Metagovernance of migration policy in the Asia Pacific region: an analysis of policy tools

ABSTRACT

This article considers how nation states steer transnational forums by examining how the Australian government has ‘metagoverned’ the Bali Process on People Smuggling, Trafficking in Persons and Related Transnational Crime, a multilateral forum with 50 plus members (nation states and international agencies) that was created in 2002 to promote policy dialogue, information sharing and practical cooperation on irregular migration in the Asia Pacific region. A ‘tools-based approach’ is used to analyse the Australian government’s metagovernance of this forum across four dimensions: nodality; authority; treasure; organisation. The overall finding is that Australia has accumulated steering capacity across each of these dimensions and that a ‘tools-based approach’ to metagovernance offers a promising way of understanding differences in steering capacity within transnational forums.

KEYWORDS:

• Metagovernance
• policy tools
• steering
• migration policy

Introduction

It has been well documented that nation states have developed ‘new’ governance arrangements to tackle cross-boundary policy problems (Legrand, 2015; Stone & Ladi, 2015). This article focuses on one such arrangement, a multilateral forum, to examine the metagovernance of migration policy in the Asia Pacific region. Specifically, it examines how the Australian Government has used procedural policy instruments to ‘metagovern’ the Bali Process on People Smuggling, Trafficking in Persons and Related Transnational Crime (hereafter referred to simply as ‘the Bali Process’). By foregrounding governing resources, the ‘tools-based approach’ to metagovernance deployed in this article also highlights how a combination of nodality, authority, treasure and organisation-based resources were used to steer this multilateral forum. Overall, the article finds that metagovernance has been exercised using different procedural tools and that these procedural tools have had important structuring effects on the Bali Process and its substantive policy instruments.

As an area of study, migration policy is well suited to examining the role of states in steering networked arrangements because migration is a cross-boundary policy problem that has been subject to a number of multilateral mechanisms operating at the global, interregional and regional levels. In 2010, the International Organization for Migration (IOM) formalised these arrangements by distinguishing between: Global Processes on Migration (international policy dialogues), Inter-Regional Forums on Migration (IRF) (dialogues that connect two or more regions) and Regional Consultative Processes on Migration (dialogues that connect different states from within an agreed geographical region) (IOM, 2021). As forums, the participants in these arrangements can be involved in a range of activities from agenda-setting, networking, consensus-building, policy coordination and knowledge exchange to knowledge production, capacity building, norm-setting and norm diffusion (IOM, 2021; Martinez-Diaz & Woods, 2009). They are typically state-led forums but they are also non-binding, advisory and consultative in nature, which makes them particularly interesting when considered from the perspective of individual nation states and the exercise of metagovernance (IOM, 2021).

This article begins to address this interest by examining one multilateral forum, the Bali Process, and how one nation state, the Australian government, has exercised metagovernance over it. The Bali Process is an IRF that was established in February 2002 under a co-Chair arrangement between the Australian and Indonesian governments. As a relatively active and longstanding multilateral forum, the Bali Process has a broad-based membership that includes both international agencies as well as approximately fifty member states drawn from Europe, Asia, the Americas and the Middle East. It has also met relatively frequently having held six ministerial conferences and more than seventeen senior officials’ meetings between 2002 and 2017. Interestingly, the Bali Process also covers a region where many countries continue to view migration policy as a matter of national security and where many nation states are not signatories to the international treaties and conventions that pertain to migrants and their families (Carr, 2016). At the same time, migration, and irregular migration in particular, has been the subject of several high-profile crises (e.g. the Rohingya refugee crisis (BBC, 2019)) which has meant that it has come to be viewed as a pressing policy concern by many governments in the Asia Pacific region.

In examining the Bali Process, this article makes several contributions. Theoretically, this article develops the literature on metagovernance and policy tools around a specific interest in the relation between access to governing resources and influence over procedural policy instruments (Howlett, 2000). This is a natural extension for the theory of metagovernance, which has been used in policy studies to examine how policy actors steer governance networks and in how hierarchy interacts with networked arrangements that have otherwise been characterised as horizontal, flexible and loosely structured in nature (Dommett & Flinders, 2015). Conceptually, this article illustrates how a ‘tools-based approach’ to metagovernance can act as an effective intermediate framework between metagovernance theory and procedural policy tools (see Howlett et al. in this special issue) particularly in how it foregrounds attention towards different governing resources. It also supports those studies which have argued that policy concepts developed at the national level can be usefully applied in transnational contexts (e.g. Mamudu, Cairney, & Studlar, 2015; Stone, 2008).

