Global Democracy in Decline?

How Rising Authoritarianism Limits Democratic Control over International Institutions

In: Global Governance: A Review of Multilateralism and International Organizations
Eugenia C. Heldt Technical University of Munich TUM School of Governance

Search for other papers by Eugenia C. Heldt in
Current site
Google Scholar
Henning Schmidtke Technical University of Munich TUM School of Governance

Search for other papers by Henning Schmidtke in
Current site
Google Scholar
Open Access


Over the past decade, rising authoritarian regimes have begun to challenge the liberal international order. This challenge is particularly pronounced in the field of multilateral development finance, where China and its coalition partners from Brazil, Russia, India, and South Africa have created two new multilateral development banks. This article argues that China and its partners have used the New Development Bank and the Asian Infrastructure Investment Bank to increase their power and to restrict democratic control mechanisms. By comparing formal mechanisms of democratic control in both organizations to the World Bank, this article shows that civil society access, transparency, and accountability are lower at the AIIB and NDB than they are at the World Bank.


Over the past decade, rising authoritarian regimes have begun to challenge the liberal international order. This challenge is particularly pronounced in the field of multilateral development finance, where China and its coalition partners from Brazil, Russia, India, and South Africa have created two new multilateral development banks. This article argues that China and its partners have used the New Development Bank and the Asian Infrastructure Investment Bank to increase their power and to restrict democratic control mechanisms. By comparing formal mechanisms of democratic control in both organizations to the World Bank, this article shows that civil society access, transparency, and accountability are lower at the AIIB and NDB than they are at the World Bank.

1 Introduction

International development finance has the potential to shape the livelihoods of hundreds of thousands of people. A large share of these financial resources is allocated by multilateral development banks (MDB s), where unelected and often loosely supervised international bureaucrats make important funding decisions.1 Consequently, democratic control—that is, formal mechanisms through which civil society organizations (CSO s) and individual citizens can scrutinize and shape MDB decision-making—is particularly important for the effectiveness and legitimacy of global governance and its institutions.2 Established MDB s, such as the World Bank, learned this lesson when the projects financed by them produced far less developmental returns than expected and even created grievances for the populations they were supposed to support. Confronted with a wave of criticism by CSO s, national parliaments, governments, and the media, the World Bank enacted, for instance, a variety of reforms with the aim of becoming more open, accountable, and responsive.3 First, the Bank opened up by engaging with CSO s in general consultations and project-specific interactions.4 Second, it became more transparent by enacting—as one of the first major MDB s—a comprehensive access-to-information policy.5 Finally, the Bank established several accountability mechanisms, including the Inspection Panel, the Compliance Advisor/Ombudsman, the Independent Evaluation Group, the Internal Audit Vice Presidency, and the Grievance Redress Service.6 With time, the World Bank and other MDB s have thus expanded democratic control mechanisms, following a trend that spans international organizations (IO s) more broadly.7

Yet recent shifts in the global distribution of power threaten to undermine this democratic progress at the multilateral level—even if critics underline the limits of these control mechanisms. In particular, the rise of authoritarian regimes, such as China and Russia, have led some observers to lament the decline of democracy on a global scale.8 This work shows how authoritarian regimes have worked to systematically narrow the space available to civil society, by restricting access to decision-making, transparency, and accountability on the domestic level. These developments present us with the question of whether this decline of democracy extends to global governance and its institutions.9 Do rising authoritarian regimes use their growing economic and political power to restrict democratic control over IO s?

We address this question by comparing formal mechanisms of democratic control established at the World Bank, which is still dominated by Western democracies, to the institutional design of two MDB s recently established under Chinese leadership. The launch of the New Development Bank (NDB) in 2015 and the Asian Infrastructure Investment Bank (AIIB) in 2016 are often described as a manifestation of China’s new power in international development and global governance more generally.10 Consequently, we presume that their design and operations are strongly shaped by Chinese preferences and normative demands.

We conceive of mechanisms of democratic control as a dimension of the institutional design of IO s, similar to membership, issue scope, or centralization that many see as the research frontier in the study of international institutions.11 Democratic control describes the formal institutional mechanisms that give stakeholders the right to scrutinize and, if necessary, correct IO decision-making and policies. Our analysis reveals that democratic control mechanisms at the AIIB and the NDB are indeed weaker than at the World Bank. Both new MDB s provide less formalized access for CSO s and are less transparent. The NDB is also less accountable to those adversely affected by its actions.

The article proceeds in five steps. First, we review the liberal constructivist literature on institutional design, suggesting that powerful states upload norms they hold dear at the domestic level to IO s. Second, we introduce the research design and three key mechanisms of democratic control, including civil society access, transparency, and accountability. Third, we present empirical findings. Fourth, we discuss potential alternative explanations for the observed differences. Finally, we summarize our findings and outline their implications for research on democratic control and global governance.

2 How Domestic Norms Shape International Institutional Design

Why do states, typically protective of their sovereignty and political influence, restrict their monopoly of control over IO s by introducing mechanisms of democratic control? In this section, we outline a liberal constructivist argument suggesting that powerful states tend to upload important domestic norms to their IO s. This argument motivates our expectation that rising authoritarian powers, such as China, are less likely to formalize mechanisms of democratic control in international institutions. This might, in turn, contribute to a decline of democracy in global governance.

Grounded in the notion that international institutions are not exclusively shaped by the specificities of cooperation problems and the characteristics of pertinent states,12 international relations scholars have shown that ideas and norms about what constitutes legitimate modes of global governance play an important role for international institutional design.13 This logic is often referred to as liberal constructivism, as it derives states’ international preferences from commitments to domestic values and institutions.14 It suggests that domestic political regimes and their ideational underpinnings shape patterns of international cooperation.15 Ranging from democratic peace and military alliance formation,16 to trade liberalization,17 over international dispute settlement,18 to mechanisms of democratic control,19 international institutions are shaped by the normative preferences of powerful states.

We translate these insights into an argument as to why rising authoritarian powers should be expected to design IO s with limited democratic control mechanisms. To be sure, we do not argue that normative considerations are the main reason why rising powers challenge extant global governance institutions and engage in institution making in the first place.20 Rather, we suggest more modestly that when rising powers have formed coalitions strong enough to create new international institutions, powerful authoritarian states are likely to restrict the formal democratic control over their new creations.

