The externalization of border control in the global South: The cases of Malaysia and Indonesia

Introduction

There is a significant body of scholarship that examines the novel approaches adopted by states around the world to control mass mobility. In particular, states in the global North increasingly govern unwanted migration from the global South by exporting models of border control under various guises of migration management and aid; outsourcing part of their control apparatus to migrant ‘sending’ or ‘transit’ countries ¹ and detaining unauthorized arrivals ‘at a distance’; expanding the reach of sovereign powers extraterritorially; and transforming border security across organizational entities nationally and internationally. From different disciplines and with varying emphases, these developments have been conceptualized as ‘externalization’ (Hyndman and Mountz, 2008; Welander, 2021; Zaiotti, 2016), ‘remote border control’ (FitzGerald, 2020) and ‘extra-territorialization’ (Dastyari and Hirsch, 2019).

Border externalization policy developments disrupt conventional distinctions between the domestic and foreign, the humanitarian and exclusionary, and the correspondence between state power and territory. These new configurations and proliferation of border control are marked by unequal global geo-political relations and epitomize what Aas (2013: 26–28, 2014: 532) theorizes as the ‘Northern penal state’, with the capacity to export its penal models and technologies of crime control; extend its security agenda and specific visions of order into the international domain; and expand its domestic interests and control functions through ‘extraterritorial forms of policing, prisoner transfers, investments in prisons abroad and such like’.

Although many studies have provided important critiques of US, European and Australian externalized border controls and their devastating impact on migrants’ lives, discussions of the actual workings of externalization and border security building from the perspective of the global South remain scarce (for notable exceptions, see Frowd, 2021; Gaibazzi et al., 2017; Stambøl, 2019, 2021; Stock et al., 2019; Vammen et al., 2021). It is crucially important to go beyond the northern analytical focus and to ‘see from the peripheries’ (Aas, 2012: 11) in order to understand the ‘situated’ meanings and practices of externalization. In this context, a comparative approach has been particularly useful in providing a much-needed corrective to universalistic claims about border control—for example, by examining the relational nature and negotiated aspects of European Union border externalization, as Turkey, Moldova and Morocco develop different strategies of ‘migration and border diplomacy’ to leverage with the EU over political resources and maximize the opportunities for increased mobility of their own citizens (Laube, 2019).

This article seeks to contribute to our criminological understanding of the divergent nature and policy outcomes of border externalization by comparing how offshore control plays out in Malaysia and Indonesia in Southeast Asia. These two geographically proximate, non-signatory countries to the 1951 UN Refugee Convention have attracted much attention in the last two decades in the context of Australia’s controversial border protection policies, which came to be known as the ‘Pacific Solution’ since 2001 (Taylor, 2021; Weber, 2006). Australia regards Indonesia and Malaysia as key transit states and has pursued a broadly similar policy to keep out Australia-bound asylum seekers and refugees. The so-called Indonesia and Malaysia ‘solutions’ of migrant interceptions, offshore processing and detention exemplify Australia’s embodiment of the northern penal state and its capacity to outsource migration control and extend its borders into the sovereign domain of southern states. Seen in this light, Indonesia and Malaysia can be understood as the ‘most similar cases’ under what Lijphart (1975) has characterized as the ‘comparable-cases approach’ in comparative research.

At the same time, the two cases (countries) differ in the roles they play under Australia’s externalization policy agenda, and how far they respond to interventionist policy making over time. Their border control outcomes exhibit much more variation and contestation than is often acknowledged in the externalization debates and reflect important intervening factors in the two countries’ internal (domestic political and economic realities; attitudes towards migrants and their control) and external (interstate geo-political relations) environment. In Indonesia, its strategic importance as a key transit state in Australia’s regional deterrence frontier, the role of the state in refugee management, and its attitudes towards refugees provide the conducive factors for Australia’s outsourcing of immigration detention. In Malaysia, the state’s attitudes towards refugees and international actors in refugee management, as well as the economic realities of migrant labour under a state-sponsored politics of exclusion, provide the constraining factors for Australia’s outsourcing of refugee processing.

Through a comparative analysis of border externalization and migration control in Indonesia and Malaysia, this article will sketch out the uneven and contingent nature of border externalization and demonstrate the importance of developing ‘empirically based concepts to make sense of non-western based realities [of political order and cultures of control] that are relevant for criminological theorizations of the relationship between state/political power, crime control and borders’ (Stambøl, 2021: 487). The discussion is based on secondary sources including a systematic review of multi-disciplinary scholarship and grey literature; publicly available English-language material from government sources (e.g. Australian Hansard Senate Committee between 2000 and 2020) and non-government sources (e.g. international organizations related to migration, human rights and border security; Malaysian and Indonesian non-governmental organizations (NGOs), legal and human rights bodies); digitalized news media and political commentaries on Southeast Asian affairs.

