Introduction
When a United Nations (UN) humanitarian mission transmits cholera to a population devastated by an earthquake, when UN peacekeepers engage in abuse, or when an International Monetary Fund (IMF) loan exacerbates social vulnerability, observers and scholars are quick to raise questions about the accountability of international organisations (IOs). Indeed, the accountability of IOs is a prominent topic of academic and advocacy debate. The debate first reflected the waves of democratisation since the late 1980s and subsequently the discontent with the power of international economic institutions and the failures of IOs to protect the populations that they are intended to serve.Footnote 1 The expansion of the European Union (EU) and the deepening of supranational rule-making similarly raised concerns about a democratic deficit and accountability.Footnote 2 These questions persist with respect to the performance of IOs, and salient events or controversies surrounding international institutions.
Academic and public debates initially focused strongly on the question of whether international agencies are accountable or not.Footnote 3 Subsequently, the literature elaborated on the concept of accountability as it applies to international governance, examining the mechanisms through which such accountability can be exercised in the absence of a broadly participatory public process.Footnote 4 More recently, studies have grappled with the accountability claims by diverse sets of external audiences, including affected populations, civil society organisations, and private donors. Such claims open new considerations about accountability as a source of authority and legitimacy of multilateral institutions.Footnote 5 Competing accountability claims can furthermore produce contradictions,Footnote 6 while there is less understanding about how such contradictions might be resolved. The literature has thus generated important conceptual insights on the nature and limitations of international accountability. However, the comparative understanding of the variable extent to which IOs have been accountable – to which actors and with what effect – remains limited. We lack an overarching theoretical framework that interrogates the conditions that shape how the different dimensions of accountability may vary and interplay across institutional settings and over time, and the implications for accountability systems and practice.
Accountability is defined here as a relation and a set of processes, whereby an IO, as an accountability giver, has an obligation to provide information and justify its actions to accountability holders, who can pass judgement and impose consequences.Footnote 7 This article aims to advance the theorising of international accountability in several ways. First, it introduces the concept of scope of accountability as a set of dimensions of accountability and their interface, in order to better understand the complex and variable dynamics of delivering accountability in practice. Such conceptualisation allows us to unpack and scrutinise the politics of accountability systems, which are both institutionalised, but also subject to contradictions, contestation, and flux, in response to evolving claims, normative expectations, or organisational failures.
Second, the article theorises how political conditions, namely the exercise of power, institutional design, and public salience of IOs, are likely to influence the ability of organisations to deliver different types of accountabilities and the contradictions that may result. More specifically, the theoretical framework stipulates how these three factors may affect tensions between and within two fundamental dimensions of accountability systems, namely access by different accountability holders and the IO’s capacity to effectively deliver multiple accountabilities in accordance with the public mandate of the organisation. This theoretical approach is, to the best of our knowledge, the first systematic attempt in International Relations (IR) to interrogate the interface between different types of accountabilities and the political conditions that shape accountability systems and outcomes. The integrative framework enables us to shed new light on persistent trade-offs and dilemmas, inviting further empirical and comparative exploration to understand the sources of both gains in accountability, as well as instances of dysfunction.
The article proceeds as follows. The first part reviews existing perspectives on the accountability of IOs. In dialogue with these perspectives, it advances the rationale for a synthesis that moves beyond binary debates of whether IOs are accountable or not, in order to identify institutional and political factors likely to increase or limit the scope of international accountability. The second part elaborates on the concept of a scope of accountability of IOs, as a basis for an integrative approach to analysing how systems of accountability work. The third part of the article, in turn, advances a set of propositions on how power, institutional structure, and peaks in salience affect the variation in accountability systems in general, and in particular the access-efficiency frontier. The conclusion discusses the implications of this integrative theoretical framework for understanding variable accountability outcomes, as well as persistent dilemmas. It outlines an agenda for expanded research on complex accountabilities and their relevance for policy and for the legitimacy of multilateral institutions.
Perspectives on international accountability: a fragmented field
Are IOs accountable? This question remains prominent in public debates, reflecting the complexities of global governance. After the end of the Cold War, pundits declared ‘the end of history’ with the ascendance of liberal democracy and celebrated the prospect of a cosmopolitan society.Footnote 8 International institutions promoted market liberalism and the virtues of democratic governance. As IOs, such as the World Bank, the IMF, and the World Trade Organization, gained power and visibility, the accountability of international agencies for the social and environmental effects of their work was questioned. The questioning has intensified ever since, with respect to the broader role of IOs in global crises on issues that include health, conflict, and natural disasters, and on issues of justice and human rights.