We start from the observation that rising authoritarian powers, such as China and Russia, share a common vision of a global political order that deviates from Western liberal internationalism.21 Rooted in domestic regimes and their ideational underpinnings, this common vision is based on the idea of a global order in which rules prevent any one state, or a group of states, from dominating the international system and imposing a specific ideology on the rest of the world.22 As a critique of the Western international system, this vision stresses the primacy of state sovereignty and questions the tendency of liberal-democratic governments and IO s to meddle in the domestic affairs of other countries.23 Part of this argument is the domestically routed rejection of democratic control over international institutions. Similar to domestic restrictions of civil society access, transparency, and accountability, authoritarian powers do not accept strong formal mechanisms of democratic control at the international level.

Following the argument that powerful states upload their norms to international institutions, we expect rising authoritarian powers to inject these normative considerations when they design new international institutions.24 As an observable implication of this effect, the AIIB and the NDB should provide fewer formal mechanisms of democratic control than the World Bank, which is largely dominated by Western democracies. As the creation and design process of both new MDB s was led by China, they should follow the model of noninterference with national governments’ decision-making. Thus, formal CSO access standards, transparency rules, and accountability mechanisms should be less demanding, if present at all.

3 Research Design

We compare the formal mechanisms of democratic control at the World Bank to those at the NDB and the AIIB at the time of their creation in 2015 and 2016, respectively. Our focused comparison of three central IO s in the field of international development finance is an important step in its own right. Studies of the institutional design of the NDB and the AIIB and their consequences for global governance are, to our best knowledge, few and far between.25

The comparison is premised on three presumptions. First, international development finance constitutes an issue area of international cooperation that manifests our phenomenon of interest—that is, the consequences of rising authoritarian powers on formal democratic control over international institutions—intensely. China and its coalition partners Brazil, Russia, India, and South Africa (BRICS) have challenged the institutional design of existing MDB s and have begun to create competing organizations of their own making.26 While these challenges have been paralleled in other issue areas, including international security and financial cooperation, they appear to be most advanced in development finance. Second, we select the World Bank from the pool of more than twenty-five extant MDB s as our baseline for democratic control mechanisms because, under the leadership of liberal democracies, it has become the focal MDB,27 providing the institutional blueprint for many MDB s created since Bretton Woods.28 Third, in the spirit of a most similar systems design, the World Bank is similar to the AIIB and the NDB in many important respects, but differs in terms of its dominant members. In contrast to some regional MDB s, such as the Latin American Development Bank (CAF), that limit their membership to specific regions, all three organizations are formally open to global membership. Similarly, the AIIB, the NDB, and the World Bank have a broad geographic scope and functional purpose, whereas other MDB s concentrate on specific regions and narrower tasks. The three selected organizations aim broadly to mobilize resources to finance a wide array of development projects.29 Other MDB s potentially relevant for our analysis due to the dominance of democratic member states fulfill more limited tasks. In this vein, the European Bank for Reconstruction and Development (EBRD), which is dominated by democratic member states, aims much more exclusively to foster the transition toward open market–oriented economies in Central and Eastern European countries.

The AIIB opened its doors in January 2016. In August 2018, it had forty-three regional members, including powerful states such as Australia, China, India, and Russia, and twenty-three nonregional members, including Western powers such as Canada, Germany, and the United Kingdom.30 Its Articles of Agreement mandate the AIIB to “(i) foster sustainable economic development, create wealth and improve infrastructure connectivity in Asia by investing in infrastructure and other productive sectors; and (ii) promote regional cooperation and partnership in addressing development challenges by working in close collaboration with other multilateral and bilateral development institutions.”31 Its financial model and basic institutional design are similar to the World Bank. Building on paid-in capital of $ 20 billion, the AIIB raises its lendable resources on financial markets, it has a three-tiered governance structure, and it implements a project and program lending approach.

Similarly, the NDB’s financial model builds on paid-in capital of $ 10 billion and its institutional design and lending approach is highly similar to the AIIB and the World Bank. The NDB became fully operational in February 2016. Its membership consists of the BRICS coalition. According to its Articles of Agreement, the NDB “shall mobilize resources for infrastructure and sustainable development projects in BRICS and other emerging market economies and developing countries to complement existing efforts of multilateral and regional financial institutions for global growth and development.”32

Empirically, our analysis focuses on the formal rules that can be observed in treaties, constitutions, protocols, and the organizations’ rules of procedure. This excludes informal practices and procedures. While we acknowledge that informal governance might affect democratic control over IO s,33 this article focuses on formal governance. These formal rules impose explicit, public commitments on states. If such rules mattered only marginally or not at all, then one would not expect to find systematic variations. Because mechanisms of democratic control are an institutional design feature, they are different from democratic influence per se, even if the two often go together. Democratic control mechanisms consist of rules that give stakeholders the formal right to information, access, and accountability. Democratic influence, by contrast, denotes the presence and activities of stakeholders in these institutional venues and other informal practices. It may, therefore, be the case that stakeholders have informal democratic influence without formal mechanisms of democratic control or that formal mechanisms do not fully translate into democratic influence. Yet informal influence is only offered voluntarily as long as it suits the needs of IO s and can be easily withdrawn.

To operationalize formal mechanisms of democratic control, we started from the democratic norms that Alexandru Grigorescu identifies as most important for the democratic quality of IO s.34 We use these norms as guidance to identify the key democratic control mechanisms of IO s (see Table 1). Specifically, we ask whether: (1) CSO s have the formal right to access the main bodies of the organizations (access); (2) whether the organizations are formally obliged to provide information to the public (transparency); and (3) whether those affected by IO activities have the right to hold them accountable for their actions (accountability).35

Table 1

Democratic influence on international organizations




Civil society access

Absent or passive participation in General Assembly


Active and indirect or active and direct participation


Nonstate voting right


Public access to information (transparency)

No access-to-information policy


Presence of incomplete access-to-information policy (missing one or more of the below elements)


Comprehensive access-to-information policy that is (1) based on a presumption in favor of disclosure, which (2) includes precise definitions of exemptions to this right to access information, which (3) include an appeal mechanism, and that (4) allow for the provision of information free of charge


Social accountability

No formal social accountability mechanism


Formal social accountability mechanism


4 Democratic Control Mechanisms at the AIIB, the NDB, and the World Bank

We proceed with the assessment of democratic control mechanisms at the three MDB s. We first present the rationale for individual indicators and respective observable implications before we analyze formal access, transparency, and accountability.

