Private sustainability standards as tools for empowering southern pro-regulatory coalitions? Collaboration, conflict and the pursuit of sustainable palm oil

The social and environmental impact of commodity production in the global south is now governed by an array of global market-driven standard-setting schemes, which interact with state-centred legal and administrative governance ‘on the ground’ in producing countries. Drawing on a case study of contested regulatory governance in the Indonesian palm oil sector, this paper investigates the effects of interactions between (northern) marketbased and (southern) state-centred regulatory authorities. Analysis shows that it is not the collaborative or conflictual character of governance interactions that matters most in shaping regulatory capacity, but rather how such interactions influence the motivations, capacities and legitimacy claims of competing regulatory coalitions within commodity producing jurisdictions. While conflictual pathways of regulatory empowerment can sometimes be productive, their effects on destabilizing power relations between elite and marginalised actors in producing countries render them distinctively vulnerable to legitimacy challenges from incumbent powerholders. This generates dilemmas for global regulators, whose efforts to influence change through strategies of empowering southern pro-regulatory coalitions are subject to challenge from competing coalitions of southern


Introduction
Increasing scale and complexity of transnational governance systems addressing social and environmental impacts of global commodity production has generated rising interest in the complex interactions between state and non-state governance. On one hand, it is widely recognised that states often depend on private and hybrid mechanisms of "new transnational governance" to help them effectively solve environmental problems (Abbott and Snidal, 2009a). At the same time, the effectiveness and legitimacy of transnational governance initiatives targeting sustainable commodity production depend on the ability of transnational actors to secure support from governments at national and sub-national levels in the countries where commodity production is located (Bartley, 2011;Pacheco et al., 2018). It is in this context that a burgeoning literature has sought to explore both synergistic and competitive interactions between state-centred, private and hybrid forms of governing authority, extending from global to local levels (Eberlein et al., 2014;Bartley, 2014;Gulbrandsen, 2014).
Across a wide range of industrial and agricultural product sectors, social and environmental impacts of business are now regulated by an array of non-state standard-setting and certification schemes, together with social and environmental standards laid down by individual companies and by both private and multilateral financiers of transnational business activity. These interact with legal and administrative systems at international, national and sub-national levels, to produce a complex sphere of regulatory governance, in which public, private and hybrid sources of governing authority compete for power and legitimacy. 1 The central aim of this paper is to deepen our understanding of the effects of market-state governance interactions on regulatory capacity in producing country jurisdictions, and the conditions under which such effects can be productive. 2 The paper's focus is further narrowed to centre on regulatory governance interactions associated with one particular functional element of the governance process-complaint handling mechanisms that have sometimes been referred to as 'fire alarm' systems of monitoring and compliance-promotion (Amengual, 2010;McCubbins and Schwartz, 1984). Complaint-handling has been widely recognised as a crucial dimension of the regulatory process, operating (at least in theory) to support monitoring, enforcement, and capacity building and learning (Eberlein et al., 2014). However, compared with rule-making, monitoring and enforcement, 'fire alarm' compliance systems have rarely been subject to systematic study through a transnational business governance lens (for partial exceptions, see Marx, 2014;Gray and van Rooij, 2017). As many regulatory scholars have highlighted, performance of each distinct functional component of the regulatory process demands a correspondingly distinctive portfolio of resources or capacities; these may encompass financial resources, social connectivity or structural position, organisational capacities and relationships, and/or distinctive claims to expertise or legitimacy (Abbott and Snidal, 2009b;Eberlein et al., 2014). A focused analysis of complaint-handling, as a distinctive yet under-researched component of the regulatory process, can therefore broaden our understanding of regulatory governance interactions across the breadth of the regulatory process.
These 'fire alarm' compliance systems are of particular importance to understanding dynamics of compliance 'on the ground' in commodity-producing locations, and as such are of particular relevance to the aims of this special issue. Because complaints about violations of sustainability standards often relate to conflicts between southern companies and southern workers or communities, a focus on complaint processes also facilitates exploration of the plurality of southern interests and actors engaging with global market-based governance initiatives on the ground-embracing, appropriating or resisting them in a variety of ways. While marginalised communities seek to draw on these complaints mechanisms as tools of empowerment in relation to both individual grievances and broader struggles to strengthen social and environmental regulation, incumbent powerholders in commodity producing countries sometimes view them as a threat, and seek actively to undermine their legitimacy and influence.
Drawing on detailed field-based research on contested sustainability governance in the Indonesian palm oil sector, the paper presents illustrations of both productive interactions that enhance the capacity of regulatory authorities, and conflictual interactions in which state and market regulatory processes compete, or even actively seek to undermine one another's capacity and authority. While much existing literature on governance interactions has stressed the productive potential of collaborative interactions (Abbott and Snidal, 2009b), this paper's analysis reveals that the productive potential of these interactions depends less on whether they are collaborative or conflictual in character, and more on how interactions influence regulatory contests at the local level, by re-shaping the motivations, capacities and legitimacy claims of competing regulatory coalitions. 3 Global regulators are therefore shown to face a dilemma as they seek to influence regulatory change in commodity producing countries: while conflictual market-state interactions involving the empowerment of southern pro-regulatory coalitions can provide productive pathways to strengthened regulatory capacity, such strategies are distinctively vulnerable to legitimacy challenges from incumbent powerholders at the local level, who resist the legitimacy of 'foreign' market-based governance systems as a means of protecting their own power and interests within local regulatory processes.

Theorizing governance interactions
Recognition of the complex, multi-level character of contemporary transnational governance has given rise to expansive bodies of scholarship examining transnational-local governance interactions (Eberlein et al., 2014). Interactions have been variously conceptualised with reference to the effectiveness or influence of global regimes (Young and Levy, 1999;Bernstein and Cashore, 2012), dynamics of legal globalization (Halliday and Carruthers, 2007;Shaffer, 2012), or the transnationalization of regulatory fields (Dingwerth and Pattberg, 2009;Djelic and Sahlin-Andersson, 2006). These bodies of work in turn have drawn heavily on broader frameworks for understanding interactions between normative and institutional orders at transnational and local levels, with diverse disciplinary origins in international relations (Acharya, 2004), anthropology (Merry, 2006), law (Berman, 2006), sociology (Bartley, 2011) and geography (Bebbington, 2003).