Finally, this article makes an important empirical contribution by studying a multilateral forum about which we know relatively little (but, see: Curley & Vandyk, 2016; Kneebone, 2014). Data is drawn from twelve interviews which were conducted in 2016 with senior officials who had occupied key positions within the Bali Process between 2002 and 2015. This level of access is unusual given the secrecy that typically surrounds diplomatic forums and the officials were selected using the purposive sampling technique (Guest, Bunce, & Johnson, 2006). Interviewees were asked about the processes and tools of the Bali Process (e.g. the role and contribution of its members, interactions within the forum and its purpose and procedures) across its full range of activities and workstreams. This line of questioning was directed towards understanding the procedural policy tools that underpin the Bali Process, rather than its substantive policy instruments. An extensive search of the Bali Process website (http://www.baliprocess.net) was also undertaken, including communiqués, announcements and press releases, publicly available reports and any relevant ‘grey material’ such as news articles. Overall, a detailed understanding of the Bali Process emerged from this data, which covers the period up to 2015 given the timeframe in which the interviews were conducted.

The article proceeds in three sections. The first section discusses metagovernance theory in transnational settings and the tools of government approach. The second section applies a ‘tools-based’ approach to metagovernance to examine the Australian government’s attempts to steer the Bali Process specifically through its use of procedural policy instruments. The third section of the article returns to the earlier discussion by reflecting on the merits of applying ‘tools-based’ approaches in transnational settings.

Metagovernance in transnational forums

Whilst decision-making capacity has traditionally been located within a nation state, globalisation and transnational policy problems have challenged traditional notions of Westphalian sovereignty (Slaughter, 2009). As new locales of decision-making have emerged, Stone (2008) has referred to the emergence of a ‘global agora’ in which: ‘authority is diffuse, decision making is dispersed and semiprivatized, and sovereignty is muddled by recognition of joint responsibility and collective action’. Whilst this suggests a potentially more disordered environment, power imbalances persist and policy actors have more or less capacity to ‘steer’ the networked interactions that characterise this emerging sphere of policymaking. In policy studies, this concern connects with the theory of metagovernance which Sørensen and Torfing have defined as a ‘higher-order’ form of governance that involves: ‘a reflexive and responsive process through which a range of legitimate and resourceful actors aim to combine, facilitate, shape and direct particular forms of governance in accordance with specific rules, procedures and standards’ (Sørensen & Torfing, 2009, p. 245).

Several ‘schools’ of metagovernance have emerged over the past 10 to 15 years. To take three examples: the public management school has examined how public managers exercise metagovernance through network design and network interaction (e.g. Erik-Hans & Edelenbos, 2007), the state centric school has explored the role and capacity of the state to steer networked interactions (e.g. Bell & Hindmoor, 2009; Fawcett & Daugbjerg, 2017; Marsh, 2011) whilst the interactive governance school has focussed on the political aspects of metagovernance (e.g. Sørensen & Torfing, 2016). Whilst the public management and interactive governance schools have typically viewed metagovernance as a means for public managers and politicians to generate effective and legitimate policy outcomes (e.g. Sørensen & Torfing, 2009), the state centric school has generally viewed it as an attempt by the state to reassert hierarchy and control over networks (e.g. Davies, 2011). So, whilst the literature has certainly broadened, the underlying intent to study how policy actors use their knowledge, resources and capacities to influence a network’s scope, processes and outcomes has remained more or less constant.

Whilst these conceptual developments are to be welcomed, metagovernance theory has overwhelmingly drawn on case studies from within nation states (e.g. Dommett & Flinders, 2015). Whilst several studies have examined metagovernance in multi-level contexts (e.g. Bossong, 2014), very few studies have examined metagovernance at levels above the nation state (for an exception, see: Holzscheiter, Bahr, & Pantzerhielm, 2016). Yet, what this review suggests is that the literature on metagovernance provides rich insights that deserve to be tested empirically in contexts beyond the national level to which they have been conventionally applied. In this article, a step in this direction is taken by combining metagovernance theory with Hood’s (1983) ‘tools of government’ framework (Hood & Margetts, 2007).

The tools of government framework has been widely applied in policy analysis (for a review, see Margetts & Hood, 2016) and governance studies (e.g. Hood & Margetts, 2007; Howlett, 2011). Its overarching argument is that government’s achieve their goals by deploying a range of policy tools that are each underpinned by a ‘principal governing resource’ (Hood & Margetts, 2007, p. 144–6). This article extends that insight to examine the relationship between access to resources and the effective metagovernance of transnational forums. As such, it builds on the work of several scholars who have already observed the value of linking metagovernance theory with the tools of government framework. For example, Knill and Bauer (2016, p. 949 and 952) have argued that policy tools provide a ‘useful starting point’ for exploring the ‘taxonomy of administrative resources’ through which ‘international public administrations’ influence global policymaking practices – but they do not test their framework empirically and important differences also exist between international public administrations and the multilateral forums that dominate in migration policy. In another study, Baker and Stoker (2012, p. 1026) successfully use metagovernance theory and the tools of government approach to examine nuclear power policy; but they focussed on a comparison between different governments in Europe and therefore remained at the level of individual nation states.