4.1 Access

Civil society access increases democratic control over IO s, assuming that CSO s are representative of society and largely independent.36 Jonas Tallberg and colleagues have shown that IO s’ openness to CSO s is driven by the normative preferences of their membership.37 The higher the weight of democratic states among IO members, the more open is the organization to CSO s, as democratic members upload their preference for civil society participation to the international level. Hence, a lower level of openness to CSO s in the AIIB and the NDB would support our argument, that China designed the new MDB s according to its normative standards. As China and its BRICS partners value national sovereignty more than democratic control, we expect them to design less open IO s. Following the conceptualization of CSO access by Tallberg and colleagues,38 we rank IO s with no or passive participation rights in main decision-making bodies as low CSO access, IO s with active and indirect or active and direct participation rights as medium access, and IO s that grant voting rights to CSO s as high access.

The World Bank provides various access routes for CSO s. Although it is clearly not the most open IO of our times, the data by Tallberg and colleagues demonstrate that the Bank has transformed from an exclusive interstate organization into an IO that provides medium access.39 The Bank has established a set of formal regulations and procedures, opening up the organization to civil society. First, the Bank’s Articles of Agreement establish an Advisory Council.40 This council is composed of independent representatives of banking, commercial, industrial, labor, and agricultural interests. It is the first inroad for CSO s in advising the Bank. Second, the Bank’s Operations Manual explains that in working with countries “the WBG [World Bank Group] engages through consultations with the private sector, civil society and other stakeholders.”41 Third, in 1993 the Bank created the World Bank Inspection Panel. Established as an independent complaints mechanism, the panel consists of civil society representatives independent of the Bank’s management.42 Fourth, in 1982 the Bank established the NGO–World Bank Committee. Composed of CSO representatives from around the world and senior Bank staff, the committee offered opportunities for direct exchange and policy dialogue.43 This standing consultative mechanism gave the Bank and up to twenty-five CSO participants the opportunity to exchange views until it was succeeded by the more thematic World Bank–Civil Society Forum in 2000. Overall, these formal mechanisms provide CSO s with medium access to the World Bank. While the Bank clearly goes beyond a shallow access, in which CSO s are limited to the role of silent observers, its various forums and consultation mechanisms do not offer the same rights and opportunities to CSO s as governmental representatives enjoy. CSO s are allowed to voice their concerns and demands, but they do not have access to formal voting in the Bank’s decision-making bodies.44

By contrast, formal CSO access to the AIIB and the NDB is marginal. Their founding treaties do not discuss the matter at all and additional regulations largely externalize access to the Bank’s borrowers. The AIIB has also created an International Advisory Panel, in which civil society is represented, and it regularly conducts informal public consultation. Yet neither the Advisory Panel nor outreach activities have formal status at the organization. Formally, the AIIB’s Environmental and Social Framework, which is a central operational policy of the organization, claims that the Bank aims to “develop new and sustainable approaches to meet Asia’s infrastructure challenges, in partnership with government, other multilateral development banks, bilateral development organizations, the private sector and civil society”45 and that “meaningful consultation is essential for the design and implementation of a Project.”46 Details on the implementation of these goals formalizing CSO access are, however, absent. The organization thus has broad discretion, and civil society can barely insist on specific access rights.

Similarly, the NDB’s Code of Conduct claims, without providing further details, that “to face challenges and solve problems of the ever evolving economic and financial situation in the world, collaboration with shareholders, stakeholders, field experts, team members of the Bank and entities is of vital importance”47 and that it is critical for its staff to “maintain constructive relationships and work to build trust and confidence with the representative government, private sector and civil society.”48 The most concrete provisions established in the NDB’s Environment and Social Framework are equally vague and do not formalize access for CSO s. The document states that the:

NDB requires the Client to conduct a meaningful consultation process, that is compliant with national laws and regulations and this Policy, and (i) engages with communities, groups or people affected by proposed projects; (ii) begins early and is carried out on an ongoing basis; (iii) provides timely disclosure of relevant and adequate information that is understandable and readily accessible, is undertaken in an atmosphere free of intimidation or coercion; (iv) is gender inclusive and responsive, tailored to the needs of vulnerable groups; and (v) enables the incorporation of all relevant views of affected people and other stakeholders into decision making.49

In sum, the AIIB and the NDB do not provide formal access for CSO s. The regulations of both new MDB s largely externalize CSO access to clients instead of formalizing access rights to the organization itself. Consequently, the level of CSO access to both organizations is low, and democratic control is thus lower than at the World Bank. Although access to the World Bank is limited to dialogue and rarely gives CSO s the opportunity to actively shape the Bank’s decision-making, these formal communication channels are more than what the AIIB and the NDB offer.

4.2 Transparency

Our second democratic control mechanism zeros in on transparency, or more specifically, public access to information about IO activities and policies. From a normative perspective, transparency is an important dimension of democratic control because it is a necessary condition for holding IO s accountable. “Unless information about an organization’s deliberations, decisions, and actions is available, it is impossible to determine if state representatives and IO officials are acting in the public interest.”50 Empirically, Grigorescu has shown that established IO s have implemented formal access-to-information policies due to the normative demand by powerful democratic member states.51 Hence, a lower level of transparency in the new MDB s would support our argument that rising authoritarian powers design less democratically controlled IO s.

We ask for the comprehensiveness of access-to-information policies to compare the transparency of the three organizations. We focus on four aspects: first, a presumption in favor of disclosure; second, precise definitions of exemptions to the right to access information; third, appeal mechanisms; and, finally, no charges for the provision of information.52 Transparency is low for IO s that do not have any formal rules on access to information. Medium transparency results from access-to-information policies that feature only some of the above elements, and high transparency is achieved if the policy includes all four elements.