Governance interactions have also been extensively theorized by scholars working with legal, socio-legal and organisational and business lenses. Legal and socio-legal scholars have developed sophisticated frameworks for analysing interactions between hard and soft law (Meidinger, 2006), while business and organisational scholars have developed rich comparative analyses of CSR-government interactions (Vallentin and Murillo, 2012;Hofman et al., 2017;Scherer and Palazzo, 2011). Meanwhile, critical political economists have highlighted the interplay between transnational and national scales in constituting the "social and regulatory relations of production" in which the power and interests of competing groups within transnational regulatory processes are grounded (Foley and Havice, 2016, p. 24). This kind of political economy lens can usefully bring critical analyses of power and interest within global production systems (Van der Ven, 2018; Levy and Newell, 2002) together with contextual analyses of the "configuration of political and economic forces" in particular places (Singh and Bourgouin, 2013 p. 34;Ponte, 2008). 4 Such investigations of transnational-local, 1 Throughout the paper, the concept of governing 'authority' is used to refer to decision-making actors or institutions that successfully claim the right to perform governance functions on behalf of a collective (Zürn et al., 2012). The concept of legitimacy is understood as "a generalized perception or assumption that the actions of an entity are desirable, proper or appropriate within some socially constructed system of norms, values, beliefs and definitions" (Suchman, 1995, p. 574). As they are used in this paper, the concepts are very closely related, insofar as both refer to a social relationship in which the actions of one entity achieve stable recognition or acceptance by another, based on beliefs that such recognition is consistent with self-interest, is morally right, and/or is simply taken for granted as a socially appropriate response. However, the terms are not used interchangeably. Characteristics of the agent or process through which decisions are generated are assumed to play a central role in generating recognition or acceptance of governing authority; in this sense, claims to authority are grounded independently from claims about the justification of specific decisions (Krisch, 2017). The concept of legitimacy is broader, insofar as legitimacy claims can relate not only to governing authority (the focus of this paper), but also to claims about specific substantive decisions or rules. 2 The paper does not take on the broader aims of either systematically evaluating the effectiveness of public and private authorities, or empirically evaluating the success with which multiple actors compete for authority and legitimacy within the governance sphere.
3 Although governing actors can exercise influence within a governance sphere in the absence of strong claims to authority (or legitimacy), increased recognition of the legitimacy of a given set of authority claims can enable, augment or sustain other sources of power. 4 Stable patterns of material and ideological power through which regulatory fields are structured at global as well as local territorial scales have been conceptualised in diverse ways, with reference variously to concepts such as hegemony (Levy and Newell, 2002), regimes (Jessop, 1997) and public-private, governance interactions, provide rich theoretical insights to inform our inquiry into the effects of market-state governance interactions on regulatory capacity and compliance in commodity producing countries, and the conditions under which such effects can be productive. We are particularly interested here in exploring the extent to which productive effects of market-state regulatory interactions depend, as has often been assumed, on collaborative mechanisms of interaction, in which market and state governing authorities direct their activities in mutually supportive ways towards shared visions of regulatory change. Such collaborative frameworks can be contrasted with those emphasising more conflictual dynamics of governance interaction, through which competing regulatory coalitions struggle for dominance within a governance sphere. Such approaches typically highlight the indirect character of market-state governance interactions, which operate through reshaping actor interests and power relations within the wider social arena in which governance processes are embedded. With the aim of understanding the conditions under which both collaborative and conflictual processes of market-state governance interaction can be productive, the following discussion draws out from these wider bodies of scholarship some specific propositions regarding both mechanisms of collaborative and conflictual governance interactions, and the conditions under which each may be expected to generate productive effects on regulatory capacity.

Collaborative governance interactions between governing authorities
Much analysis of collaborative governance interactions has focused on institutionalised cooperation between market-and state-based authorities. Numerous studies have examined coordinated interactions through formal organisational collaborative mechanisms such as multistakeholder schemes (Abbott and Snidal, 2009b), or other formalised cooperative arrangements (Amengual and Chirot, 2016). More weakly institutionalised forms of regulatory coordination have been variously conceptualised with reference to concepts such as 'regulatory orchestration' (Abbott and Snidal, 2009a) or 'enrolment' (Braithwaite and Drahos, 2000, pp. 494-5)-broadly invoking the capacity of institutional or societal actors to coordinate or appropriate the powers and capacities of others to further their own desired regulatory objectives (Amengual, 2010;Locke et al., 2013;Overdevest and Zeitlin, 2014). Such collaborative interactions have commonly been characterised as vehicles for generating productive effects on regulatory capacity by bringing together complementary functional capacities of state and non-state regulatory actors, each of which is viewed as having distinctive "competencies" (Abbott and Snidal, 2009b, p. 46) or "comparative advantages" (Amengual, 2010, p. 405) in performing regulatory functions. 5 Regulatory scholars have also identified a number of ways in which complementary interactions may occur in the absence of active institutionalised coordination (Cashore et al., 2011). Focusing on the interplay between state-and market-based labour regulation, Amengual (2010) has documented processes of 'regulatory co-production', in which state and private regulatory processes support one another indirectly via a more or less conscious division of regulatory labour between public and private actors, taking advantage of differing, and at least partially complementary, regulatory capacities and resources. Others have highlighted the dynamic effects of governance interactions on actor motivations and capacities, for example where the layering of market and state regulatory schemes increases the visibility, status and influence of regulatory bodies vis a vis other parts of government (Perez, 2011), facilitates capacity-building in relevant state agencies (Kolben, 2011), or contributes more broadly to mutually reinforcing socialization effects and increased density of compliance and enforcement mechanisms (Jordan et al., 2015).
Collaborative interactions of these varied forms are viewed as being strongly associated with productive outcomes for regulatory capacity-producing positive effects by bringing together diverse competencies of different actors, facilitating dynamic processes of learning or capacity building, or exercising joint effects on the incentives facing regulatory targets and/or prevailing norms and discourses within the wider sphere. However, it is often at least implicitly recognised that such productive interactions are only likely to emerge under certain conditions: where there are aligned regulatory objectives, complementary functional capacities, and in the case of institutionalised forms of collaboration, mutual knowledge and recognition of interacting market and state authorities and actors motivated to initiate collaborative processes.

