the culpability of the couple may be shared equally between the “active perpetrator” and the “abettor.” Over a 15-month period, husband-and-wife pair Lim Choon Hong (the employer) and Chong Sui Foon starved their domestic worker who ended up losing 40% of her weight, turning the scales at 29 kg when the abuse came to light. The legal responsibility the first respondent bore for the victim made him an “abettor” instead of a mere bystander in their migrant domestic worker’s state of food deprivation (PP v. Lim Choon Hong, 2017).

Despite the manifold legal provisions in Singapore for sentencing employers found guilty of abusing their migrant domestic workers, reaching the safe haven of the law is no small feat. In order for an investigation into abuse to even commence, migrant women need to seek help or lodge a complaint against their employers or agents. However, the challenges with requiring migrant women to lodge complaints are multiple: migrant women may have a limited awareness of their rights and entitlements and limited language proficiency, face threats of deportation or further abuse, and fear being discredited (Dodgson, 2016). As the courts have recognized, “many maids come from a background which carries with it a natural fear that the police are working for the rich, and are reluctant to seek their protection even when the opportunity presents itself” (Lee, 2002). To lodge a complaint is to put their livelihoods at risk, as errant employers may deny the right of their migrant domestic worker to seek a transfer of employer, repatriate her suddenly, or lodge an unverified complaint in MOM’s “feedback system” (HOME & TWC2, 2017). By the time migrant women decide to lodge a complaint or manage to seek help, repeated occasions of abuse may have already occurred. This illustrates how the clusters of social protection available to migrants are often interrelated: strong informal social connections, such as supportive friends or family in Singapore who can direct victims of abuse to necessary resources, as well as the assistance of a third sector to render help throughout the reporting process may be the preconditions for migrant domestic workers’ access to the legal justice and social protection proffered by the law. In the context of Singapore, where the migration regime renders low-waged migrant women as easily deportable subjects, more transnational forms of social protection are limited, and women’s recourse to justice may reside in localized networks or institutions bounded by the nation–state.

Even after migrant women report abuses, many cases reported to the police, agencies, and MOM are settled out of court (Yeoh et al., 2004). Perhaps because of the problem of evidence, case workers at HOME observe that issues related to “verbal abuse, denial of rest days, poor living conditions, excessive working hours, intrusive employer surveillance, and the confiscation of phones or unreasonable restriction of mobile phones are generally not viewed as ‘valid claims’” that migrant women who file such complaints will likely be repatriated (HOME & TWC2, 2017, p. 11). HOME’s case manager observes that MOM only prosecutes the “most egregious” cases of food deprivation (Zaccheus, 2017) when up to 8 in 10 migrant domestic workers who sought help with the organization did not get enough food (HOME & TWC2, 2017). Migrant women have also had to remain in the country for several months to over a year to aid in investigations, which may deter them from seeing the investigation to completion.

Civil Society Action

Over the past two decades, the concern about persistent abuses migrant domestic workers face has galvanized Singapore’s civil society action and the formation of migrant organizations to fill gaps in services, advocate for the rights of migrants, and improve their social protection (Yeoh & Annadhurai, 2008; Yeoh et al., 2015). As scholars have noted, civil society performs a particularly important mediating role when migrant women are unable to negotiate with their employers or the state and when power relations are highly unequal (Koh et al., 2017a).

Migrant organizations in Singapore tend to be migrant-serving, where locals and other long-term residents represent migrant interests, rather than migrant-led, where migrants are at the vanguard of advocacy work (Yeoh & Annadhurai, 2008). First, the Registration of Societies Act stipulates that the majority of Committee Members of “any society whose object, purpose or activity [ . . . ] is to represent persons who advocate, promote, or discuss any issue relating to any civil or political right” must be Singapore citizens (Ministry of Home Affairs, 2018). Considering that public assemblies without a formal permit are also illegal (Yeoh et al., 2015), migrant women as noncitizens face tremendous difficulty organizing themselves into a formal collective in Singapore. Second, due to their precarious social and legal status, migrant women can be easily repatriated by employers or the authorities if they are perceived as causing trouble. As a migrant rights activist from HOME explains, “You make too much noise, you get sent back.” Third, the host government often has little incentive to engage with the concerns of migrants who are excluded from the electorate and whose presence is considered temporary. For these reasons, migrant-serving rather than migrant-led organizations prevail in Singapore.

Among the migrant-serving organizations, there is a larger number of service-oriented organizations compared with advocacy-oriented organizations. Civil society in Singapore has been characterized as an illiberal political terrain with limited room for activism (Koh et al., 2017a), particularly in the case of migrant worker-related activism (Piper, 2006; Yeoh & Annadhurai, 2008). Civil society is expected to seek incremental and reformist changes palatable to incumbent ideologies rather than challenge the regime’s fundamental premises (B. H. Chua, 2000). As such, service-oriented migrant organizations—which operate social spaces for migrant women and equip them with skills such as cooking and financial literacy (Yeoh & Annadhurai, 2008)—form the larger part of the migrant civil society landscape.