The scholarly literature has been divided in its treatment of the theoretical and practical prospects of achieving greater accountability by IOs.Footnote 9 Democratic theory rests on the fundamental concern that accountability ought to be exercised ultimately by the people and more precisely by a political community that has granted authority to public institutions and is affected by those who govern.Footnote 10 Yet, as Kriesi summarises, ‘The difficulty of a democracy beyond the nation-state is that there is no corresponding community of solidarity at the global level….’Footnote 11 In the absence of an understanding of such a community, the possibility of ultimate accountability to the people – and of sanctioning through direct public participation in a political process – is largely absent with respect to international decisions.Footnote 12 Thus, from the perspective of democratic theory, there are inherent limitations to the prospect of international accountability.Footnote 13
More optimistic readings on democratic accountability emphasise that the participatory quality and responsiveness of global governance institutions can be expanded through overlapping decision-making forums.Footnote 14 Such overlapping institutions and networks link, for instance, EU institutions with domestic agencies, parliaments, and sub-national constituencies, and provide channels through which domestic actors could demand accountability from both national and supranational institutions.Footnote 15 Nonetheless, from a democratic theory perspective, the concern is that transnational networks may privilege horizontal accountability to peer organisations, experts, and advocacy groups, rather than responsiveness to people or national legislatures. This could reinforce a technocratic turn in multilevel governance and ‘policy drift.’Footnote 16 The public accountability of unelected transnational organisations as ‘surrogate’ accountability holders is further questioned.Footnote 17
Another line of theorising has proposed pluralistic approaches to understanding international accountability. Departing from a delegation model of states as the primary principals of IOs, Grant and Keohane identify a plurality of mechanisms for eliciting information, justification, and (to a lesser extent) imposing consequences.Footnote 18 Scholars of the EU further identify multiple dimensions along which accountability practices and deficits can be evaluated with respect to expectations for public accountability.Footnote 19 Pluralist conceptualisations of accountability have provided a basis for documenting the accountability structures and mechanisms of different IOs,Footnote 20 as well as further conceptual elaboration on specific accountability dilemmas, such as the implications of the proliferation of competing claims and mechanisms,Footnote 21 or the limitations of transparency and reputational mechanisms.Footnote 22
Finally, multiple perspectives have elaborated on the normative underpinnings of accountability as it relates to international agencies. Works in the social constructivist tradition identify mechanisms through which normative pressure and agency have contributed to redefining the nature of IO accountability, for instance, through the diffusion of norms for democratising representation and internal procedures,Footnote 23 or through the politics of demanding accountability for the rights of affected populations and environmental justice.Footnote 24 Scholars working in the constructivist tradition have furthermore conceptualised bureaucratic pathologies as fundamental failures of international agencies to be accountable for their constitutive norms and purpose.Footnote 25 In turn, legal scholarship has interpreted international human rights law to advance arguments that IOs ought to be accountable for the human rights and integrity of the populations that they serve.Footnote 26 In her work on pluralistic layers of accountability, Hirschman argues that international human rights principles can be applied to IOs as standards for accountability to the extent that they can be treated as ‘customary international law,’ and given their centrality to the norms of the multilateral system and the obligations of member states to comply with such standards.Footnote 27 Significantly, the substantial literature on legitimacy as a fundamental conduit of the authority of international agencies clarifies the relevance of democratic norms and expectations for procedural and output-based accountability, as referent points for legitimate governance.Footnote 28 In other words, normativity has profoundly shaped and reshaped the understanding of what legitimate objects and standards of IO accountability are.
Where do these perspectives leave us? The literature embarked by debating whether IOs are accountable or not, and has made significant advances in providing foundations for thinking about international accountability and documenting accountability mechanisms. However, despite these advances, the understanding of the accountability of international agencies remains fragmented. Each conceptual perspective works within its apparatus to gain depth. As in the parable of sensing the parts of the proverbial elephant rather than seeing the full complex creature, different aspects are theorised and cases examined, with a predominant and perhaps understandable interest in dysfunctions. However, less attention has been given to how these multiple dimensions interface to shape systems of accountability. Importantly, there is limited overarching theorising to explain variation in accountability systems across institutions and change over time. Yet, such analysis is of critical importance to inform public debates and the relative trust in international governance. The IR literature ought to be better equipped to help address more fine-grained questions about variable accountability practices. What combination of accountability mechanisms are most relevant in the context of different IOs? Who are the relevant accountability holders, and according to which standards? Can we consider the trade-offs, contradictions, and complementarities within accountability systems in a more systematic manner?
In dialogue with existing research, the following sections seek to articulate a new integrative theoretical approach that provides a framework to analyse how different dimensions of the scope of accountability of IOs interact in specific contexts. The framework further stipulates a set of political conditions that can help to account for variation and change in the scope of accountability systems along two core dimensions, namely broadening the access to accountability systems by internal and external accountability holders and their ability to effectively elicit accountability, in ways that are aligned with the public mandate of the organisation.
The scope of IO accountability
The scope of accountability of IOs includes the range of accountability givers (those subject to accountability constraints) and accountability holders (those with authority and/or capacity to demand accountability), and the set of mechanisms through which accountability holders can elicit information, justification, and potentially impose consequences, in reference to the norms and standards associated with the mandate of the organisation. The agency as a collective body and its staff are accountability givers, while member states, intergovernmental forums, and other internal or external forums, such as civil society, comprise the accountability holders. Thus, public accountability is a fundamentally constitutive relationship between a public agency and a set of accountability holders, of which the scope and normative standards may vary and evolve in world politics.
Figure 1 attempts to capture, in a schematic manner, the complexity and inter-connectedness of different layers in systems of IO accountability. The following discussion unpacks the three core dimensions – who are the accountability holders, what are the objects and standards of accountability, and through what mechanisms they are pursued – in order to illuminate how their variation and interface produce accountability outcomes that are more complex and variable than previously recognised. While questions such as to whom agencies are accountable, for what, and how are a staple in accountability studies of bureaucracies at large,Footnote 29 the novelty of the scope of the accountability conceptual framework is its synthetic focus on complexity and interdependencies within accountability systems. It offers an integrative approach that highlights the interface between different types of accountabilities and frequent contradictions that have implications about who gets to exercise accountability and to what effect. Such conceptualisation challenges existing debates and assumptions of accountability as a binary or static feature of international institutions. The scope of accountability is thus not necessarily fixed or finite, but rather a dynamic system of multiple dimensions whose working is fundamentally influenced by politics. It can deliver multiple accountabilities, while at the same time create and potentially address contradictions and distinct dilemmas.
