Abstract
In the context of Asia, understanding migration governance needs to transcend statism to encompass the ‘middle space’ of migration. Unlike migration linked to settlement in liberal democratic states of the West, a large part of low-skilled migration in Asia – predominantly circular, feminized, and contractual—is brokered by private recruitment agencies. In adopting migration brokers as a methodological starting point, we make the case for bringing the migration industry into the fold of global migration governance analysis. Based on interviews with employment agencies deploying Indonesian domestic workers to Singapore from 2015 to 2016, we argue that migrant-destination states in Asia devolve responsibility for workers to the migration industry to order migration flows and circumvent formal cooperation with origin countries. Comprehending migration governance in Asia requires grappling with the co-constitutive governance of the state and migration industry and its interdependent dynamics, which we illuminate through the theory of strategic action fields.
1 Introduction
In tandem with the feminization of labor migration in Asia, large numbers of women migrate to be employed as domestic workers in the region and beyond (Kaur, 2007). Key to their mobilities within Asia is the migration industry, a range of brokers from licensed recruitment agencies to informal recruiters, who move, match, and place domestic workers with employers. Despite their importance in migration governance in Asia, the migration industry has been relatively neglected in the global migration governance literature.Um, there are ex-workers in the pool that we still have. Luckily my pool, I still got a lot. There are some agencies which get hit. This is their problem. But to us, I'm fine, ok. This is business. The more messy it is, it's better for us. If it is so straightforward, like air ticket, all the agents will die.
–Singapore placement agent
When asked about how his business fared when the Indonesian government was rolling out a new initiative,1
The Kredit Usaha Rakyat Tenaga Kerja Indonesia (KURTKI) scheme that grants prospective migrant domestic workers from Indonesia low-interest loans from five government-appointed banks encountered teething issues in its initial implementation.
This short vignette provides a glimpse into the topic of this article: the contested dynamics of the state and employment agencies in governing the migration of domestic workers to Singapore. Moving away from the methodological nationalism of global migration governance and the vilification of brokers, we ask what the role of the migration industry is in governing migration. Despite a growing recognition that private actors such as brokers and firms produce orderly flows alongside the state, these accounts belie the contestations and strategizing that governing the migration industry and governing through it entails. As such, we draw on the theory of strategic action fields to unveil the ongoing negotiations within the migration industry as well as with the state. Conceiving of the migration industry and state as fields nestled within a broader field of migration governance guards us against overstating the power of the state as well as understating its responsibility.
2 Global migration governance
Notwithstanding the multiple genealogies, connotations, and uses of ‘global governance’ across various fields (Rose, 1999; Grugel and Piper, 2007; Betts, 2008), global migration governance can be broadly defined as the norms, regimes, and organizations that regulate states’ responses to international migration (Betts, 2011) that is regarded as both ‘necessity and impossibility’ (Newland, 2010). States are urged to coordinate their migration governance efforts to facilitate ‘orderly migration’, protect the well-being of migrants, prevent a ‘race to the bottom’, and ensure that the interests of origin and destination states are met (Ghosh, 2000; Hamada, 2011; Hollifield, 2012; Abella, 2013; Woods et al., 2013). However, compared to regulated flows of goods and capital, global migration governance has broadly been characterized as ‘a missing regime’, ‘fragmented’, and ‘piecemeal’ (Hollifield, 2000; Betts, 2008; Ghosh, 2008; Grugel and Piper, 2011; Lavenex and Panizzon, 2013; Castles, 2014; Pellerin, 2014). Even with the involvement of a normative institution like the International Labor Organization (ILO) and an intergovernmental organization like the International Organization for Migration (IOM), migration lacks an overarching multilateral regulatory framework as well as a single authoritative body, reflecting the general reluctance of states to relinquish ‘sovereignty’, or control over their national borders, to a supranational authority (Hollifield, 1992; Newland, 2010; Hamada, 2011; Abella, 2013; Pellerin, 2014).
Betts (2011) identifies five layers of global migration governance: formal multilateralism, ‘embedded governance’ in migration-related issue areas, regional coalitions and dialogues, bilateral arrangements, and unilateral migration policies. To states, regional agreements, policy networks of government bureaucrats, and dialogue platforms that are informal, network-based, and non-binding, often serve as palatable alternatives to formal multilateralism (Martin, 2008; Betts, 2010; Newland, 2010). These layers of global migration governance are unevenly distributed over issue areas; other than the international refugee regime, which is the most formalized of the migration regimes (Betts, 2010; Koslowski, 2011), other areas of migration governance are deemed to have weak or missing regimes. Adopting a liberal approach, scholars have applied a public goods framework to make sense of uneven constellations of global mobility regimes across issue areas (Rai, 2004; Betts, 2008; Hollifield, 2011; Koslowski, 2011; Woods et al., 2013). For instance, since the benefits of refugee governance are non-excludable and non-rival, refugee governance approximates a global public good, which explains the institutionalization of the refugee regime as mentioned earlier. In contrast, since the benefits of labor migration governance accrue selectively to the states of origin and destination, it is a non-public good in which regional and bilateral levels of governance prevail. Given the disparate interests and the power asymmetry between origin and destination states, which tend to be policy ‘takers’ and ‘makers’, respectively (Betts, 2011; Hugo, 2013; Woods et al., 2013), the current labor migration regime is ineffective if not absent (Koslowski, 2011; Hollifield, 2012). Besides the lack of consensus amongst states, immigration also tends to be a divisive public issue, which further hinders the construction of an effective labor migration regime (Hollifield, 2004; Geddes et al., 2012; Martin, 2015).