Advocacy-oriented organizations, on the other hand, express standpoints that may compete with government policy, and seek to plug gaps in social provision through influencing policy agenda and shaping public opinion (Yeoh & Annadhurai, 2008). The entrance of migrant advocacy organizations has been fairly recent in Singapore’s history. Following the death of a 19-year-old Indonesian domestic worker Muawanatul Chasanah in 2002, The Working Committee 2, which eventually became TWC2, brought migrants’ rights issues to the forefront of public consciousness (Gee & Ho, 2006; TWC2, 2018). HOME, another prominent migrant worker advocacy organization, branched out from the Archdiocesan Commission for the Pastoral Care of Migrants and Itinerant People, and can be considered both service- and advocacy-oriented. In relation to tackling abuse, HOME operates a helpdesk for migrant domestic workers seeking assistance and recourse and a shelter for vulnerable migrant domestic workers who have suffered abuse and exploitation and are seeking legal resolution. Simultaneously, these organizations engage state agencies in closed-door dialogue to resolve specific abuse cases and pressure government officials to enact policy reform (Yeoh & Annadhurai, 2008). Increasingly, both TWC2 and HOME acknowledge that service provision is crucial to their work in advocacy; the outcomes in casework form the basis of how advocates frame proposals for policy change, and effectiveness in long-term change depends on the immediate, short-term work of aiding migrant women.

A significant advancement in social protection for migrant domestic workers over the past decade has been civil society’s campaign for a rest day. In 2012, the MOM announced that the EFMA would be amended to enshrine migrant domestic workers’ entitlement to a weekly rest day. The importance of having a weekly day off cannot be overstated for migrant domestic workers; Jolovan Wham, one of the frontrunners in the joint civil society campaign for a mandatory day off beginning in 2008, explained that there are numerous resources available for migrant domestic workers, but the trouble is often being able to access them. As mentioned earlier, Singapore, like many other migrant destination countries, has been (and continues to be) reluctant to ratify international and regional conventions that protect migrant rights (Yee, 2009). Particularly in Singapore, where human rights are construed as a transnational “Western-centric” discourse, the notion of human rights is often insufficient in persuading states to extend social protections to migrants (Basok, 2009; Yeoh et al., 2015); in fact, advancing it in advocacy work may spark an irascible state response and render the cause unpopular among the general public. In this light, the work of civil society activists in vernacularization, or the adapting of human rights discourses for local communities, was crucial in pressuring the state to extend a weekly rest day to migrant domestic workers (Koh et al., 2017a; Levitt & Merry, 2009; Merry, 2006). The transnational social protection open to domestic workers in terms of Singapore’s third sector is thus circumscribed, as migrant advocacy organizations must carefully manoeuvre away from any explicit identification with cross-border advocacy networks in order to succeed, and migrants themselves, as mentioned earlier, are precluded from formal organization.

The impact of the rest day legislation, however, has not reached its full potential. Since the legislation includes provisions that allow employers to financially compensate migrant women for working in lieu of a “mutually agreed upon” number of rest days a month, the mandatory rest day law can in effect be undermined. One TWC2 (2015) study estimated that only 41% of migrant domestic workers had a weekly day off, and a second (Paul 2018) found that 44% of migrant women who arrived after 2014 did not have a rest day during the first few months of repaying their loans. The underspecification of what a rest day entails in terms of hours also means that some employers expect migrant women to complete chores before they leave or after they return (HOME & TWC2, 2017). Nonetheless, these statistics point to an estimated 10% more women having access to rest days compared with before the legislation was amended. Importantly, it provides a tangible signal to placement agents and employers that giving migrant domestic workers a weekly day off is a recognized norm and not a matter of employers’ discretion.

Conclusion

As Levitt et al. (2017) argue, social provision for migrants emanating from host nation–states is often limited, and migrants must themselves patch together informal sources of transnational social protection to meet their needs. Comprising the market, the state, the third sector, and the characteristics of individual migrants and their networks, people on the move are embedded in resource environments that transcend nation–state boundaries, and draw on various resources for social protection. Yet in Singapore, the ability of countries of origin as well as the influence of the third sector in promoting better work conditions for migrants remains circumscribed. As Marti (2019, p. 1346) has argued, Singapore has not only “resisted the diffusion of norms and initiatives regarding migrant workers developed at the international and regional levels into the national level,” but also disabled “the ‘multi-layering’ of the governance of [migrant domestic workers] within the nation state” by constraining the spaces available for civil society action. The resource environment available to migrant domestic workers, therefore, is one where the power of the receiving state is paramount, and transnational options shrink as a result.

In this context, adopting an expansive, transnational view of social protection needs to be accompanied by an understanding that the resource environment that social protection provisioning depends on is deeply structured by a hierarchical distribution of power and responsibility. For migrant domestic workers in Singapore, formal social protections in the vehicle of the instruments of international human rights and labor standards or via the discourses of regional integration essentially stop at the border. Once in Singapore, migrant domestic workers are expected to cope with precarious work conditions and unequal power relations vis-à-vis employers, agents, and the state, with mainly the safety net of the law to protect themselves from the worst forms of abuse. While the legal system has been constantly refreshed over the past decade to enhance its deterrence value (such as by increasing penalties specific to crimes against migrant domestic workers) and ensure its continuing relevance (such as giving attention to psychological distress and bystander responsibility), this system of protection is not easy to reach.

Beyond legal protection from abuse, given a highly constrained resource environment, the case for advancing social protection for migrant domestic workers depend on whether migrant advocacy groups are able to negotiate for incremental change (such as the day off legislation) within the fraught terrain of shifting state–civil society relations (Yeoh & Annadhurai, 2008). As L. J. Chua (2012, p. 722) put it, this is an exercise in “pragmatic resistance,” where “the goal is to stay alive and advance with skirmishes, rather than court demise with open warfare declared on grander principles.” While significant work has been accomplished with regard to strengthening the remit of legal protection for migrant domestic workers facing egregious conditions of abuse, noteworthy barriers stand in the way of mounting any fundamental challenge to the way that migrant labor is organized in Singapore. As this study demonstrates, the host society treats migrant domestic workers as transient labor with the right to freedom from abuse in the workplace, but stops short of construing them as real workers for whom it bears responsibility for in relation to long-term social provision.