Figure 1. The scope of accountability of IOs: an integrative framework.
The upper inner layer of Figure 1 captures the recognition that member states are the most direct accountability holders of IOs, as stipulated by delegation theory and the nature of IO authority as public, intergovernmental institutions.Footnote 30 Indeed, member states and intergovernmental forums are the only accountability holders with supervisory authority vested by the legal instruments establishing IOs. They also have formal authority to create, modify, and approve administrative instruments through which accountability is sought and delivered within organisations (upper layer of Figure 1). However, because accountability is a constitutive relationship between a public agency and its different accountability holders, it is a more complex and contested process than implied by the strict assumptions of delegation models that principals ought to be able to control the agency. In some organisations, member states elicit formal accountability through multiple forums (e.g., Assemblies, committees, and executive bodies), whereas in other cases, an executive body (e.g., Board of Directors) is the main supervisory forum, with implications about what kinds of mechanisms different states have access to (Figure 1). Moreover, not all principals are alike. IOs are likely to be most strongly accountable to principals who have the political and financial power to demand answerability and to impose consequences via resource allocation, membership withdrawal, or appointments of senior staff. With respect to member states that are recipients of funding and projects, there is often a dual relationship in which IOs hold recipient states accountable for implementation, while also being accountable to them as part of intergovernmental forums.Footnote 31 Some IOs manage mechanisms such as conditionality, reporting, or periodic reviews (e.g., in development and human rights organisations, respectively) for keeping states accountable for international commitments and reciprocity.
Furthermore, the strict assumption according to which IOs are accountable largely to states has been challenged with time on several theoretical fronts and in practice. Democratic theory distinguishes between a delegation logic of accountability and accountability for affectedness by public institutions.Footnote 32 In IR, affective accountability by IOs has long been compromised as a consequence of the presumed role of the state as an accountability holder internationally and accountability giver domestically. At the same time, as Woods observed, organisations such as international financial institutions expanded their work ‘to include some of the most basic decisions about the budget and economic structure [of recipient countries]… Yet while their responsibilities have increased their accountability has not.’Footnote 33 This quote is illustrative of the beginning of a normative shift to press for, legitimate, and institutionalise a normative stipulation that IOs ought to be accountable to affected populations. Such articulation has taken place since the 1980s both through political struggles and advocacyFootnote 34 and through theoretical articulation and interpretation of human rights law.Footnote 35 Many organisations now recognise the norm of ‘social accountability,’Footnote 36 and some, but not all, have adopted formal administrative mechanisms (e.g., The World Bank Inspection Panel; redress mechanisms) intended to implement the principle (Figure 1, lower-left side).Footnote 37
In addition, the rapid proliferation of non-state actors, including non-governmental watchdogs, but also advocacy and private organisations, foundations and public–private partnerships working directly with IOs has expanded the range of external forums that can de facto make accountability claims vis-à-vis IOs. Such actors could also impose consequences through means of resources and reputation (Figure 1, lower-right side). These developments have produced turbulence in accountability systems that reflect the overlaps and fluidity of authority in contemporary global governance.Footnote 38 In sum, the scope of IO accountability in terms of who gets to be an accountability holder varies across organisations and over time. It entails multiple and often contested relations in response to evolving politics, norms, and practice, while nonetheless maintaining (at least in the present) the hierarchy of accountabilities towards intergovernmental forums (Figure 1).
The second dimension in the scope of IO accountability comprises the instruments that accountability holders can deploy to elicit transparency, answerability, and passing judgement or consequences. Such mechanisms occupy a central place in terms of achieving accountability outcomes with respect to the norms and standards of the organisation, being the essential link between to whom IOs are accountable and for what (Figure 1). Drawing on the existing literature that has elaborated pluralist typologies of international accountability mechanisms,Footnote 39 Table 1 summarises the types of mechanisms that constitute this layer in the scope of IO accountability. The typology is expanded here to distinguish between formal and informal mechanisms. Formal mechanisms are those encoded as part of the legal and administrative apparatus of the organisation. Informal mechanisms typically respond to incentives and social processes, for instance, competition for resources, reputation, informal influence, and vetting by peer organisations. They are not necessarily mandated by or implemented through administrative rules and can apply both internally and, in particular, to external accountability audiences.
Table 1. Accountability how: a typology of mechanisms *
* Examples of instruments and processes are illustrative, not a comprehensive list
Behind the seemingly static typology in Table 1, there are a myriad of politics that drive their workings. Formal mechanisms intend to reduce information asymmetries with respect to intergovernmental forums and elicit accountability through evaluation, program approval, financial oversight, external oversight functions, and budgetary decisions, among others. However, organisations differ considerably in the extent to which they rely on supervisory structures that are broadly representative of member states and have deliberative functions. The UN and the World Health Organization (WHO), for instance, rely on Assemblies for policy deliberation, review of the work of organisations, and intergovernmental oversight. At the same time, core policies at the WHO are vetted by an executive body with strong influence by donor countries, while the UN Security Council decisions depend on the veto power of permanent members. In other institutional settings, for instance, the World Bank and the IMF, primarily executive bodies provide supervisory oversight. Administrative mechanisms of internal oversight and evaluations are crucial in terms of the flow of information and learning. They are intended to improve performance and transparency and support oversight by supervisory bodies. At the same time, the proliferation and complexity of such administrative mechanisms could create risks of proceduralism that may reduce substantive accountability and insulate organisations from external scrutiny.Footnote 40
External audiences, such as peer and advocacy organisations, have access primarily through informal mechanisms (Figure 1 and Table 1). Overall, however, with the increase in connectivity, a plurality of non-state actors, and their roles as funders of international programs, the exposure of IOs to market-based, peer, and reputational mechanisms has increased. Over time, non-state actors have also gained formal procedural access to IOs, although to a variable extent across IOs and with limited ability to directly influence formal accountability.Footnote 41 As a consequence, the legitimation of hybrid authority in such fields critically depends on functioning accountability structures that attempt to reconcile the public purpose of IOs and their growing engagement with non-state actors.Footnote 42 There is a certain fluidity that exists in practice between the functioning of formal accountability structures that pertain primarily, although not exclusively, to intergovernmental forums, and informal accountability mechanisms that can be activated more broadly by external audiences (Figure 1).
