Introduction

Although little has been written about the political roles of so-called transit states in contemporary securitized migration management, it seems to be widely assumed that transit states follow the orders of their more powerful neighbors, who seek to deter irregular movements outside of their respective jurisdictions. It is also often anticipated that, if foreign aid is tied to providing infrastructure, covering the costs of offshore processing of refugees, and enabling policy capacity, it creates dependencies in those states, which are unable (or unwilling) to dedicate more of their own domestic budget to the prevention of people smuggling (Curley and Vandyk 2017). In line with their perceived hierarchy of interests, destination states or communities of states, such as the European Union, seek to impose their interests on their neighbors, hoping to ensure that their neighbors comply with their migration agenda by offering aid and other incentives (Andersson 2014; Choplin 2012; Yildiz 2016). This article, however, argues that, despite providing lucrative funding, material incentives, and other support to combat people smuggling, destination countries are unable simply to impose their strategies upon neighboring transit states, but may face open refusal and more subtle forms of noncompliance. This article demonstrates in particular that Australia's outsourced policies to prevent asylum seekers’ irregular departure from Malaysia and Indonesia did not meet the expectations of the respective governments in Indonesia and Malaysia, which favored mutually beneficial cooperation on irregular migration. More importantly, we spell out how Australia's externalized border and asylum policies created a number of detrimental outcomes for the regional collaboration in Southeast Asia.

This article is based on ethnographic fieldwork in Indonesia and Malaysia; while Hoffstaedter spent more than a year researching asylum seekers in Kuala Lumpur in 2015, Missbach conducted altogether 16 months of multisited fieldwork on people smuggling networks in Jakarta, Nusa Tenggara Timur, Makassar, and Batam between 2013 and 2016. While the prime focus of our respective research projects was directed at asylum seekers and refugees and their agency to overcome immobilization, in this article we make use of interviews with state officials from both countries in order to analyze the state perspective and thus compare the different responses of the Indonesian and Malaysia governments toward the Australian externalized border and asylum policies in recent years.

Australia depends on the active and passive support of its neighbors in Southeast Asia for its anti-people-smuggling measures in the region. Thus, Australia has sought bilateral and multilateral agreements with its neighbors in Southeast Asia to target people smuggling in the region through border control and law enforcement, offshore processing of refugee claims, and wider agreements on managing irregular migration (Curley and Vandyk 2017; Gammeltoft-Hansen 2011). There are, of course, several regional forums that take an interest in migration issues, first and foremost, the Bali Process on People Smuggling, Trafficking in Persons and Related Transnational Crime (Bali Process) headed by Australia and Indonesia, which we will address in this article in some detail, and the Association of Southeast Asian Nations (ASEAN).¹ Generally speaking, both Indonesia and Malaysia have a genuine interest in managing migration and safeguarding their borders and thus are interested in regional cooperation. Irregular migration for them, however, means more than just asylum seekers and refugees.

The existing bilateral collaborations, although promoted to the collaborators as balanced and mutually beneficial, are often dominated and led by Australia. While Australia perceives itself as a regional leader (Phillips 2017a), its neighbors perceive it as a bully (Megalogenis 2019). Rather than concentrating on mid- and long-term collaborations, Australia has prioritized short-term deterrence measures to combat the irregular movement of migrants. Australia seeks to outsource its “asylum seeker problem” by focusing on combating transnational crimes committed outside its national territory. Both Malaysia and Indonesia have at times complied with Australia's demands, but this article seeks to highlight instances where they have prioritized their own interests and intends to counter the widespread view of the trouble-free cooperation between a seemingly powerful destination state and supposedly complacent transit states.

Two main arguments are offered. First, this article questions the role of Indonesia and Malaysia as docile executors of border and migration policies formulated outside their national jurisdictions. Having retraced several disagreements over humanitarian and security-related responsibilities for asylum seekers and refugees, we show that Malaysian and Indonesian responses to Australian demands for the restriction of the onward movement of transiting migrants have varied widely over the last decade. We contest Australia's role as senior partner in charge of these bilateral relationships by showing that the relationships are not as heavily weighted in Australia's favor as widely assumed. Both Malaysia and Indonesia, because of their presumed role as a bulwark against Australia-bound migrants, have gained more political leverage in determining the nature of the relationships. Unlike the countries that surround the European Union and depend on EU aid, Malaysia and Indonesia are not dependent on Australian aid, which has decreased significantly over the past few years, and are, therefore, less receptive to interventionist policy making unless it also serves their own domestic interests. Highlighting domestic migration discourses in Indonesia and Malaysia and focusing on internal discussions of a number of conflicting policy interests help to flesh out a more nuanced corrective narrative about the often tense bilateral relationships of both countries with Australia, as well as the domestic priorities of both countries with regard to managing irregular migration.