Border externalization

The term ‘externalization’ has been used by policy makers, the media and scholars from various disciplines to refer to the extension of bordering practices, even though the term itself is rarely discussed in debates outside the northern context (Gaibazzi et al., 2017: 10). Scholars point to its use as a ‘policy tool to reduce the economic, political and social costs of “unwanted immigration” for receiving states’ and to shift the ‘moral responsibility for migrants’ well-being from states in destination countries to actors on the national or local levels in origin countries or transit spaces’ (Stock et al., 2019: 2). Others highlight the capacity of northern states to manipulate the borders in time and space through various migration management strategies. These include ‘anticipatory regimes of regulation and control’ that involve ‘a range of disciplinary, punitive and militaristic technologies aimed at preempting arrival at the physical border’ (Wilson and Weber, 2008: 125); ‘crimmigration’ strategies that blur conventional distinctions between criminal law and immigration law (Stumpf, 2016); and the use of ‘spatially disaggregating techniques of governance … to categorize, contain and control’ migrants (McNevin, 2014: 302). By outsourcing border enforcement to jurisdictionally ambiguous offshore spaces and immobilizing prospective asylum seekers before they reach sovereign territory to make protection claims, northern states that claim to uphold liberal democratic values and norms of human rights have been criticized for circumventing their responsibility under international conventions and contributing to the global refugee crisis (Gammeltoft-Hansen, 2011; Gibney, 2004). Some externalized border controls may be more visibly violent, involving mass drownings and abuses of migrants as in the EU–Libya context (Villa et al., 2018; Zaiotti, 2016). Other measures may be more mundane but no less harmful, as in the administrative immigration detention and the everyday ‘politics of exhaustion’ (Welander, 2021) under the British ‘juxtaposed border arrangements’ with France (Bosworth, 2020).

Externalization policy developments signal and reinforce the political and economic dominance of migrant receiving states and reflect longer histories of unequal exchanges and exploitation. There is a considerable body of literature that examines how northern states impose their migration control priorities and domestic security concerns on southern states by strategically ‘link[ing] ideas of security, risk, and punishment, with humanitarianism, belonging, and development’—or what Bosworth (2017a: 58) has theorized as ‘penal humanitarianism’. Border externalization can be understood as part of the growing entanglement of logics of securitization, managerialism and humanitarian aid. In the British context, for example, the UK government has routed extensive funds and development assistance in Africa in the name of security and state building ‘to bolster human, technological and infrastructural capacity to control borders in major sending and transit countries … [or] tying foreign aid to the sending state’s cooperation on returns’ (Aliverti and Tan, 2020: 207). Taken together, these critical insights provide a compelling critique of contemporary immigration control and make an important contribution to the analysis of the intersections of migration and criminal justice within criminology (see, for example, Bosworth, 2017b).

However, what is often overlooked in the externalization debate is that the processes and outcomes of externalization are neither homogeneous nor one-sided. As Vammen et al. (2021) argue in relation to Europe’s exportation of border control to Africa, externalization is not a one-way street, imposed by an all-powerful Europe on a passive Africa. States may have different agendas and interests to assist in border control or to resist it, or they may appropriate particular discourses and policy measures for their own regards to migration management, policing priorities or political and economic gains. For example, Vogt (2020: 54, 60) argues that Mexico’s willingness to carry out America’s ‘dirty work’ to curb unauthorized Central American migrants crossing its territory is ‘deeply embedded within trade and security relationships between the two countries’ and has to be considered in the context of Mexico’s demand for cheap immigrant labour under ‘exploitative capitalism’ and the ongoing security concerns related to its war on drugs. Similarly, critics suggest the British government’s latest plan to send asylum seekers to Rwanda for offshore processing and resettlement provides the UK ‘a political and economic foothold’ in a region that is rich in resources; at the same time, the plan’s £120 million development aid provides ‘short-term financial benefits’ to Rwanda and arguably elevates its ‘international profile as an engaged partner in global migration and refugee governance’ (Nair, 2022). As such, the interstate relationships in border externalization must be understood as ‘historically constituted, mutually beneficial—at least to some parties—and ultimately tenuous’ (Vogt, 2020: 51).