In realising this connection with metagovernance theory, Hood’s (1983) original formulation has been selected because its ‘resource-based categorization’ of policy instruments makes it particularly well suited to examining differences in steering capacity (see Howlett et al. in this special issue; Howlett & Ramesh, 2016). In this respect, procedural policy tools play a particularly important role because they provide the ‘structure’ within which a policy is formulated and implemented. This extends into a consideration of both the specific procedural tools (or mix of procedural tools) within any particular governance arrangement (e.g. inter-departmental committees, multi-year strategic plans, commissioned reports, fact-finding missions, audits, mission statements, public consultations and advisory committees) as well as the ends that they seek to achieve (e.g. policy formulation, information gathering, partnerships, accountability and regulation) (see Bali et al. 2021 and Howlett et al. in this special issue). As such, procedural policy tools foreground important processual issues as they relate to questions about access to governing resources (a concern of the tools of government framework) and the capacity of policy actors to steer complex governance arrangements (a concern of metagovernance theory). In what follows, this relationship between influence over procedural policy instruments, access to governing resources and a metagovernor’s ‘steering capacity’ is explored. It is informed by the expectation that access to governing resources provides the conditions through which policy actors can exert greater levels of influence over procedural policy instruments and therefore enhanced metagovernance capacity over networked arrangements. Taking each governing resource in turn, this expectation is briefly explored in further detail below.

Nodality refers to the relative centrality of one’s position within social and information networks and the improved access to information or knowledge that this is likely to engender as a result. Those at the centre of information networks are likely to exhibit higher metagovernance capacity because of their greater capability to store, monitor and disseminate information (Hood, 1983, p. 21). Metagovernors may try to promote their nodality by publishing data, information, recommendations and advice in an attempt to persuade or motivate policy actors to act differently. In transnational forums, national interests may not align so this raises important questions about who possesses nodality, what information is shared and how it stored, processed and so on. It is also likely that the capacity to persuade others will depend on multilateral support, rather than unilateral action. Ultimately, however, metagovernance through nodality requires a metagovernor who is at the centre of information flows and who is able to use their central position within information networks to their advantage.

Authority is a resource possessed by those with legitimate legal or official power, which can include both ‘low-constraint tokens’ that ‘approve or recommend’ as well as ‘high-constraint tokens’ that ‘order or forbid’ (Hood, 1983, p. 55). In making this distinction, Hood (1983, p. 62, 71) noted that it is often the case that governments can ‘ … do no more than offer certificates in the hope that they will have an effect’ where ‘the difference between betokening authority and dispensing information starts to disappear’ and ‘obedience or compliance is not automatic or readily forthcoming’. In transnational forums, authority is much more likely to be exercised through low-constraint tokens that are ‘self-consciously official’ but which are not legally binding (e.g. standards, targets, certificates, approvals and codes of good behaviour) (Hood, 1983, p. 55). Authority therefore becomes important in transnational contexts to the extent that a metagovernor can reasonably expect that their decisions will be viewed as legitimate and that they will be acted on.

Treasure is a property of having a command over money and resources that can be readily exchanged. Policy actors who command treasure generally have a higher metagovernance capacity because they can use it to develop their own organisational capacity (e.g. by hiring policy experts or establishing specialised units) or they can use it to try and influence others (e.g. through grants and subsidies). Members that possess treasure, or those who can create dependent financial relationships between themselves and other members, are more likely to have higher steering capacity. However, treasure and its influence is also contingent on other factors such as whether a policy actor possesses the resources that another actor needs, the size of those resources, their availability and the ease with which they can be deployed. Treasure also takes on a particular form in multilateral forums which often rely on self-generated funding or the willingness of members to make voluntary contributions (Reddy, Mazhar, & Lencucha, 2018).

Finally, organisation refers to the stock of people, skills, land, buildings, and technology that a metagovernor can call upon to provide administrative, epistemic, logistical and organisational support. This is similar to what Knill and Bauer (2016) have called ‘bureaucratic authority’, which includes the possession of technical expertise and professional authority over a particular policy domain. Treasure is different from organisational capacity because the former is about fungible assets whereas the latter is about the human capital and physical infrastructure that a metagovernor can use in an attempt to steer a network in one direction rather than another.