Second, although Australian policy makers continue to view Indonesia and Malaysia as transit countries or corridors, in reality both countries now resemble cul-de-sacs, as they have become de facto (in)voluntary destination countries for “immobilized” asylum seekers, refugees, and other migrants wanting to stay there (Missbach and Phillips, this volume). Malaysia has always been a destination country for, especially Rohingya, refugees from Myanmar, while other ethnic minorities from Myanmar and refugees from farther afield have placed their hopes on a swift resettlement process through the UN High Commissioner for Refugees (UNHCR) or on an eventual onward journey to Indonesia and Australia. Malaysia was, until the Syrian crisis, the largest resettlement post in the UNHCR system, with approximately 10,000 refugees resettled a year to the US, Canada, Australia, and some European countries. However, the closure of Australia's borders to spontaneous arrivals by boat and the unwillingness of third countries to resettle recognized refugees from Indonesia and Malaysia have caused a substantial backlog of immobilized asylum seekers and refugees in both countries. This backlog consists of those awaiting their processing and resettlement, and others who have undergone the status determination process but have been rejected. Although “screened-out” asylum seekers are subject to deportation, the actual number of rejected asylum seekers who are deported is extremely low, partly because deportation is considered too expensive and administratively difficult.² Immobilized migrant populations have more or less voluntarily created homes while they are in transit. These homemaking practices force Malaysia and Indonesia to cope with a range of related issues, such as their health, education, and welfare. Both the Malaysian and the Indonesian governments have come to understand the catch-22 situation they are in: the more they comply with Australia's demands for them to prevent irregular onward migration of asylum seekers and refugees, the greater the responsibility they carry, as they must host refugees and asylum seekers for longer periods of time.

Redefining the State of Transit in Malaysia and Indonesia

Both Indonesia and Malaysia functioned as transit hubs for refugees long before the term “transit state” was applied to such countries (Missbach and Phillips, this volume). After the fall of Saigon in 1975, tens of thousands left Vietnam and, later, Cambodia to seek protection in the region, including in Indonesia and Malaysia, before most were eventually resettled in France, Australia, the US, and Canada or repatriated to Vietnam (Tran 1995).³ Malaysian and Indonesian government representatives continue to take pride in the goodwill former governments showed by hosting Vietnamese and Cambodian refugees from the late 1970s to the 1990s in their territories.⁴ Nevertheless, the terminology and concept of transit have only recently made their way into the language of policy circles in Southeast Asia. During the early stages of our respective fieldwork in Indonesia and Malaysia, most policy makers showed confusion on hearing the terminology of transit, which has now become part of their standard lexicon.

Since there are, so far, no definitions from Southeast Asia for what constitutes a transit state, as a starting point we rely on Kimball's definitions that have arisen out of other geographic contexts, despite a number of inherent shortfalls, as indicated in the introduction to this special section. Kimball bases her understanding of what defines a transit state on four criteria (geography, migration flow, function, and state response). First, transit states must border a fully developed country; second, transit states must show a higher rate of emigration than immigration; third, transit states must function as primary staging grounds for migrants who intend to travel on to a nearby desired destination country; and fourth, over time transit states adopt and enforce more restrictive migration and border policies (Kimball 2007: 12).

When applying these four criteria to Indonesia, it becomes apparent that Indonesia does not fully meet them (Missbach 2015: 152–154). First of all, Indonesia, an archipelago of more than 17,000 islands stretching over 5,000 kilometers from west to east, has no land border with Australia; it is, nevertheless, an immediate neighbor of Australia, but is across the sea that surrounds them both. The jurisdiction of maritime zones and borders is more complex than land borders.

With regard to the second of Kimball's determinants, inbound and outbound migration, Indonesian migration rates clearly show an emigration surplus. The net migration rate in 2016 was estimated to be −1.2 migrant(s)/1,000 population (Index Mundi 2017). In 2015 there were at least six million Indonesians working overseas, mostly in Malaysia, Singapore, Hong Kong, and the Middle East (IFRC 2015). Indonesia grants work rights to expatriates with specialist skills but remains unwilling to grant residency rights to asylum seekers and refugees currently within its territory, fearing that they would “overstay their welcome.”⁵

With regard to Kimball's third criteria, between 1998 and mid-2013 Indonesia has served as a staging ground for more than 55,000 asylum seekers undertaking irregular journeys by boat (Phillips 2017a). Asylum seekers who can apply for tourist, student, or other types of visas to enter Australia by air usually do not come via Indonesia, but fly straight from their countries of origin or a neighboring country. However, people fleeing Afghanistan, Pakistan, Somalia, Myanmar, and Sri Lanka have little hope of ever being granted a visa to enter Australia and resort to irregular journeys, the last leg of which is by boat from Indonesia.

