The promises and perils of international institutional bypasses: defining a new concept and its policy implications for global governance

ABSTRACT The rapid proliferation of international institutions has been a defining feature of postwar global governance architecture. Many international institutions overlap, playing similar or identical governance roles. Contributing to the existing research literature on this phenomenon, we offer the new concept of ‘international institutional bypass’ (IIB). Just like surgeons grafting new pathways around blocked arteries in coronary bypasses, global governors have increasingly responded to clogged international institutions by working around them. After presenting a definition of IIBs, we articulate why this concept may prove useful to work on global governance, how it is different from existing concepts, and discuss possible policy implications. Specifically, when faced with competition from an IIB, existing dominant institutions either fight back, shape up, or do nothing, which may result in displacement, merger, or co-existence. Our analysis both informed and drew upon a series of case studies published in this symposium issue of Transnational Legal Theory journal.


I. Introduction 1
The rapid proliferation of international institutions has been a defining feature of the postwar international architecture. Since the end of the Second World War, the international system has seen the creation of thousands of international treaties and organisations that have established rules governing a multitude of issues spanning international security, human rights, trade, health and the environment.
At the same time, many of the institutions created in the postwar era actively overlap in their subject matter, such that multiple institutions may play a governance role with respect to the same substantive issues. For instance, the World Health Organization (WHO) is not the only organisation governing global health-others include the Global Fund to Fight AIDS, Tuberculosis & Malaria, Gavi-The Vaccine Alliance, and the United Nations (UN) Joint Programme on HIV/AIDS (UNAIDS). Global security is not the sole prerogative of the UN Security Council; rather, security is also governed by the North Atlantic Treaty Organization, the International Civil Aviation Organization, and the European Security & Defence Policy. Similarly, functions relating to international refugee protection are performed by the UN High Commissioner for Refugees (UNHCR), the International Organization for Migration, and the European Convention on Human Rights.
Faced with the reality of institutional proliferation and overlap, scholars have developed a number of concepts to explain this phenomenon, such as regime shifting, forum shopping, fragmentation, differentiation, institutional choice/adaptation, and regime complexity. 2 These concepts have undoubtedly advanced our understanding of the current international system. They also help to capture at least part of the motivation that states have to establish new regimes: one cannot deny that the rational interests of states and the emergence of new issues will play at least some role in explaining why particular international institutions emerge.
Despite these numerous concepts, we still believe that there is space for at least one more concept within this broader literature on institutional proliferation and overlap: the international institutional bypass (IIB). Just like surgeons grafting new pathways around blocked arteries in coronary bypasses, global governors are increasingly responding to clogged international institutions by creating new ones that work around them, rather than reforming existing structures.
This concept offers at least one distinct advantage over those concepts already found in the research literature on international institutional proliferation and duplication, namely that it describes attempts to fix deficiencies in an existing system. In so doing, the IIB concept focuses on a functional yet relational dynamic of institutional innovation, revealing a particular institution's relationship with the dominant institution that performs the same function. This focus on the functional and relational dimensions of reforms facilitates a better understanding of the dynamics of global institutional change and the evolution of the international system by allowing the identification of commonalities across domains. Going further, we believe this new concept will provide practical insights for guiding the design of future global governance architecture and international policy.
The concept of an institutional bypass-that is, creating a parallel institution that performs at least part of the same function of the dysfunctional institution-has already proven useful for analysing institutional reforms undertaken at the domestic level. 3 In previous publications, we argued that it is potentially applicable to the global governance context as well. 4 To explore this possibility further, we brought together an interdisciplinary group of researchers who specialise in different aspects of global governance and we challenged them to either assess or apply the IIB concept within their respective areas of expertise. 5 Among other things, these discussions revealed at least one major challenge with the use of institutional bypasses in the international context: while the possibility of 'bypassing' existing institutions may be counterintuitive in the domestic sphere, it may be closer to the norm, rather than the exception, in global governance. Domestic institutional bypasses are counterintuitive because sovereign states are often assumed to hold a monopoly in the provision of certain services and the performance of certain functions at the domestic level, as is notably the case with policing. 6 Thus, recognising that states sometimes choose to operate parallel institutions, and in some cases even make them compete with each other, may constitute a paradigm shift for those concerned with institutional reforms and good governance in the domestic sphere. However, this same assumed monopoly does not exist in global governance, where researchers have extensively documented and theorised on the related phenomena of institutional proliferation and duplication. 7 Accordingly, a question arises as to what value, if any, the concept of institutional bypasses can provide to the existing global governance literature. In a previous work, we attempted to provide an answer to this question by presenting a preliminary definition of IIBs and articulating why this concept may prove useful to work on global governance. 8 This current article builds on that earlier work, confirms the utility of defining IIBs in the way we previously did, and discusses some of the possible policy consequences of IIBs. Our analysis both informed and drew upon a series of case studies that are described in separate articles published in this symposium issue of Transnational Legal Theory journal.