Australia’s Pacific ‘non-solution’

The Australian model of securitizing unauthorized migration as threats to state sovereignty and outsourcing border controls has been subject to critical scrutiny in criminology, migration and critical border studies, legal and refugee studies (see, for example, Grewcock, 2013; Hirsch, 2017; Minns et al., 2018; Weber and Pickering, 2011). Most accounts refer to the 2001 incident involving asylum-seeking migrants on board a sinking Indonesian fishing vessel, their rescue by the Norwegian vessel MV Tampa, and the subsequent Australian hard-line response during the Howard government’s election campaign, as a pivotal moment in its development of what became known as the ‘Pacific Solution’ of offshored, outsourced and militarized border control (Grewcock, 2014; Taylor, 2021; Weber, 2006). Since then, the Australian government has pushed its borders offshore (including designating the Australian mainland as an ‘offshore excised place’ for the purpose of seeking asylum); contained ‘irregular maritime arrivals’ in third countries; and adopted a progressively punitive approach to ‘stop the boats’ under various renditions of the Pacific Solution while simultaneously framing border control as ‘regional cooperation’ to clamp down on human smuggling and ‘to save lives’ (Hirsch, 2017; Little and Vaughan-Williams, 2017; Weber, 2007).

Australia’s sustained policy to create a ‘ring of steel’ (Dastyari and Hirsch, 2019: 435) has been heavily criticized for the everyday violence of its border policing practices; inflicting ‘a scale of systemic abuse’ on migrants akin to ‘state crime’ (Grewcock, 2013: 11); turning arbitrary immigration detention into ‘a global brand’ (Global Detention Project, 2022); and for exploiting vulnerable and poorer states that reflect ‘a more direct neocolonial relationship’ (Grewcock, 2014: 75). In particular, the Australian funded and controlled offshore detention facilities (known as ‘regional processing centres’ in Australian official terminology) in its former controlled Pacific Islands states of Papua New Guinea (Manus Island) and Nauru have been widely condemned not only for their inhumane conditions and routine human rights abuses against migrants (Amnesty International, 2016; UNHCR, 2018), but also for the carceral regime that is ‘imperial in form’ and reiterates patterns of ‘historical violence, control and dispossession’ (Giannacopoulos and Loughnan, 2020: 1118, 1123).

Successive Australian governments have used political pressure, development aid, cash contributions, and the provision of infrastructure to leverage some of the most impoverished countries in the region to play specific roles in Australia’s externalization regime. In the case of Nauru, the world’s smallest island nation and bankrupt when the Pacific Solution was first brokered in 2001, it has been transformed into a remote processing outpost in Australia’s offshore refugee operations with its entire economy based on ‘managing humans for profit’ (Morris, 2021: 691). The bilateral agreements allow for the forced transfer of asylum seeking migrants to Nauru for protection claims processing by Australian (and later local) officials, indefinite detention and, after 2012, for refugee resettlement. In exchange, Nauru receives substantial aid contributions and revenue from the ‘enduring processing’ deal, which in turn reconfigure and perpetuate its ‘colonial infrastructural forms’, dependency and injustice (Morris, 2021: 688).

Australia’s outsourcing of border control is not uncontested. In 2001 against a fragile post-coup environment in Fiji, Australia’s offer of money in return for setting up a regional processing centre was strongly opposed by religious and regional bodies and later withdrawn amid criticisms of ‘the Pacific region continuously becoming a dumping ground for the benefit of industrialized nations’ (‘Australia accused of “human trafficking”’, CNN, 26 October 2001). ² In 2010, Australia again proposed to establish a regional processing centre in the newly independent state of Timor Leste (ranked 147 out of 187 in the 2011 UN Human Development Index) in exchange for funding for major infrastructure projects. The proposal was blocked in part owing to domestic political struggles in Timor Leste after years of Indonesian occupation. Its parliament voted against it while Timorese NGOs and civil society groups signed a joint statement to oppose it. ³ In 2016 amid sustained international criticisms, local resentment and signs of Papua New Guinea’s ‘growing assertiveness in its relationship with Australia’ (Wallis and Dalsgaard, 2016: 323), the Papua New Guinea Supreme Court found the offshore detention of asylum seekers in the Manus camp unconstitutional. This led to the immigration detention camp’s formal closure and Australia’s substantial compensation to those unlawfully detained (Dastyari and O’Sullivan, 2016; Giannacopoulos and Loughnan, 2020).