Kimball's fourth determinant, relating to the implementation and enforcement of restrictive immigration and border policies, is especially relevant to Indonesia and will, therefore, be foregrounded in this article. To date, Indonesia has not signed the 1951 Convention Relating to the Status of Refugees or the 1967 Protocol, and has no legislative framework for the protection of asylum seekers and refugees. However, over the last decade Indonesia has produced new regulations and policies that seek to prevent irregular entry, residence, and exit of asylum seekers and migrants. At times it has sought to exercise stricter border controls, but because of the enormous cost of comprehensive control, a lack of political commitment, and widespread corruption among border patrol officers and immigration authorities, its extensive sea borders remain porous. It has been suggested that Indonesia's Law on Immigration (2011) and the newly enforced restrictions on the mobility of asylum seekers reveal a level of Australian interference (Connery et al. 2014a; Mathew and Harley 2016; Taylor 2005). Although the extent of the Australian interference remains questionable, political pressure has not automatically resulted in full acceptance and implementation by Indonesian officials.

Despite some obvious deviations, Indonesia fits Kimball's characterization of a typical transit state better than Malaysia does. Malaysia is a difficult case, especially in terms of Kimball's first two criteria—bordering a fully developed country and net emigration. Malaysia's high economic development status has made it a magnet for labor migration from other Asian countries. In 2017 Malaysian Home Ministry figures recorded over 1.7 million legal foreign workers in Malaysia (Nasa 2017); other sources estimate that undocumented foreign workers in Malaysia would double or even triple this number. Malaysia's importance as a transit country for migrants on their way toward Australia is a result of its relatively lax visa regime, which allows travelers of many countries to enter without visas. This regime, intended to attract tourists, has made Malaysia a first place of protection and asylum for many fleeing their home countries. Thus, Kimball's third criterion of transit countries functioning as staging grounds is crucial in understanding Malaysia's role in transit migration in Southeast Asia.

Many asylum seekers and refugees enter Malaysia legally as tourists and wait, work, and sometimes register with the UNHCR before moving on to Indonesia, the secondary staging ground for boat journeys to Australia. Now that boat journeys from Indonesia to Australia have largely ceased, refugees tend to remain in Malaysia. This, crucially, is consistent with Kimball's fourth criterion—that over time transit states adopt and enforce more restrictive migration and border policies. Like Indonesia, Malaysia has ceded to Australian demands by limiting the visa-free entry of nationals from Sri Lanka, Pakistan, and Afghanistan. The periodic increase in immigration control, immigration crackdowns, and the visibility of immigration efforts are predominantly influenced by domestic politics. The Malaysian state periodically seeks to demonstrate its ability to control irregular migration to Malaysia and especially the migrant labor force (Kassim and Zin 2011).⁶ Human trafficking and efforts to combat it have been a focus, and some reforms, notably to the victim protection system, were passed in 2015. Like Indonesia, Malaysia has not signed the 1951 Convention Relating to the Status of Refugees or the 1967 Protocol. No domestic legislation has addressed the rights of refugees and asylum seekers in Malaysia, and their policing remains under the purview of the existing immigration laws that deem them illegal immigrants and thus subject to the penalties in those laws. A large part of the Malaysian government's reluctance to implement legislation on this matter is their fear of creating a pull factor that would attract many more refugees from the region, especially the refugee camps along the Thai-Burma border. Indeed, ever since the mass transit of Indochinese refugees through Malaysian refugee camps to the West, the Malaysian government has sought to portray itself as a transit state that periodically, and on purely humanitarian grounds, will provide sanctuary to some refugees, based on the understanding that the international community will resettle those Malaysia does not intend to integrate over time (Hoffstaedter 2017).⁷

Most asylum seekers and refugees currently residing in Indonesia consider themselves in transit, not least because Indonesia offers them no legal options to make their residence permanent and jobs in the informal economy are hard to come by. Asylum seekers and refugees in Malaysia are in the same legal limbo, but many want to stay in Malaysia, as they can find opportunities to work in the informal economy (Hoffstaedter 2014). Others, particularly those from Afghanistan, Sri Lanka, and Iran, find it easy to enter Malaysia, the first destination on their journey. They often do not register with the UNHCR in Malaysia and seek out opportunities to travel to Indonesia. Thus, both Indonesia and Malaysia serve as transit countries and destination countries, depending on how asylum seekers and refugees view their future in these countries. For instance, Rohingya, Malaysia's largest refugee population, see it as a destination country, considering it a Muslim country where they can practice their religion freely after decades of oppression in Myanmar. In 2017, the Malaysian government began to raise their plight with ASEAN and the Organisation of Islamic Cooperation. For the second-largest refugee population, the Christian Chin from Myanmar, Malaysia remains a transit country where they wait for resettlement in the US or Australia.