Finally, the object of IO accountability (for what) is, in turn, highly interdependent on who the accountability holders are and on organisational mandates and norms that provide the fundamental standards against which accountability ought to be assessed (Figure 1, outer layer). Surprisingly, the variable scope along this dimension has been explored to a more limited extent in IR. As public agencies, IOs are ultimately accountable for their performance in implementing their mandate and programs, and for following procedural norms of fairness, impartiality, and effective and transparent management of public funds.Footnote 43 In addition, there are expectations regarding the personal qualities of public servants in terms of competence, integrity, and management skills. The management of international agencies also ought to be accountable for providing a safe working environment for its employees. At the same time, the intergovernmental authority of international agencies implies that they are accountable for responsiveness to the preferences of states, the predominant focus of delegation models. However, the preferences of states often diverge with respect to policy and organisational priorities and sometimes may contradict broader public purpose or send conflicting mandates to IO staff. Should the IMF be accountable for engaging in policies to address climate change? When the organisation took a proactive stance on climate, some of its Board members expressed strong dissent.
Thus, different conceptualisations of the nature of international agencies and their accountability holders may privilege different objects of accountability, which could produce contradictory signals and accountability outcomes. Furthermore, the delivery of accountability for specific organisational mandates also depends on appropriate mechanisms, through which internal and external accountability holders can advance claims. The assessment of performance itself depends on the audience; it is in the ‘eye of the beholder.’Footnote 44 This could create gaps in accountability when satisfactory outcomes for some accountability holders have negative implications for others.Footnote 45 For example, the reforms for which environmentalists pressured the World Bank in the 1980s and 1990s were met with variable support and opposition by different member states, reinforcing a lacklustre responsiveness.Footnote 46 When such contradictions are exposed, for instance, through advocacy pressure or legal action, the very objects and standards for accountability can evolve. The institutionalisation of principles, such as the right not to be harmed, the right to access information on and safety from environmental harms, or the right to health and access to medicines, are examples of the evolving scope of the objects of IO accountability.
In summary, the scope of accountability concept seeks to provide a novel tool for understanding accountability as a complex and systemic phenomenon, with interdependent dimensions that can expand in terms of access, mechanisms, and standards for accountability. The scope could be narrower (largely determined by states on the governing board of a formal IO) or broader (including multiple intergovernmental forums, non-governmental organisations [NGOs], and private donors). This conceptual framework provides an approach for critically exploring variation and changes in systems of accountability over time. It also raises the question if a broader scope of accountability means that an accountability system is likely to be more effective. The answer is conditional. We cannot always assume a positive complementarity across mechanisms. As the discussion illustrated, tensions and potential trade-offs are frequently inherent in the politics of accountability. The following section articulates a set of theoretical propositions on the political conditions that are likely to shape the scope of IO accountability, focusing more specifically on the access to and effectiveness of accountability systems.
Arbiters of accountability: power, institutional design, and salience
Under what conditions are IOs likely to be more or less accountable for their decisions, operations, and practices? While the literature has engaged in debates on whether or not IOs are accountable and the nature of accountability mechanisms, we lack an integrated theoretical elaboration on the conditions that influence who can elicit accountability, for what, and what mechanisms are triggered or blocked.
Drawing on theories of democratic accountability and international institutions, this section advances a theoretical framework that stipulates that the exercise of power, the relative decentralisation of institutional structures, and organisational salience are key conditions that shape the scope of IO accountability in terms of who has access to holding IOs accountable and for what, and to what extent they can effectively pursue accountability claims in practice. The discussion and propositions focus more specifically on how such conditions affect outcomes with respect to the interface between access to accountability systems by different accountability holders and efficient delivery of accountability, which are critically important dimensions of accountability systems for several reasons. First, both access by accountability holders and the efficacy with which they can elicit accountability are necessary for effective accountability practice. Theories of democratic control state that access by a plurality of audiences via formal and informal mechanisms is important for the deliberative quality of governance (informing and explaining) and the activation of supervisory mechanisms that could impose consequences. Access is a necessary but insufficient condition, however, as accountability holders should also have the capacity and political will to impose consequences and/or change. Second, as already illustrated, the extant literature has documented multiple potential contradictions and dysfunctions that occur along the access-efficacy frontier but has not theorised the political conditions likely to shape different outcomes. Finally, critical questions about the procedural and democratic quality of governance and justice in terms of who gets to hold IOs accountable and for what purpose depend greatly on how tensions along this frontier are resolved. The three political factors identified by the theoretical framework are discussed in turn.
The exercise of power
Rendering a public agency accountable is in itself an exercise of power by formal principals or other legitimate accountability forums. Power has been defined as the capacity of an actor to influence outcomes and advance strategic goals.Footnote 47 Constructivist and critical perspectives further distinguish between direct ‘compulsory’ power and more diffused forms of ‘structural’ and ‘institutional’ power relations that are embedded in social structure and subjective meanings.Footnote 48 Propositions 1 and 2 stipulate how the differential manifestations of power are likely to affect the scope of accountability and its access-effectiveness outcomes.