The World Bank was one of the first IO s to adopt a public information policy in the early 1990s.53 In its current version, this policy is comprehensive. Thus, we classify the Bank as having a high transparency standard.54 The policy reveals the Bank’s presumption in favor of disclosure as it “allows access to any information in its possession that is not on a list of exceptions,”55 it provides a clear list of exceptions to this general rule,56 it has a clearly defined appeal mechanism,57 and it does not list charges for the provision of information.

At first glance, the AIIB and the NDB seem to take transparency even more seriously than the World Bank as both organizations’ founding treaties explicitly note that the transparency of operations will be promoted.58 Yet their access-to-information policies do not fully live up to this promise. Because both MDB s have implemented only incomplete policies, we classify their transparency as medium.59 Specifically, the AIIB explicitly presumes disclosure and does not list charges in its policy. However, the list of exceptions to this presumption includes vague and broad formulations60 such as the provision that the organization will not disclose information “if doing so would misuse its resources and facilities … jeopardize its credit worthiness [or] the international character of the Bank.”61 These formulations give the organization scope for interpretation and, thus, make it difficult to question its decisions. Furthermore, the policy does not formalize an appeal mechanism.62

The NDB’s policy establishes a clear appeal mechanism and does not list charges. Yet the policy does not reveal a presumption in favor of disclosure, as it states that “the Bank is guided by an underlying presumption that information concerning the Bank’s activities will be made available in a timely manner to the public in the absence of an appropriate reason for confidentiality” and that “the Bank reserves the right to determine whether or not to disclose requested information.”63 The policy also does not provide a precise list of exceptions. First, some formulations, such as the rule that “information that when disclosed might endanger or harm or adversely affect the relations between the Bank and its member countries” will not be disclosed, are vague and can be interpreted broadly.64 Second, the policy explicitly states that “in case of exceptional circumstances and with the approval of the BoD [Board of Directors] the Bank reserves the right not to disclose information that would otherwise normally be disclosed under this policy.”65

Overall, all three MDB s have adopted access-to-formation policies, signaling that transparency to the public is on their agenda. Yet the World Bank’s policy is clearly most comprehensive. It discusses all of the above-mentioned elements while neither the AIIB nor the NDB have a comprehensive policy. As a result, the two new MDB s offer less democratic control than the World Bank does in the dimension of transparency.

4.3 Accountability

Our third and final mechanism of democratic control is accountability. From a normative perspective, accountability and democratic control are twin issues because a crucial feature of representative democracy is that those who govern are held accountable by the governed.66 This accountability relationship includes three components: “first, standards that those who are held accountable are expected to meet, second, information available to accountability holders, who can then apply the standards in question to the performance of those who are held to account, and third, the ability of these accountability holders to impose sanctions—to attach costs to the failure to meet the standards.”67

Empirically, the mechanisms by which IO s are held accountable are manifold. Depending on the nature of accountability holders, the nature of the conduct, and the nature of the obligation, various types of accountability mechanisms exist.68 With regard to the democratic control over IO s, we focus on formalized “social accountability”; that is, accountability mechanisms that enable individual citizens, CSO s, and other private interest groups to have their complaints about IO activities reviewed by an independent body.69 In accordance with our theoretical argument, case studies on why IO s have begun to establish mechanisms, such as the World Bank Inspection Panel, reveal that democratic member states are the key drivers behind this institutional innovation.70 Hence, a lower level of social accountability at the AIIB and the NDB would support our argument. To compare the social accountability mechanisms of the three MDB s, we use a twofold ranking. IO s without formal accountability mechanisms are ranked low in this dimension. IO s with formal accountability mechanisms are ranked as highly accountable.

According to Ruth Grant and Robert Keohane,71 the World Bank is a highly accountable IO when it comes to member states’ control over their activities. This is also the case for its social accountability mechanisms. Confronted with pressure from CSO s and the U.S. Congress, the World Bank has established two separate accountability mechanisms, which provide civil society actors with the opportunity to file complaints against the Bank’s activities.72 First, the World Bank Inspection Panel is an independent semijudicial mechanism through which impacted communities or individuals can file complaints about projects financed by the International Bank for Reconstruction and Development (IBRD) and the International Development Association (IDA). Created in 1993, the panel consists of three independent members, appointed by the Executive Board. It has the power to make preliminary assessments of complaints and to recommend complaints for a full investigation. Second, for complaints related to projects supported by the Bank’s International Finance Corporation (IFC) and Multilateral Investment Guarantee Agency (MIGA), the Complaints Advisor/Ombudsman (CAO) was created in 1999. The CAO is also independent of the Bank’s management and has a separate bureaucratic structure and staff. Similar to the Inspection Panel, it has the power to accept complaints from individuals or groups affected by IFC or MIGA projects and to make recommendations on how to improve these projects. Overall, these two mechanisms establish the high social accountability of the World Bank.

A similar diagnosis holds for the AIIB. The organization’s Project-Affected People’s Mechanism (PPM) was still in draft stage at the time of this writing. Its basic contours are largely similar to the World Bank’s social accountability mechanisms. Under the PPM, individuals and groups who are adversely affected by a lack of the AIIB’s compliance with its own Environmental and Social Policy may file a complaint with the Compliance, Effectiveness and Integrity Unit (CEIU). If eligibility criteria are fulfilled, the CEIU has the power to evaluate the complaint and to propose remedies, including consultations between affected individuals and the AIIB, information sharing, and the redesign of projects, to the Board of Directors, which takes the final decision.73 This formalizes a high standard of social accountability at the AIIB.

By contrast, the NDB is not socially accountable. Although the issue is discussed in its Environment and Social Framework, where the organization “requires that the client establish and maintain a fair and effective grievance redress mechanism to receive and facilitate timely resolution of affected peoples’ concerns and grievances about the client’s environmental and social performance at project level,”74 this provision does not constitute a mechanism by which the NDB can be held accountable. Rather, it shifts the burden of accountability to the organization’s clients and, thus, insulates the organization from its outside world. Consequently, the NDB is less socially accountable than the World Bank and the AIIB.