The contestation also comes in other forms. After the offshore processing arrangement with Malaysia was derailed (discussed below), Australia signed a bilateral agreement with Cambodia (traditionally an origin country in forced migration and one of only two Southeast Asian signatory countries to the UN Refugee Convention) in 2014 to resettle ‘voluntary’ refugees from Nauru. In exchange, Cambodia would receive around AU$40 million overseas development assistance package and up to AU$15.5 million for the actual resettlement (Gleeson, 2019). ⁴ The refugee resettlement deal was supported and facilitated by the International Organization for Migration (IOM) ⁵ but widely criticized by international human rights organizations over Cambodia’s track record of human rights abuses and corruption. The expectations of multiple key actors were also very different from the everyday reality of resettlement of the migrants (Gleeson, 2019). Evidently, forcing people to leave is difficult and persuading other countries to take them is costly. After the Cambodia deal expired in 2018, it was dubbed ‘arguably the most expensive and ineffective way of relocating one single person’. ⁶

While these developments may not be interpreted as policy failures by the Australian government, they are a stark reminder of the limits of the northern penal state and the contingent nature of the outsourcing of some forms of sovereign power that underlie border control ‘at a distance’. Taking into account the frictions, frustrations and the leveraging in border control politics provides an important corrective to an all-encompassing externalization thesis. Southern states are not simply passive recipients of externalization policies; instead, they respond to the incentives and pressures differently depending on a number of intervening factors, which in turn lead to different outcomes over time as the case studies of Indonesia and Malaysia reveal.

Mobility and control in the Malay Archipelago —the historical context

Border control has not always been regarded as a policy priority in Indonesia and Malaysia. Indeed, both countries have a rich cultural tradition and a long history of cross-border flows of people and goods (Ford and Lyons, 2012). As Garcés-Mascareñas (2015: 129) contends, in pre-colonial times, before artificial boundaries were created between the Dutch East Indies and the British Straits Settlements, ‘the Malay Archipelago (which includes Indonesia and Malaysia) was a free, open migration area with seas, islands, and shallow, narrow waterways, which facilitated the movement of merchandise and people’. Against this background, the uneven development of outsourcing of border control in Indonesia and Malaysia needs to be understood in light of their relative strategic importance to Australia’s border security regime, their divergent attitudes towards migrants and control as well as their domestic political realities and socio-economic exigencies.

Indonesia, which became embroiled in a bitter decolonization war before gaining independence from Dutch colonial rule in 1949, is a vast archipelagic state of over 17,000 islands. The country has a violent history of extrajudicial mass killing of (alleged) communists and incarceration without trial of leftists and other political suspects in notorious colonial-era prisons and forced labour camps in the 1960s and 1970s (Robinson, 2018). Even after the dramatic fall of Soeharto’s authoritarian government in 1998 and Indonesia’s tumultuous transition to become the world’s third largest democracy, elements of the state apparatus of repression, state-sanctioned violence and routinized corruption have taken roots in the society.

Malaysia, a British colony until 1957, underwent a more stage-managed decolonization even though it, too, has a chequered history of exclusion and repression directed at the non-Malays, communists and others deemed ‘undesirable’ especially during economic downturn. Scholars note the persistent influence of ‘authoritarian neoliberialism’, patronage politics associated with criminal elements in civil society, and a deeply ‘ethnicised national politics’ that favour Malay indigenous groups across socio-economic and political life (Hoffstaedter, 2014: 875; Juego, 2018: 55; Lemière, 2019).

Indonesia and Malaysia have been at the centre of forced migration in a region where recognition and protection of refugees have long been politically controversial. Crucially, the two countries share a number of commonalities as well as key differences in their response to refugee issues. They are both non-signatory countries to the UN Refugee Convention and generally view refugees as the responsibility of northern resettlement countries. During the protracted Indochinese refugee crisis, they provided temporary refuge to the massive influx of Vietnamese ‘boat people’ under large-scale regional processing and camp regimes from the late 1970s until the 1990s (Hedman, 2008; Missbach, 2013; Prabaningtyas, 2019). Then as now, Malaysian law does not recognize the formal status of refugees. They are regarded as ‘illegal’ migrants and subject to boat pushbacks, arrest, detention, and (corporal) punishment for immigration offences. Refugees are generally invisibilized within Malaysian society except as a very select group, such as the Rohingya from Myanmar. However they, too, have been subject to periodic boat pushbacks, as Malaysia constantly changes its approach towards particular refugee groups from ‘a position of indifference to a posture of burden-sharing state of international migration or defender of fellow Muslims on the international stage’ (Ansems de Vries, 2016; Hoffstaedter and Perrodin, 2018: 197).