Harmonizing migration governance, however, is far from a magic bullet. Institutional gaps as well as implementation problems hamper the effectiveness of migration governance efforts (Koser, 2010; Lavenex et al., 2016). Piper et al. (2016) contend that we have not grasped how global regulations translate into national policies, especially with the limited institutional capacities and the weak political will of resource-poor countries. Scholarship on global migration governance has also been criticized for its methodological nationalism, though the remedy has been to turn the spotlight on civil society and transnational activist networks that contribute to governance roles where states have failed (Grugel and Piper, 2011; Rother, 2013; Schierup et al., 2015).
2.1 The role of the migration industry in migration governance
Refraining from older theories of International Relations that fixate on states and formal institutions, we treat ‘global governance’ as the outcomes of interactions and interdependence among a range of political actors with transnational implications (Rose, 1999; Bevir and Hall, 2011). Geiger (2013) argues that the state has never been the sole actor in moving people across borders, even though the view that only ‘states’ can effectively regulate mobility still prevails, presumably because states have monopolized the authority over mobility (Torpey, 1998). Therefore, we make the case for focusing on the ‘middle space’ of migration (Lindquist et al., 2012) for a fuller grasp of labor migration governance. This means attending to the recruitment practices of brokers, ranging from village-level recruiters to licensed multinational agencies in varying degrees of proximity to the state and migrants (Lindquist, 2010, 2012; Lindquist et al., 2012; Farbenblum, 2017). Several labels and units of analysis have emerged over the years, such as: ‘merchants of labour’ (Martin, 2005), ‘migration industry’ (Hernandez-Leon, 2005; Sorensen and Gammeltoft-Hansen, 2013), ‘migration infrastructure’ (Lindquist et al., 2012; Xiang and Lindquist, 2014), and ‘irregular migration control market’ (Lopez-Sala and Godenau, 2016), each with slightly different connotations and a different range of actors. We eventually use ‘migration industry’ in our discussion of licensed private employment agents in Singapore, but first preface the use of the term by examining the broader terrain of arguments about the role of private actors in migration governance.
Private actors in the migration industry are taking on larger roles in migration governance. Some of the most conspicuous and provocative instances of this are states in the West, outsourcing migration control (such as the detainment and the deportation of migrants and the guarding of borders) to private industries for the purported neoliberal gains of shifting costs, strengthening control, and avoiding blame (Gammeltoft-Hansen, 2013; Menz, 2013; Lopez-Sala and Godenau, 2016). While these examples of the migration control industry may be contrasted with the facilitation industry which moves people across borders, this distinction is disrupted when we examine the work of the private employment agents in Singapore. Licensed and law-abiding employment agencies both facilitate and control legal migration, and by doing so blur any easy distinctions between state and market (Lindquist et al., 2012; Gammeltoft-Hansen, 2013; Lopez-Sala and Godenau, 2016). Despite this, states commonly erect a strict boundary between themselves and the migration industry, which becomes a detached lightning rod of blame and criticism (Xiang, 2012; Gammeltoft-Hansen, 2013; Surak, 2013b).
It comes as no surprise that the state and migration industry exist in varied and complex configurations. Sorensen and Gammeltoft-Hansen (2013) argue that the ‘statist structures such as immigration policies, labor market regulation, visa requirements, border control etc. almost always remain an essential backdrop for understanding how these migration industry actors emerge and function’. Surak (2013b) demurs that the state is only a backdrop and argues that East Asian developmental states act variously as platform, principal, or piggybacker in relation to the migration industry, both operating together in ‘symbiotic alliance’. In a similar vein, Tseng and Wang (2013) examine how the Taiwanese state depended on economists and demographers to create a rationale for a guest worker scheme, and continues to depend on employers and brokers to surveil migrant guest workers and to enforce their return. Finally, Xiang (2012) makes a compelling case that brokers are crucial to a ‘complex structure of governance’ in China within which they produce legality and legibility. States capitalize on brokers as key mediators of capital, information, and the movement of migrants in order to govern these flows (Lindquist et al., 2012).
Without discounting the crucial insight that states and migration industry govern migration collectively, we contend that a focus solely on how the migration industry contributes to governance belies the contestations and strategizing around governing migration through brokers. A fuller account of the role of the migration industry in governance involves examining how it governs as well as how it is itself governed, both of which involve contention, resistance, and negotiation. Therefore, we apply Fligstein and McAdam’s (2011) theory of strategic action fields to illuminate the dynamics between the state and the migration industry in governing migration. Drawing on Bourdieu’s and institutional theories of fields as well as the social movement literature, they explain that a strategic action field is a ‘meso-level social order where actors (who can be individual or collective) interact with knowledge of one another under a set of common understandings about the purposes of the field, the relationships in the field (including who has power and why), and the field’s rules’ (p. 3). Strategic action is the endeavor by which social actors sustain stable social worlds through mobilizing other actors. Instead of treating the relationship between the state and the migration industry as established and unchanging, or the migration industry as an unproblematic extension of state apparatus, the theory of strategic action fields illuminates that conflict and change are endemic to reproducing the order of a field, whether stable, crisis-ridden, or moving between both ends of a continuum (Fligstein and McAdam, 2011). Actors in the field are constantly locked in contest: they make moves that other actors then have to construe and respond to within the constraints of a range of possible actions. The migration industry and the state can thus be conceived as separate fields, the former subordinate to the latter, as well as nested within a broader field of migration governance. We now turn to the migration governance efforts in Asia before zooming in on the case of domestic worker migration to Singapore.