5 Alternative Explanations

In sum, the empirical data are in line with our theoretical argument. The World Bank has stronger mechanisms of access, transparency, and accountability than the NDB. Similarly, access and transparency are also more developed at the World Bank than at the AIIB while accountability is similar.

One might object that these results could be shaped by factors other than dominant members’ regime type. First, from a historical institutionalist perspective,75 it could be argued that a comparison of the World Bank with its layers of formal democratic control built up over decades and the newly founded AIIB and NDB is futile. The two new MDB s have not been subject to member state and civil society pressure that made the World Bank establish mechanisms of democratic control. In this view, the observed variance would be the result of different levels of experience and expertise in designing democratic control mechanisms.

A second alternative explanation could focus on the role of international bureaucrats in the design of IO s. Assuming that democratic control mechanisms at IO s are of low salience to states, design decisions could be shaped by international bureaucrats working at extant IO s.76 In this view, seasoned international bureaucrats are often asked for input on the institutional design of new IO s because of the expertise and experience they have gathered.77 The observed variance could thus be shaped by the different experiences that international bureaucrats involved in institutional design have made at their home institutions.

To probe the explanatory power of these alternative explanations, we examine the involvement of international bureaucrats in the design of the AIIB and the NDB. If the historical institutionalist objection holds, these actors should not play a role in the design process. As a result, the experience gathered by World Bank officials and others would not play a role and our comparison would indeed be inappropriate. If the observed outcomes are shaped by international bureaucrats instead of member states, we should find that international bureaucrats from institutions with weak mechanisms of democratic control were involved in the design of the AIIB and the NDB.

First, public statements of MDB representatives suggest that the design of the AIIB and the NDB was assisted by MDB s with more demanding democratic control mechanisms. Most importantly, the World Bank itself was either directly involved in the design process or the Bank’s mechanisms of democratic control were explicitly considered. The AIIB president acknowledged, for instance, that during the design stage, “the World Bank has provided advisory and technical support on a wide range of topics, including … institutional governance.”78 Similarly, World Bank representatives noted that the 2016 cooperation agreement with the NDB “formalizes the World Bank Group–NDB knowledge partnership that has accompanied the establishment of the NDB.”79 Even more directly, the AIIB explicitly references the World Bank’s accountability mechanisms in its own accountability rules80 and explains that access-to-information policies of various MDB s, including the World Bank, were considered “to benefit from the experiences of long-standing MDB s.”81

Second, senior management positions at both new MDB s are filled with professionals that have acquired experience with stronger mechanisms of democratic control at other MDB s. At the AIIB five of eight senior management positions are held by individuals who have previously been at the World Bank, Asian Development Bank (ADB), and African Development Bank (AfDB). The AIIB’s International Advisory Board lists a number of former members of the World Bank’s senior management. Finally, the first managing director of the AIIB’s Compliance, Effectiveness and Integrity Unit, which is central to the organization’s transparency and accountability mechanisms, held similar positions at the ADB, where mechanisms of democratic control are strong.82 At the NDB, three of the five senior management positions are occupied by BRICS nationals who have previously worked at the World Bank or the ADB.

Overall, the AIIB and the NDB may be young MDB s, but their institutional design has been accompanied by international bureaucrats with long years of experience with the democratic control of MDB s. Both organizations therefore had access to relevant expertise, including the design of democratic control mechanisms. Especially, the AIIB reflects the influence of this design expertise in many dimensions of its institutional design.83 Yet both the AIIB and the NDB decided to design less demanding mechanisms of democratic control than those of the World Bank. This suggests that founding members decided deliberately against formal democratic control mechanisms at their new international institutions. Thus, both alternative explanations do not seem to have the same strong explanatory power as our theoretical argument.

6 Conclusion

Rising authoritarian powers’ challenge to global economic institutions has the potential to spur profound changes in global governance. This article contributes to our understanding of these consequences by mapping the formal democratic control mechanisms at the AIIB and the NDB. Since both organizations were recently created under Chinese leadership, they have the potential to become important contenders to the Bretton Woods Institutions and, thus, to undermine the standards of democratic control developed at the World Bank and other established MDB s. Our conclusion summarizes the findings of the analysis and expands on their implications for democracy in global governance.

Our argument can be summarized in two principal conclusions. First, the democratic control mechanisms established at the AIIB and the NDB differ substantively from the World Bank’s institutional design. Overall, access, transparency, and accountability mechanisms at the World Bank are clearly more developed than at the AIIB and the NDB. Although both new MDB s highlight in their formal documents their openness to CSO s, transparency, and accountability, formal institutional reality reveals a different picture. CSO access is not formalized and is largely delegated to AIIB and NDB clients, whereas the World Bank has established a set of formal access mechanisms. While these World Bank mechanisms do not allow CSO s to vote and to actively shape the Bank’s decision-making, the AIIB and the NDB do not even ensure consultation and exchange. Similarly, their access-to-information policies are less comprehensive than the World Bank’s, as they give member states the final say about which documents will be public and which ones will be classified. Finally, the World Bank and the AIIB have established formal social accountability mechanisms, whereas the NDB does not give individuals or groups affected by its projects the right to file complaints. Overall, democratic control mechanisms at the World Bank are clearly more developed than at the NDB. To a lesser extent, this is also true for the AIIB, where formal rules on CSO access and transparency are weaker than at the World Bank while social accountability mechanisms are similar.