II. What is an international institutional bypass?
An IIB is a type of institutional reform that has the following six characteristics 9 : (1) it keeps the dominant institution in place; (2) it creates an optional alternative pathway through which to discharge functions performed by the dominant institution; (3) it has at least one distinctive feature that aims at addressing a perceived dysfunction in the dominant institution; (4) it has effects in the same international regimes or domestic legal orders of the dominant institution; (5) it is compatible with the requirements of the international regimes or domestic legal orders within which the bypass is operating; and (6) it is separated from the dominant institution's governance structure.
As the above requirements make clear, the IIB concept is relational as it focuses on the way in which two institutions-the pre-existing (dominant) institution and the new (bypassing) institution relate to each other. An IIB assumes the existence of a dominant institution performing a function, but does not try to directly modify, change, or reform that dominant institution. Instead, the IIB is designed independently of and without directly disrupting the operation of the dominant institution, which remains in place. This differentiates the IIB from attempts at directly promoting reforms in international institutions, even as the existence of the IIB may indirectly affect the functions and performance of dominant institutions.
In order to qualify as an IIB, the alternative institution must offer a choice to actors between it and the dominant institution. This means that the IIB must perform at least one of the same functions that the dominant institution still performs, all the while presenting at least one distinctive feature that serves to distinguish it from a mere replica of the dominant institution. Accordingly, the IIB may either compete with the dominant institution by attempting to sway actors to use it, or collaborate with the dominant institution by offering a supporting alternative pathway that may reduce overflow and increase overall performance. In either situation, the distinctive feature of the IIB must be aimed at resolving one or more perceived dysfunctionalities of the dominant institution.
Going further still, when applied in international contexts, the institutional bypass concept can be used to describe two distinct ways of actually addressing perceived dysfunctionalities in international institutions. The first is what we call a 'horizontal IIB', where an international institution bypasses other international institutions. The second is what we call a 'vertical IIB', where an international institution bypasses domestic or regional institutions or vice-versa. In the case of a horizontal IIB, the effects need to be observed within the same international regime. 10 Two examples of horizontal IIBs are the New Development Bank and the Asian Infrastructure Investment Bank, which serve as bypasses of the World Bank and the Asian Development Bank, respectively. 11 By contrast, a vertical IIB is created when a dominant institution and the bypass are located at different jurisdictional levels, being it international, transnational, regional or domestic. This notion includes a domestic or a regional institution that tries to bypass an international institution, or vice-versa. One typical example of a vertical IIB is the case where the bypassing institution, despite not being an international institution, produce effects within the international regime where the bypassed international institution operates. For instance, regional swap lines and regional central banks can be conceptualised as bypasses of the International Monetary Fund (IMF). 12 Another example of a vertical IIB is when an international, regional or transnational institution produces effects within a domestic jurisdiction where the bypassed institution operates. For example, regional institutions in West Africa, and particularly those providing health services, can be framed as bypasses of domestic institutions that serve the same functions. 13 In vertical IIBs, an international institution can thus be either the dominant institution being bypassed or the alternate international institution that is bypassing the dominant domestic institution.
Neither of these scenarios is meant to fully encompass those institutional arrangements that can serve as alternatives to dominant institutions but do so in ways that are not compatible with the requirements of the international regimes or domestic legal orders within which the bypass is operating. An IIB, as we defined it, excludes illegal activities that offer alternatives to dominant institutions and would otherwise meet all the other criteria of an IIB. An IIB must also comply with the rules, norms, and principles of the system in which it is operating. In defining the concept in this manner, we are cognisant that many international regimes lack an overarching normative system, meaning that a bypass may not need to actually comply with this requirement in practice. However, many IIBs will be housed within or have effects in existing international regimes or domestic legal orders. In such cases, it is a requirement that the IIB not violate any other rules, norms and principles within those regimes or legal orders. This requirement excludes organised crime and terrorist activities. However, there is a grey zone of activities that are not clearly illegal, as they may not be sanctioned or legally recognised by a formal legal order, but are not prevented by it either. These would be included in the concept of IIB.
Finally, another feature of the IIB concept is that it must be separate from the governance structure of the dominant institution. Arrangements that meet this requirement can take many forms. The creation of a brand-new institution to perform the same function of the dominant institution is perhaps the clearest example. It is possible, however, that the alternative institution materialises by repurposing another pre-existing institution. For instance, a long-existing institution may start to perform a new set of functions that overlap with those already performed by the dominant institution. In this case, the long-existing institution becomes an IIB when actors start to use it as an optional alternative pathway to the dominant institution.