Malaysia and Indonesia continue to treat asylum seekers and refugees on the basis of their experience with the Indochinese and the 1989 Comprehensive Plan of Action that led to the resettlement of all asylum seekers and refugees after Malaysia and Indonesia had provided them temporary refuge on the understanding that the international community would take full responsibility for their well-being and resettlement elsewhere (Robinson 2004). Based on this experience, Malaysia and Indonesia consider themselves generous for providing temporary acceptance but place responsibility for anything beyond that with the UNHCR and its international funders. Both countries reject any financial responsibility for the care of asylum seekers, citing their lack of capacity and of domestic support for refugees and refugee rights. Nonetheless, the Malaysian government has shown a mixed response, invoking humanitarian reasons to support some refugees, mainly Muslims from Southeast Asian countries, while ignoring others (Hoffstaedter 2017). Malaysia has fully integrated some refugee populations, providing citizenship and settlement services, as in the case of Moro refugees in East Malaysia in the 1970s and Cham Muslims in West Malaysia in the 1980s. In Indonesia it is theoretically possible to gain citizenship after being a lawful resident for ten years, but that requires giving up any claim for resettlement elsewhere. Since the relatively friendly reception of the Indochinese between the late 1970s and the early 1990s, the treatment of asylum seekers and refugees in Malaysia and Indonesia has changed considerably, not least due to ongoing pressure from Australia and its externalized border and asylum policies.

Successes and Failures of Anti-People-Smuggling Efforts in Indonesia

Despite lingering tensions between Indonesia and Australia over diplomatic issues, Australian engagement to combat people smuggling has penetrated many Indonesian government institutions. Preventing asylum seekers from leaving Indonesia has been at the core of Australia's externalized border policies. Under Prime Minister John Howard, direct disruption campaigns, such as the sabotaging of boats, were used in Indonesia (Howard 2003: 44). After the Lombok Treaty (“Agreement” 2006), signed by the Indonesian and Australian governments in 2006, became the basis of the Indonesia-Australia anti-people-smuggling collaboration, rather than intervening directly in Indonesia, Australian governments concentrated on building the capacity of Indonesian authorities and funding their countersmuggling activities (Connery et al. 2014b; Phillips 2017b). For instance, Australia provided equipment for detecting fraudulent documents and other biometric devices (Nethery and Gordyn 2014).

When the number of asylum seekers crossing to Australia started to increase again in 2009, the Indonesian police set up a task force with 12 regional branches in people-smuggling hot spots to combat people smuggling (Spinks et al. 2013). The Australian Federal Police (AFP) supported the task force with office facilities, vehicles, investigation kits, and new patrol boats. Some of this equipment, although gratefully accepted, has not been used to its full extent; for example, police officers in Kupang, Eastern Indonesia, lamented that the new boat was of no use because no provisions were made for additional fuel, as observed by Missbach during previous fieldwork in Nusa Tenggara Timur in 2012.

At times, more than 20 AFP officers were posted to work side by side with the Indonesian anti-people-smuggling task force, to coordinate activities aimed at preventing people smuggling at sea, and to share information and intelligence data, particularly in the apprehension of organizers of people-smuggling operations (Connery et al. 2014a). Because a comparatively small number of Indonesian police officers were posted to Australia, some Indonesian officials became concerned that the large number of Australian police in their country was affecting Indonesian sovereignty. Nevertheless, in 2011 Indonesia and Australia, in conjunction with other members of the Bali Process,⁸ agreed on a Regional Cooperation Framework to intensify regional collaboration and enable practical arrangements for member states to combat people smuggling (Phillips 2017b).

Over the last decade Australia has provided training for Indonesian partner institutions, including the police and immigration authorities, to improve immigration intelligence and anti-people-smuggling law enforcement. Most of the training workshops were run in conjunction with the International Organization for Migration (IOM). By 2013 more than 30,000 Indonesian immigration, police, and army officers, prosecutors, and local government officials had taken part (IOM 2014: 3). IOM also distributed printed materials that addressed how the Indonesian police should intercept, investigate, and respond to people smuggling. To prevent Indonesian fishermen from becoming involved in people smuggling, Australia has financed several information campaigns to target remote locations in Indonesia that have been departure points for asylum seekers heading to Australia (McNevin et al. 2016).