Proposition 1. The exercise of asymmetric power by a subset of principals is likely to influence institutional equilibria of accountability systems in ways that secure more effective accountability of IOs in their favour. However, such capacity to demand strong accountability may lead to reduced access to accountability by less powerful accountability holders and forums, particularly in contexts of competing objectives.
Proposition 2. Asymmetric power could also be used to reform accountability systems towards a greater scope of access and responsiveness for the mandate objectives and norms of the agency.
In IR, both institutionalist and realist theories stipulate that power is likely to broadly affect the nature of institutional equilibria,Footnote 49 particularly their distributional implications.Footnote 50 Thus, powerful states substantially influence, in the first place, both institutional structure and the objects of accountability that are negotiated. They can also deploy informal and formal mechanisms to effectively elicit accountability and change when dissatisfied with IO performance.Footnote 51 Projecting control takes place through concrete mechanisms such as voting quotas or a veto in formal structures, budgetary decisions, and informally through shifting resources, or increasing voluntary and project-based financing for specific programs (Figure 1). However, while such deployment of power may increase accountability to governing Boards or to specific donors or client countries with strategic significance,Footnote 52 it may de facto undermine accountability for broader mandate objectives and access to accountability by other accountability holders.Footnote 53 Critics of international financial institutions in the early 2000, for example, argued that the strong accountability of IOs to large shareholders on their Boards and informally to the financial market compromised their accountability for affectedness to the governments receiving structural adjustment programs, in effect disenfranchising populations that bear the social costs of austerity.Footnote 54 Scholars have documented the significant capacity of the United States as the largest shareholder of international financial institutions, along with other significant donors, to demand accountability and impose consequences.Footnote 55 Woods and Narlikar have argued that such strong and asymmetric accountability ‘magnifies and highlights the gaps in the institution’s accountability to all other stakeholders.’Footnote 56
Nonetheless, powerful states on organisational boards can also use their leverage to broaden organisational accountability and access by accountability holders. Such political dynamics are illustrated by expanding accountability systems for affectedness with respect to the environmental and social impacts of international financial institutions.Footnote 57 The creation of the World Bank Inspection Panel and new accountability mechanisms for grievance and redress in multilateral agencies provide another set of examples of institutional changes, underwritten by powerful shareholders and supervisory forums, towards broader access.Footnote 58
More diffused forms of structural and institutional power could affect the fundamental relations of who gets to be an accountability holder and giver, the objects of accountability, and the capacity to make accountability claims. For instance, through mutually constitutive categories, such as ‘donors’ and ‘recipients,’ a set of countries that are formally accountability holders of IOs become accountability givers with respect to resources received from donors or via international agencies. Moreover, when donors prioritise accountability mechanisms that emphasise competition, and cost-efficient and quantifiable results, they elicit a high level of targeted accountability, which could, however, shift resources and accountabilities away from broader mandate objectives and the very populations IOs are intended to serve.Footnote 59 Power can thus enhance the efficient delivery of accountability through supervisory, financial, and administrative mechanisms and informal influence, but can also undermine access primarily through institutional and structural politics. Structural theories of power direct our attention to paths of disempowerment that may compromise accountability for affectedness and the implementation of core normative mandates.Footnote 60 Such perspectives suggest that both recognition of affectedness and increased capacities of accountability holders are important conditions for more effective access to accountability.
Institutional structure
The significance of institutional structure in establishing multiple, overlapping mechanisms for achieving accountability of public agencies cannot be overstated. Beyond elections or centralised enforcement, the delivery of accountability by public agencies in domestic politics relies on institutional checks and balances, as well as on political oversight within governance bodies.Footnote 61 Democratic theory further emphasises the role of civic associations, and professional and advocacy organisations as horizontal ‘fire alarms,’Footnote 62 which foster knowledge and social capital to demand and gain public accountability.Footnote 63 Institutional structure, therefore, underwrites in important ways how public accountability works with respect to domestic as well as international agencies. Taking such considerations into account, Propositions 3–5 stipulate the implications of decentralised institutional structure for the access-efficacy frontier in IO accountability systems.
Proposition 3. Decentralisation of the institutional structure of an IO is likely to increase access by a broader set of accountability holders.
Proposition 4. Decentralisation of institutional structure is further likely to strengthen the efficacy of mechanisms for procedural and deliberative quality (informing and providing justification) as part of accountability systems.
Proposition 5. The implications of institutional decentralisation for imposing consequences and efficient delivery of accountability may be ambivalent and dependent on the capacity of supervisory bodies to reduce information asymmetries and arbitrate consequences in favour of the public mandate and standards of the organisations.