Metagoverning the Bali Process

Having detailed the key features of a ‘tools-based’ approach to metagovernance, Table 1 summarises how the Australian Government has metagoverned the Bali Process by using its access to the four governing resources outlined above. Specifically, it details the various ways in which Australia has influenced the direction and form of different procedural policy instruments. In addition, Table 1 maps the substantive policy instruments through which the Bali Process has sought to influence migration policy at the regional level onto the procedural policy instruments that the Australian Government has used to metagovern the processes by which these have been developed.¹

Table 1. Metagoverning the Bali Process

Governing Resource	Substantive Policy Instruments	Procedural Policy Instruments
Nodality	Public communiques and statements, which are more or less binding (e.g. co-Chair statements and Ministerial Declarations) Publicity materials (e.g. website, press releases and social media)	Australia metagoverns by influencing: the overall design and governance of the Bali process (e.g. instigation of the co-Chair arrangement, the tiered governance structure and the Steering Group; exercising its power to convene meetings, cast formal vetos and perform an overall leadership role) the membership of internal committees and networks (e.g. by chairing or participating in committees) the provision and distribution of information (e.g. by drafting communiques, playing a lead role in preparing public statements and having oversight of information flows)
Authority	Non-binding cooperation frameworks, which codify certain principles, values and ‘common understandings’ (e.g. the Regional Cooperation Framework) Bilateral and multilateral agreements aimed at implementing overarching frameworks such as the Regional Cooperation Framework	Australia metagoverns by influencing decisions about: how to design and implement non-binding cooperation frameworks (e.g. via its role on the Ad Hoc Group)
Treasure	Project funding (e.g. by funding international agencies to implement in-country programs and initiatives) Grants (e.g. by providing financial support to certain members so that they can participate in the Bali Process) In-kind support (e.g. by offering expertise, seconding personnel and hosting workshops)	Australia metagoverns by influencing decisions about: whether to provide or withdraw funding for projects which projects should receive funding in negotiation with other members (e.g. via playing a key role in the lobbying and diplomatic backchannelling that takes place amongst the small number of member states who are willing and able to finance their ‘pet projects’)
Organisation	Establishment of a Regional Support Office in Bangkok to build organisational capacity (e.g. by provisioning for logistical, administrative and operational supports) Record keeping (e.g. the preparation of Annual Reports and strategy documents) Establishing expert working groups, training networks amongst senior officials, as well as conferences and workshops	Australia metagoverns by: using its position within networks to influence member states that there is a need for greater institutional capacity seconding personnel from the Australian Public Service to manage the Regional Support Office

Nodality

The Australian government has ensured its nodality within the Bali Process by creating a tiered membership structure and placing itself at the centre of its governance arrangements. The Bali Process’s tiered membership structure is reflected in Australia and Indonesia’s role as joint co-Chairs, which forms the first level in its membership structure. This has given both governments representation at the ministerial and official levels in all key forums. Australia also created a new post of Ambassador for People Smuggling Issues, which was a significant move because it confirmed its commitment to regional cooperation and ‘established a basis for consistent engagement [in the Bali Process] from relatively senior Australian government officials’ (Carr, 2016, p. 325).

As co-Chairs, Australia and Indonesia are also automatically granted representation across the various forums within the Bali Process and have automatic veto power over its motions and outcomes. They also co-manage the Regional Support Office (RSO), which provides organisational assistance to the Bali Process (see below). These factors combine to place the Australian government at the centre of various information networks at both the political and official level. As one senior official noted, there is: ‘ … no question that Indonesia and Australia would see themselves as having the lion’s share of the power and the determination in respect to the Bali Process’. Further, many interviewees identified the co-Chair role as a very deliberate attempt by the Australian government to secure their nodal position within the Bali Process.

Alongside the co-Chairs, a further 50 member states and 10 international agencies also belong to the Bali Process but they are officially divided into two categories: full members and observer members. In keeping with the Bali Process’s non-binding nature, applying to be a member does not imply a significant level of commitment which is one reason why all of the major source, transit and destination countries in the Asia-Pacific region have joined. Applying to be a member also follows a fairly informal process with the co-Chairs taking a decision after input from other member states. As one senior official explained:

It’s not put through any collective vote as it were, it’s usually an expression of interest of the government to the co-Chairs that they would like to be a member, which is then put forth at the senior officials meeting … this usually occurs after some bilateral or offline discussions have taken place between the co-Chairs themselves and other countries.

International agencies have also been crucial members of the Bali Process since its inception, particularly the IOM and the United Nations High Commissioner for Refugees (UNHCR). As one senior official commented:

… the Bali Process is unique in the fact that international organisations are also members. When you look at [it], it’s a state-led forum. But, dynamics change when international agencies have a role … Part of the challenge is meeting the expectations of a broad range of stakeholders who are at different stages of development.

Another senior official argued that international agencies were ‘not going to compromise their principles but, they recognize the practicalities; and they see that by working with [nation] states’. So, whilst the Bali Process is led by nation state governments it is a forum that also relies on support from international agencies. This includes secretariat support from the IOM and project implementation support, through in-country programs and initiatives, which are funded by the Bali Process but which are carried out by international agencies.