Contested interactions between competing regulatory coalitions
A second broad analytical lens for conceptualising public-private governance interactions has focused on more overtly political (and often conflictual) dynamics of interaction, through which competing regulatory coalitions struggle for dominance within a governance sphere. Such a lens is common amongst both critical political economists interested in the power struggles embodied in systems of global sustainability regulation (Foley and Havice, 2016;Ponte, 2008;Orsato et al., 2013), and political-institutional scholars who foreground competing interests and contestation as key drivers of regulatory politics (Bartley, 2014;Cashore et al., 2011;Fransen, 2011). Such perspectives explore how engagement of transnational market-driven regulatory systems with state authority can indirectly influence state regulation, by reshaping the interests and power of key domestic actors, and the normative and material power structures in which their regulatory activity is embedded. This orientation entails a shift of focus beyond interactions between governance authorities themselves, and towards contests between wider regulatory coalitions (Bartley, 2011;Ponte, 2008).
Scholars focused on the embedding of public-private governance interactions in wider regimes of state and business power have often been highly sceptical of the potential of market-state governance interactions to support strengthened sustainability regulation. The rising significance of private regulatory governance vis a vis state authority has often been argued to close off political space for pro-regulatory groups such as trade-unions, for example by creating parallel structures of worker representation that are delinked from more radical agendas or organising capacities (e.g. see discussion of such views in Amengual, 2010). More broadly, critical scholars have often viewed private governance schemes as potential instruments for legitimizing hegemonic power relations (Bloomfield, 2012;Moog et al., 2015), particularly in the presence of hostile local governments (Varkkey, 2013). Those critical scholars focused on regimes of state, business and social power in particular places have highlighted how global regulatory schemes can be systematically undermined by their "attempts to institutionalise an order that provides for a distribution of benefits that is not in line with local constellations of power and interest" (McCarthy, 2012(McCarthy, , p. 1885Mikler, 2018;Bebbington et al., 2017).
While recognising the significance of such risks, it is also possible to identify potential mechanisms through which conflictual market-state governance interactions can generate productive effects on regulatory (footnote continued) cumulative concatenation of power re-distributions between competing actors and networks (McAdam et al., 2003), economic, geographical and institutional shifts in the structure and governance of global value chains (Kaplinsky, 2000;Gereffi et al., 2001), and/or broader processes of contestation and accommodation at the level of wider regimes (Levy and Newell, 2002). 5 Abbott and Snidal (2009b) also stress the importance of information, ideas and the agency of regulatory entrepreneurs in shaping the extent to which such functional complementarities are fully exploited in practice. In contexts where such conditions are absent, the potential for governance capacity to be eroded as a result of public-private interactions has also been acknowledged, as in the context of service delivery in fragile states (e.g. Joshi and Moore, 2004).

K. Macdonald
Ecological Economics 167 (2020) 106439 capacity: by enabling pro-regulatory actors and coalitions to access new sources of power, or reconstituting the interests and identities of existing powerholders in ways that reshape how their power is exercised in pursuit of regulatory change. First, productive effects of conflictual governance interactions may result from the reconstitution of interests and identities of elite actors, in ways that bolster political support for strengthened regulatory capacity. Transnationalization of economic governance has been shown to produce significant effects on the interests of both business and government elites, who may acquire self-interested reasons to support strengthened state regulation on particular issues (Schamis, 1999;Shadlen, 2017). Perceived interests may also be reshaped as a result of complex processes of learning, socialization and ideational change-reconstituting not only beliefs about interests and appropriate behaviour, but also dominant discourses concerning organisational legitimacy and expertise (Ponte and Cheyns, 2013;Risse et al., 1999).
It is also possible to identify mechanisms through which productive effects may result from more overtly conflictual dynamics of governance interaction. One important potential pathway for such productive effects is associated with redistributions of power between competing regulatory actors and coalitions. Direct transfers of resources and knowledge between members of transnational coalitions can provide an important means through which power in domestic governance arenas is redistributed (Bernstein and Cashore, 2012, p. 59). Transnational regulatory schemes may also reconfigure the structural positions and relationships of different actors within social networks, redistributing networked forms of power (Kahler, 2015), or reshape broader structural power relations within the governance sphere. Structural power shifts may in turn facilitate increased organisational density and capacity amongst pro-regulatory social groups who otherwise lack favourable opportunities to collectively organise-supporting subsequent claimmaking processes directed at state institutions (Auld et al., 2015;Grugel and Peruzzotti, 2010;Rodríguez-Garavito, 2005). Such organisational sources of power may be of particular salience in the case of 'fire alarm' regulatory systems-where affected groups are expected to have the collective capacity to 'ring the alarm' in cases of regulatory standard violations.
While it is therefore possible to identify a number of propositions about pathways through which conflictual market-state governance interactions may sometimes generate productive effects, the conditions under which such pathways are likely to emerge have not been so explicitly theorized. These will be further explored through the following case study.

Governance interactions in the Indonesian palm oil sector
In what follows, these varying propositions regarding the effects of collaborative and conflictual mechanisms of market-state governance interaction are explored with reference to a case study of the Indonesian palm oil sector. Empirical data collection and analysis have been organised to reflect the multiple levels or scales at which market-state governance interactions play out-encompassing a macro-level analysis of the governance sphere, together with analysis of micro-and mesolevel interactions between particular actors, networks or governing authorities within the wider sphere (Eberlein et al., 2014).
At the macro level, the focus on governance interactions in the Indonesian palm oil sector enables a broad view of a particular geographically demarcated 'slice' of the governance sphere as a whole, which is constituted over the multi-sited geographical spaces in the global south and north within which the regulation of 'sustainable' palm oil is contested. The goal, however, is not to present a comprehensive aerial view of all interactions within the governance sphere (if such a task would even be possible), but rather to zoom in on mesolevel interactions (at varying spatial and temporal scales) involving specific regulatory actors within the Indonesian state, and two prominent market-based governance mechanisms present in the Indonesian palm oil sector: the Roundtable on Sustainable Palm Oil (RSPO) and the Compliance Advisor Ombudsman (CAO) of the World Bank Group.
Analysis of micro-mechanisms of interaction involving specific companies, communities and regulatory networks is facilitated by more detailed exploration of one protracted series of interconnected complaints concerning disputes between a number of Indonesian communities, and the Singapore-based palm oil production and trading company Wilmar. Wilmar is one of the world's largest producers and processors of palm oil, 6 and a major player in Indonesia. It is also a prominent RSPO member, and has received some financing for its operations from the IFC. Wilmar's prominence within the palm oil sector and the RSPO meant that Wilmar's protracted conflicts with Indonesian communities had wider significance for regulatory struggles throughout the palm oil sector, making this case a particularly significant one to investigate. Analysis of this case draws on field research carried out by the author and other collaborators in Indonesia (and relevant transnational locations) between 2012 and 2015, involving 63 interviews and focus groups with 179 individuals.