Propositions 3–5 capture the complexity of the constitutive relations between institutional structure and the objectives of access, deliberative quality, and efficacy that are essential to accountability systems. The nexus between institutional structure and the often competing objectives of accountability thus deserves further elaboration. To begin with, the degree of institutional decentralisation varies considerably across IOs in two distinct ways. Within an organisation, it involves the devolution of decision-making and accountability functions across multiple forums, each with a different degree of representation. On a territorial and jurisdictional level, the supervisory functions can be more centralised in the headquarters of an organisation, or they can be more distributed with offices in countries of operation and geographical regions. Such institutional variation reflects the political dynamics at specific historical junctures as well as the nature of organisational mandates.Footnote 64
For instance, the UN Security Council and the Bretton Woods institutions were established with structures seeking to assure that those with power also have the responsibility and means of control to underwrite economic stability and support peace. Such institutional designs sought to correct the factors behind the failure of the League of Nations, which lacked such mechanisms to act in the face of crises. In the UN system, organisations such as UNICEF and WHO have club-like supervisory bodies dominated by donor countries (e.g., the UNICEF Executive Board and the WHO Executive Board), which, however, co-exist with other broadly participatory supervisory forums, such as the General Assembly and the World Health Assembly, respectively, and their specialised committees. Greater devolution of supervisory forums provides opportunities for more meaningful access and oversight by the broad membership of the organisation, and importantly, platforms for deliberation, explaining, and rendering of accounts as important procedural grounds for accountability processes and democratic legitimation (Proposition 4).Footnote 65 Figure 1 illustrates the co-existence of multiple possible accountability forums, which in practice vary substantially across organisations.
An accountability episode of the reinstatement of UNICEF as a permanent UN agency in 1953 is illustrative of the significance of broader access. According to the UN historian Richard Jolly, the Fund was reinstated as a permanent agency despite scepticism by powerful donors and competing UN agencies, in great part because of deliberations in the Executive Committee and in the UN General Assembly. The latter provided a forum for developing country representatives to argue that UNICEF’s efficacy in reaching children in the reconstruction of post-World War II Europe was urgently needed for children in the context of decolonisation and development.Footnote 66 While access to deliberation through overlapping forums does not necessarily mean equal responsiveness to accountability claims, it increases opportunities for checks and balances and a broader scope of account rendering.
Decentralisation also takes the form of devolving some level of authority to IO offices at the regional and country levels. Structures vary between traditional high centralisation at the Headquarters level in agencies such as the IMF and the World Bank, and historically more decentralised structure of the United Nations Development Program (UNDP) or UNICEF, for instance, with representation on the ground and accountability mechanisms vis-à-vis national governments and vertically to supervisory bodies. Cross-jurisdictional decentralisation makes international agencies more ‘visible’ to and accessible to societies affected by their work. It facilitates the activation of two-level accountability mechanisms through reputation and peer accountability, or transactional networks of NGOs (Figure 1).Footnote 67 It can improve access by domestic government agencies and parliaments, which implies greater domestication of the objectives for which IOs are held accountable, albeit through softer mechanisms such as informing, explaining, and in some instances formal evaluation.
However, while both cross-jurisdictional decentralisation and layering of accountability forums are likely to expand access in the scope of IO accountability, the implications for the efficacy of accountability practice are more ambiguous (Proposition 5). On the one hand, cross-jurisdictional decentralisation has historically contributed to both broader and stronger accountability by organisations such as UNDP and UNICEF to national governments and societies.Footnote 68 In addition, decentralisation and access by informal accountability forums and reputational mechanisms have so far been the primary mechanisms for signalling abuses of power in the field, be it by peacekeepers, health workers, or humanitarians employed or mandated by IOs.
On the other hand, decentralisation can create challenges for efficient and equitable rendering of accountability. Cross-jurisdictional decentralisation increases the distance and informational asymmetries for supervisory bodies, potentially reducing their ability to effectively control the bureaucracy and field operations. Expanding access by different stakeholders beyond member states may thus create the so-called ‘problem of many hands,’ whereby proliferating lines of accountability could in fact result in less accountability through processes such as blame-shifting, competing accountability claims, and reduced clarity on who is ultimately responsible for an outcome.Footnote 69 Legal scholars have raised concerns about fragmentation, when multiple accountability holders, such as public or private donors, partnerships, or NGOs, gain strong accountabilities for a multiplicity of narrowly specified objectives, potentially detracting from internal administrative mechanisms and from accountability for the broader mandate of the organisation.Footnote 70 Fragmented accountabilities could ultimately privilege elite networks with leverage, rather than the publics at large. In other words, while decentralisation may expand the access dimension and deliberative quality in the scope of IO accountability, it may not be a sufficient condition for the effectiveness of accountability systems as a whole. The efficacy of broadening the access to accountability depends also on the extent to which the mechanisms facilitated through decentralisation strengthen the capacity of public accountability holders to demand explanations and make claims for affectedness and change.Footnote 71 Furthermore, mechanisms that enable supervisory bodies to reduce information asymmetries and manage competing accountability claims in accordance with the public purpose and norms of the organisation, remain critical factors for delivering both broader and more effective accountability in more decentralised contexts.
Salience
Finally, creating or seizing peaks in salience around the mandates of IOs is another factor likely to be part and parcel of demanding greater accountability. Proposition 6 stipulates the impact of such events on the scope of accountability of IOs.
Proposition 6. Political processes that increase the salience of IOs are likely to result in greater access by accountability holders and recognition of affected populations as such (accountability to whom and for what); with less direct or significant impact on the instruments for efficient rendering of accountabilities.