Australia’s nodal position is also reflected in the internal governance structures that make up the Bali Process. These arrangements have evolved into a structure that currently consists of five parts. The first part is the Ministerial Conference, which meets biannually and which includes representation from the Foreign Ministers in each of the member states. The six-member Steering Group comes next. It sits directly below the Ministerial Conference, as the highest official-led forum within the Bali Process, and is made up of four nation states (Australia, Indonesia, New Zealand, and Thailand) and two international organisations (the UNHCR and IOM). The Steering Group identifies ways to facilitate deeper regional cooperation and ensures that the Bali Process is able to deliver against its commitments. The Senior Officials Meeting (SOM) sits below this structure. It monitors and takes stock of progress against agreed priorities and projects. Alongside the SOM, the Ad Hoc Group (AHG) exists to develop operational procedures and practices aimed at enhanced regional cooperation. Established in April 2009, the AHG’s membership is decided by the Steering Group and is limited to the ‘most-affected countries’, which included sixteen nation states and three international agencies in 2019. The AHG is also responsible for implementing the Regional Cooperation Framework (RCF) (see below) through a series of bilateral and multilateral agreements. One senior official explained the relationship between these different governance structures as follows:

The Steering Group will meet and discuss whatever. They then set the strategic direction, review and provide advice and guidance on what needs to take place. All the working groups, the RSO and the like then report to that Steering Group.

As such, Australia occupies a key nodal role within the Bali Process as both a co-Chair and key member of the six member Steering Group.

Decisions about what topics get discussed in which forums is another way in which nodality, through procedural policy tools, gets to be exercised in and through the Bali Process. Here, a formal process exists in parallel with an informal one. The formal process allows any member to submit a proposal to the RSO for consideration by the AHG. Once the member has support from the AHG, the proposal will be forwarded to the SOM and then onto the Steering Group, which makes a final determination prior to endorsement by the Ministerial Conference. In reality, this means that the co-Chairs are able to maintain tight control over the agenda setting process through their management of the RSO and membership of the Steering Group, which makes the final determination about which ideas from the AHG it will accept and which it will reject.

Importantly, however, informal bilateral talks run in parallel with this process. What this effectively means is that only those proposals that have undergone an extensive process of diplomatic back-channeling will make it to the AHG. As one senior official explained:

Within a regionally concentrated mechanism it’s probably a bit looser in terms of trying to discuss informally [and] engage with counterparts, than a bilateral meeting would be. So, it complements bilateral efforts but it does it in a way that allows more latitude I guess you could say - whether it’s in the form of discussions within the margins or topical discussions that are put forward and the sharing of ‘lessons learnt’.

Alongside these internal processes, the Bali Process also seeks to exert nodality by issuing public communiques which come in one of two forms: co-Chair statements that are written by the co-Chairs and which members are instructed to ‘peruse’; and ministerial declarations, which are fully negotiated statements that speak on behalf of member states and their governments. Whilst co-Chair statements have been in use since the beginning of the Bali Process, ministerial declarations have only been in use since March 2015 though their introduction has been viewed by many as an important extension of the forum’s official power. Australia was also an important advocate for the introduction of ministerial declarations arguing that they needed members who were willing to make public their commitment to implement certain actions. As one senior official explained:

The ministerial declaration represents an important change from a co-Chair statement because it’s inclusive of the full membership … A declaration speaks on behalf of your government, which means that you’re going to be much more actively engaged in what it’s saying.

In terms of influence, Australia has always played a key role in drafting both sets of communiques and has been able to do so because of its nodality in the structures through which these statements are prepared. It has also played a crucial role in the negotiation of these statements. Whilst the co-Chair statement essentially summarises what was discussed at each meeting and any agreed actions, the ministerial declarations have generally required far more work; but Australia has played a crucial role at each and every stage in their development.

Alongside formal communiques, the Bali Process has also sought to raise its profile and reputation amongst the media and external audiences through its website, social media accounts, media events, press releases, reports and publications. Australian public servants seconded to the Bali Process have initiated and led many of these initiatives at the same time as they have contributed to the organisational support, capacity and expertise necessary for their successful implementation. Thus, Australia has maintained nodality in the Bali Process via its role as a co-Chair and the key position that it has occupied within the forum’s tiered membership structure, its centrality within its agenda setting processes, in the role that it has played in the drafting of its public communiques and through the support that it has provided in raising its profile and reputation amongst external stakeholders and the media.

Authority

Authority comes from being viewed as a legitimate legal or official power. As an IRF, the Bali Process has no legal power. As Alexander Downer (cited in Larsen, 2006, p. 43), the former Australian Minister for Foreign Affairs, has noted:

[t]he Bali Process doesn’t legislate, the Bali Process doesn’t force anybody to do anything. What it does do is provide a framework, it provides context and it provides priority.

