Safe migration programme interventions: regulating migrants through anticipation, traceability, and re-embedding

Safe migration programme interventions: regulating migrants through anticipation, traceability, and re-embedding

Sverre Molland

Molland, Sverre. 2022. Safe migration programme interventions: regulating migrants through anticipation, traceability, and re-embedding. MoLab Inventory of Mobilities and Socioeconomic Changes. Department ‘Anthropology of Economic Experimentation’. Halle/Saale: Max Planck Institute for Social Anthropology.

Download via Doi: 10.48509/MoLab.4477

In 2018 the United Nations’ (UN) promulgated its global strategy for migration governance, titled the Global Compact for Safe, Orderly and Regular Migration, which elevates safe migration to a policy principle, objective, and strategy.[1] Safe migration has been in vogue within the aid sector since the mid-2000s, especially amongst the UN’s International Labour Organisation (ILO) and International Organisation for Migration (IOM), as well as several NGOs that work with migrant populations. Drawing on my recent research on safe migration programmes in the Mekong region in Southeast Asia (with specific focus on Thailand, Laos, and Myanmar), I show how these programmes draw on anticipation (i.e., programmes intervene prior to forecasted danger, risk, and peril within labour migration) and traceability (where programmes attempt to facilitate – as opposed to control mobility – through connectivity). These temporal and spatial qualities, in turn, precondition safe migration programmes to become dependent on large networks of migrants in Thailand. Although rarely part of aid agencies’ programme design, safe migration programmes re-embed labour migrants as social agents through programme implementation. Hence, safe migration programme delivery presents a case study for how emergent migration governance generates new spatial, temporal, and social relations.

Safe migration

Many aid programmes have adopted the emerging discourse of safe migration given its explicit focus on facilitating (as opposed to hindering) migration flows through policy interventions that enhance migrants’ safety. This uptake has occurred at the backend of several years of anti-trafficking interventions which often resulted in highly territorial interventions (e.g., border control measures, including deportations, justified in the name of anti-trafficking).

Building on earlier guest worker programmes and circular migration models, safe migration extends a focus on legal migration pathways (which emphasises the use of passports and work permits) by considering how migrants’ dispositions and social relations may pre-empt exploitation and other forms of misadventure whilst migrants are deployed for work in a host country. This includes aid agencies providing aspiring migrants with training on specific knowledge and desirable behaviours that may bolster their safety (which can range from cultural etiquette training, information on visa regulations and labour rights, and the importance of learning the local language). Such interventions transform “would-be migrants”[2] into mobile subjects with behavioural dispositions that are deemed conducive to safe migration outcomes. The temporal dimension of such interventions is notable: to ensure safety in migration, interventions take place in advance. This is unsurprising as the very concept of safety is inherently proactive rather than reactive.

Such training can also include advice on how social relations mediate safety whilst migrating. Who facilitates migration (informal broker, licensed recruitment agency, friend, or family member) may be just as important in structuring a migrant’s safety, as opposed to whether the migrant possess legal documents per se.[3] This is where safe migration programmes’ temporal qualities (i.e., manufacturing safety through awareness raising prior to migration) mesh with their de-territorial characteristics. Safe migration programmes provide migrants with a range of services and assistance through their migration process in the form of pre-departure, post-arrival training, and a range outreach services and support mechanisms in the migrants’ destination country. Hence, rather than a policy framework which attempts to “anchor” movement (e.g., through border control measures), safe migration policy models target migrants as they move through space (as opposed to being confined by it). In addition to anticipatory interventions (which are typically offered in source countries in the form of pre-decision and pre-departure training) aid programmes must find ways to connect with migrants regardless of their location in the host country to assist with various problems migrants may encounter (wage theft, health problems, extortion, labour abuse etc).

How then do safe migration programmes operationalise these temporal and spatial dimensions through interventions, and activities? Safe migration’s anticipatory characteristics can take the form of simulation, such as role play during pre-departure training.[4] However, amongst the numerous safe migration programmes I have studied in the Mekong region, a far more dominant method pertains to the operational reach of programmes, especially in migrants’ destination country (Thailand). Programmes must connect, target and trace migrants through their labour migration cycle. This presents a challenge for aid programmes and governments alike. Formal aid programmes (NGOs, UN agencies, government authorities) do not have the reach - due to migrants’ mobility and spatial dispersal - nor the trust, given a general fear of deportability[5], to easily access migrants. In other words, there is an inherent scalar tension between regulating migration as an aggregate phenomenon and the need for intervention at an individual level. Governments and aid programmes therefore move beyond conventional hierarchical organisational modes and engage a range of rhizomic relations in their work which in effect re-embeds migrants within policy interventions. My research suggests that such re-embedding takes three forms: (a) proliferation of brokerage, (b) state emulation and contestation premised on patronage and reciprocity and (c) digital technologies.

Re-embedding migrants

To access migrants, some aid programmes establish relationships with informal migrant associations. Migrant associations are ubiquitous, especially amongst Burmese labour migrants, throughout Thailand. Migrant associations are made up of and administered by labour migrants who often have considerable migration experience, yet often do not hold any official registered license to operate by Thai authorities. Migrant associations vary in purpose and size, ranging from self-organising committees in migrant dormitory compounds (adjacent to factories), religious-based associations (often in association with Thai Buddhist temples), to migrant groups that span provincial and even national levels. Some associations operate as unofficial pseudo unions, assisting migrants with a range of problems ranging from visa status, wage-theft, and work accident compensation claims. At times, migrant associations assist with organising strikes involving hundreds (even thousands) of migrant workers. While they possess unrivalled access (and trust) within migrant communities, aid programmes provide something that migrant associations do not possess: official status (and skills) to mediate with state bureaucracies (e.g., in compensation claim cases). Migrant associations trade unprecedented access to migrant populations in exchange for protection given formal aid organisations' legal status and ability to navigate Thailand’s state bureaucracy (e.g., advance work accident cases through Thai labour courts). Patron-client relations between formal aid programmes and informal migrant associations ensues, transforming the latter into key intermediaries between migrants and the formal aid sector.