III. Contextualising the concept of international institutional bypass
Many of the existing concepts in the international relations and international law literatures can be used to describe institutional proliferation and overlap in global governance. However, we believe that this admittedly densely populated terrain could benefit from at least one more concept, namely the IIB. Not only can the IIB be distinguished from existing concepts in the literature, but it also captures a particular dimension of the dynamics of global governance that remains largely untheorised: relational institutional innovation.
This dimension emerges from the focus of IIBs on specific attempts to correct and perfect those existing global governance mechanisms that were designed to perform certain functions and achieve particular goals. The interaction between IIBs and these dominant institutions may in turn produce a wide range of outcomes that are at least partly dependent on the dominant institution's reactions to the IIB. These outcomes include, but are not limited to: the dominant institution disappearing and being replaced by the IIB; the dominant institution shaping up and making the bypass redundant; and the dominant institution and the IIB continuing to co-exist and cooperate with each other in one form or another.
By focusing on the relational outcomes of these functional institutional innovations, the concept of IIB is thus narrower than others like regime complexity and fragmentation, both of which also aim to capture the proliferation and overlap of international institutions and regimes. The former (regime complexity) is defined as 'an array of partially overlapping and non-hierarchical institutions governing a particular issue area', 14 and results in the latter (fragmentation) when it 'reduces the clarity of legal obligation by introducing overlapping sets of legal rules and jurisdictions governing an issue'. 15 While both of these concepts rightly account for the seemingly everincreasing number of institutions, rules, norms and principles that exist at the international level, not everything that contributes to increased complexity or fragmentation is an IIB. Any two international institutions that govern a particular issue may contribute to regime complexity or fragmentation, but to qualify as an IIB, one of them must address a perceived dysfunctionality by providing an alternative means of performing the same function as the other. Thus, the World Trade Organization (WTO) and the 1992 UN Convention on Biodiversity are not bypasses of one another since they do not allow for alternative pathways through which to deal with plant genetic material, but rather set out ostensibly contradictory norms as to its proprietary nature. 16 By contrast, regional trade agreements may be perceived as increasing complexity and even promoting fragmentation in the international trade system, but some of them could potentially be classified as IIBs of the WTO.
In other words, the IIB concept can be understood as having a more particularised focus than both regime complexity and fragmentation, focusing on the trees rather than the forest. 17 Accordingly, the concept of IIB hones in on one or more particular institutions and their relationship with what is perceived to be the related dominant institution. At least two other concepts in the international relations literature have a similar focus, namely institutional adaptation and institutional choice. Adaptation starts from the premise that the decentralisation inherent in regime complexity allows for greater adaptability, especially 'when adaptation requires complex changes in norms and behavior'. 18 This notion is closely related to regime flexibility, according to which complexity allows rules to be more easily adapted to different conditions and the needs of different actors. 19 An IIB would be used when adaptation has not happened and when the existing institutional arrangement is not adequately performing its function. In this sense, the IIB is often a second-best solution, becoming especially relevant after attempts to reform the dominant institution have failed. By contrast, the concept of institutional choice is very close to the IIB concept. This similarity is because choice assumes that actors are more likely to work within existing institutions the more suitable they are 'for addressing the cooperation problem at hand', and are increasingly likely to 'choose costlier and riskier strategies that transform the institutional landscape' the more those institutions are deficient. 20 While an IIB assumes the existence of choice, the main difference between the two concepts is the fact that IIB focuses not on cooperation, but on the specific function that institutions perform and the goals they were designed to achieve (eg, reducing carbon emissions, responding to pandemics, or providing financial liquidity to sovereign countries in times of crisis). As a result, an IIB may encompass other cooperative solutions, as in the case of horizontal bypasses, but it may also choose domestic or regional solutions (vertical bypasses) that require less or no cooperation in order to perform the same function. In either case, the IIB assumes the existence of a dominant institution that is not necessarily mandated by institutional choice.
A similar distinction can be made between IIBs and what the literature on regime complexity calls 'rivalry' among institutions. Rivalry is a form of competition between two or more institutions in which one or more of them attempts to displace the others within a given regime. Beyond assuming the existence of a dominant institution, the concept of IIB is agnostic to the underlying intent or strategic motivation behind an IIB's creation and may or may not generate direct competition among institutions. Indeed, while some IIBs may attempt to displace the dominant institution, others may act in a complementary, supplementary or even synergistic fashion.