However, the Bali Process does possess ‘official power’. Most notably, it has extended its official authority by introducing a RCF. Announced in March 2011, the RCF was developed largely at the behest of the Australian Government as ‘an inclusive but non-binding regional cooperation framework’ (Co-Chairs’ Statement, 2011). As a document, it consists of five ‘core principles’ (e.g. ‘ … asylum-seekers should have access to consistent assessment processes, whether through a set of harmonised arrangements or through the possible establishment of regional assessment arrangements’) and eight ‘considerations’ (e.g. ‘arrangements should promote human life and dignity’) (Co-Chairs’ Statement, 2011). Supporting the RCF’s introduction, the former Australian Prime Minister, Kevin Rudd (cited in Douglas & Schloenhardt, 2012, p. 15), argued that it would create an ‘umbrella under which participating states [can] pursue common objectives through practical arrangements at a bilateral or subregional level’.

The RCF proposal was initially developed under the auspices of the Bali Process via an UNHCR-led forum (Curley & Vandyk, 2016). In this form, the RCF was a framework that included both general opportunities for regional support and practical arrangements for specific situations. It recognised that irregular migration was both a local and global issue that required a regional response and provision was also made for a set of ‘common understandings’ aimed at ensuring a fairer distribution of responsibility for refugee protection (UNHCR, 2010). Whilst Australia supported the provisions coming out of this forum, it also sought to incorporate additional commitments in the negotiations that followed. These included commitments to target the smuggling of migrants through border control and law enforcement, a provision for bilateral and regional processing agreements and a clause to consider the sovereignty of parties in any such agreement (Curley & Vandyk, 2016).

These additional commitments were viewed as strategically important by the Australian government who were wary that many members of the Bali Process were not signatories to the international agreements and treaties that exist on the treatment of irregular migrants, including their protection-related principles. Whilst the RCF was therefore a small and non-binding step, the Australian government nevertheless welcomed its development and viewed it as going some way towards codifying certain norms in the region (Nethery & Gordyn, 2014). In terms of official power, the RCF was therefore an agreement that was actively supported and endorsed by the Australian Government. After securing additional commitments, it was also an agreement that the Australian Government viewed as representing an important development for the Bali Process and its role within the region.

Treasure

Publicly available data on the Bali Process, its overall budget, who funds it, how it is funded and the projects that it supports is virtually non-existent, though official documents do confirm that it receives a mix of financial contributions, both targeted and in-kind, from its members (e.g. Bali Process, 2011b). In interviews, senior officials estimated that Australia provided ‘around 90%’ of core funding – in other words, the lion’s share of its operational funding comes from Australia. This is separate from the financial support that Australia provides to developing countries so that they can attend the Bali Process. Whilst many countries would not be able to participate without this support (e.g. some Pacific Island states), ensuring broad based participation in the Bali Process is also in Australia’s interests because it helps to lend legitimacy to the claim that it is the region’s pre-eminent forum for addressing issues related to irregular migration.

Beyond core funding, members of the Bali Process typically provide targeted financial support and ‘in-kind’ contributions (e.g. hosting a Bali Process workshop or seconding an official to the RSO on a time-limited basis to work on a specific project). Financial support that is targeted is usually tied to specific projects (including in-country projects) that have been approved by the Bali Process and its members. Australia has been an important financial contributor in this respect having supported various projects that the IOM has been contracted to implement. One senior official explained the process by which projects get considered in the following way:

What we do is we’ll develop a draft concept note, then we’ll talk to a range of stakeholders … Then, we’ll invite a small group of states to come and work on it … They’ll come and meet on two or three occasions, review our work, provide input and comment. Then, we’ll take the outcome of that work to the broader membership … have input and comments put in, and then we’ll take it up to the Ad-Hoc Group … as a product that has been reviewed by the entire membership.

As such, the Bali Process’s lack of financial autonomy has meant that members with money have been able to exert a very high degree of control over which projects get funded and implemented. Alternatively, members who disagree with the way in which a project has been implemented can cut it short by withdrawing their support. This has meant that the co-Chairs, and Australia in particular, have often had to lobby members to fund projects either directly (e.g. through the SOM) or indirectly (e.g. through diplomatic back channeling) – a dilemma that is not uncommon even in more established intergovernmental organisations (e.g. Reddy et al., 2018). As one senior official explained, officials connected with the Bali Process were often like: ‘Oliver with an empty bowl, going around to countries like Korea, Japan, US, Canada, Australia … seeking funding on proposals for activities that have been ticked off by the senior officials meetings, but that didn’t come with funding.’ In turn, however, members can also use the cover provided by the Bali Process to support projects that their governments would not have funded bilaterally. Australia has funded various projects on this basis but it has also played an important role in trying to persuade other countries to contribute financially.

This mismatch between activities and the funding to support them has also been documented in the Bali Process’s Strategy and Forward Work Plan in which only half of the 20 projects listed are ‘funded’ with a further six ‘partially funded’ and four not having received any funding at all (Regional Support Office, 2015, p. 10). The Work Plan (Regional Support Office, 2015, p. 8) concludes that:

Effective implementation … requires all RSO stakeholders, including Bali Process Members to participate to achieve desired outcomes. Each member is therefore encouraged to consider the contribution that each government and international agency can make in supporting the stated aims of this work plan through in-kind expertise, secondment of officials and/or funding.