Australian perceptions of its collaboration with Indonesia in combating people smuggling was, at least until late 2013, optimistic, as it was assumed that Indonesia was sincerely assisting the Australian navy to prevent maritime arrivals (Maley and Taylor 2013). A closer look at the interception of asylum seekers in Indonesia in recent years reveals a greater reluctance. Indonesian law enforcement officers soon realized that the more successful they were in preventing the onward migration of these asylum seekers, the more Indonesia became responsible for their whereabouts. From January 2012 until September 2013, the Indonesian police arrested about 12,790 “irregular migrants” (records do not differentiate between registered asylum seekers, recognized refugees, and undocumented migrants) (Missbach 2013). Arresting these asylum seekers did not always immobilize them efficiently. Given the level of overcrowding in Indonesia's detention centers, breakouts from detention centers were frequent. Moreover, many police officers became too disheartened to make arrests, preferring to let asylum seekers pass through Indonesia, often without even demanding bribes (Missbach 2015). Hence it is not surprising that between early 2012 and June 2013 about 30,310 asylum seekers arrived in Australia, most of them coming from Indonesia.

The Indonesian police, supported by their Australian colleagues, made an effort to arrest those who organize smuggling operations out of Indonesia. In 2012, the Indonesian police arrested 103 Indonesians and six foreign nationals suspected of people smuggling and brought 36 people-smuggling cases to the courts, followed by a further 37 cases in 2013 (Missbach 2016b). While the Indonesians arrested were usually employed as drivers and boat crew, the foreign nationals were the recruiters and managers of the people-smuggling operation, but not necessarily the main organizers. The majority of those arrested and prosecuted were just low-level drivers and boat crew, whose imprisonment did not interrupt people-smuggling networks, as more drivers and boat crew could be easily recruited from other parts of Indonesia (Missbach 2016a).

In parallel, with its efforts to “stop the boats,” Australia has sought to establish a formal mechanism for accommodating and processing asylum seekers in Indonesia. Since 2001, a Regional Cooperation Arrangement between Australia and Indonesia has been in place, which, with funding from Australia, provides for IOM to care for asylum seekers and refugees in Indonesia while allowing them access to the UNHCR's refugee status determination process (Nethery et al. 2013). Australia's role in this arrangement is primarily as provider of funding. Some incentives offered by Australia, however, were not accepted because of the expectation that they would disfavor Indonesia. For example, for many years Australia's proposal to build a regional processing center for asylum seekers in Indonesia, where their claims for international protection could be dealt with by Australian immigration officers, was rebuffed by successive Indonesian governments (BBC News 2010). Offers to build more detention centers were also rejected. Thus, it has become apparent that Australia's financial incentives to Indonesia do not guarantee Indonesia's compliance.

Indonesia's overall willingness to cooperate began to change in late 2013 after Tony Abbott was elected prime minister of Australia and introduced Operation Sovereign Borders as a whole-of-government response under the leadership of the military, exposing a zero-tolerance approach to the arrival of asylum seekers by boat. Intercepted asylum seekers were either returned to Indonesia or, if they had departed from their country of origin (Sri Lanka and Vietnam), subjected to on-water screening. If found to have prima facie protection claims, they were sent to Nauru or Papua New Guinea; if not, they were handed back to authorities in the country of origin. This unilateral approach was not linked to the already established Bali Process, so Australia's uncompromising approach generated new tension. For example, in November 2013, the Indonesian government refused to take back a boat that the Australian navy had intercepted. Since then the Indonesian government has repeatedly complained about the lack of consultation regarding asylum seeker turnbacks and stated that it would not accept any policy that ran counter to Indonesia's interests (Alford 2013).