Intermediaries do not only take the form of associations. Several aid organisations employ outreach workers, who often are themselves experienced migrants. Leveraging their pre-existing connections and know-how, they are, like migrant associations, highly effective in connecting with migrants. Yet, many outreach workers have former careers as smugglers and migrant brokers (which sometimes NGOs are unaware of). Outreach work requires the ability to straddle different social worlds and engage with deplorable social actors (e.g., predatory brokers, uncooperative employers, and policy officers). This may include the ability assist migrants recuperating “fees” imposed by unscrupulous employers, mediating disputes at workplaces, collect evidence from migrants that can be used in court cases against exploitative employers, or delicately arranging safe passage for migrants who do not have their formal documents in order. Brokers fit such a role perfectly. As one former outreach worker (who later launched a career as a labour migration broker) told me: “if you don’t enter the tiger’s cage, you won’t get the cub!”

Brokerage and patronage are also evident in how relations form between state authorities and migrant associations. Paradoxically, government bodies in both sending (i.e., Myanmar) and receiving countries (i.e., Thailand) engage (as opposed to hinder) such practices, despite migrant associations lack of legal status and tendency to hold an oppositional stance against authorities. Like aid organisations’ limited reach amongst migrants, state authorities tolerate, and even engage these associations as they can leverage them in implementing policy. Given migrant groups’ unrivalled connections with migrants, it is far more efficient for labour and immigration officials to allow migration associations leeway in assisting migrants as this brings orderliness and predictability to policy implementation (it is far easier negotiating with factory workers through the mediation of a few bilingual migrant association representatives as opposed to hundreds of disgruntled, unruly migrant labourers). Migrant groups also present unparalleled abilities in transmitting policy directives (such as updates on new visa processing systems) to migrant constituents. Hence, contestation (in the form of strikes, compensation claims and advocacy against abuse of migrants) goes hand in hand with state emulation (migrant associations extend policy announcements to migrants); a dynamic that comes most to light through the use of social media, which in turn underscore tempo-spatial qualities of such practices.

In contrast to many formal aid organisations who bankroll ineffective social media campaigns (such as developing migration assistance apps with minimal uptake), several migrant associations deploy Facebook and text messaging apps (such as Line) through their communication with migrants. At times, this is deployed to assist and trace migrants directly (e.g., using GPS location functionality of smartphones to locate the whereabouts of an abused domestic migrant worker who cannot leave her workplace in fear of the employer reporting her to immigration authorities for deportation). More commonly, Facebook serves as a broadcast for both advocacy and financing purposes. With daily updates on migration policy, stories of successful compensation claims and (at times) thinly veiled critiques of authorities’ failure in cracking down on abusive employers and corrupt officials, social media present emergent forms of connectivity which reinforce the aforementioned tension between state emulation and contestation. They also serves the purpose of financing migrant associations (who receive donations from migrants). These digital networks are enormous. For example, one of my key informants (a member of one of the Myanmar migrant associations) has more than one million followers on Facebook. Although such media presence presents challenges for state authorities who are weary of critical social media commentary, such large-scale social media networks also serve important purposes for both Thai and Myanmar governments: they serve as conduits for policy (e.g., broadcasting new rules for obtaining a Thai work permit).

These examples share in common how migrants are transformed from objects to instruments of policy. Safe migration’s anticipatory (coaching migrants on safe migration as a preventive strategy to address labour exploitation) and de-territorial logics (e.g., tracing and connecting with migrants through outreach work and GPS enabled smartphone capability) only become possible through incorporating migrants into safe migration aid delivery, whether this is in the form of UN and NGO funded aid delivery or government-led migration policies. Migrants’ networks and social relations become extensions of the migration policy apparatus.

[1] United Nations. 2019. Global Compact for Safe, Orderly and Regular Migration. New York: United Nations.

[2] Biao, Xiang. 2014. “The Would-Be Migrant: Post-Socialist Primitive Accumulation, Potential Transnational Mobility, and the Displacement of the Present in Northeast China.” TRaNS: Trans -Regional and -National Studies of Southeast Asia 2(2): 183–99.

[3] Although it is well documented within migration literature (and often privately acknowledged by aid officials) that legal migration status and labour exploitation may co-exist, the claim that legal migration status protect migrants is commonplace within safe migration discourse and policy circles more broadly. See: Molland, Sverre. 2021. Safe Migration and the Politics of Brokered Safety in Southeast Asia. London: Routledge.

[4] Simulation as policy implementation is arguably most developed within the armed forces, but is also deployed in the humanitarian sector and in preparedness exercises relating to epidemiology. See: Lakoff, Andrew. 2008. The Generic Biothreat, Or How We Became Unprepared. Cultural anthropology 23(3): 399–428.

[5] De Genova, Nicholas P. 2002. “Migrant ‘Illegality’ and Deportability in Everyday Life.” Annual Review of Anthropology 31(1): 419–47.

This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (, which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.

Go to Editor View