By contrast, Indonesia has developed a range of domestic laws and policies that provide partial recognition of refugees within the country. Kneebone et al. (2021) documented the competing legal and policy framework and normative struggles around refugee management from a general indifference to the long-term presence of refugees to the continuing tensions between a securitization and humanitarian assistance approach under the 2016 Presidential Regulation Concerning the Treatment of Refugees (see also Afriansyah and Zulfa, 2018). This limited recognition of refugees provides an important backdrop to Indonesia’s ‘inconsistent and ad hoc’ approach to migration control:

vacillat[ing] amongst a permissive laissez-faire attitude that allowed thousands of asylum seekers to pass through Indonesia freely; a hasty and heavy-handed use of incarceration in an overcrowded IDC [immigration detention centre] system in keeping with the interests of Australian government funders, and a pragmatic shift to [alternatives to detention].

A tale of two externalizations

In the contemporary context, Australia regards Indonesia and Malaysia as key transit states and has pursued broadly similar externalization policies in order to shield itself from asylum seekers and refugees. It has strategically persuaded both countries to revise their relatively lax visa regime and limit the visa-free entry of particular foreign nationals (e.g. from Pakistan, Iran and Afghanistan), and incentivized them to deter, intercept and contain unwanted migrants (Dastyari and Hirsch, 2019; Nethery and Gordyn, 2014). According to Hirsch (2017: 71):

Over the last decade, Australia has spent over $13 billion on regional cooperation to prevent the movement of irregular migrants. This includes support to improve and share intelligence on people smuggling, increased use of biometric data, the use of anti-migrant-smuggling campaigns, increased intelligence and monitoring of ports, the harmonisation of restriction border policies across the region, and the development of a joint Assisted Voluntary Return (AVR) programme with IOM.

Significantly, Australia has targeted Indonesia owing to its geographical proximity and strategic importance in Australia’s regional deterrence frontier and made a sizeable investment in the country as the ‘final bulwark’ against Australia-bound asylum seekers. This includes substantial funding for immigration detention capacity building (discussed below); donations for infrastructure and equipment (e.g. patrol boats, surveillance aircraft, detection equipment); and various kinds of technical assistance and training for local enforcement officials in raids and intercepting asylum seeking migrants in the last leg of their boat journey to Australia (Dastyari and Hirsch, 2019; Hirsch, 2018; Missbach, 2018; Nethery et al., 2013).

However, Australia’s exportation of border controls is neither a linear nor stable process. While Indonesia and Malaysia have long-standing links with Australia across policy areas and an interest in maintaining good neighbourly relations, their cooperation with Australia is highly contingent, shaped by mutual concerns as well as suspicion and an increasing pull to ‘look North’ economically and politically (Funston, 2013; Lindsey and McRae, 2018). In the Indonesian context, the bilateral relationship in border control, and in criminal justice more broadly, has been characterized by an ‘uneasy mix of political and policy interests’ and ‘reciprocal exchanges’ when there is a core of mutual concerns, for example, after the Bali bombings in 2002 (McKenzie, 2018: 163–172). Throughout the various renditions of the Pacific Solution, there are examples of joint agency cooperation and higher-level strategic agreements in some aspects of externalized border control and Indonesia’s resistance in others—such as protesting against Australia’s tow-backs of asylum seeker boats and rejecting the proposal from successive Australian governments to establish a refugee processing centre in Indonesia (Missbach, 2018; Missbach and Hoffstaedter, 2020).

In the Malaysian context, the government has been receptive to Australia’s ‘ever widening zone of extended border security that stretches through Southeast Asia and Melanesia’ (Hoffstaedter, 2011), for example, by allowing Australian Federal Police to be posted at Malaysian airports (as in Indonesia and Sri Lanka). However, the bilateral cooperation rarely extends to refugee issues largely because of Malaysia’s official non-recognition of temporary refugee presence and a corresponding absence of formal policy and refugee management structures at the national level. As I argue below, these factors provide the backdrop to the variations in the policy outcome of outsourcing immigration detention and refugee processing in Indonesia and Malaysia.

Offshore immigration detention in Indonesia

One key variation in Australia’s externalization policy outcome is the proxy funding of offshore immigration detention to contain asylum seekers and refugees in Indonesia, which is absent in Malaysia. As the Australian Immigration and Citizenship Secretary explained the use of overseas development assistance in offshore immigration detention to the Senate Committee in 2010:

One issue that we have seen is that the Indonesian law enforcement authorities have been very active in helping to identify and intercept boatloads or groups of people en route to Australia but have not had the facilities in which to accommodate those people in a secure way … [the additional funds are] a ramping up of the arrangements to try and assist Indonesia to prevent, detect and hold people so that they are processed in Indonesia … [in] an overall expectation that that would suppress the number of people coming to Australia. ⁷