Second, this variation in the organizations’ institutional design is explained by a combination of three principal factors. The rise of the authoritarian countries, most notably China, is clearly a necessary condition for the creation of the AIIB and the NDB. As China and its partners in the BRICS coalition were discontented with their restricted control over the World Bank’s decisions and policies, they began to create MDB s that are similar in terms of purpose and technical profile but that they can control more directly.84 Finally, the normative demands of powerful authoritarian states help to explain why democratic control mechanisms in these new MDB s are comparably weak. The World Bank was driven to engage in democratic reforms, opening up the Bank to CSO s and making it more transparent and socially accountable by the pressure of powerful democratic member states, especially the United States. Yet China and other rising authoritarian powers do not subscribe to this Western preference for democratic control over IO s. Their vision of global governance is strongly based on national sovereignty and noninterference, giving states full control over global governance institutions. Consequently, the rise of authoritarian powers not only contributes to a decline of democracy on the domestic level, but it also seems to result in less democratically controlled IO s. Further research, extending to democratic control mechanisms of IO s in other issue areas will be needed to establish the generalizability of our argument. The nature of the underlying cooperation problems that IO s are designed to address could constitute an important scope condition.85 Tallberg and colleagues have, for instance, shown that IO s are more likely to give access to CSO s when they engage with issue areas that demand local activity or that are characterized by noncompliance incentives.86

These results have important implications for the debate on democracy in global governance and its potential decline.87 Many have argued that the legitimacy of global governance institutions needs to be bolstered by democratic reforms.88 According to this notion, opening up IO s to CSO s, making them more transparent and more socially accountable, is required to improve global governance’s legitimacy in the era of politicization and rising populist nationalism. Our results speak to the empirical viability of this normative vision and yield a pessimistic verdict. Assuming the continued rise of authoritarian powers, their design of less democratically controlled global governance arrangements in competition with existing institutions is likely to continue. Since the democratic legitimacy of these new institutions is lower than in Western-dominated IO s, the democratization of global governance as a whole may face a difficult future. Furthermore, our results might signal the emergence of a renewed debate on the status of democracy as an “authoritative principle of international life.”89 If a coalition of increasingly powerful states refuses to accept democratic principles as a normative standard to be respected in international institutional design, democracy might be relegated to the less important status that it had during the Cold War.


This article presents data generated in the context of the research project “Delegation of Power to International Organizations and Institutional Empowerment over Time” (DELPOWIO) funded by the European Research Council under the European Union’s Seventh Framework Programme for Research and Technological Development, Grant Agreement No. 312368. The authors would like to thank Matthew Clarke, Robert Csehi, Orfeo Fioretos, Julia C. Gray, Tamar Gutner, Eric Helleiner, Chris Humphrey, André Isidro, Christian Kreuder-Sonnen, Louis Pauly, Omar Serrano, Catherine Weaver, Bernhard Zangl, all the participants of the ECPR panel in Oslo 2017, of the workshop “Bretton Woods@75: Legacies and Innovations in Global Economic Governance” in Munich/Raitenhaslach 2017, of the ISA panel in San Francisco 2018, and the two anonymous reviewers for their insightful comments on earlier versions of this paper.


Babb 2009; Park 2016.


Scholte 2011.


Weaver 2008; Park 2017.


Tallberg et al. 2013, 62.


Tallberg 2016, 1170.


Buntaine 2015.


Tallberg et al. 2013; Grigorescu 2015.


Plattner 2016; Diamond, Plattner, and Walker 2016.


Kreuder-Sonnen and Zangl 2016.


Weaver 2015.


Lake and McCubbins 2006, 341.


Koremenos, Lipson, and Snidal 2001.


Lenz and Viola 2017; Grigorescu 2015.


Tallberg, Sommerer, and Squatrito 2016, 63.


Farrell and Newman 2014.


Bueno de Mesquita et al. 1999.


Mansfield, Milner, and Rosendorff 2000.


Keohane, Moravcsik, and Slaughter 2000.


Tallberg, Sommerer, and Squatrito 2016.


Morse and Keohane 2014; Urpelainen and Van de Graaf 2015.


Laidi 2012.


de Coning, Mandrup, and Odgaard 2015.


Castaneda 2010.


Kreuder-Sonnen and Zangl 2016.


For notable exceptions, see Chin 2014; Cooper and Farooq 2015.


Vestergaard and Wade 2013.


Nielson, Parks, and Tierney 2017.


Humphrey 2017; Park and Strand 2016.


Serrano Oswald 2018.


AIIB 2018a.


AIIB 2014, para. 1.


NDB 2014, para. 1.


Stone 2013.


Grigorescu 2015, 10.


See also Kreuder-Sonnen and Zangl 2015.


Steffek and Nanz 2008.


Tallberg, Sommerer, and Squatrito 2016.


Tallberg et al. 2013, 25–28.


Tallberg et al. 2013.


World Bank 2012, para. V sec. 6.


World Bank 2014, 5.


Park 2010.


Nelson 1995, chap. 3.


Tallberg et al. 2013, 62.


AIIB 2016, para. 6.


AIIB 2016, para. 13.


NDB 2016a, 7.


NDB 2016a, 11.


NDB 2016b, 10. An almost identical statement is included in AIIB 2016, para. 13.


Tallberg 2016, 1180; see also Grant and Keohane 2005; Scholte 2011.


Grigorescu 2007.


Grigorescu 2007, 628.


Tallberg 2016, 1170.


For a similar conclusion, see Donaldson and Kingsbury 2013, 123.


World Bank 2015, 9.


World Bank 2015, 9.


World Bank 2015, 18.


NDB 2014, para. 15; AIIB 2014, para. 34.


The fact that the AIIB and the NDB implemented such policies early on suggests that democratic control mechanisms are not necessarily a matter of long years of experience.


AIIB 2018b, para. 5.1.1.


AIIB 2018b, para. 9.


For a similar assessment, see de Jonge 2017.


NDB 2017, 3, 9.


NDB 2017, 8.


NDB 2017, 10.


Grant and Keohane 2005, 29.


Buchanan and Keohane 2006, 426.


Grant and Keohane 2005; for typologies, see Bovens 2007; Heldt 2018.


Bovens 2007, 457.


Nielson and Tierney 2003; Park 2010.


Grant and Keohane 2005, 37.


For an overview, see Heldt 2018.


AIIB 2018c.


NDB 2016b, 11.


Fioretos 2011; Rixen, Viola, and Zürn 2016.


Johnson 2014.


Johnson and Urpelainen 2014.


AIIB 2015.


World Bank 2016.


AIIB 2018c, 2.


AIIB 2018d, para. 6.


Grigorescu 2007, 644.


Gutner 2018; Serrano Oswald 2018.


Zangl et al. 2016; Vestergaard and Wade 2013.


Koremenos, Lipson, and Snidal 2001.


Tallberg et al. 2014.


Gratzke and Naoi 2011; Keohane, Macedo, and Moravcsik 2009.


Scholte 2011.