To make matters worse, while the Australian government was struggling to mend the bilateral relationship that had suffered substantially from Abbott's megaphone politics, it was revealed late in November 2013 that Australia had engaged in espionage in Indonesia, tapping the phones of the president, his wife, and several other eminent political figures (Tanter 2014). President Yudhoyono was outraged and demanded an apology that Abbott refused to give. Therefore, Yudhoyono ordered all people-smuggling and intelligence collaboration to cease. Australia resorted to applying more unilateral measures to stop the boats, which further undermined the level of trust in the bilateral relationship. Most significantly, Australia admitted that it had breached Indonesia's territorial integrity on at least five occasions when the navy had returned asylum seeker boats to Indonesian waters without prior consent from or collaboration with the Indonesian government (Australian Senate Inquiry 2014). Jakarta protested these intrusions vehemently, stating that “the government of Indonesia deplores and rejects the violation of its sovereignty and territorial integrity by the Australian vessels” (Salna and Osborne 2014). As is not uncommon among postcolonial nations, issues of sovereignty and territorial integrity are of paramount importance to Indonesia. Despite eventually signing a Joint Understanding for the future implementation of the Lombok Treaty in August 2014, including a new “code of ethical conduct” and intelligence protocol, on which Indonesia had insisted, the anti-people-smuggling collaboration was never restored to the extent that it had operated previously (Curley and Vandyk 2017; Missbach 2018). Moreover, Indonesian officials were generally disappointed when Australia announced in November 2014 that it would stop resettling refugees from Indonesia altogether.⁹ Many more diplomatic issues have increased the trust deficit between Indonesia and Australia, and mutual suspicion runs high on both sides. For example, Tedjo Edhy Purdijatno, Indonesia's coordinating minister for political legal and security affairs, threatened Australia: “If Canberra keeps doing things that displease Indonesia, Jakarta will surely let the illegal immigrants go to Australia. … There are more than 10,000 [asylum seekers] in Indonesia today. If they are let go to Australia, it will be like a human tsunami” (Doherty 2015). Considering the many efforts and payments made, one unintended outcome of the long-standing anti-people-smuggling collaboration has been an increase in mutual suspicion between Indonesia and Australia. The end of boat crossings from Indonesia to Australia together with the decrease of resettlement of refugees from Indonesia has put asylum seekers and refugees in stasis, a situation that Armelle Choplin (2012: 166) describes as “from thoroughfare to cul-de-sac,” coining the term “post-transit.”

Malaysia Case Study: Swap Deal

The relationship between Malaysia and Australia is also marked by suspicion and has endured its fair share of misunderstandings over the years. Yet, despite their sometimes tense political relationship, Australia and Malaysia have a long history of cooperation in the areas of defense and security. Formal defense cooperation was strengthened by the 1992 Malaysia-Australia Joint Defence Program, while more recently the two countries have cooperated in response to emerging threats such as international terrorism (Snyder 2015). As a member country of the Bali Process, Malaysia has worked alongside Australia to deal with people smuggling and transnational crime, culminating in the 2009 establishment of the Malaysia-Australia Working Group on People Smuggling and Trafficking in Persons. The working group seeks to address people smuggling by improving cooperation between the two countries in relation to closer intelligence sharing, legal cooperation, and capacity building (Taylor 2012).

The Australia-Malaysia refugee “swap deal” had its origins in the Regional Cooperation Framework agreed to at the Fourth Bali Process Regional Ministerial Conference on People Smuggling, Trafficking in Persons and Related Transnational Crime in March 2011. The framework aimed to support the development of practical arrangements in response to people smuggling and the movement of refugees in the region, and while the primary focus of the framework is border control, some principles for the protection of refugees and asylum seekers were also incorporated (Taylor 2012). However, the arrangement was predicated on Australian domestic needs rather than regional needs, and its focus was never to provide a framework for a regional refugee burden-sharing mechanism.

In May 2011, then Australian Prime Minister Julia Gillard announced that the Australian and Malaysian governments were to sign a bilateral agreement in order to stop people smuggling to Australia by removing the “product” that the smugglers sell, namely, the opportunity to apply for asylum in Australia. The swap deal would have had future asylum seekers arriving in Australia by boat transferred to Malaysia for the processing of their claims, also removing any possibility of their resettlement in Australia. The Australian government anticipated that this would remove any incentive for asylum seekers to travel to Australia by boat, thereby preventing future boat arrivals (Spinks 2011).

At the center of the proposed agreement was the transfer of eight hundred asylum seekers from Australia to Malaysia for refugee status determination. In response Australia would, over a four-year period, resettle four thousand recognized refugees currently living in Malaysia. The deal was to be fully funded by the Australian government and provision for the four thousand refugee places was announced a few days later in the 2011/12 budget. This arrangement also portrayed Malaysia as a transit country that could stage status determination, but under the understanding that this was funded by Australia and that Australia would continue and increase resettlement of refugees from Malaysia. The total cost of the program was estimated to be nearly AUD 300 million over four years.