3 Migration governance in Asia
Unlike migration linked to settlement in the case of liberal democratic states in the West, a large part of low-skilled labor migration in Asia is circular, feminized, and governed by temporary employer-bound employment contracts (Martin, 2009; Kaur, 2010; Hugo, 2013; Piper et al., 2016). These jobs that low-skilled labor migrants perform are employer-driven and sector-specific like construction and manufacturing for men, and – in greater numbers – domestic work for women (Ahsan et al., 2014; Piper et al., 2016). While Europe transited from a guest worker system in the 1950s and 1960s to allow settlement and family reunification, Asia is unlikely to follow suit (Battistella, 2002; Surak, 2013a). Instead of the state-administering guest worker programs as in the case of Europe, recruitment has been left to private recruitment agencies and an ecosystem of intermediaries in Asia, as has been the norm for the past few decades (Agunias, 2009; Martin, 2009; Kaur, 2010; Lindquist et al., 2012; Hugo, 2013; Ahsan et al., 2014). Beginning with the oil crisis in the 1970s that initiated substantial labor migration to the Gulf States, the numbers of recruitment agencies and their sub-agents across Asia have soared exponentially, mooring regimes of transnational circular migration (Lindquist et al., 2012; Xiang, 2013; Ahsan et al., 2014; Jones, 2015). The burgeoning numbers of brokers have been associated with the tightening of migration controls as migrants come to depend on brokers to navigate the labyrinth of paperwork from proliferating government regulations (Lindquist et al., 2012).
Many have been quick to point out the dangers of delegating the recruitment, transport, matching, and placement of low-skilled labor migrants to profit-oriented brokers (Kaur, 2010; Xiang, 2013). Profit-making recruiters are accused of manufacturing irregularity by bypassing state regulations, imposing onerous debts on migrants that lead to ‘debt bondage’, deceiving migrants about the terms of employment, and inflicting emotional or physical violence on migrants. Extreme cases of exploitation shade into categories of trafficking and forced labor, though less provocative instances of exploitation like withholding passports are supposedly the norm in Asia (Jones, 2015). Piper et al. (2016) lay the blame for these migrants’ precarity on the governments of origin and destination countries for abetting the illicit operations of the migration industry and under-enforcing the rights of migrants. However, scholars problematize the ‘a priori vilification’ of brokers who straddle the Manichean dichotomies of legality/illegality, formality/informality, profit-orientation/altruism (Lindquist et al., 2012), and encourage agnosticism in considering the ethics of a broker (Lindquist, 2015). The ample diversity of actors within the migration industry and settings they traverse guards us against typecasting brokers and compels us to pay close attention to the empirical context of brokerage.
In recognition of the tremendous power at the hands of the migration industry to help and to harm migrants, regulating recruitment has risen on the agendas of international organizations like ILO, IOM, and the Asian Development Bank Institute (ADBI) (Wickramasekara and Baruah, 2017). The ILO transited from advocating for the abolition of profit-making private recruitment agencies to regulating them through the Convention concerning Private Employment Agencies since 1997 (Agunias, 2009; Farbenblum, 2017). More recently, these international organizations have advocated licensing, regulating, and incentivizing ethical recruitment in the industry (Gordon, 2015; Jones, 2015; Tayah, 2016). Governing the migration industry of recruitment agencies and their sub-agents, however, is notoriously complex, given the opacity of cross-border transactions and the challenges of enforcing extraterritorial regulations (Wise, 2013; Pittman, 2015; Farbenblum, 2017). Further, many countries of origin in Asia have not ratified key ILO instruments and United Nations (UN) human rights treaties relevant to recruitment, not to mention the countries of destination. The Association of Southeast Asian Nations (ASEAN) has expressed interest in improving recruitment processes, but lacks the teeth of binding regulations and leverage over destination states. As discussed earlier, bilateral labor agreements are more amenable to states than multilateral cooperation, varying from formal government-to-government (G2G) agreements to less formal Memoranda of Understanding (MoU), stipulating the terms of recruitment, remittances, and return. Origin countries such as Philippines and Bangladesh, as well as destination countries like the Republic of Korea, Malaysia, and Thailand have actively forged bilateral agreements with partner states (Jones, 2015). However, destination countries like Singapore are unwilling to sign G2G agreements for migration and recruitment regulation, and so depend on regulating the market for low-skilled labour migrants such as domestic workers (Devasahayam, 2010; Kaur, 2010).
Of the labor migration flows within Asia, we suggest that the case of Indonesian domestic workers in Singapore deserves further scrutiny. Along with cities such as Taiwan and Hong Kong, Singapore occupies a particular niche within the Asia-Pacific region as a relatively wealthy and extensively governed country of destination for Indonesian migrant domestic workers. At a glance, the governance mechanisms in these oft-compared cities are fairly similar, calibrating the number of incoming migrant workers to specific labor market niches through finessed guest worker schemes (Lu, 2011; Surak, 2013b). However, the relative market positions of all three destinations fall into a more obvious hierarchy when the labor conditions of migrant domestic work are taken into account. The most desirable labor destination of the three is Hong Kong where labor laws stipulate a minimum wage for migrant domestic workers, mandate a weekly rest day, and enshrine their right to unionize (Bell and Piper, 2005; Tayah, 2016). Compared to Taiwan where wages are similar to Hong Kong, Singapore offers lower wages and lacks regulated days off (Basu, 2012). Singapore usually serves as a ‘stepping stone’ or ‘training ground’ for women ultimately aiming for better salaries in Hong Kong or Taiwan and elsewhere (Paul, 2017). As such, its precarious labor market positioning vis-à-vis competing destinations persistently raises concerns within the Singapore government and its citizenry if it is adequately attracting qualified migrant domestic workers to its shores.