That being said, IIBs should also be distinguished from actor-led attempts to search for more favourable normative systems. The main difference is that the IIB concept is trying to capture institutional innovation and creativity, which sets it apart from other important concepts such as regime shifting and forum shopping. Regime shifting is 'designed to reshape the global structure of rules' by turning to parallel regimes where alternative priorities exist. 21 By contrast, forum shopping is defined as an activity 'where actors select the international venues based on where they are best able to promote specific policy preferences '. 22 In practice, the concept of IIB covers situations that fall somewhere in between these two existing concepts: on the one hand, IIBs do not necessarily turn to parallel regimes or seek to directly undermine the dominant institution; on the other hand, IIBs must also ultimately interact with the dominant institution in a manner that goes beyond the particular choices of individual actors.
These differences are made all the more apparent in light of the requirement that an IIB have effects within the same normative order as the dominant institution. On its own, this requirement is sufficient to distinguish IIBs from forum shopping, since the latter relies on an individual actor's choice between two or more separate jurisdictions or regimes. However, this requirement to have effects within the same normative order as the dominant institution also means that IIBs simply create an optional alternative pathway that may compete with existing institutions within the same regime rather than effecting a full regime shift in the traditional sense. The IIB can thus be seen as a response to what some authors have termed 'regime viscosity'-that is, the degree of internal friction that exists within a single regime complex and which increases the costs of regime shifting. 23 In summary, the IIB differs from existing concepts in the literature on international institutional proliferation and overlap in a number of important respects. More broadly, however, the IIB concept can be distinguished from many existing concepts on the basis of its underlying objective. Indeed, the IIB does not aim to address one of the most controversial debates in the global governance literature-that is, the question of why proliferation is taking place at all. Instead, the IIB assumes that there is an intention to fix dysfunctions within the existing system, regardless of possible underlying strategic motivations, and adopts a functional view based on performance outcomes once the IIB is in place. Specifically, it focuses on the relational dimension of this functional approach, asking whether there are long-term institutional transformations, changes, or innovations as a result of the complex interactions between the dominant institution and the IIB. This functional and relational view raises a broad range of descriptive, normative, and policy issues, to which we turn next. law and international relations to analyse whether and how the IIB concept might be useful across various fields of study. These fields include corporate social responsibility, finance, health, and labour rights, viewed across bilateral, regional and multilateral governance structures. While the differences among these fields require a great deal of caution in drawing generalised inferences for policy purposes (as we discuss in the next section), they also provide fertile ground to 'test' if the definition provided in this introductory framing article is capable of accurately capturing what scholars observe on the ground. Indeed, many of the constitutive elements of an IIB were challenged in the case studies included in this symposium issue. One such challenge is reconciling the concept with the possibility of coexisting but incompatible legal orders. In other words, can IIBs account for the insights of legal pluralism? 24 While affirming its explanatory value, Victor Ramraj suggests that the IIB concept should be expanded to include the possibility of co-existing legal orders that, although incapable of formal reconciliation or incorporation into a clear hierarchy, are capable of working together in practical terms. This approach, he argues, would allow the concept to avoid adopting a state-centric approach to law and reject the assumption that state-derived law is the only legitimate point of reference for assessing IIBs. Adopting such a legal pluralist understanding of IIBs, according to Ramraj, can facilitate the discussion of normative considerations such as what compatibility might look like, including how competing normative ideals can co-exist and who ought to make such decisions. For example, the Equator Principles III-a process to identify and mitigate environmental and social problems associated with large-scale infrastructure projects to be employed by those seeking financing from International Financial Institutions (IFI)-present a possible normative conflict insofar as the minimum standards required by the IFIs may prove higher than those established by domestic political actors. This, in turn, can challenge the practical capacity of domestic political actors to make trade-offs between long-term sustainability and short-term economic growth.

IV. Challenges to the concept of international institutional bypass
A second challenge is the intent of IIBs vis-à-vis their respective dominant institutions. Rohinton Medhora's discussion of the monetary unions, regional financial arrangements, and central bank swap lines as bypasses of the IMF highlights the significance and changeability of institutional intent in understanding whether and when an institution can be considered an IIB in the strictest sense. The definition of an IIB that we outline above stipulates that IIBs do not try to directly modify, change, or reform the dominant institution. Particularly with respect to regional financial arrangements (RFAs), Medhora argues that a significant driver of their creation was 'dissatisfaction with the domination of the IMF in official global finance'. 25 In this sense, there is some ambiguity as to whether a regional institution such as the Chiang Mai Initiative-the creation of which was driven by a deep dissatisfaction with the IMF's approach to the 1997-1998 Asian financial crisis-could truly be considered an IIB. Excepting occasional rhetoric to the contrary, RFAs such as the Chiang Mai Initiative do not seek to replace the IMF as the dominant institution, but they do seek to introduce heterodox intellectual approaches to addressing financial crises (in many cases successfully) and, in turn, influence how the IMF does business. At the same time, despite a certain degree of hostility in method, the four RFAs Medhora discusses work with the IMF and in certain cases rely on information and requirements it produces.