The agreement between the two countries came into effect after both governments signed a document entitled Arrangement between the Government of Australia and the Government of Malaysia on Transfer and Resettlement (hereafter referred to as the Arrangement) on 25 July 2011. The purported statutory basis for the agreement was provided by an Instrument of Declaration issued by the Australian minister for immigration on the same day, which identified Malaysia to be a “declared” country under section 198A of the Migration Act of 1958 (Australian Senate 2011: 6). The Arrangement document outlined a framework by which the swap deal would operate and included logistical details of the transfers. Asylum seekers to be transferred to Malaysia would be the first eight hundred to either arrive in Australia by boat or be intercepted at sea after 25 July 2011. While Australia would choose who to send after a medical and national security check, Malaysia had to provide consent and approval for the transfers and therefore had the power to veto transferees (Department of Immigration 2011). Once in Malaysia, transferees would have the opportunity to have their claims for asylum heard by the UNHCR. In return, the four thousand refugees to be resettled in Australia would have to have been registered with the UNHCR in Malaysia prior to the signing of the Arrangement and would have to meet Australia's legal requirements for resettlement in Australia. The Arrangement also detailed two joint commitments: that asylum seekers would be treated with dignity and respect and in accordance with human rights standards; and that special procedures would be developed by the two countries to deal with the special needs of unaccompanied minors and other vulnerable individuals.

While Australia initiated the swap deal, there is some evidence that Malaysia was not merely a passive stakeholder in the process. Media reports from June 2011 suggest that Malaysia tried to make sure the agreement was acceptable (Star Online 2011). Malaysia put forward several proposed amendments to the draft agreement, including the veto power over who is transferred to Malaysia and the removal of all reference to human rights (Cannane 2011). Later that year the Malaysian home minister acknowledged that the refugee swap deal “is just a small part of the bigger picture. The whole reason we are doing this is to send a message to the syndicates not to look at Malaysia or Australia as a country of destination or transit anymore” (New Straits Times 2011). This statement suggests that Malaysia may have participated in the deal not simply in response to Australia's request but out of self-interest, as a potential strategy to address the large number of asylum seekers in transit that it hosts.

The announcement and subsequent signing of the swap deal generated widespread public debate in both Malaysia and Australia. In Malaysia, opposition politicians, civil society actors, and lawyers were at the forefront of challenging the validity of the agreement and Malaysia's role in its implementation. However, in Malaysia the opposition, like the judiciary, remains weak, and the government prevailed with its narrative that focused on tackling the regional people-smuggling problem with a novel approach that Home Minister Hishammuddin Hussein claimed was ahead of its time (Lee 2011). On 17 August 2011 the Australian Senate referred the agreement to the Legal and Constitutional Affairs References Committee for inquiry into the proposed implementation of the Arrangement (Australian Senate 2011). Much of the controversy related to concerns for the protection and human rights of asylum seekers. While both governments made a commitment to respect human rights standards and the principle of nonrefoulement, the agreement was nonbinding and contained no requirement for Malaysia to adopt any new international or domestic legal obligations. This was particularly significant given that Malaysia is not a signatory to the Refugee Convention or indeed to many other key human rights conventions (Foster 2012). Despite its stated commitment, therefore, Malaysia was not obliged to respect the human rights of transferees (Lowes 2012). Additionally, both the adequacy of the determination process available to asylum seekers and their treatment in Malaysia were questioned, while the lack of detail regarding the special procedures that would be developed to protect vulnerable asylum seekers, including unaccompanied minors, was also a concern (Foster 2012). Many of these same issues were highlighted in the Australian Senate committee inquiry, which concluded in October 2011 that the Australian government should not proceed with the implementation of the Arrangement “due to the obvious flaws and defects in that arrangement” (Australian Senate 2011).

After the original announcement of the swap deal in May, it was anticipated that all subsequent boat arrivals would be transferred to Malaysia for processing. In reality, however, this did not occur. Although the first asylum seekers identified for transfer underwent a preassessment for the transfer to Malaysia, where they could then apply for refugee status determination with the UNHCR, the transfer did not proceed. A challenge at the Australian High Court was launched on 7 August 2011 by two Afghan citizens, an adult male (referred to as M70) and an unaccompanied 16 year old boy (M106), who had arrived at Christmas Island by boat on 4 August 2011, as well as a number of other asylum seekers similarly affected by the swap deal (Lowes 2012). The plaintiffs’ main argument focused on Malaysia's lack of legal obligations toward asylum seekers and refugees (Foster 2012) and, consequently, the lawfulness of the minister of immigration's written declaration that Malaysia was a “declared” country under the Migration Act that would provide effective procedures and protection to asylum seekers transferred there for processing (Lowes 2012). The court granted the plaintiffs a temporary injunction until the matter was determined by the High Court (Foster 2012). On 31 August 2011 the High Court ruled by a margin of six to one in favor of the plaintiffs, finding that any arrangement needed to incorporate legally binding protections to ensure that asylum seekers’ human rights are properly protected (Lowes 2012).