These destination states rely heavily on Indonesian domestic workers, who comprise 47% of Hong Kong’s migrant domestic work force (Asian Migrant Centre, 2015) and 79% of Taiwan’s migrant caregivers (Chou et al., 2017). While official statistics by nationality are not readily available for Singapore, it is often estimated that more than half of the migrant domestic workers in the nation-state are from Indonesia. Geopolitically, Indonesia is often perceived as the tacit leader within the ‘leader-less’ ‘ASEAN Way’, although its recent approach has been to wield ‘soft power’ rather than to act aggressively (Rattanasevee, 2014). In comparison to the Philippines – a competing ‘labour brokerage state’ (Rodriguez, 2010) – Indonesia’s migration infrastructure is vast and sprawling, its uneven administration of regulations compounded by contestations among government authorities (Palmer, 2016). Lindquist (2012) argues that the centralization of migration control along with the fragmentation of labor recruitment opens up a space for informal brokers who traverse the boundaries between legality and illegality when navigating complex bureaucratic processes.
In light of Indonesia’s position as a significant country of origin and an ASEAN heavyweight, Singapore’s relatively precarious perch within Asia’s migration industry, proximity to, and common membership in ASEAN increase the stakes of a diplomatic relationship with Indonesia, which we suggest Singapore establishes through the lubrication of the migration industry. In scrutinizing the case of Singapore’s migration governance of domestic workers, we hope to illuminate the dynamics of the state devolving governance to a governed migration industry that renders formal cooperation obsolete, and conversely, how the lack of formal cooperation spurs the migration industry to fill the gaps in governance. Key to our argument is the theory of strategic action fields that is attentive to contestation and conflict in the production and reproduction of an ordered migration governance.
4 Methodology
Besides semi-structured in-depth interviews with employment agents that place and recruit domestic workers (n = 28), we also interviewed key stakeholders (n = 6) in the domestic worker industry such as the industry association, non-governmental organizations (NGOs), embassy personnel, and representatives from the MOM from June 2015 to August 2016. Employment agencies were recruited through a mixture of purposive stratified and snowball sampling techniques, based on their retention rate, placement volume, and type of accreditation. We use interviews with MOM officers to triangulate and inform our understanding of their rationale behind governance measures and Singapore Parliament Records as a window into legislative considerations as well as a longer history of the industry.
Part of the fieldwork also took place in Indonesia in November 2015 where interviews were conducted with key stakeholders (n = 13). These include recruitment company owners and managers, training center managers, ex-domestic workers employed as administrative staff in recruitment companies, brokers at various levels in Indonesia’s migration infrastructure, and NGO representatives.
5 Regulatory frameworks governing domestic worker migration to Singapore
Singapore’s structural reliance on migrant domestic workers is well established (Huang and Yeoh, 2003; Yeoh, 2006). Beginning with Singapore’s export-oriented manufacturing in the 1960s and subsequent economic restructuring in the 1980s, higher rates of female labor force participation created a ‘care deficit’ and a gap in social reproduction. With the numbers of Cantonese amahs2 Amahs are a group of migrant women, hailing from China’s Guangdong province, who played crucial roles in Chinese and expatriate households as domestic workers, nannies, and cooks from the 1930s to the 1970s. They worked mostly in Singapore, Malaysia, and Hong Kong (Gaw, 1988).
As with many other countries in Asia, Singapore deploys a ‘revolving door’ policy that enshrines migrant domestic workers’ return at the end of their employment contracts. In June 2016, there were 237,100 migrant domestic workers in Singapore, the majority of whom came from the Philippines, Indonesia, and Myanmar (MOM, 2016a). These women are brought in on ‘Work Permits for Foreign Domestic Workers’ for two-year, employer-specific work contracts, subject to the employment relationship proceeding smoothly. For employers, the Work Permit conditions require them to put up a S$5000 security bond as a guarantee for adherence to the conditions of the Work Permit and buy personal and medical insurance for each worker, amongst other things. Abella and Kouba (2016) characterizes Singapore’s admission system as ‘trust the employer’ to denote the large role that employers play in migration governance. For workers, the Work Permit prohibits them from marrying Singaporean citizens and permanent residents without MOM’s approval and becoming pregnant over the duration of the Work Permit, as monitored through medical health screenings every half a year (MOM, 2016b). These Work Permit conditions, as well as the offences for violating them, are legislated under the Employment of Foreign Manpower Act (EFMA). The state demarcates migrant domestic work as an informal economy through boundary markers like excluding these workers from the Employment Act and declaring homes as private spaces that cannot be regulated (Kaur, 2007). In place of the Employment Act, the EFMA stipulates the entitlements and well-being of migrant domestic workers (MOM, 2016c).