Indonesia has used its immigration detention facilities to detain foreigners without valid passports or foreign criminals awaiting deportation in the past. However, observers suggest that without Australia’s generous proxy funding to shore up its immigration detention capacity over the years, ‘it is unlikely that Indonesia would [have detained] thousands of transit migrants’ (Missbach and Adiputera, 2021: 495; Nethery et al., 2013: 96). Under the 2001 Regional Cooperation Arrangement (RCA) between Australia and Indonesia, Australia has channelled a total of AU$388 million through the IOM to finance the refurbishment, extension, maintenance and staff training of Indonesia’s immigration detention centres, and to manage and provide care for intercepted asylum seekers and United Nations High Commissioner for Refugees (UNHCR)-recognized refugees who had been detained (Dastyari and Hirsch, 2019; Hirsch, 2018; Nethery et al., 2013). Immigration detention is a precondition to accessing post-detention assistance by the International Organization for Migration (IOM), which includes food, basic medical support and cash allowance. So-called ‘alternatives to detention’ community accommodation are also funded and supervised by the IOM with mixed local results (Missbach, 2017, 2020).

The role of the IOM has been pivotal to facilitating the externally funded immigration detention system in Indonesia. Scholars note that the IOM operates as ‘the face of migration management qua extra-territorial border security for hire’ (Watkins, 2020: 1121), ‘creating the impression of a rights-based humanitarian UN body while still maintaining its core role in conceptualising, proposing, and implementing migration control activities’ in immigration detention, anti-smuggling information campaigns, and ‘voluntary’ returns of asylum seekers and refugees (Hirsch and Doig, 2018: 686). ⁸ In 2016 the IOM Indonesia reported operational expenditure of AU$49 million, making Indonesia one of the largest missions and the IOM’s eighth-most expensive field location in the world. Almost 90% of IOM Indonesia’s annual funding came from Australia (Hirsch and Doig, 2018).

As Nethery and Gordyn (2014: 190) argue, ‘incentivized policy transfer’ works when migration control is presented in Indonesia’s ‘own national interests’, enabling Indonesia to ‘strengthen its relationship with Australia at little political or financial cost’ and be seen by its domestic constituency as ‘making a rational decision without pressure from its wealthier neighbour’. Indonesia has never funded services for refugees under its national budget but rather relied on external funding. In this regard, recasting matters of sovereignty and security building under a cloak of humanitarianism makes Australian funded immigration detention politically palatable in Indonesia and, more importantly, facilitates its ‘abdication of state responsibilities’ in refugee management (Missbach and Adiputera, 2021: 502).

Nonetheless, Indonesia’s willingness to act as the gatekeeper for Australia-bound asylum seeking migrants is highly contingent. This contingency is most evident when bilateral relations are frosty (e.g. when Indonesia threatened to release ‘a human tsunami’ of asylum seekers on Australia during a death row dispute involving several Australians in Indonesia) ⁹ and when Australia intensified its boat tow-backs and significantly reduced resettlement options in Indonesia after 2014, leaving more refugees ‘trapped in limbo’ for longer periods. ¹⁰ Against this background and faced with persistent international criticisms about immigration detention (Human Rights Watch, 2013), the Indonesian government became more open to the normative and policy shift towards decarceration for refugees. By the time Australia withdrew its IOM funding in 2018 ostensibly to stop attracting more asylum seekers to Indonesia, the Indonesian government had abolished the use of immigration detention centres for asylum seekers and refugees, releasing them from detention into locally funded community shelters under the 2016 Presidential Regulation and, in the process, ‘further removes the Indonesian state from responsibility for refugees by shifting responsibilities on matters concerning foreign policy to the local-government level’ (Hirsch, 2018; Kneebone et al., 2021: 446; Missbach and Adiputera, 2021).

The paradox of ‘non-existent’ refugees in Malaysia

Unlike the proxy funding of offshore immigration detention in Indonesia that has spanned almost two decades, Australia has not been able to extend its offshore refugee operations to Malaysia even though the country itself makes heavy use of immigration detention. The so-called ‘Malaysia Solution’, which was essentially an incentivized arrangement to establish a regional processing centre in Malaysia in 2011, never materialized. The arrangement was intended to be a political ‘solution’ to the growing domestic criticisms of immigration detention on the Australian mainland and Christmas Island and Australia’s limited options after the offshore processing deal was rejected by Timor Leste. Under the arrangement, Australia would transfer 800 asylum seekers who arrived ‘irregularly’ by boat for their refugee status determination in Malaysia. In exchange, Australia would accept up to 4000 refugees recognized by the United Nations High Commmisioner for Refugees (UNHCR) who were awaiting resettlement in Malaysia over four years. In addition, Australia would pay AU$95,000 per person for the 800 asylum seekers to be processed in Malaysia and AU$54,000 each for the 4000 refugees to be sent to Australia, that is, a total of AU$292 million (Lopez, 2011).