The High Court decision effectively annulled the bilateral refugee swap deal. As a result, the asylum seekers due to be transferred to Malaysia would have their claims for asylum heard in Australia. Undeterred, the Australian government attempted to circumvent the High Court's ruling with the submission of legislation to amend the Migration Act. The proposed amendment was to allow the government and minister of immigration to decide which countries could be used for offshore processing. Widespread political, as well as public, opposition to these moves meant they failed to be passed in Parliament (Lowes 2012). In October 2011, the Australian prime minister announced that, while the Malaysia Arrangement would remain government policy, it could not be implemented without legislative change. While Australia would honor their commitment to accept four thousand recognized refugees from Malaysia, there would be no increase in the quota, and these places would be taken from the existing refugee intake (Gillard and Bowen 2011). For those asylum seekers deemed irregular maritime arrivals, the government signaled its intention to use mandatory detention for the purposes of health and security checks, but then allow access to other tools for managing pressure on detention centers such as access to bridging visas and community release (Karlsen 2012).

Despite the failure of the swap deal, there appears to have been hardly any adverse consequences on bilateral relations between the two countries. Within the Malaysian media some criticism was leveled at the Australian courts for failing to show sufficient respect for Malaysia's commitment to meet its obligations to asylum seekers (New Straits Times 2011). Overall though, the failure of the Arrangement appears to have had little impact on the two countries, which have continued to cooperate in areas of border protection and people smuggling under the framework of Operation Sovereign Borders. Following a visit to Malaysia by the Australian minister for immigration in October 2013, it was announced that the Working Group on People Smuggling and Trafficking would be replaced by an Australian-Malaysian Joint Working Group on Transnational Crime, while the two countries restated their commitment to joint Australia-Malaysia operations at Malaysian air, land, and sea borders. At the same time Malaysia also announced that it would stop issuing visa-on-arrival arrangements for Iraqi and Syrian nationals, following an earlier move in 2010 to stop issuing visas on arrival for a range of nationals from the Indian subcontinent, including Afghan nationals. Further cooperation includes the 2015 gifting of two retired Australian patrol boats to Malaysia, to be stationed in the Straits of Malacca—identified as a transit point for asylum seekers from the Middle East heading to Australia—to try to curb people smuggling in the region (Dutton 2015). More recently, media reports from 2016 indicate that the Australian government may again be pursuing an asylum seeker deal with Malaysia (News.com.au 2016), although no formal announcements have been made in this respect. Such agreements reiterate the Malaysian position of being a reluctant country of first asylum, instead demanding they be treated as a transit country that would see refugees depart in time through resettlement and return to their homeland. Australia acknowledged this position even in the Malaysia swap deal by agreeing to the transferees only being in Malaysia for processing temporarily, and in return bringing a much larger caseload of refugees from Malaysia for resettlement, thereby taking on Malaysian demands that the international community share responsibility.

Conclusion

This article has challenged the common assumption that transit countries—for the right price—tend to be willful implementors of externalized border and asylum policies (Andersson 2014; Curley and Vandyk 2017; Choplin 2012; Kimball 2007; Yildiz 2016). Australia, as we have shown, did provide substantial financial and material support for combating people smuggling in the region in order to prevent the departure of asylum seekers from Indonesia and Malaysia. Yet, as we have argued, the success of those measures remains questionable. Not only did the financial incentives have little impact on reducing irregular maritime journeys, more importantly, they impacted negatively on the overall bilateral relations, particularly in the case of Indonesia. Ignoring the domestic political interests related to irregular migration in Malaysia and Indonesia has earned Australia substantial criticism on being blatantly geared toward its own political interests. In particular, Australia's policy under its Operation Sovereign Borders on turning back asylum seeker boats to Indonesia has severely undermined mutual trust.

Moreover, Australia's fixation on “stopping the boats” has ignored important changes of perception in Malaysia and Indonesia vis-à-vis the international expectations of them to help ease the global refugee crisis. By now Indonesia and Malaysia have become de facto destination countries, whether or not they accept this fact, and must move beyond identification solely as transit countries and cope with their post-transit realities. Australia's strategic encounters, however, also need to factor in this change. Australia's unilateral and bilateral approaches had negative diplomatic consequences for the whole region and seriously undermined broader regional cooperation focused on irregular migration. As long as they remain primarily premised on Australian interests and driven by Australian funding, these unilateral approaches and bilateral arrangements risk many pitfalls. The focus on unilateral action over regional instruments, such as the Bali Process, does not take seriously regional partners’ domestic issues with irregular migration, nor does it respect wider international norms and laws. Not least, as the Papua New Guinea and Nauru refugee detention and resettlement deals have shown, by depending on regimes that are unstable and corrupt, Australia puts itself in a vulnerable position and diminishes its ability to speak out as a regional democratic and human rights leader (Hoffstaedter 2013). Therefore, the way forward must surely include forging more multilateral arrangements involving source, transit, and destination countries.