5.1 Regulating the migration industry in Singapore
In comparison to the history of domestic work in Singapore, the history of the employment agency industry is less solidly documented. Extant work on recruitment practices of agencies in Singapore has mostly been derived from domestic workers’ accounts (Anggraeni, 2006; Platt et al., 2013). Wong (1996) traces the start of the employment agency to Singaporean male tourists, mostly small-time businessmen, who headed to Manila in the late seventies for sex tourism and brought women back to Singapore to work as domestic workers. These offenders who were fined were also encouraged to apply for an employment agency license, and became the forerunners of the employment agency industry. More ‘respectable’ recruiters entered the industry in the 1980s and sought prospective domestic workers from more diverse countries in conjunction with greater state regulation.
Continuing the expansion of the migration industry, there was a 30% surge in the number of licenses issued to Key Appointment Holders (KAH) of employment agencies from 2013 to 2016. Intensified by the spatial clustering of employment agencies with similar niches, a large majority of employment agencies engage in price competition. The landscape of employment agencies is highly polarized. In April 2017, 11 large employment agencies each placed upwards of 1,000 migrant domestic workers while 776 agencies formed a large majority of agencies that placed between 1 and 199 workers a year (MOM, 2017).
Employment agencies that recruit and place migrant domestic workers are regulated under the Employment Agencies Act. In 2010, the then Minister of State for Manpower explained the need for amending the Employment Agencies Act in context of a seven-fold increase in the number of employment agencies between 1984 and 2009, a surge in complaints to MOM about the malpractices of employment agencies, and a lack of distinction in the caliber of employment agencies (Singapore Parliament Reports, 2010). Subsequently, the Employment Agencies Act was amended in 2011 to weed out errant agencies and improve the professionalism of the industry, which entailed increasing penalties for malpractices, mandating knowledge of relevant laws for licensed employment agents, and moving accreditation conditions into legislation (Singapore Parliament Reports, 2011a). These changes entailed the following for employment agencies who place migrant domestic workers: employment agencies that place large volumes of workers are required to put up a security deposit of between S$40,000 and S$60,000; licensed employment agents need to provide ‘acceptable food and accommodation’ while migrant domestic workers are under their charge; a directory with indicators of the employment agency’s performance would be published on MOM’s website; and MOM would develop a new voluntary accreditation scheme in tandem with the industry (Singapore Parliament Reports, 2011b). The amendments to the Act reveal the preferred tools at MOM’s disposal for governing the industry, including key measures like licensing employment agencies, publishing statistics of the industry’s performance, penalizing aberrant behavior, and collaborating in the formulation of an accreditation scheme.
6 Discussion
6.1 Devolving the placement of migrant domestic workers to employment agents
The state depends on the migration industry to produce governable legal subjects. In order for a prospective Indonesian domestic worker to enter Singapore, an employment agent in Singapore usually has to work with an Indonesian counterpart to facilitate the worker’s journey. Figure 1 shows the brief stages of a typical recruitment pathway of a prospective domestic worker. Usually, an informal broker accompanies the prospective worker to a government center a distance away. Once she signs a letter of placement, completes a medical check-up certifying her health, and purchases insurance, she stays at a training center for 40 days while the staff at the training center prepares her ‘biodata’ and passport. While she is still at the training center, agents and employers in Singapore interview the worker over Skype. Once the worker is matched with an employer in Singapore, she is sent for pre-departure training before leaving for Singapore.
The typical recruitment pathway of an Indonesian domestic worker
As illustrated by the brief account of an Indonesian worker’s recruitment pathway, it is evident that the migration industries in countries of origin and destination play crucial roles in governing workers in working to both control and facilitate migration flows. First of all, intermediaries are essential to turning the prospective migrant into a ‘paper migrant’ through producing paperwork. Preparing documents alone is onerous for prospective workers in Indonesia (Patunru and Uddarojat, 2015). They require, amongst others, an identity card, a family card, a letter of consent from the family and village head, birth certificates, education certificates, a passport, insurance, and a medical certificate. Therefore, a vast majority of prospective migrant domestic workers depend on informal brokers to visit government offices long distances away and navigate the labyrinthine application processes, not to mention the subsequent chain of licensed intermediaries that prepare further documentation at training centers. Besides the set of documents required for the migrant’s legal exit from Indonesia, she also must have a temporary Work Permit, or an In-Principle Approval (IPA), for her legal entrance into Singapore. Again, the employment agents in Indonesia and Singapore play crucial roles in preparing and processing the paperwork, especially in cases of new hires. Following Scott’s (1998) work on the state’s demand for legibility, several scholars have emphasized the importance of paperwork in constituting the legal placement of a domestic worker (Alpes, 2013; Berg and Tamagno, 2013). Xiang (2013) notes that with the ‘infrastructural turn’ of labor migration in China, ‘to be a legal migrant was to be a paper migrant’. In addition, the systematic relations between multiple documents often require an ‘expert in small-scale bureaucracy’ (Lindquist et al., 2012) like the employment agent to navigate. As the quote in the introduction suggests, an industry that is messier and tougher to navigate benefits brokers. The state depends on the migration industry of employment agencies in Singapore to administer the system of paperwork crucial to the legal entrance of migrant domestic workers through coordinating with their Indonesian counterparts.