Palm oil is a highly globalized domain of production and trade in which transnational, market-driven systems of standard-setting, audit and complaint-handling have achieved significant visibility and penetration. It is also a sector that has been the target of protracted environmental campaigns by European NGOs and their local collaborators, who have pressured European financial institutions and food retailers to strengthen commitments to social and environmental production standards throughout their supply chains (Diprose et al., 2019;Pye, 2010). Indonesia is a major player in the global palm oil industry, producing, together with Malaysia, 85% of global palm oil production (Teoh, 2010, p. 5;Colchester, 2011). Palm oil is viewed by the Indonesian government as integral to the nation's development, both as a major employer, and an important export earner (Jiwan, 2009in Teoh, 2010. The social and environmental impacts of the palm oil sector have however attracted significant controversy. Although the sector provides a source of employment, this is usually characterised by low wages and weak labour standards, and there has been significant concern regarding the exclusion of many smallholders from economic opportunities created by the sector's development (Jiwan, 2012, pp. 70-72). Expansion of land used for oil palm plantations has also been associated with extensive deforestation and associated environmental impacts (Pacheco et al., 2018;Jiwan, 2012, pp. 59-65). Particular scrutiny has been placed on pervasive conflict with local communities surrounding land acquisitions (Obidzinski et al., 2012). 7 While palm oil supply chains have become highly transnationalized, practices of both production and regulation are deeply embedded in local regimes of extraction within Indonesia, in which clientelist relations linking bureaucratic, political and business actors shape patterns of power and interest at national and sub-national scales (McCarthy and Zen, 2010;Pacheco et al., 2018). Palm oil governance in Indonesia occurs in the context of a highly decentralized legal and administrative structure, with responsibilities for different aspects of social and environmental governance diffused between multiple Ministries and Agencies (Djogo and Syaf, 2004 also carried out by authorities across a range of other portfolios, including those responsible for licencing and regulation of agricultural plantations, environmental regulation, and management of land conflicts through law and order institutions (particularly police and military). A range of channels within the Indonesian state provide communities with formal opportunities to seek redress in cases where alleged violations of social and environmental standards have occurred. The court system is occasionally used by communities seeking redress for land-related grievances. Communities also sometimes bring grievances to national 'conflict desks', housed in both the National Land Agency (BPN) and the Ministry of Forestry, which are empowered to investigate and mediate disputes. Conflict-handling teams also exist within the environmental regulation portfolio, 8 and are sometimes constituted by District Heads and Provincial Governors on a largely ad hoc basis to address company-community conflicts that arise in their jurisdictions. 9 Particularly where disputes are complex and protracted, Indonesia's National Human Rights Commission (Komisi Nasional Hak Asasi Manusia, usually known as Komnas HAM), has sometimes also become involved in investigating and mediating specific complaints.
There is significant variation in the capacity and willingness of government regulators in different portfolios and at different levels of government to support social and environmental regulation and facilitate remedy in the case of complaints. State-based complaint-handling mechanisms suffer significant challenges of resourcing and authority, while judicial processes are widely perceived to offer little hope of positive remedies for communities, in the context of weak legal recognition of customary rights to land, a weak culture of legal compliance, and close political alliances between companies and many subnational government officials (Gillespie, 2011;Ito et al., 2014). Conflict-handling desks and associated complaint-handling processes at national and sub-national levels are extremely under-resourced, their capacity to conduct independent investigation is limited, and they often lack authority to determine whether compensation or enclaving of certain areas of land from a concession will occur, or to enforce any remedies that may be agreed (Institute for Policy Analysis of Conflict, 2014). 10 It is within this fragmented and deeply contested state-centred governance environment that transnational market-based governance systems have come to play a significant role. The first of these examined here is the complaint-handling system operated by the Roundtable on Sustainable Palm Oil (RSPO), a private multi-stakeholder certification system that fits well with Cashore's original conception of non-state market-driven (NSMD) governance (Cashore, 2002). The RSPO is a multi-stakeholder governance scheme that sets social and environmental standards for business activity in the international palm oil sector, and certifies companies against these standards. Over 20% of palm oil is RSPO certified globally, and around 10% of Indonesian companies operating in the palm oil sector are signed up to the RSPO, accounting for around 24% of Indonesian palm oil exports (author selfreference, 2016, pp. 48-49). 11 The RSPO is legally constituted in Switzerland, while the Secretariat is currently based in Kuala Lumpur, with a satellite office in Jakarta. Its membership, governing Board and issue-based working groups are organised around cross-sectoral representation of palm oil businesses and NGOs. 12 The pervasive presence of company-community disputes in the palm oil sector has meant that development of a complaints system has become an important component of the RSPO's overall regulatory strategy. The RSPO has established a formal Complaints System that incorporates a Dispute Resolution Facility, designed to facilitate the mediation of individual conflicts, and a Complaints Panel, which is empowered to adjudicate disputes arising from complaints, and provide recommendations to the RSPO Board on appropriate remedies. 13 The second transnational regulatory mechanism examined here is the Compliance Advisor Ombudsman (CAO), which is an independent accountability and grievance handling body designed primarily to manage conflicts associated with business activity funded by the World Bank's private sector lending arms, the IFC and MIGA. The CAO is institutionally housed within a major inter-governmental organisation, and cannot be categorised as a non-state regulatory mechanism. Nonetheless, like NSMD regulatory mechanisms, the CAO helps to regulate transnational business activity by drawing on the agency of non-state as well as governmental actors, and its influence is grounded primarily in non-binding forms of authority (cf. the concept of 'transnational new governance' in Abbott and Snidal, 2009a), backed in significant part by market leverage associated with IFC lending. 14 As its name suggests, the CAO comprises three elements: a Compliance auditor, which assesses the IFC's adherence to its own social and environmental policies; an Advisory arm, which advises the World Bank Group on how IFC/MIGA's social and environmental performance can be improved; and an Ombudsman arm, which provides recourse for people affected by IFC or MIGA projects, and facilitates mediation between companies, communities and other affected parties. 15 While the RSPO has a physical presence in Indonesia, and has handled numerous land disputes arising there, the CAO's role is more localised, only becoming involved in specific cases where the relevant company has received World Bank Group finance, as was the case for the company Wilmar.