Salience can be defined as ‘the extent to which people cognitively and behaviourally engage with a political issue.’Footnote 72 With respect to IOs, one of the chief concerns of democratic accountability theorists has to do with the limited salience of international institutions for domestic audiences. Thus, even if elected governments were to be efficient intermediaries of accountability in the long chains of delegation, the publics would not have sufficient knowledge of, or interest in distant bureaucracies to hold them to account.Footnote 73 However, political changes such as transnational advocacy mobilisation, the quantum increase in the speed of information exchange, and the density of transnational actors around IO mandates have created waves of accountability politics and salience events, challenging the domestic–international divide.Footnote 74
Empirical analyses have demonstrated that salience events, which are frequently measuresd as peaks in media coverage and visibility, account for changes and variations in the responsiveness of domestic agencies, controlling for other political factors.Footnote 75 Such events often motivate greater political control given the broader public resonance of salient issues related to agency mandates and performance, and the political implications of agency slippage.Footnote 76 If anything, the significance of salience events for accountability politics at large has grown in the contemporary ‘media age.’Footnote 77 In international politics, multiple types of events can contribute to greater salience of IOs. These range from advocacy actions designed to demand accountability by states, corporate actors or IOs themselves, institutional failures and scandals, or external crises such as natural disasters, wars, or humanitarian crises. The COVID-19 pandemic is just the most recent reminder of how crises bring WHO advisories into everyone’s living rooms, along with related politics and demands for greater accountability.
Salience events could increase the exposure of IO to greater scrutiny and accountability through several mechanisms. To begin with, such events focus attention and reduce the informational distance via-a-vis publics, external audiences, supervisory bodies, and in some instances legislative bodies and financing agencies at the domestic level. The visibility of IOs as material symbols of advocacy discontent furthermore exposes their work more readily to the public at large and thus to de-legitimation pressures.Footnote 78 Strategies of advocacy coalitions to galvanise attention by exposing unacceptable human suffering in vivid and personal termsFootnote 79 further reinforce the expectation that salience events are likely to contribute to formal recognition of norms for affectedness, expanding the scope of accountability in terms of who is recognised as accountability holder and for what. Salience events could thus create fluxes in demand for greater accountability, for example, sometimes towards non-state and sub-state actors, as in instances of environmental and social concerns, and at other times towards states, as evidenced by the COVID-19 pandemic and the accountability scrutiny over the WHO by the United States and other states. Therefore, the overall expectation is that salience events can contribute to expanding access in the scope of accountability of an IO, as more audiences and mechanisms of those mapped out in Figure 1 become engaged in processes of access to information, rendering account, contestation, and deliberation.
However, the implications of salience events for the efficient delivery of accountabilities are again more ambiguous. Public reputational mechanisms are typically activated during salience events, which could result in financial consequences if public or private donors withdraw support. Response to salience events that trigger reputational, market-based, and supervisory instruments could also lead to institutional reforms demanded by supervisory bodies. This implies that intergovernmental principals and supervisory bodies have to act in agreement and/or through financial power to impose consequences or help to craft new institutional mechanisms to address gaps in accountability. The creation of the World Bank Inspection Panel illustrates political dynamics triggered by normative pressure and salience that ultimately resulted in leveraging political power and implementing institutional change for broader access to accountability and affectedness.Footnote 80 Salience events triggered by advocacy campaigns for access to medicines ensued a similar political dynamic, contributing to the institutionalisation of new institutional norms and mechanisms both at the WHO and through public–private partnerships.Footnote 81
At the same time, public reputational pressures could also elicit largely public relations responses by the agencies, deflecting responsibility, while action by governing bodies remains limited.Footnote 82 Scholars and practitioners have also noted the phenomenon of ‘principal drift,’Footnote 83 whereby supervisory bodies could fail to act or simply shift resources without advancing substantive change. In other words, when reputational pressure capitalises on salience peaks and de-legitimation to instigate change, the effective delivery of accountability remains highly dependent on empowering access to accountability mechanisms and the capacity and willingness of principals to demand and underwrite change.
In summary, the preceding discussion suggests that the exercise of power, institutional structure, and salience shape the scope of accountability of IOs in an integrative and dynamic way through multiple pathways. The propositions articulated here reveal that different modalities of power and institutional design are both constitutive elements of the accountability system and key factors for enabling their activation and variable outcomes. Such outcomes range from accountability asymmetries to, alternatively, broadening the access to and even the efficacy of accountability. The series of six propositions elaborated here imply that the variation in accountability outcomes is likely to reflect their interplay, and how such factors may enable or detract from relational accountability processes between an international bureaucracy and its different accountability holders in a given context. While path dependency and structural power could lock in institutional features or asymmetries in accountability systems at a specific point in time, different forms of power could also be instrumental in demanding change and realising institutional reform. Salience events and new technologies bring the work of IO closer to the public view, however, with variable implications for updating accountability systems in response to salient issues. Accountability systems are therefore inherently political, dynamic, prone to contradictory demands, and, importantly, subject to evolving practices. The concluding section considers the implications of the theoretical framework proposed here for further research on accountability and rethinking accountability dilemmas in IR.
Conclusion
The accountability of governance institutions is a democratic ideal that is often difficult to attain and to match against the full range of its expectations. In world politics, the challenge of the public accountability of international agencies is further compounded by long chains of delegation that constitute their authority and competing expectations by diverse accountability holders on the nature of accountability that ought to be delivered. Accountability to some actors may detract from or conflict with accountability to others. Mechanisms and objectives of accountability may produce contradictory outcomes. Periodically, explosive revelations of failures or impunity for abuses of power and harms committed by international operations create peaks of attention to accountability. Contestation, even if it is politically uncomfortable for organisations, is thus part of accountability processes. Nonetheless, despite contradictions and flaws, the norm of accountability of IOs is important for the quality and legitimacy of the multilateral order and its ability to deliver on its norms and objectives.