Edefe Ojomo questions what it means for an IIB to 'address dysfunctionalities', pointing to the fact that regional institutions in West Africa may be addressing the immediate needs of individuals but may not be helping counties to develop domestic capacity. In her case study of the Economic Community of West African States (ECOWAS), she shows that a regional bypass may be circumventing existing domestic institutions to provide particular goods or services to those in need. Specifically, the West African Health Organization (WAHO), a regional institution established by ECOWAS in 1987, has proven effective at addressing health system issues, at least in part, by engaging directly with domestic stakeholders such as hospitals, clinics and pharmaceutical companies to delivers goods and services directly to citizens. WAHO often coordinates the distribution of limited healthcare resources-a function typically performed by governments, even when governments lack the capacity to perform this function effectively. WAHO also acts as a regional representative with global health organisations such as the WHO, Médecins Sans Frontières (MSF), and Gavi-The Vaccine Alliance. WAHO's strategy has proven effective at addressing unmet needs resulting from deficiencies in domestic institutions. But this strategy may also stunt the development of domestic institutional capacity in the long term, blur lines of accountability with respect to governance, and raise questions regarding state legitimacy. 26 In her examination of the roles played by IIBs in the ongoing attempts to credibly enforce human rights standards on transnational corporations (TNCs), Oonagh Fitzgerald raises questions about their effect on domestic institutional development. She details the limitations of the (former) dominant institution-the United Nations Commission on Human Rights (UNCHR)-as well as the role of IIBs such as the Kimberley Process for Conflict Diamonds, the UN Global Compact, and the Guiding Principles on Business and Human Rights (Guiding Principles) in both exploring non-legal options for normalising human rights norms and bringing about the UNCHR's successor, the United Nations Human Rights Council (UNHRC). In so doing, she suggests that IIBs may well further the public good, particularly in the short term, but they 'may also serve to distract from and delegitimise efforts to address the real issues and to delay achieving real solutions'. 27 For example, discussion, debate, and adoption of nonbinding standards like the Guiding Principles, may well have delayed any meaningful change by giving the appearance of action, particularly on the part of the TNCs themselves, thereby dissuading concerned parties and observers from fully considering the issue of enforceability. Her analysis raises the troubling suggestion that IIBs may be reflective of the lack of political will to effect real change. This problem may be particularly acute regarding TNCs specifically, and human rights generally, where there is research to suggest that not all states are interested in creating effective institutions. 28 More broadly, it reminds us that well-functioning institutions, much like high levels of social capital, 29 are not necessarily socially or morally desirable.
Each of these articles embrace but also challenge one or more of the six elements we propose as constitutive of an IIB. However, in our view, none of those challenges is so fundamental as to threaten the validity of the concept. Indeed, each author suggests that the concept is useful in furthering the understanding of international law and international relations within the particular areas they discuss. To us, this implies that the concept has value as long as it is not deployed in a highly rigid manner. Rather, although the defining characteristics of an IIB should be maintained, on a case by case analysis particular elements may need to be modulated to capture the particular nuances of each scenario in which any given IIB is operating.

V. Policy implications of international institutional bypasses 30
Once an IIB is implemented, its objective of being more effective than the dominant institution means that it can be considered a potential competitor, as it performs the same function. Broadly speaking, dominant institutions can respond in three ways to competition from an IIB: (a) fight back and boycott the bypass (attack); (b) attempt to 'shape up' and compete with the bypass (defend); or (c) do nothing (retreat). Each of these reactions may, in turn, generate at least three types of more permanent institutional changes: (i) one of the two competing institutions may disappear (displacement); (ii) one may be absorbed into the other or a new institution created from a combination of the two institutions (merger); or (iii) both institutions may continue to exist either by dividing tasks or merely because each of their institutional capacities alone is not enough to meet demand (co-existence). We explore each of these hypotheses and their consequences next.

Possible responses
a. Attack: fighting back against the bypass One possible response of a dominant institution to a new bypassing institution is for the former to fight back and resist the implementation or operation of the IIB. A motivation for this response to the bypass is the perception that it represents a threat to the dominant institution's ongoing viability. This response can be initiated either before or after the IIB is created. Fighting back before the bypass is established would aim at stalling or preventing the IIB's approval or implementation. Strategies to accomplish this outcome include lobbying against its creation, attempting to block the allocation of resources necessary for implementation, and withholding information or cooperation necessary for the IIB to function effectively. A dominant institution may also engage in ex post resistance by sabotaging the IIB's capacity to perform its functions efficiently such as by employing tactics analogous to anticompetitive practices. A dominant institution might also fight back through a negative public relations campaign, designed to affect the image of the IIB. Such campaigns may include attacking the legitimacy of the IIB's sponsoring entities, making appeals to potential supporters, or alleging the IIB is an attempt to advance the agenda of particular interest groups to the detriment of others' interests.