These state-and market-based governance systems come into interaction at multiple spatial-temporal scales in the governance of regulatory complaints in the Indonesian palm oil sector-generating both collaborative and conflictual interaction dynamics.

Collaborative governance interactions in pursuit of sustainable palm oil
We begin by examining collaborative interactions, finding that productive effects of market-state regulatory interactions were present only in relation to some isolated examples of micro-and meso-level 8 Law enforcement teams based at Provincial level are responsible for handling complaints relating to alleged violations of environmental standards. These teams are given quite significant powers to conduct fact finding investigations, carry out mediations, and enforce regulations. 9 These are often supported by staff drawn variously from the Land Agency, Forestry and Plantation Ministries, and/or other relevant departments. 10 See also interview with government officials, Pontianak, February 2013.
Such institutional barriers to redress are compounded by multiple layers of disadvantage experienced by communities involved in the specific disputes we examined, including lack of access to basic financial and organisational resources, vulnerabilities associated with concerns about livelihood and physical security, fear of reprisals from company or state actors, challenges in establishing and sustaining agreed approaches to collective action within communities, challenges in accessing information and understanding of available grievance handling processes, and difficulties acquiring the forms of written 'evidence' to support their claims that are typically granted recognition within formal complaint processes. Such barriers confront community efforts to bring complaints through both governmental and non-governmental redress processes. 11 The figure of 24% has been calculated based on 2017 figures from RSPO Impact Update 2017 https://rspo.org/toc/RSPO-Impact-Update-Report-2017_ 221117.pdf p.42 and https://www.statista.com/statistics/971491/total-exportvolume-of-crude-palm-oil-indonesia/. 12 Seven groups are formally represented: oil palm growers, processors or traders, consumer goods manufacturers, retailers, banks and investors, environmental or nature conservation NGOs, and social or developmental NGOs. 13 http://www.rspo.org/en/system_components_and_terms_of_reference. 14 The intention is not to compare the regulatory governance processes associated with each of these market-based regulatory mechanisms, but rather to understand their interaction in the context of the overarching governance sphere. 15 http://www.cao-ombudsman.org/about/whoweare/index.html. (2020) 106439 governance interactions. Overall, the theorized conditions for governance interactions to generate productive effects (involving the presence of complementary functional capacities, and mutual recognition by market and state regulators of each other's legitimacy) proved more difficult to establish than established literature might lead us to expect. Both the RSPO and the CAO have demonstrated deeply ambiguous approaches to cooperation with the Indonesian government in the management of regulatory complaints. Despite the fact that the CAO and RSPO were managing some of the same complaints that had been handled at various times by Komnas HAM, the National Land Agency, and both Provincial and District governments, none of these government processes was directly coordinated with CAO or RSPO mechanisms. The absence of institutionalised collaboration can be attributed in significant part simply to the absence of established relationships between these market and state bodies, undermining their knowledge of each other's roles and capacities, and their capacity and motivation to initiate active collaboration. Motivation to collaborate was also undermined by concerns on the part of both the RSPO and CAO about the limits of their own mandates, as foreign organisations, to collaborate directly with local governments. While the RSPO has publicly affirmed its intention to operate as a support and complement to government-led complaint-handling systems (author self-reference, 2016), RSPO staff have expressed concerns about the absence of a clear mandate to engage in such collaborations. 16 The CAO has similarly articulated potential reasons for limiting direct engagement with local government that relate to concerns about their legitimacy, including a desire to retain independence, and concerns about mission creep. 17 Deeper constraints to collaborative potential were also associated with the absence of strong complementarities in functional capacities, in the presence of close parallels between capacity gaps experienced by government and market-based regulators. The RSPO has relatively weak enforcement capacity, despite possessing significant market leverage over participating companies (Wiggs, 2016). The RSPO's contribution to the mediation of specific disputes has been further limited by persistent deficits in financial and human capacity-mirroring gaps of government complaint handling systems (Jonas, 2014, p. 28;Grassroots, 2013, p. 25). Like the RSPO, the CAO's enforcement capacities rely on soft forms of moral authority, together with market-based leverage, limiting its ability to induce compliance with desired regulatory outcomes. 18 Thus, while the overlaying of market and state regulatory mechanisms created some thickening of functional capacity within the governance sphere as a whole, these complementarities remained limited.

K. Macdonald
There are some rare examples in which productive collaborations were established at micro and meso scales, but these depended on very particular conjunctures of enabling conditions. One notable example of market-state collaboration involved the CAO's role in managing a dispute in Jambi Province between a number of local communities, and a subsidiary of the company Wilmar. After a prolonged mediation facilitated by a local NGO had run aground, the CAO established what they referred to as a Joint Mediation Team, involving direct collaboration between the CAO and selected participants from both the provincial government, and the Batang Hari District government.
Local communities and NGOs worked together with international NGOs in initiating a complaint to the CAO, and pressuring the CAO to investigate the case at the local level-bringing the CAO into direct engagement with provincial and district governments. Government recognition of the potential value of CAO involvement then depended in part on the presence of complementary functional capacities between CAO and government authorities. The district and provincial government officials tasked with managing this dispute were not experienced in handling disputes, lacked adequate staffing, and were open to drawing on the resources and expertise offered by specialist CAO mediators. According to one local government official: "Many parties had tried, but not resolved [the conflict] … Maybe we missed something, we don't understand. So if there is another party that is more professional in mediation, we really welcome that… [and] the biggest role is from the CAO, because they have more experience. They understand how this mediation process works". 19 At the same time, by enlisting government participation in mediation processes, the CAO mediation team was able to benefit from the specialized knowledge of government actors regarding the intricacies of local land management practices, the legitimacy that government involvement helped confer on CAO processes, 20 and government support for implementation of negotiated agreements. 21 Establishment of this Joint Mediation Team was however a rare moment of collaboration amidst a much longer history of conflicting and weakly coordinated interventions. And in fact despite the unusually strong commitment to institutionalised collaboration that was established in this case, these joint efforts proved insufficient to arrive at a mediated resolution to the conflict, as government efforts to support resolution at the provincial level were ultimately undermined by strong support for the company's land claims by key actors within the district government. 22 Throughout these dynamics of governance interaction, not only was collaboration approached with great caution by both transnational and government actors, it thus proved highly vulnerable to the vicissitudes of local politics.