Furthermore, the state enlists the migration industry to participate in the production of ‘legal’ migration flows not only through the production of ‘paper migrants’ but also through displacing risks and responsibilities to private employment agents. For example, the state expects agents to verify the ages of the women entering Singapore to work in order to avoid the ramifications of being held accountable for the trafficking of underaged women into Singapore. The state minimizes this risk in two ways. First, as mentioned, agents are held responsible for ensuring the workers’ legal entry into Singapore, and are threatened with warnings or the revocations of their licenses should they fail (MOM, 2010). Second, the state sets the minimum legal age for entry to work as domestic workers at 23; while ostensibly this is meant to ensure more ‘mature’ workers, agents and other observers conjecture that this is a ‘safety net’, so that even if women under 23 do slip through, they are presumably ‘unlikely’ to be minors who are younger than 18. Agents respond strategically to the environmental constraints of the field as set out by the state (Fligstein and McAdam, 2011). Knowing that they are expected to perform their due diligence in verifying workers’ ages in an industry where documentation is frequently falsified, agents develop a number of informal techniques to insulate themselves against the risk of losing their licenses, such as through unexpectedly asking a worker what year she was born and gauging her hesitation before she answers, or through avoiding hiring women who look ‘too young’, no matter what her documentation says. This double layer of governance not only shows how states draw the migration industry into the production of a regional migration regime through the displacement of risk, but also shows how agents actively intervene in and shape migration flows in strategic response.
States also depend on the migration industry to administer a debt-financed migration regime. Many of these documents and paperwork require upfront payment, which prospective migrant domestic workers often lack the capital to make. Besides the costs of paperwork, which may include fees for speeding up administrative procedures, migrants also need to bear the fees for their medical check-up, insurance, brokers’ services, and most significantly, mandatory training. The migration industry is crucial to the debt-financed migration regime that enables the movement of domestic workers. Women looking to work as domestic workers need not pay for their migration costs upfront like their male counterparts do (Lindquist, 2010), but repay their debts through the first 7–9 months of salary deductions. In the meantime, the debt of a worker that snowballs along the recruitment pathway is passed along the chain of intermediaries until it reaches the employer. Goh et al. (2016) catalogue the series of transactions in much greater detail and explore the various consequences of the employment relationship splintering. Again, employment agencies anchor the debt-financed migration system by facilitating the multidirectional cash and paper transactions that make the movement of migrant domestic workers possible. The bridging of the lack of capital at source with eager sources at destination through the coordinating work of the migration industry mobilizes women for domestic work in Singapore and frees the state from bearing the financial strain and responsibility of bad debts.
6.2 Who mediates? The contested field of the migration industry
The elective affinity in interests of agents in making money and states in making order (Xiang, 2012), points to a co-constitution of the state and migration industry. Employment agencies by-and-large produce paperwork, bear debts, and harness matches in order to successfully place workers with employers (Goh et al. 2016). Yet even in what appears to be a stable devolution of governance between the state and the migration industry is a ‘constant jockeying of positions’ (Fligstein and McAdam, 2011) among employment agents and actors in the field. Even the most stable of strategic action fields is constantly changing. Although states ostensibly seek to outsource the risks and responsibilities for moving migrants onto employment agents as described in the previous section, this move is contested by a number of actors. Foregrounding these contestations guards us against taking the role of employment agents in mediating employment relationships and moving migrants for granted.
Against the backdrop of the Work Permit regime, employment agencies and a series of actors vie for authority to mediate conflicts in relationships as well as legitimize the pathways through which migrants enter the country. Given their role in placing migrant domestic workers with employers, employment agents are bound to the employment relationship through contractual terms that clarify the conditions and costs of a replacement. If an employment relationship does not work out, employment agents need to mediate between the worker and employer, distributing debts and fees, and transferring the worker to a new employer or sending her home as they judge fit. This right to mediate conflicts is contested by NGOs, para-governmental organizations, and associations who provide competing platforms. NGOs such as the Humanitarian Organisation for Migrant Economics (HOME) dispute the role of conventional employment agents in adjudicating between employers and migrant domestic workers given the asymmetry of power and the likely collusion between employers and employment agents. Para-governmental organizations like the Centre of Domestic Employees (CDE), a fairly recent ‘challenger’ (Fligstein and McAdam, 2011) initiated by the National Trades Union Congress, asserts its own role to mediate conflicts as a ‘neutral’ arbiter, implying that NGOs – otherwise what Fligstein and McAdam would define as relative ‘incumbents’ in the field –would be partisan and unsuitable. The Association of Employment Agents (Singapore) (AEAS), the industry association, counters the charge of partisanship by arguing that para-governmental and consumer protection bodies lack industry knowledge of debates, the intricacies of contracts, and the thorniness of an employment agent’s work. As an industry association, the chairperson contends, it is in the best position to govern its accredited members.
Apart from the authority to mediate conflicts, employment agents also struggle to define and legitimize the pathways through which migrants enter the country. Building on Weber’s seminal definition, Bourdieu argues that a state is the ‘ensemble of fields that are the site of struggles in which what is at stake is […] the monopoly of legitimate symbolic violence i.e. the power to constitute and to impose as universal and universally applicable within a given "nation"’ (Bourdieu and Wacquant, 1992). In this case, MOM effectively delineates the terms of a regular entrance to Singapore, as outlined in the earlier section. However, MOM also allows for direct hiring, which allows migrant domestic workers to enter regularly without a regular exit from their origin country. This opens up a zone of jurisdictional ‘greyness’ between the regulations of Singapore and the transnational regulations of countries of origin in which the migration industry operates (Wise, 2013). While the legal or regular paths defined by the state are fairly well established, licit paths require the acceptance by the majority of actors as legitimate (van Schendel and Abraham, 2005), inviting the competing claims of employment agents within the migration industry.