The bilateral agreement was pitched as part of the Australian government’s overall approach to break the human smugglers’ business model by removing the ‘product’ that the smugglers sell, namely, the opportunity to apply for asylum in Australia. ¹¹ The Malaysian Home Minister described it as ‘a pioneering and cutting edge solution to tackle people smuggling’. ¹² While the deal was supported by the IOM, it was opposed by human rights and refugee advocates in Malaysia and Australia, who regarded the deportation of 800 asylum seekers as a gross violation of their rights and the principle of non-refoulement under the UN Refugee Convention. ¹³ A coalition of Malaysian lawyers, the Malaysian Bar Council and human rights groups called on Australia to scrap the refugee exchange deal given Malaysia’s poor human rights record. ¹⁴ In the end, it was the High Court of Australia that halted the transfer arrangement by ruling it unlawful (Foster, 2012; Kneebone et al., 2011; Lowes, 2012; O’Sullivan, 2011). ¹⁵

Unlike local Indonesia and Timor Leste’s resistance to the regional processing centre proposals, the offshore processing arrangement in Malaysia was thwarted by a judicial decision in Australia and thus reflects what Marmo and Giannacopoulos (2017: 1) characterized as the ‘cyclical power struggle’ between the judicial and executive will on migration issues. The arrangement was negotiated in haste under a cloak of secrecy at a time of major domestic political upheaval in Malaysia. ¹⁶ Some observers suggest the government might have agreed to the refugee exchange ‘out of self-interest, as a potential strategy to address the large number of asylum seekers in transit that it hosts’ (Missbach and Hoffstaedter, 2020: 73) or to improve Prime Minister Najib Razak’s ‘personal international profile to compensate for declining domestic popularity’, even though foreign policy is generally ‘a non-issue’ in Malaysian politics (Lopez, 2011). Notwithstanding some attempts to ‘reboot’ the bilateral cooperation on joint border security operations under a transnational crime control framework in 2013 ¹⁷ and brief speculation of a refugee resettlement deal with the Razak government in 2016 ¹⁸ during the biggest corruption scandal (1Malaysia Development Berhad scandal) in Malaysian history, Australia has neither targeted nor made any inroad with its offshore border control policy in Malaysia.

Compared with Indonesia’s strategic position in Australia’s maritime jurisdiction and border security, Malaysia holds a relatively lower geo-political importance in Australia’s construction of national interest and ‘stop the boats’ regional deterrence paradigm. Furthermore, Malaysia’s contradicting realities of official non-recognition of its vast population of refugees and asylum seekers represent key barriers to Australia’s outsourcing of refugee operations and the role of international actors on the ground. All ‘irregular’ migrants are considered illegal under Malaysian law even though exceptions are sometimes made for select groups such as Rohingya on discretionary grounds (Hoffstaedter, 2014, 2019). As Ansems de Vries (2016: 882) suggests, what is distinctive about migration control in Malaysia is the ‘combination of an absence of nationally managed procedures and structures with extensive international governance within a rather informal and contingent environment’. The government tolerates the UNHCR’s presence to document and conduct refugee status determination, but it does not recognize the refugee agency’s formal presence in the country even though Malaysia has one of the largest urban populations of asylum seekers and refugees and hosts one of the UNHCR’s busiest status determination operations in the world (Ansems de Vries, 2016).

At the end of 2021, there were at least 180,000 refugees and asylum seekers registered with the UNHCR in Malaysia compared with around 13,000 in Indonesia. ¹⁹ The majority of these people lead a precarious existence in urban squatter dwellings (and previously in makeshift jungle sites) and are subject to ‘arbitrary arrests, immigration raids, extortion and abuse by local gangs as well as by police and immigration officials’ (Hoffstaedter, 2019; Hoffstaedter and Perrodin, 2018: 188; Nah, 2010). They receive limited material assistance from the UNHCR and other non-governmental and civil society organizations, which are very small in size or lack autonomy (Hoffstaedter and Perrodin, 2018; Kudo, 2013; Nah, 2011).