Contested interactions between competing regulatory coalitions
In this highly fragmented governance environment, conflictual dynamics of interaction in which competing regulatory coalitions struggle for dominance were found to play a central role. By altering patterns of power and alliance amongst domestic regulatory actors, such interactions have the potential both to support and to undermine pro-regulatory coalitions-understood here as those actors seeking to strengthen social and environmental regulatory capacity. The following discussion explores the pathways and conditions under which such interactions have the potential to empower both elite coalitions promoting regulatory change through top-down processes, and more marginalised coalitions seeking to challenge the power bases of established southern elites. Analysis of the case also reveals the 16 Interview with RSPO staff, Jakarta, February 2013. 17 Interview with CAO staff, Washington DC, June 2012. 18 Even though the IFC is usually a minority lender, its participation in a given investment project can play an important role in catalysing other sources of private finance, increasing the strategic value of its involvement, and associated leverage. 19 Interview with Provincial government official, Jambi, February 2013. 20 Roundtable on Sustainable Palm Oil, "Complaints System Components & Terms of Reference," www.rspo.org/publications/download/ 33dc47007811e3d. Similarly, in a CAO-led mediation with Wilmar in Sambas, NGOs reported that the support of local parliamentarians was important in convincing the company to engage in the CAO process, lending the legitimacy and support of elected officials to bolster the authority of the CAO complaint handling process (interview with NGO officials, Pontianak, February 2013). 21 For example, in the community dispute with Wilmar located in Sambas, the CAO attempted to embed the eventual negotiated agreement in government processes, by establishing a monitoring committee involving representatives of local government officials and parliamentarians. 22 In April 2013, Wilmar sold its majority share of the local subsidiary PT AP to Prima Fortune International Ltd. and PT Agro Mandiri Semesta, neither of which have financial links to the IFC/MIGA or are members of the RSPO. This sale also coincided with a shift towards control over the dispute from the provincial level (where the farmers' unions supporting some of the community claimants had significant political support), to the district level, where alliances with the company were much stronger. This led to an agreement being signed with a group of local adat (customary) leaders to grant the concession area to the company, and the eviction of the local communities (IPAC, 2014).
countervailing strategies of regulatory resistance that such interactions can provoke, particularly when such strategies disrupt and unsettle established orderings of power at the local level.

Building elite pro-regulatory coalitions
Elite pro-regulatory coalitions have played an important role in market-state governance interactions in the Indonesian palm oil sector-both shaping these interactions, and shaped by them. The RSPO's position as a focal-point within the regulatory governance sphere has helped it coordinate processes of policy dialogue, learning and agendasetting amongst wider networks of business, NGO and government actors on issues such as Free Prior and Informed Consent of communities in the case of new land clearing activities, protection of High Conservation Value (HCV) land, and recognition of collective land title. Such interactions have played a particularly important role in supporting preventative strategies to tackle systematically recurring patterns of complaints concerning land use conflicts.
Although the government has no formal role within RSPO, such coalitions have sometimes made indirect contributions to shaping government policy, as 'progressive' positions on key issues developed within the RSPO and its networks have been subsequently taken to government via the leadership of key companies. The potentially productive effects of such interactions can be seen for instance in dynamics surrounding development of Indonesian policy in relation to strengthened protection of HCV land. RSPO standards require protection of HCV land, but there have been several instances where RSPO certified companies who refrained from developing some areas of their concessions for palm oil then had the land taken back by district governments to give to other non-RSPO members, who were willing to convert the land for palm oil production. Senior managers of several large RSPOaffiliated companies have mobilized on this issue, working together to influence politicians or senior government officials in specific sub-national jurisdictions where sympathetic government actors have been identified, and sometimes calling on the RSPO to support their efforts. 23 Particularly productive links have been built between RSPO members and members of the Central Kalimantan Provincial government (RSPO, 2011). Such coalition-building and advocacy efforts have sometimes been intensified following specific campaigns or complaint-handling processes, such as in the high-profile case of the company Golden Agri Resources (GAR), whose contributions to deforestation were targeted by sustained NGO campaigns, leading to changes in the company's internal policies, and its engagement in wider policy coalitions seeking strengthened protections for HCV land (REDD Monitor, 2012;Grassroots, 2013, p. 22).
There are also examples of such coalitions feeding into broader efforts to influence government policy at the national level. For example, the international Palm Oil Innovation Group (POIG) was formed in 2013 in the context of debates surrounding potential reforms to RSPO standards concerning issues of deforestation, carbon stocks, biodiversity and social relations, which POIG members wanted to extend. 24 The group aimed to influence both broader corporate practices in the sector, and relevant areas of government policy. Subsequently, the Indonesia Palm Oil Pledge-involving the Indonesian Chamber of Commerce (KADIN) and a number of major companies involved in the palm oil sector-was developed as a collective commitment to lobby the government to codify key elements of the pledge into law-particularly with regard to protections for high conservation value areas of land. 25 One of the advantages of such elite-focused dynamics of marketstate governance interaction is that the conditions required to establish productive interactions are relatively accessible within prevailing configurations of market and political power. Market pressure exerted by powerful investors and buyers in global supply chains helps to motivate engagement of key companies in pro-regulatory coalitions on specific issues, while the powerful position of these companies within local business and political networks increases the likelihood that their efforts will translate into regulatory strengthening at the local level.
Nonetheless, because such interactions are driven primarily at micro and meso scales, their impact is very uneven, and they are highly vulnerable to reversal in response to macro political shifts. In June 2016 it was announced that the Indonesia Palm Oil Pledge was to be officially terminated, amidst a push from the administration of President Joko Widodo (who is from a forestry background) to assert visible government control over the terms of initiatives designed to combat deforestation-a trend that has continued since that time. This political shift once again demonstrates the fragility of such issue-specific coalitions, and their vulnerability to often unpredictable dynamics of local political change, within the fragmented and pluralist governance environment of the Indonesian palm oil sector.