Both employers and employment agents in Singapore can apply for an IPA letter with varying degrees of adherence to the regulations of source countries, provoking varied responses across the migration industry depending on their positions. MOM has revamped the online portal to be as user-friendly as possible (Abella and Kouba, 2016), nudging employers to hire workers without the mediation of employment agents in Singapore. This has incited ire from several employment agents who accuse MOM of ‘spoil[ing] market’ and undermining their roles in brokering the entrance of domestic workers. However, not every employment agent in Singapore has as grim a view of the direct hiring route. An ethical agent whom we interviewed capitalizes on the ‘direct hiring’ route and facilitates the departure of women through one of Indonesia’s multiple ports of exits in order to circumvent the 40-day training mandated by The National Agency for the Placement and Protection of Overseas Indonesian Workers (BNP2TKI). He insists that bringing in workers without the hefty costs and onerous loans of exit regulations is most humane. In contrast, given their sizeable placement volumes, larger investments at stake, and ease of scrutiny, the ‘incumbent’ large agencies feel pressured to comply with the regulations of source countries than the rest.
6.3 Moves and counter-moves in the field of migration governance
In the negative space produced by the Singapore authorities’ laissez-faire approach to migration governance, employment agents have taken up the burden of securing market access amidst a prevalent fear of sources ‘drying up’, especially from Singapore’s anticipated lack of adherence to the toughening regulations of origin countries. The Singapore state’s limited engagement with authorities in origin states seems inextricably intertwined with its caution in avoiding diplomatic missteps with its ASEAN neighbors (Piper, 2005) as well as its silence about the terms, conditions, and standards of migrant domestic workers’ employment (Huang and Yeoh, 1996, 2003). Nonetheless, the leadership vacuum produced by the Singapore state’s reticence has also led to the industry association’s decision to take on particular functions of the state, such as negotiating with government authorities in countries of origin.
First, employment agents perceive MOM as a regulator of a free market concerned only with marshalling an adequate supply to meet demand, with an agent lamenting that MOM engages only minimally with source country authorities:
According to the employment agent, MOM is unwilling to engage with authorities in countries of origin as long as the demand for migrant domestic workers is met. Workers from various countries are interchangeable in MOM’s strategy of exploring one new country of origin after another, from the Philippines and Indonesia to Myanmar. Most recently, MOM approved Cambodia as a ‘source country’ for domestic workers (Au-Yong, 2016), though the pilot project to recruit workers has been widely regarded in the industry as a failure (Tan, 2014).Employment agent: So far about the maid policy, MOM is like playing a hands-off la. It’s like so long you all got the maid, I’m not going to care about it. […]
Interviewer: So they’re very reactive ah? They don’t really take the lead or…
Employment agent: No, for them they look at the total volume. So long, ok, well… bringing in Filipino is a lot of problem, oh but Myanmar is coming in. OK, number tally. So long somebody got maid, I’m not going to care about what is complain, quality, or things like that. I’m not going to just care about it. Ya, so at the moment of course the cycle has been from Filipino to Indonesian, Indonesian to Myanmar. Ya. So what’s next, you see? What’s next? So everybody start to panic, what’s next already?
A recent debate in Parliament confirms the priorities and strategies of the Singapore government. Indonesian authorities have spoken of a ‘Domestic Worker Road Map’, a plan to stop sending women to work as migrant domestic workers abroad, which has surfaced alongside calls for live-out arrangements (Arshad and Seow, 2016). When asked about Indonesia’s request for live-out arrangements for domestic workers, the Minister of Manpower (Singapore Parliament Reports, 2016) said:
The minister appeals to the lack of a formal engagement with Indonesian authorities to justify maintaining the current migration regime for domestic workers from Indonesia, mentions that employment agents have continued facilitating the entrance of workers to Singapore, and emphasizes meeting the ‘supply’ of workers as a priority.The Indonesian authorities have not formally informed MOM of its plans for non-live in FDWs [foreign domestic workers]. Our employment agencies have also not reported disruption to the supply of live-in foreign domestic workers from Indonesia. We continue to welcome Indonesian FDWs into Singapore. […] Apart from Indonesia, there are currently 11 other approved sources of FDWs. MOM will continue to monitor the adequacy of our FDW supply.
This set of understandings about the ‘rules’ in the field (Fligstein and McAdam, 2011) concerning the Singapore government’s role is acknowledged and challenged by the Indonesian embassy counsellor, who construes the government’s decision to allow direct hiring as part of its construction of domestic work labor migration as a free market, permitting employers to hire workers without the mediation of employment agencies. In contrast, the Indonesian law mandates the involvement of employment agencies as a means of protecting workers, as is the case in other origin countries (Farbenblum, 2017). The Indonesian embassy counsellor explains:
The state’s passivity has opened up room for what Fligstein and McAdam (2011) call ‘episodes of contention’, where actors employ creative forms of action vis-à-vis one another in conditions of ambiguity about rules and power relations. For example, AEAS, in particular, plays a decisive role in coordinating and cohering migration governance efforts with origin countries. The industry association has collaborated with MOM in the past to devise accreditation schemes and regulatory efforts, but has recently decided to forge ahead with securing agreements with governments of origin countries without the involvement of MOM. The president affirms that many of her meetings with other countries of origin occur at the ministerial level as well as with industry associations in origin countries. AEAS recently signed a MoU with the Indonesian National Board for Placement and Protection of the Indonesia Overseas Workers (BNP2TKI) to collaborate on a Household Service Workers Industry Scheme (HIS) that involves an ‘approved cost structure, a new bank finance scheme for HSW and a qualifying programme for employment agencies on both sides’ (AEAS, 2016) negotiated in advance of upcoming changes in regulations. Without delving into the details of what the new scheme entails, it bears noting that the industry association stepped into the leadership vacuum created in the reticence of MOM. This example of the industry association coordinating migration governance with origin states in response to the Singapore government’s reticence makes the case for treating industry associations and the migration industry as political actors who directly participate in the global governance of migration.MOM [has] two kinds for deployment – direct hiring and through the agency. Singapore is free market, right? You can choose. And then Singapore government cannot push, ′You must [use] the agency′. I mean, force the employer. I know it is a free market, but in our regulation, it's – uh, undang-undang [laws in Bahasa Indonesian] is means the law right, the public Law, 39/2004, regarding the deployment and protection of Indonesian domestic worker. […] They must hire the Indonesian maid through the agency. And then the Indonesian agency will be send to the Singapore agency.