Clearly, the Bali Process relies on the treasure provided by its members and there is no doubt that its most important contributor in this respect is Australia. It provides almost all of the funding to cover its operating costs, supports many countries to attend the Bali Process and funds many of the projects that take place under its auspices.

Organisation

Organisation is a property of the possession of a stock of people, skills, land, buildings, and technology. A key turning point in this respect came in September 2012 when an informal secretariat was launched. Known as the RSO, its purpose is to ‘support and strengthen practical cooperation on refugee protection and international migration’ with a view towards acting as a ‘support facility and focal point for Bali Process Member States’ (Bali Process, 2011a). It has pursued this aim by promoting activities that: facilitate information sharing on refugee protection and international migration; support capacity building and the exchange of best practices; encourage the pooling of common technical resources; and that help ensure the successful implementation of joint projects by providing logistical, administrative and operational supports (Bali Process, 2011a).

Headquartered in Bangkok, the RSO is staffed with public servants seconded from the Australian and Indonesian civil service with additional support from the UNHCR and IOM. Other member states have also seconded officials to work on specific projects typically for between six months and three years at a time. Importantly, however, there is a heavy reliance within the RSO on staff seconded from the Australian Public Service. As such, whilst some organisational capacity has built up around the RSO, much of this backing and expertise continues to depend on the Australian Government and its willingness to support the Bali Process in this way.

In terms of functions, the RSO plays an important role in providing both front end and back end support. Front end support primarily refers to the RSO’s liaison role. As one senior official explained:

A lot of the speaking, going out, representational work, if it’s not at a diplomatic level, will fall to [the RSO] … . [The RSO] communicate directly with Kemlu, Indonesian Foreign Affairs and with Australian Foreign Affairs to talk about [the work of the RSO] to provide advice and guidance. If there’s a policy statement, [the RSO] can best address that policy statement through operational action.

In contrast, ‘back end’ support refers to the RSO’s role in overseeing the implementation of agreed actions. These activities are listed in the RSO’s Forward Work Plan, which describes itself as ‘an operational and action-oriented framework … [that] clearly sets out the goals for each area and commits to concrete action against these goals’ (Regional Support Office, 2015, p. 2). As a planning mechanism, the Forward Work Plan prioritises different actions and distributes its resources towards those ends (Regional Support Office, 2015, p. 2).

Initiatives in the Forward Work Plan are organised around four goals: early detection, prevention, protection and migration management. Examples of recent projects that have been undertaken by the RSO include: a regional data management initiative; a voluntary repatriation capacity building and support project; a pilot study on information exchange and data analysis on irregular migration by sea; and the organization of a regional roundtable on irregular movements by sea (Regional Support Office, 2015). These ‘major projects’ exist alongside a number of other ‘smaller initiatives’, including an agreement to establish a roster of experts on migration management and the development of a set of best practices on refugee protection. The RSO has also advised member states on various matters (e.g. the harmonisation of national legislation, information campaigns and tenders for research projects), organised technical training and assistance for law enforcement personnel and has published various ‘policy guides’ (e.g. in areas such as human trafficking and induction training for frontline border officials). Australia has played a crucial role across all of these activities in seconding the staff that have helped to support this work and the organisational capacity that underpins it.

The RSO has also supported a number of networking and information sharing activities at the official level, including working groups, technical workshops and training networks. The working groups typically meet once a year to provide senior officials (drawn from the AHG) with a place to meet and discuss the latest developments in migration policy (e.g. trafficking in persons and people smuggling networks). In addition, technical workshops have been organised (e.g. in areas such as the ‘immigration aspects of airport security’ and ‘irregular movement by air’) as well as several ‘training networks’ (e.g. the Regional Immigration Liaison Officer Network which shares information on the irregular movement of people by land and air). The Australian government has supported these initiatives both directly through financial contributions as well as in-kind by providing the expertise and personnel to run them.

These activities suggest that the Bali Process, through the RSO, is developing some bureaucratic autonomy but its scope is also significantly circumscribed. Members have also disagreed on the broader question of whether the Bali Process should move more towards a greater institutionalisation of its presence in the region. Member states have remained generally cautious but some international agencies have supported the idea (e.g. Feller, cited in Kneebone, 2014, p. 610). Nevertheless, Australia clearly plays a crucial role in terms of providing organisational support both in terms of seconding staff to the RSO and in providing expertise to support its operations.

Discussion and conclusion

The Bali Process was created to promote greater joint working by member states and international agencies in the Asia Pacific region. Whilst it has repeatedly asserted the interdependencies that exist between its members not all of its members participate on an equal footing. In particular, the Australian government has used its access to different governing resources to exercise its metagovernance over procedural policy instruments; and, by doing so, it has been able to exercise influence over the Bali Process and the substantive policy tools that underpin it.