Empowering societal pro-regulatory networks
It is also possible to find isolated examples in which more marginalised societal coalitions supportive of strengthened social and environmental governance have developed new capacities or sources of leverage through their interactions with market-based transnational governance. If societally-driven 'fire alarm' compliance systems are to operate in ways that support strengthened social and environmental regulation, significant efforts are often needed to overcome the structural disempowerment of many prospective complainants; some specific CAO and RSPO interventions have attempted directly to counter these structural power imbalances. For example, the CAO has sometimes offered special clinics and workshops to assist community preparation for the mediation process, logistical support such as transport to help community groups attend mediation sessions, and technical support on specific issues arising in the course of individual disputes. However, provision of such support has been constrained in significant ways by concerns about possible perceptions of partiality in their dealings with conflicting company and community parties. 26 RSPO and CAO complaint handling processes can sometimes also support pro-regulatory societal coalitions more indirectly, to the extent that they provide opportunities for NGOs and communities to develop new or strengthened networks and alliances with one another, 27 or with members of elite coalitions. For example, some Indonesian NGOs reported that their ability to access both Wilmar company representatives and local government officials was enhanced by the personal and organisational networks they developed through engagement with CAO and RSPO processes, and the greater visibility and credibility that these mechanisms lent to structurally marginalised community and NGO voices. 28 Such interactions can have cumulative effects on the organisational capacity and density of pro-regulatory societal coalitions, in turn strengthening the capacity of these groups to direct claims towards government complaint mechanisms. For example, one initiative led by the Dutch NGO Oxfam Novib, together with the Indonesian NGO Sawit Watch, aimed to build the capacity of communities and their supporters to utilize both state-and market-based grievance mechanisms-sending expert teams of local pro bono lawyers to advise communities on the best ways of pursuing remedy in the case of specific grievances, and in some cases assisting communities with documentation of their claims. 29 23 Interview with RSPO staff, Jakarta, February 2013. 24 NGO staff, Jambi, September 2012. 29 For each individual complaint, the pro bono lawyers would advise communities as to whether cases should be taken to RSPO's dispute settlement facility, to the courts, or to other government channels, such as Komnas HAM, the police or local government mediation (interview, NGO staff, Jakarta, February In isolated instances, such interactions can generate productive effects-supporting community capacity to engage in broader claimmaking. However, the enabling conditions for such effects are difficult to establish: while marginalised pro-regulatory coalitions are highly motivated to promote strengthened regulatory capacity, significant and sustained shifts in power between networks of business and state actors would be needed to facilitate more sustained empowerment of communities seeking redress. As a result, such interactions have usually lacked sufficient institutional support to be sustained over time at the level of the wider governance sphere. Perhaps the best that could be said is that these processes can potentially support the pluralisation of authority and power within the governance sphere, thereby increasing attendant uncertainty about distributions of power and alliance. Some saw this kind of pressure-albeit weak-as one of the more accessible and feasible means available for pursuing regulatory change, in the face of apathetic or politically hostile governments. 30

Anti-regulatory coalitions and the politics of (de)legitimation
Another important challenge confronting efforts to deploy conflictual market-state governance interactions as a basis for strengthening social and environmental regulatory capacity involves managing the risk of countervailing pressures from more structurally powerful southern actors who regard their own interests or power as being threatened by global agendas of sustainability regulation. Particularly where market-based regulatory mechanisms have been employed as means of challenging established power relations at the local level, they have often encountered strong resistance from government and business coalitions, who are heavily invested in the sector and benefit from the status quo.
One important strategy used to resist RSPO efforts to bolster its governing authority has involved discursive efforts to portray transnational regulatory bodies as threats to both substantive and procedural dimensions of national sovereignty (Cashore and Stone, 2012). According to one producer representative: "the definition of sustainability must be defined from our own perspective … as a sovereign country with our own policy … [and] in accordance with our own law". 31 Such sensitivities surrounding the RSPO's legitimacy as a foreign organisation have significant implications for its capacity to draw on external market and political leverage as a basis for promoting sectoral change. According to one observer: "this is a huge challenge, to avoid any perception that the North is trying to increase pressure to the South -Indonesian and Malaysian speakers always bring this up [at RSPO forums]". 32 In view of such sensitivities, collaborative initiatives or influencing efforts have tended to be more successful where they have emerged around areas of shared interest -or at least where they have been successfully represented in that way. The RSPO sometimes attempts to avoid challenges to its legitimacy by highlighting subordination of its own authority to that of sovereign host state governments-demanding sensitivity when engaging with government on relevant policy issues. According to RSPO staff we spoke with, there was a clear understanding that "because we are not elected, we are strictly voluntary … we don't interfere in government legislation or rules and that sort of thing". 33 This creates significant dilemmas, to the extent that the RSPO has a clear mandate from its own constituencies to promote RSPO standards, which in some respects differ from or go beyond those promoted or institutionalised by relevant governments.
Strategies of resistance from countervailing regulatory coalitions have also been associated with complex, and sometimes somewhat obscured, dynamics of regulatory competition or rivalry, which the RSPO endeavours to manage with great delicacy. In both Indonesia and Malaysia, there have been prominent attempts to establish governmentcontrolled rivals to the RSPO. In Indonesia, development of the Indonesian Sustainable Palm Oil system (ISPO) has been ongoing since 2011, and a Malaysian Sustainable Palm Oil (MSPO) certification scheme was established in 2014. Such initiatives were catalysed in significant ways by external pressure from international NGOs and regulatory frameworks, including those within the RSPO. For example, one of the ISPO's early promotional pamphlets contextualised the organisation's creation by stating that "In recent years, Indonesia's palm oil sector has faced various problems, including never-ending allegations by NGOs, … demands from the members of RSPO and conditions imposed by the EU" (Indonesian Palm Oil Commission, 2011, p. 4).