The integrative theoretical framework elaborated in this article suggests that the accountability of international institutions is a multilayered phenomenon, rather than a subject that can be assessed on a single binary scale. The article therefore advances argument for a deeper interrogation of the scope of organisational accountability and how its different dimensions vary and interplay to shape practices in specific institutional contexts. It treats accountability not as a finite category, but as a system of multiple accountabilities that is fundamentally political. Institutional mechanisms intend to address expectations by member states as primary principals and by other accountability audiences and to create cumulative gains. However, they are also subject to competing claims that could create trade-offs and dysfunctions. Such complexity and contradictions are not unique to international institutions and have been elaborated in the study of public agencies and democratic accountability at the domestic level.Footnote 84 The primary difference in multilateral contexts is the absence of generalised mechanisms that supersede the politics of multiple accountabilities, for instance, through universal elections or constitutional courts.
The theoretical framework proposed here provides a new tool to critically examine the multiple dimensions of accountability systems and the role of power, institutional structure, and salience in shaping the differential scope of IO accountability. Moreover, these three factors are unlikely to operate in isolation or independently of each other, as the structure may codify certain power relations and salience can ultimately prompt institutional change. In other words, accountability systems in multilateral institutions are both highly structured by pre-existing and contemporary power relations and prone to politics and dynamic evolution over time and across contexts.
The integrative approach to holding IOs accountable has important implications both for further research and for policy and outcomes. First, it invites stronger attention to evaluating systems of accountability as such, and their fit or misfit with respect to advancing different types of accountabilities in accordance with public objectives and standards embedded in organisational mandates (Figure 1). Such analyses can be conducted internally within organisations as a means of identifying persistent blockages or gaps, which ought to be addressed to improve performance. New social science research can in turn illuminate the politics of the variable scope of IO accountability across institutional settings, as well as within organisations and their performance over time. The literature has documented multiple specific cases of practice and malpractice, as well as tensions between formal mandates and actual performance. However, there is less generalisable empirical understanding of how political conditions such as asymmetric power, institutional design, and salience may drive the functioning of accountability systems and delivery of accountability to different audiences.
Second, the theoretical framework invites further interrogation of recurrent accountability dilemmas that may appear to be particularly acute in distant international agencies. Scholars have argued, for instance, that the proliferation of accountability claims and mechanisms by both internal and external audiences may lead to fragmentation, reduced clarity of who is ultimately responsible for outcomes, and agency insulation,Footnote 85 ultimately resulting in less accountability or ‘multiple accountabilities disorder.’Footnote 86 The theoretical argument challenges the inevitability of pathological outcomes, given the inherent complexity of accountability systems. It clarifies how a set of political conditions, namely the interplay of different forms of power (Propositions 1 and 2) and institutional design (Propositions 3–5), can substantially influence the outcomes of multiple relational processes between accountability holders and accountability givers. Such factors are thus likely to significantly shape the extent to which the dual expectations for access to and efficacy of accountability are attained, and the retooling of accountability systems when contradictions and gaps occur. As a core element of legitimate governance, the scope of accountability of international agencies has multiple dimensions, including transparency, access, and deliberation, as well as learning as a form of consequence, not only sanctioning. Therefore, questions of effective principal control, on the one hand, and access and procedural quality, on the other, would not necessarily be seen as competing claims in a world of finite accountability.
Similarly, while there is a rich literature on the politics of contestation, salience, and the rise of external accountability demands, it is less clear under what conditions such salience events and reputational pressures will produce changes in the scope of accountability in some organisations, but not in others, and how durable and significant such changes are (Proposition 6). If accountability norms have formally expanded to address access and affectedness in different institutional settings, but claims remain difficult to make, it is critically important to establish how the relative institutionalisation of mechanisms for access (Proposition 3) and the empowerment or persistent disempowerment of accountability holders (Propositions 1 and 2) interface in different agency contexts to account for variation in outcomes. New comparative research across institutions and along inter-temporal frames can further improve the understanding of the contradictions, advances, and limitations in international accountability. Accountability or lack thereof is not a fixed feature of international governance, as many would like to assume. Achieving accountability is a political process, which by its very nature is a work in progress, riddled by competing claims, while being highly dependent on functioning and adaptive institutional practices.
Finally, the research agenda proposed in this article implies that from a policy perspective, IOs can be made more accountable. This requires deeper discussion within IOs of the sources of contradictions and deficits within accountability systems, as well as institutional mechanisms and politics needed to address them. Such responsibility rests with states, the agencies themselves, and public political processes that can influence what accountability platforms are available and to whom, and how the exercise of power and the salience of international decisions shape accountability practice.
Acknowledgements
I would like to acknowledge with gratitude the feedback provided by the anonymous reviewers and the editorial team of International Theory, as well as by Julie Billaud, Gian Luca Burci, Andrew Clapham, Kristina Daugirdas, Gergana Dimova, Cedric Dupont, Moira Faul, Giada Giacomini, Tamar Gutner, Eugenia Heldt, Stephanie Hofmann, Nina Kiderlin, Robert Keohane, Sandra Lavenex, Simon Montfort, Suerie Moon, Ioannis Papadopoulos, Latchezar Toshev, Simonetta Zarrilli, Agnese Zucca, and participants in the 2022 American Political Science Association (APSA) meeting in Montreal; the financial support of the Swiss National Science Foundation (SNF; Project No. 100017_182159); the research assistance of Agnese Zucca, Simon Montfort, Zuzana Hudakova, Antoine Pawletta, Christian Jiang, and Arne Olsen; and the research and editing suggestions of Nina Kiderlin. Agnese Zucca, Anouk Pasquier, and David Welch provided suggestions and support for the design of Figure 1. I am also grateful for the administrative support of the Centre for International Environmental Studies (CIES) and the Hoffmann Centre for Global Sustainability (HCGS) at the Graduate Institute, Geneva.
Open access