Regardless of how successful a dominant institution is in adopting this strategy, the fact that it was adopted may increase the cost of implementing an IIB. By creating obstacles to the operation of a bypass, the dominant institution could make it costlier to pursue the project. Moreover, if the boycott is effective, it can destroy a potentially successful bypass by depriving it of resources, or by exploiting information asymmetries that make stakeholders apprehensive about supporting or resorting to it. While boycotts are likely to have similar implications domestically and internationally, there are systemic differences in outcomes, as we discuss in Section V.2 (a). b. Defend: shaping up to compete with the bypass The success and popularity of an IIB might force the dominant institution to mimic the IIB or to seek other improvements to its own processes. This response can happen when the current institution is inspired by the success and efficiency of the bypass, when it fears for its existence, or some combination of the two. In these circumstances, shaping up can be a natural outcome of a competitive process, in which the dominant institution reacts to the perceived threat coming from the IIB. Here, a major strategic dilemma must be confronted: for dominant institutions to shape up in the face of competition, there must be negative consequences for its failure to do so (eg, loss of financial support, political influence, status, effectiveness). However, these same consequences, at least if announced at the initiation of the bypass strategy, may instead intensify attacks on the bypass, rather than encouraging shaping up.
There are different defensive strategies that a dominant institution might adopt in order to shape up. A dominant institution may shape up simply by mimicking the IIB. This is a relatively cheap alternative, but it may not be possible in cases where the bypass offers a unique configuration of services that the dominant institution cannot easily replicate. Even if mimicking is possible, the best a dominant institution can hope to achieve is to match the IIB. This may prevent further decline in demand allowing the dominant institution to 'catch up' to the bypass. A dominant institution might also be inspired to innovate to surpass the IIB. The question would then become how long the two institutions would be able to sustain the competition before stakeholders, funders or states weighed in on the matter. c. Retreat: not reacting to the bypass A dominant institution may not react to an IIB for two principal reasons. First, the dominant institution may not perceive the IIB as a threat, whether it is a threat or not. Second, there may be obstacles preventing the dominant institution from undertaking any meaningful action or reaction to oppose the IIB. In the absence of effective opposition to the IIB, it is possible that additional resources will become available and it will gain status and grow in importance. With additional political and financial support, the IIB is likely to expand, and if it does so successfully it could establish itself, and may well render the dominant institution obsolete.

Possible outcomes
For each of the responses to an IIB, multiple forms of equilibria can be imagined. Some of these may generate a series of lasting institutional changes, which include: (i) displacement, where an institution eventually loses its ability to compete successfully, becomes redundant, and eventually fades off; (ii) a merger, in which the two institutions are restructured to become one; and (iii) co-existence, where both institutions remain operational but divide responsibility over the functions they perform.
Any of these scenarios may or may not be preceded by significant and fruitful institutional innovations on both sides. The institution that remains at the end of the competitive process could potentially make productive use of its institutional innovations. Thus, competition among international institutions in global governance has the potential to generate fruitful outcomes. The risk is that before one of these equilibria develops, there may be a long period during which stakeholders are confronted with redundancy, greater costs and coordination problems. The lengthy co-existence of two institutions could lead to a waste of time, money, and human resources if either institution adopts anticompetitive practices or if they together maintain unnecessary surplus capacity that can serve more users than there is demand for their services.

a. Displacement: either the dominant or bypassing institution disappears
While there is no guarantee that displacement will take place in any particular situation, the possibility that the dominant institution might become redundant and shut down highlights the ultimate appeal of bypasses in reducing obstacles to international institutional reforms. Whereas the dysfunctional dominant institution may be a strong and powerful entity before the creation of a bypass, it may turn into a less relevant institution over time. As a result, it may become incapable of resisting reforms once the bypass has evolved successfully to take over the 'market' for certain functions.