The effects of such regulatory interactions are mixed. To the extent that such rival initiatives entrench recognition of the legitimacy of sustainability discourses within key producing country locations, they could be interpreted as contributing to complementary forms of regulatory layering-and indeed the RSPO and the ISPO are eager to publicly highlight complementarities between them, and their shared commitment to promotion of a more sustainable palm oil sector (Yulisman, 2014). According to one proponent of ISPO: "Although I said we're different from RSPO, we're not trying to stop them, we have no intention to undermine them or to stop RSPO. So our companies can be certified by anyone else too". 34 At the same time, however, the creation of the ISPO has been used by some actors as an explicit means of resisting the authority of the RSPO-instead 're-centring' authority in the state (Bartley, 2014). According to an ISPO representative: "We're a sovereign country. We're in a much higher position than any organisation could ever be … They're trying to undermine Indonesia, saying that we have the regulations but we're not implementing them properly. So the Ministry of Agriculture decided to have its own organisation. We are the biggest producers, exporters and consumers of palm oil so we should control ourselves. We don't need outsiders -NGOs -to control us". 35 This sentiment resonated strongly with comments from an NGO heavily involved in the Indonesian palm oil sector, who observed: "the government also sees that palm oil is contributing significantly to the national economy, and globally Indonesia has become the largest palm oil exporter since 2007. So it creates a kind of nationalistic pride that as a major producer of palm oil, Indonesia has to have its own standard". 36 In practice, co-existence alongside the ISPO generates some additional complexities for the RSPO's operations-challenging its status as a central focal point for networking and agenda-setting around standards in the sector. In the judgement of some observers, existence of the ISPO also makes it easier for firms to exit the RSPO when they have disagreements with RSPO policies or procedures. Some interpreted the business association GAPKI's exit from RSPO in favour of engagement with ISPO as evidence of such competitive dynamics, and indeed according to one participant in GAPKI: "[a key] consideration why we withdrew [from the RSPO] is that the Indonesian government now already issues ISPO, and this is mandatory for palm oil producers in Indonesia". 37 Such competitive dynamics can limit the RSPO's influence over broader processes of policy dialogue in significant ways, given the political weight of GAPKI and its connections with government. 38 (footnote continued) 2013). 30

Lessons and conclusions: from collaboration to constructive conflict in public-private regulatory interactions?
What key lessons can we can draw from the Indonesian palm oil sector, with regard to our central questions about the effects of marketstate governance interaction on regulatory capacity in the global south? Significantly, neither collaboration nor conflict wholly captured the spectrum of interactions that were documented. While some collaborative regulatory interactions generated productive effects on regulatory capacity, conflictual interactions also had productive potential, though this was sometimes strongly resisted-generating a complex and continually evolving interplay between collaborative and conflictual dynamics, at multiple sites and scales of interaction. Both productive and competitive interactions were thus shown to operate in parallel, reflecting the non-unitary structure of state authority, the pluralism of transnational regulatory processes, and the resulting complexity of associated governance interactions.
In the most positive examples of collaborative interactions, transnational regulators have interacted with, and enhanced the capacities of, governments, NGOs, communities, and companies-occasionally establishing formal mechanisms of collaboration, and often lending some indirect assistance to the development of new coalitions inside and outside of the state, in support of regulatory change. Overall, however, such governance interactions demonstrated limited capacity to destabilize dominant constellations of power at the local level, in ways that would have more meaningfully empowered pro-regulatory coalitions. Moreover, visible efforts by transnational regulators to influence national and sub-national regulations proved highly vulnerable to (de)legitimation politics, through which business and government actors at the local level sought to resist transnational regulatory agendas by invoking concerns about national sovereignty.
From the perspective of this special issue, these dynamics are particularly significant insofar as they highlight the plurality of competing southern responses to global market-based governance initiatives, and the way in which global governance mechanisms can be appropriated or resisted by southern actors in the context of wider local struggles over contested regulatory change.
This analysis also has broader implications for wider scholarship on market-state governance interactions insofar as it leads us to question widespread assumptions about the distinctive value of collaborative governance interactions as a basis for generating productive effects on social and environmental regulatory capacity. As we have seen, conflictual interactions are also capable of supporting productive effects, to the extent that they can contribute to motivating, empowering and legitimizing the agendas of pro-regulatory coalitions within commodity producing jurisdictions.
Nonetheless, as we have also seen, while conflictual pathways of regulatory empowerment can sometimes be productive, their effects on destabilizing power relations between elite and marginalised actors in producing countries render them distinctively vulnerable to legitimacy challenges from incumbent powerholders. Such dynamics of regulatory contestation between competing coalitions in commodity-producing locations led transnational regulators to face a challenging dilemma: while exercise of meaningful influence often demands political logics of action, which can shift power balances between competing regulatory coalitions, maintaining legitimacy as external actors requires that they avoid being perceived as engaged in overtly political activities-containing them within more technocratic, depoliticized roles (Auld et al., 2015). Such dilemmas may be expected to be particularly intense in contexts where neo-colonial critiques of global norms carry significant discursive weight (Randeria, 2007;Benhabib, 2009;Bernstein and Cashore, 2012, p. 602).
In managing such dilemmas, transnational market-based regulators need to make complex judgements about their potential leverage in relation to host states, the relative strength of potential allies and opponents within different parts of the state, and political sensitivities about engagement with sovereign governments. While there can sometimes be significant opportunities to forge productive collaborations or divisions of labour, the political space for such interactions can very quickly shrink, in response to either unrelated dynamics of local political change, or concerns amongst local actors about threats to their positions of established power or privilege. Such calculations play out quite differently for private regulatory bodies of different kinds, in different national and sub-national contexts, at different points in time, and at different stages of the regulatory policy cycle. Making such judgements calls for a contextually sensitive and highly adaptable approach.
The above analysis has focused on one relatively neglected element of the regulatory process-the operation of 'fire alarm' mechanisms of complaint handling. Nonetheless, many of these findings have potentially significant implications for broader debates on governance interactions, which have often focused primarily on regulatory processes of standard-setting, monitoring and enforcement. Of particular salience are the paper's findings regarding the importance of political contestation surrounding the legitimacy of private regulators. This emphasis goes beyond analysis of conflicts between competing policy objectives, as highlighted for example by the recent analysis by Pacheco et al. (2018) examining antagonisms between public and private palm oil governance. It also stresses the significance of deeper conflicts over power and legitimacy within governance processes-not only regarding specific contested policy outcomes, but also concerning who has a right to participate in policy making processes in the first place.
Overall, what then might we conclude in relation to our central animating puzzle, of how to move beyond small pockets of strong private governance, to support more broad-based strengthening of regulatory capacity for sustainable commodity production (Cashore et al., 2011)? While conflictual governance interactions can be productive to the extent that they empower pro-regulatory coalitions, the conditions for such productive interactions to occur can be difficult to establish, as established power hierarchies in southern jurisdictions prove persistent, and regulatory coalitions seeking to challenge them provoke intensified contestation over the underlying legitimacy of transnational interventions. In seeking to strengthen the impact of sustainable commodity governance, we must therefore search for answers not only in an analysis of interactions amongst regulators themselves, but also through sustained attention to the (often unintended and unpredictable) effects of these interactions on broader configurations of social alliances and power-across the plural sites of authority and social organisation within which transnational regulatory governance processes are constituted.