Clark 2007, 153–154.


  • AIIB (Asian Infrastructure Investment Bank). “Articles of Agreement” (Beijing: AIIB, 2014).

  • AIIB. “AIIB President-Designate Jin Liqun and World Bank Group President Jim Kim Deepen Cooperation.” 2015., October 2015.

  • AIIB. “Environmental and Social Framework” (Beijing: AIIB, 2016).

  • AIIB. “Members and Prospective Members of the Bank” (Beijing: AIIB, 2018a).

  • AIIB. “Policy on Public Information” (Beijing: AIIB, 2018b).

  • AIIB. “Enhancing AIIB’s Accountability: The Project-Affected Peoples’ Mechanism” (Beijing: AIIB, 2018c).

  • AIIB. “Background Paper on the Public Information Interim Policy Review” (Beijing: AIIB, 2018d).

  • Babb, Sarah. Behind the Development Banks: Washington Politics, World Poverty, and the Wealth of Nations (Chicago: University of Chicago Press, 2009).

  • Bovens, Mark. “Analysing and Assessing Accountability: A Conceptual Framework.” European Law Journal 13 (4) (2007), 447–468.

  • Buchanan, Allen, and Robert O. Keohane. “The Legitimacy of Global Governance Institutions.” Ethics and International Affairs 20 (4) (2006), 405–437.

  • Bueno de Mesquita, Bruce, James D. Morrow, Randolph M. Siverson, and Alastair Smith. “An Institutional Explanation of the Democratic Peace.” American Political Science Review 93 (4) (1999), 791–807.

  • Buntaine, Mark T. “Accountability in Global Governance: Civil Society Claims for Environmental Performance at the World Bank.” International Studies Quarterly 59 (1) (2015), 99–111.

  • Castaneda, Jorge G. “Not Ready for Prime Time: Why including Emerging Powers at the Helm Would Hurt Global Governance.” Foreign Affairs 89 (5) (2010), 109–122.

  • Chin, Gregory T. “The BRICS-Led Development Bank: Purpose and Politics beyond the G20.” Global Policy 5 (3) (2014), 366–373.

  • Clark, Ian. International Legitimacy and World Society (Oxford: Oxford University Press, 2007).

  • Cooper, Andrew, and Asif Farooq. “Testing the Club Dynamics of the BRICS: The New Development Bank from Conception to Establishment.” International Organisations Research Journal 10 (2) (2015), 39–58.

  • de Coning, Cedric, Thomas Mandrup, and Liselotte Odgaard, eds. The BRICS and Coexistence: An Alternative World Order (London: Routledge, 2015).

  • de Jonge, Alice. “Perspectives on the Emerging Role of the Asian Infrastructure Investment Bank.” International Affairs 93 (5) (2017), 1061–1084.

  • Diamond, Larry, Marc F. Plattner, and Christopher Walker, eds. Authoritarianism Goes Global (Baltimore: Johns Hopkins University Press, 2016).

  • Donaldson, Megan, and Benedict Kingsbury. “The Adoption of Transparency Policies in Global Governance Institutions: Justifications, Effects, and Implications.” Annual Review of Law and Social Science 9 (1) (2013), 119–147.

  • Farrell, Henry, and Abraham L. Newman. “Domestic Institutions beyond the Nation-State: Charting the New Interdependence Approach.” World Politics 66 (2) (2014), 331–363.

  • Fioretos, Orfeo. “Historical Institutionalism in International Relations.” International Organization 65 (2) (2011), 367–399.

  • Grant, Ruth W., and Robert O. Keohane. “Accountability and Abuses of Power in World Politics.” American Political Science Review 99 (1) (2005), 29–43.

  • Gratzke, Erik, and Megumi Naoi. “Multilateralism and Democracy: A Dissent Regarding Keohane, Macedo, and Moravcsik.” International Organization 65 (3) (2011), 589–598.

  • Grigorescu, Alexandru. “Transparency of Intergovernmental Organizations: The Roles of Member States, International Bureaucracies and Nongovernmental Organizations.” International Studies Quarterly 51 (3) (2007), 625–648.

  • Grigorescu, Alexandru. Democratic Intergovernmental Organizations? Normative Pressures and Decision-Making Rules (Cambridge: Cambridge University Press, 2015).

  • Gutner, Tamar. “AIIB: The Role of International Experts in Institutional Design.” In International Studies Association 2018 Annual Convention (San Francisco: 2018).

  • Heldt, Eugénia. “Lost in Internal Evaluation? Accountability and Insulation at the World Bank.” Contemporary Politics 24 (5) (2018), 568–587.

  • Humphrey, Chris. “He Who Pays the Piper Calls the Tune: Credit Rating Agencies and Multilateral Development Banks.” Review of International Organizations 12 (2) (2017), 281–306.

  • Johnson, Tana. Organizational Progeny: Why Governments Are Losing Control over the Proliferating Structures of Global Governance (Oxford: Oxford University Press, 2014).

  • Johnson, Tana, and Johannes Urpelainen. “International Bureaucrats and the Formation of Intergovernmental Organizations: Institutional Design Discretion Sweetens the Pot.” International Organization 68 (1) (2014), 177–209.

  • Keohane, Robert O., Stephen Macedo, and Andrew Moravcsik. “Democracy-Enhancing Multilateralism.” International Organization 63 (1) (2009), 1–31.

  • Keohane, Robert O., Andrew Moravcsik, and Anne-Marie Slaughter. “Legalized Dispute Resolution: Interstate and Transnational.” International Organization 54 (3) (2000), 457–488.

  • Koremenos, Barbara, Charles Lipson, and Duncan Snidal. “The Rational Design of International Institutions.” International Organization 55 (4) (2001), 761–799.

  • Kreuder-Sonnen, Christian, and Bernhard Zangl. “Which Post-Westphalia? International Organizations between Constitutionalism and Authoritarianism.” European Journal of International Relations 21 (3) (2015), 588–594.

  • Kreuder-Sonnen, Christian, and Bernhard Zangl. “Varieties of Contested Multilateralism: Positive and Negative Consequences for the Constitutionalisation of Multilateral Institutions.” Global Constitutionalism 5 (3) (2016), 327–343.