In most circumstances, the likelihood of displacement seems higher for domestic institutional bypasses than international bypasses. In the domestic sphere, intentional bypasses 31 are likely to lead to at least some duplication of tax-funded services. This situation, in turn, can impose a great deal of pressure on national governments to reduce redundancy, decreasing the likelihood of co-existence as a result of clearer lines of accountability. In contrast, in the international sphere, an IIB's financial and political resources may be coming from a different source than those of the dominant institution, such as from a different group of countries or from INGOs. This diversity in sources of support may reduce pressure on international institutions for retrenchment, which in turn can result in institutional proliferation since 31 Intentional bypasses are those that are conceived of and implemented with the intent of addressing one or more specific dysfunction(s). They are brought about in a centralised and coordinated fashion -often solely or primarily by a single actor-in an attempt to provide users with a functional alternative to the dominant institution. This type of bypass usually occurs in the wake of a crisis that creates a window of opportunity for reform. The State of Rio de Janeiro's creation of a special police unit (Unidade de Polícia Pacificadora or UPP) tasked with the policing of the region's large number of informal settlements (favelas) is a good example of this type of bypass.
new institutions are created without old ones being dismantled. The problem, however, is not unique to the international sphere. In the domestic sphere, spontaneous bypasses may also be subject to the same problem. 32 Consider, for example, the case of low-cost private schools in India, which, despite their efficiency, have yet to cause enough pressure on political actors for them to modify the dysfunctional system of public schools, resulting in their co-existence. 33 It is also possible that displacement will not result in the disappearance of the dysfunctional institution. As Oonagh Fitzgerald notes in this symposium issue, the allocation of attention, time, and resources to an IIB may be a strategic move on the part of those to be regulated in order to dilute or disrupt meaningful and long-lasting change. 34 In other words, financial and political support may not be solely based on an institution's capacity to innovate or promote positive institutional change. It is not always clear why funders and supporters would decide to favour one institution over another. This might lead to investments that are not reflective of an institution's capacity to offer promising, self-sustaining or desirable innovations. This could create the risk of allowing a dysfunctional institution to survive longer than it would otherwise and potentially displace the more functional institution.
Last but not least, there are cases in which displacement may not be possible or desirable. As Edefe Ojomo points out in the context of global health, there may be limited state capacity to respond to regional or other extranational IIBs, but this does not imply that such state functions should be permanently transferred to international, transnational or regional bodiesespecially if there are long-term benefits that can be achieved from building this capacity within a country. This challenge, in turn, raises questions about the role of IIBs in such cases: would they be simply generating longterm pain for short-term gain? 35 b. Merger: the dominant and bypassing institutions unite In the case of a merger, the dominant institution and the IIB become a single entity. One of the rationales for a merger, as discussed earlier, is the possibility that the dominant institution might incorporate the innovations embodied in the bypass and begin to perform the functions with the same higher level of 32 Spontaneous bypasses are the result of scattered and uncoordinated actions, at least initially, that collectively provide a good or service more effectively and/or efficiently that the institution formally tasked with doing so. Bypasses of this type tend to exhibit a more organic approach to reform, lacking clear leadership or strategic planning. The widespread use of private security services by business and individuals for services that are traditionally considered the responsibility of publicly funded police forcesas is the case in much of Latin America, South Africa, and, indeed, certain areas of the US-is a good example of this type of bypass. See generally, Prado and Trebilcock (n 3) ch. 5. 33 Ibid, ch. 5. 34 Fitzgerald, this volume. 35 See Chasukwa and Banik (n 26). quality and effectiveness. In such a case, there may be a strong incentive to promote a merger: eliminate redundancy. In the cases where an IIB operates within the same governance framework as the dominant institution -as is the case with many regional financial institutions and the IMF-there is not only the possibility of a merger, but a streamlining exercise may become effectively the most desirable arrangement, as Rohinton Medhora suggests in his contribution to this symposium issue. 36 On the other hand, in some other cases a merger may not be a viable solution, particularly where the dominant and bypassing institutions were created by or under the auspices of separate governance structures. The Accord on Fire and Building Safety discussed in Victor Ramraj's article provides a clear example of the potential difficulties in such a situation. 37 A merger can be conceived of as a potentially positive solution to the redundancy problem, especially in cases in which the dominant institution has not improved enough to surpass the IIB, but there are still significant obstacles to eliminating it from global governance. There are, however, two important risks. The first risk is that the restructuring process can result in a badly designed merger. This outcome may simply drag down what could otherwise be high-performing institutions. The second risk is that the merger can lead to an initially positive outcome but may progressively become less effective with the disappearance of incentives to improve in the absence of institutional competition. This is not an exclusive risk of a merger but may also happen if the IIB begins operating alone after the dominant institution fades and disappears.
A merger can also be a strategy that a dominant institution can use to prolong its survival. Thus, the merger may be conceived as a solution of last resort for a dominant institution that foresees its own termination and is incapable of either generating a successful boycott of the IIB or shaping up in response to the competition. In these cases, the outcome of the merger is unclear. As is the case with an efficient corporate merger, it is possible that the less efficient entity will become more efficient as it is incorporated by the bypassing institution and is forced to adopt its more efficient modus operandi. However, there is also a risk that the dominant institution will simply diminish the IIB's effectiveness. This outcome will largely depend on the terms of the merger and which personnel, procedures, methods, and modus operandi will prevail: those adopted by the IIB, those used by the dominant institution, or a combination of both. c. Co-existence: the dominant and bypassing institutions divide tasks The last hypothesis we analyse is the possibility that the dominant institution will preserve its existence by retaining a set of services while downloading others to the IIB. In this scenario, a division of tasks may be structured in such a way that functions do not overlap, whereby each institution performs an exclusive set of functions. Alternatively, it can be structured in a way that both institutions perform the same functions, such as in situations where neither institution is individually capable of meeting the full demand for that service or function. In the former, the IIB would no longer be an IIB.