One of the respondents astutely observes the fraught relationship between MOM and AEAS and suggests that the state’s reticence is deliberate: ‘At the moment we should keep AEAS—at the moment. Because government don’t want to talk. They want a mouthpiece. But they don’t want a strong mouthpiece.’ In leaving AEAS to coordinate with origin countries, Singapore authorities allow for the promise of elevated standards in the industry, which may be required for continued market access. However, any effort on the part of AEAS to address the concerns of Indonesian authorities are merely ‘informal’ and ‘private’, leaving room for Indonesian domestic workers to enter Singapore for work as long as they have their papers in order, hence benefiting employers who cannot yet meet such labor standards. Such an arrangement allows the state to refrain from making formal commitments and hide behind a ‘corporate veil’ (Gammeltoft-Hansen, 2013) if these efforts fail, as in the case of a ‘diplomatic embarrassment’ (Piper, 2005). In 1995, Flor Contemplacion – a Filipino domestic worker – was sentenced to death for killing another Filipina domestic worker and the three-year-old Singaporean son of her employer, after Singapore rejected an 11th-hour appeal from the then-Philippines president, Fidel Ramos. Her execution severely strained ties between Singapore and the Philippines and have since made Singapore authorities more wary of a diplomatic misstep in engaging with origin country governments. Without the endorsement and authoritative backing of MOM, the migration industry continues to be flooded with piecemeal initiatives that fail to endure. Nonetheless, given that the interest of origin states in sustaining remittances limits how forcefully they push for the protection of their citizens in destination states (see Tigno (2014) in the case of the Philippines), the role of the migration industry could prove helpful in flexibly accommodating the demands of origin states within labor destinations.
Considering the ‘moves and counter-moves’ of the migration industry, the Singapore authorities and the origin states according to the strategic action field of migration governance reveals the dynamics of co-constitutive governance. Reading MOM and state authorities as adopting a hands-off approach, the migration industry has taken on a larger role in accommodating the conditions of origin states. Perhaps because the migration industry perceives a possible ‘ban’ from origin states as threatening their resource dependencies, the industry association is motivated to engage with their conditions of deploying migrant domestic workers. This in turn makes it easier for Singapore authorities to pay little attention to the demands of origin states, perhaps until their political standing and legitimacy are called into question (Piper et al., 2016; Koh et al. 2016).
7 Conclusion
Using the theory of strategic action fields to examine the dynamics of the migration industry and the states to Singapore, we have spotlighted the contestations and strategizing in the governance of migrant domestic workers. Despite a recognition that the states benefit from neoliberal gains like greater control, avoiding blame, and shifting costs, these accounts of the role of the migration industry in migration governance seem staid and settled. Employment agents on whom the placement of migrant domestic workers with Singaporean employers is devolved produce paperwork, bear risks and responsibilities, and administer a debt-financed migration regime. In the ongoing, negotiated production of such a status quo, employment agents as well as other actors in the field jostle for authority to mediate employment relationships as well as define the licit pathways into the country. These contestations take place against the setting of the Singapore state's permissive visa regime as well as reluctance to intervene. Finally, in constructing an account of the shifting dynamics and speech acts of the migration industry as well as Singaporean authorities, we make the case that the migration industry plays a crucial role in accommodating and adapting to the demands of origin state authorities. Not only does the migration industry take on devolved governance, it also directly participates in diplomacy efforts with origin countries.
In contrast to theories of International Relations that bestow on states an exaggerated capacity of and coherence in governing migration, we argue that the contestations of the migration industry and the Singapore state within a strategic action field produce migration governance. On the one hand, we have foregrounded the interpreting, strategizing, and initiative of the migration industry in elevating standards for migrants instead of locating power and autonomy exclusively within states. On the other hand, we have considered the interpenetration and contestations between the state and the migration industry instead of erecting a strict boundary, which Xiang (2012) argues is what enables destination states to blame intermediaries for violating migrant rights and preserve their own moral authority. The analytic of a strategic action field accomplishes both ends.
Acknowledgements
We are grateful to all the respondents who generously gave their time to participate in the interviews, as well as Irna Nurlina and Jamie Lai for assisting with interview transcriptions. We would like to acknowledge the support of the Asia Research Institute (ARI) at the National University of Singapore. This work was supported by UK Aid from the UK government through the Migrating out of Poverty Research Programme Consortium; however, the views expressed do not necessarily reflect the UK government’s official policies.