Successive Malaysian governments have avoided adopting any measures that would amount to a formal recognition of refugee existence among a large pool of irregular migrants in the informal economy, visa overstayers, sojourners and people who have remained undocumented for historical reasons. The informality of a temporary refugee presence serves a number of government interests. Politically, as Hoffstaedter and Perrodin (2018: 193) contend, refugees can be used to support different government rhetoric and policies—‘at times used as a more tolerable population than other undocumented migrants, they can also be portrayed as “radical others” and as such become scapegoats depending on government needs, with no repercussions for the Malaysian body politic’.

Economically, Malaysia’s development strategies have been inextricably linked to its use of cheap, disposable migrant labour, including large numbers of Indonesian migrant workers, in the informal economy. There were at least 2.2 million documented migrant workers in Malaysia in 2017 (out of a total population of around 32 million) with another estimated 2 to 4 million additional undocumented migrant workers in the country. ²⁰ Irregular migrants lead a generally invisibilized existence while also being subject to periodic crackdowns especially during economic crises. ²¹ During the mass raids at plantations, factories, construction sites or in urban neighbourhoods, hundreds of migrants are arrested by immigration and police officials and the controversial government-allied paramilitary group ‘People’s Volunteer Corps’ (popularly known as RELA). ²² Against this distinctly contradictory and politically contingent environment of migration control characterized by an absence of formal recognition of refugees and international actors in refugee management and heavy reliance on migrant labour alongside state-sponsored mass crackdowns on migrants, Malaysia presents a very limited opportunity for Australia’s outsourcing strategy for migration management.

Conclusion

Australia has long been understood as an embodiment of the northern penal state with the capacity to export its domestic security agenda into the international domain; outsource parts of its border control to southern states; and extend its sovereign strength further offshore. The results are often devastating for migrants who are intercepted at sea or held in various forms of onshore/offshore immigration detention. Yet, we know relatively little about the crucial links between migration management and transit/third countries and the multi-layered interests, especially from a global comparative perspective.

Significantly, Australian’s border externalization is highly uneven and contingent in nature. This is evident from the catalogue of (un)successful incentivization; the use and disuse of offshore processing and detention; and the expansion and contraction of space for interventionist policy making in particular aspects of outsourced refugee operations over time. Economic and political vulnerability of the southern states may be an important intervening factor here. The impoverished and former Australia controlled Pacific Islands states of Nauru and Papua New Guinea seem to be generally more accommodating to Australian requests than Malaysia and Indonesia, but vulnerability alone does not explain why Fiji and Timor Leste have been able to reject Australia’s outsourcing proposals. Further comparative analyses of how and why some Pacific Islands states but not others contest Australia’s externalization can shed light on southern states’ agency, the contrasting conditions of control and the long-term impact of offshored migration management in a region dominated by offshore finance designed as part of state strategies to optimize their positioning in the global political economy (Sharman, 2010).

States that are characterized as ‘transit’ in cross-border flows of people are often perceived as both the problem and the solution to irregular arrivals. Yet, from the perspective of the global South, the divergent outsourcing processes and outcomes in the two cases of Malaysia and Indonesia are a stark reminder of the political leveraging and the complex interaction between the pre-existing domestic and external factors in shaping the situated meanings and realities of border security building. As Missbach and Phillips (2020: 23) argue:

those unwilling or reluctant gatekeeper transit states must be taken seriously, as … more and more transit countries are aware of their bargaining power and are beginning to prioritize their political and economic interests linked to migration over the interests of other countries.

In Indonesia, its high strategic importance as the ‘final bulwark’ in Australia’s regional deterrence frontier; its partial recognition of refugee rights and long-lasting indifference to their long-term presence; and its abdication of state responsibilities in refugee management to external funders, have provided the conducive factors for Australia’s proxy outsourcing of immigration detention for nearly two decades. In Malaysia, its relatively low strategic importance in Australia’s regional deterrence frontier; its formal non-recognition of refugees and international actors in refugee management; its heavy reliance on migrant labour in the informal economy; and state-sponsored periodic crackdowns under its ethnicized politics of exclusion, provide the constraining factors for Australia’s outsourcing of refugee processing.

Taking into account the variations, contestations and adaptations in the control of unwanted migrants provides a crucial corrective to an all-encompassing externalization thesis and contributes to a de-centring of northern-centric notions of the state’s role in border control. Methodologically, future debates over border externalization need to involve a more systematic comparison of migrant sending and transit countries; the roles, decision making and everyday experiences of state and non-state actors and local communities, which may have different understandings of (il)legality and control; and a more sustained effort to identify and systematize the differences among them. It is high time that criminologists shift the locus of analysis to the southern contexts and foreground the intersections between migration management and the continuities and discontinuities of power relations across the North–South divide.