In the latter, the IIB remains an IIB, but it is not clear that the IIB could survive on its own. Some institutional bypasses tend to combine or amalgamate the services provided by several dominant institutions. A domestic example is Poupatempo, a bureaucratic reform in Brazil's State of São Paulo that implemented insights from New Public Management in an attempt to streamline service provision, conceptualise citizens as clients, and provide single-window service delivery. 38 A recent international example is the Africa Centres for Disease Control & Prevention (Africa CDC), which was inaugurated in 2017 as a specialised agency of the African Union in order to streamline, coordinate and integrate the technical services provided by so many other global health actors working across Africa, perhaps most notably WHO's Regional Office for Africa. Where this is the case, cream-skimming (ie, offloading functions that are regarded as unimportant, resource-intensive, or not otherwise rewarding) is generally not possible as the bypass is effectively covering the field.
In contrast, some bypasses parallel only part of a larger institution. The Unidades de Pronto Atendimento (UPAs)-emergency care centres-established by the Brazilian government are a clear domestic example of this type of institutional bypass. Intended to address excessive demand and resulting wait times and budgetary stresses relating to the provision of health services, the UPA initiative was designed to serve as a first stop for Brazilians making use of the public health care system addressing less serious and routine concerns and referring patients requiring more substantial care to the correct institution. Because UPAs are always open at all hours of every day, access to primary healthcare services has been greatly improved while at the same time hospitals and other healthcare institutions can devote more resources to their core competencies. 39 The same can happen in the international sphere. For example, Gavi-The Vaccine Alliance was created in 2000 as a public-private partnership to facilitate more efficient procurement and deployment of life-saving childhood vaccines than what WHO had previously been able to achieve. Over two decades, Gavi has reached 760 million children and prevented 13 million deaths through several key institutional innovations, 40 including leveraging economies of scale, pooling country demand for vaccines, guaranteeing long-term purchasing, negotiating lower prices from manufacturers, co-financing with national governments, market shaping, and strengthening global vaccine delivery platforms. 41 This success has allowed WHO's Department of Immunization, Vaccines and Biologicals to focus on facilitating research and development, setting technical standards, regulating vaccine quality, developing policy guidance on vaccine use, and advising national health authorities-all of which are services that Gavi now relies upon for its own efforts. 42 For both the UPAs and Gavi, the bypassed institution essentially conceded carriage (formally or otherwise), allowing both institutions to co-exist while specialising in different dimensions of service delivery. The division of tasks may be a strategic decision on the part of the dominant institution to use the IIB to its own benefit, transferring a block of services that it does not want to perform to the bypass, while continuing to function by providing other services. This possibility would require significant political power, foresight, and strategic thinking by leaders within the dominant institution. In this case, the division of labour would be combined with cream-skimming, as described earlier. This division of labour might be proposed either at the time of the creation of the IIB or once the IIB has demonstrated initial success and has been identified as a threat. If the balance of power is more favourable to the IIB than to the dominant institution, the off-loading of undesirable tasks could operate in the opposite direction.
This division of labour can also be temporary, being used as a strategic move by supporters of the IIB facing mounting resistance from the dominant institution (such as boycotts, for instance). If they fear that the resistance may eliminate the IIB, the division of labour may help offset some of this negative reaction. If cream-skimming occurs in favour of the IIB while keeping unimportant or less popular functions with the dominant institution, there may be a stronger reaction. If the division of labour does not generate a negative reaction, this strategy might facilitate the fading away of the dominant institution, while at the same time muting opposition that could potentially harm or even eliminate the IIB.

VI. Conclusion
In this introductory framing article, we introduced the IIB concept, presented its constitutive elements, and explained how it is distinguishable from other related concepts in the specialised literature on international institutional proliferation and overlap. We then provided a map of the important insights offered by the various authors who contributed to this symposium issue of Transnational Legal Theory journal and the challenges that each author has posed to the IIB concept. Despite these challenges, the articles published in this volume all suggest that the IIB concept may be useful for informing future decisions on global governance architecture. We ended this framing article by providing a typology of the possible ways in which IIBs may promote change in global governance. This typology may offer some hypotheses that can be tested in future case studies and large-scale empirical analyses.