  • Laidi, Zaki. “BRICS: Sovereignty Power and Weakness.” International Politics 49 (5) (2012), 614–632.

  • Lake, David A., and Mathew D. McCubbins. “The Logic of Delegation to International Organizations.” In Delegation and Agency in International Organizations, eds. Darren G. Hawkins, David A. Lake, Daniel L. Nielson, Michael J. Tierney (Cambridge: Cambridge University Press, 2006), 341–368.

  • Lenz, Tobias, and Laura Anne Viola. “Legitimacy and Institutional Change in International Organizations: A Cognitive Approach.” Review of International Studies 43 (5) (2017), 939–961.

  • Mansfield, Edward D., Helen V. Milner, and B. Peter Rosendorff. “Free to Trade: Democracies, Autocracies, and International Trade.” American Political Science Review 94 (2) (2000), 305–321.

  • Morse, Julia C., and Robert O. Keohane. “Contested Multilateralism.” Review of International Organizations 9 (4) (2014), 385–412.

  • NDB (New Development Bank). “Agreement on the New Development Bank” (Fortaleza: NDB, 2014).

  • NDB. “Code of Business Conduct and Ethics” (Shanghai: NDB, 2016a).

  • NDB. “Environment and Social Framework” (Shanghai: NDB, 2016b).

  • NDB. “Information Disclosure Policy” (Shanghai: NDB, 2017).

  • Nelson, Paul J. The World Bank and Non-Governmental Organizations: The Limits of a Political Development (Basingstoke: Macmillan Press, 1995).

  • Nielson, Daniel L., Bradley Parks, and Michael J. Tierney. “International Organizations and Development Finance: Introduction to the Special Issue.” Review of International Organizations 12 (2) (2017), 157–169.

  • Nielson, Daniel L., and Michael J. Tierney. “Delegation to International Organizations: Agency Theory and World Bank Environmental Reform.” International Organization 57 (2) (2003), 241–276.

  • Park, Susan. “Designing Accountability, International Economic Organizations and the World Bank’s Inspection Panel.” Australian Journal of International Affairs 64 (1) (2010), 13–36.

  • Park, Susan. “The Multilateral Development Banks: Innovation or Stagnation?” In China-US Relations in Global Perspective, ed. Bo Zhiyue (Wellington, New Zealand: Victoria University Press, 2016), 108–125.

  • Park, Susan. “Accountability as Justice for the Multilateral Development Banks? Borrower Opposition and Bank Avoidance to US Power and Influence.” Review of International Political Economy 25 (6) (2017), 776–801.

  • Park, Susan, and Jonathan R. Strand, eds. Global Economic Governance and the Development Practices of the Multilateral Development Banks (London: Routledge, 2016).

  • Plattner, Marc F. “Is Democracy in Decline?” Journal of Democracy 26 (1) (2016), 5–10.

  • Rixen, Thomas, Laura Anne Viola, and Michael Zürn, eds. Historical Institutionalism & International Relations: Explaining Institutional Development in World Politics (Oxford: Oxford University Press, 2016).

  • Scholte, Jan Aart, ed. Building Global Democracy? Civil Society and Accountable Global Governance (Cambridge: Cambridge University Press, 2011).

  • Serrano Oswald, Omar Ramon. “The New Architects: Brazil, China and Innovation in Multilateral Development Lending.” Public Administration and Development (2018). DOI 10.1002/pad.1837, 1–12.

  • Steffek, Jens, and Patrizia Nanz, eds. Emergent Patterns of Civil Society Participation in Global and European Governance: A Cure for the Democratic Deficit? (Basingstoke: Palgrave Macmillan, 2008).

  • Stone, Randall W. “Informal Governance in International Organizations: Introduction to the Special Issue.” Review of International Organizations 8 (2) (2013), 121–136.

  • Tallberg, Jonas. “Transparency.” In The Oxford Handbook of International Organizations, eds. Jacob Katz Cogan, Ian Hurd, and Ian Johnstone (Oxford: Oxford University Press, 2016).

  • Tallberg, Jonas, Thomas Sommerer, and Theresa Squatrito. “Democratic Membership in International Organizations: Sources of Institutional Design.” Review of International Organizations 11 (1) (2016), 59–87.

  • Tallberg, Jonas, Thomas Sommerer, Theresa Squatrito, and Christer Jönsson. The Opening up of International Organizations: Transnational Access in Global Governance (Cambridge: Cambridge University Press, 2013).

  • Tallberg, Jonas, Thomas Sommerer, Theresa Squatrito, and Christer Jönsson. “Explaining the Transnational Design of International Organizations.” International Organization 68 (4) (2014), 741–774.

  • Urpelainen, Johannes, and Thijs Van de Graaf. “Your Place or Mine? Institutional Capture and the Creation of Overlapping International Institutions.” British Journal of Political Science 45 (4) (2015), 799–827.

  • Vestergaard, Jakob, and Robert H. Wade. “Protecting Power: How Western States Retain the Dominant Voice in the World Bank’s Governance.” World Development 46 (1) (2013), 153–164.

  • Weaver, Catherine. Hypocrisy Trap: The Rhetoric, Reality, and Reform of the World Bank (Princeton: Princeton University Press, 2008).

  • Weaver, Catherine. “The Rise of China: Continuity or Change in the Global Governance of Development?” Ethics and International Affairs 29 (4) (2015), 419–431.

  • World Bank. “International Bank for Reconstruction and Development Articles of Agreement” (Washington, DC: World Bank, 2012).

  • World Bank. “World Bank Group Directive: Country Engagement” (Washington, DC: World Bank, 2014).

  • World Bank. “Bank Policy: Access to Information” (Washington, DC: World Bank, 2015).

  • World Bank, “World Bank Group, New Development Bank Lay Groundwork for Cooperation.” 9 September 2016.

  • Zangl, Bernhard, Frederick Heußner, Andreas Kruck, and Xenia Lanzendörfer. “Imperfect Adaptation: How the WTO and the IMF Adjust to Shifting Power Distributions among Their Members.” Review of International Organizations 11 (2) (2016), 171–196.

Content Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 3460 894 112
PDF Views & Downloads 3611 1550 128