Whilst this analysis has highlighted Australia’s pivotal role in the Bali Process, this is not to say that it does not face its own set of challenges. Australia’s role as a co-Chair in the Bali Process places it in a nodal role that transcends any particular governance structure but it also shares this responsibility with Indonesia, international agencies and other members in the Steering Group, SOM and AHG. In recent years, some members have proposed that the chairs rotate to help disperse the perceived concentration of authority that currently resides in the co-Chairs. However, Australia has successfully managed to deflect those demands by reaffirming its joint role in the region with Indonesia and by pointing to the support that it provides to the Bali Process, which far exceeds that which any other member has been willing to commit.

Whilst Australia has contributed to the Bali Process’s official authority by promoting it as the region’s premier forum, it still important to note that it remains a non-binding forum. Its members are not legally bound by its decisions and any decisions that it takes still need to be implemented by its members. Here, multi-country expert working groups play their role but it is clearly a constrained and circumscribed one. Australia has also resisted calls for a more formalised secretariat by arguing that it would prevent the Bali Process from being able to respond to fast moving events. Whilst Australia may try to metagovern the Bali Process, it clearly cannot single-handedly determine its outcomes let alone those within the field of migration policy more broadly construed.

Nevertheless, what this article has clearly shown is that, whilst member states often celebrate transnational forums as exemplars of regional interdependency and collaborative working, there is often a difference in power and influence between the members who belong to them. Whilst Australia has certainly used procedural policy instruments to steer the Bali Process, what this analysis has also shown is that metagovernance often operates differently at the transnational level. Whilst nodality between policymakers often functions in much the same way as it does in a national setting, the Bali Process also clearly operates at one removed from direct ministerial oversight. As with most other IRFs, it is a forum designed and run by officials. At the level of authority, the Bali Process works on the basis of official authority in which expertise and information are far more important than legal rules or formal power. Treasure also takes on a different meaning compared with national contexts. Without access to any of its own financial means, the focus shifts to who contributes what resources. Finally, with respect to organisation, whilst the RSO has augmented the Bali Process’s organisational capacity it nevertheless works on a small budget with a temporary staff seconded mainly from the Australian Public Service.

Whilst this summary of the Bali Process shows that it has built capacity across four different dimensions, it is worth restating that this is not the same as arguing that the Bali Process is an ‘effective’ regional response to irregular migration. Further, the Bali Process has evolved since its early days, and various crises in the intervening period, have both highlighted its shortcomings as a regional forum at the same time as they have prompted members to reaffirm its importance. Nevertheless, the focus of this article has been on policy processes and procedural instruments, rather than the efficacy, or otherwise, of the Bali Process and its substantive policy tools.

Further empirical work and theoretical refinement is required before we will know whether metagovernance theory can be usefully applied in other transnational contexts. This case study has applied it to an IRF responsible for migration in the Asia Pacific, which is a region that has its own set of distinct characteristics. However, the particularities of the Asia Pacific region do not preclude the application of metagovernance theory to other regional and transnational settings, including both inter-regional arrangements (such as the Asia-Europe Meeting) as well as other transgovernmental networks and regimes (such as the Financial Action Task Force). By extension, this would require further application of the tools of government framework to better understand the similarities and differences in how procedural policy tools are deployed at the transnational level. As it stands, the aims of this article were both more modest but also more fundamental: to use a ‘tools-based’ approach to metagovernance to show how the Australian Government has used procedural policy instruments to steer the Bali Process. By linking steering capacity in multilateral forums to a nation state’s access to different governing resources, this article has also shown how substantive policy instruments are structured by the procedural tools that underpin them. This finding suggests the need for more research on multilateral forums where procedural policy tools often play a crucial role in the formulation and development of transnational policy outcomes.

Acknowledgments

My thanks to the senior officials and diplomats who agreed to be interviewed as part of this project. I would also like to thank the participants in the Melbourne-NUS joint workshop on Policy Tools as well as Azad Singh Bali, Mike Howlett, Gerry Stoker, Diane Stone and the anonymous reviewers for their feedback on earlier drafts of this article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Paul Fawcett

Paul Fawcett is Associate Professor of Public Policy and Chair of the Political Science Discipline in the School of Social and Political Sciences at The University of Melbourne. His research interests lie in governance theory, depoliticisation and anti-politics. Some of this research has been published in journals such as Policy Sciences, Administration & Society, Policy & Politics, Journal of Public Policy, among others.

Notes

¹ A few caveats are warranted. First, a policy tool can rely on multiple governing resources, and that the primary resource (nodality, authority, etc) that the tool uses to give effect is listed here. Second, often tools are used in conjunction or layered in a ‘policy mix’ to give effect to stated goals (Bali et al 2021; Howlett et al. this issue). The aim of this table, however, is to clarify the primary substantive and procedural tools that are deployed.