3.1 Introduction
Almost one year into the COVID-19 pandemic, a ray of hope shone in the form of a set of vaccines that could be potentially rolled out quickly enough to save millions of lives. But rich countries soon cornered the market, empowered by rights under pre-existing contractual arrangements, and enabled by exceptional access to capital in a world ruined by lockdowns.Footnote 1 As some of the richest countries effectively hoarded vaccines, and some of humanity’s most vulnerable groups were essentially left to die, one of the clearest examples of lack of solidarity at a global scale was crystallizing before our eyes.Footnote 2 From the Siege of Melos to the Siege of COVID, in 2021, the rich of the world did what they could, and the poor suffered what they must.
At a bare minimum, international organizations might be expected to play a central role of tackling such challenges to inter-state coordination.Footnote 3 And yet, global COVID-19 vaccine governance was characterized by the sidelining of international organizations, mainly for three reasons: first, perhaps as a result of the capital intensive nature of the challenge, much of the multilateral decision-making was privatized, creating in effect a kind of hybrid regulatory regime in which private players, like the Bill and Melinda Gates Foundation, played a pivotal role, shadowing international organizations and their public powers.Footnote 4 Second, COVID-19 arrived at a peak state-sovereigntist moment, in which the likes of Donald Trump and Jair Bolsonaro were able to play their nationalist tune without much control, undermining multilateralism.Footnote 5 And, third, even when international organizations like the WHO, the World Bank, or the WTO strove to act, each of their individual mandates seemed too narrow to tackle the different dimensions of the vaccine challenge – hence, the move to private actors, with deeper pockets and more flexibility.
While leadership in cash-strapped international organizations seemed to have little option but to accept an expanded role of well-funded private actors, and to juggle the tension with Trump and his allies, they were quick to realize that interaction among organizations was something that they could do something about. Enter, thus, the Multilateral Leader’s Task Force on COVID-19, a joint initiative of the IMF, the World Bank, the WHO, and the WTO, aimed at supporting ‘the roll out of COVID-19 tools by leveraging multilateral finance and trade solutions, particularly for low and middle-income countries’.Footnote 6
The intuition behind the Task Force seems reasonable. In a fragmented institutional landscape, attention to inter-institutional interaction, and strategies of coordination, seem paramount. And yet, this challenge is often overlooked by the law of international organizations, which famously focuses on the legal relation between the organizations and its member states, while ignoring its interaction with other actors – such as other organizations.Footnote 7 This chapter intervenes in that debate, by exploring the role of law in organizational interaction. In contrast with recent work on international institutional law that seeks to overcome functionalism and make legal sense of interaction,Footnote 8 this chapter argues that interaction among international organizations is a legally constituted phenomena, in two specific senses. First, law constitutes the space of the interaction (that is, their ‘organizational ecosystem’). Second, law provides the background norms for organizational autonomy and the vocabulary for the decoupling of the organization’s practice and its formal goals. Such a decoupling through institutional law allows international organizations to flexibly interact with each other and adapt to external pressures. Thus, in its dual role, international law provides the building blocks of interaction, playing a crucial role before the need to ‘regulate’ interaction even appears.
3.2 The Varying Size and Density of the Space of International Organization Interaction
Interaction among international organizations is the rule, and not the exception, in global governance. It defines a wider sphere of power that goes beyond the independently exercised powers of discrete institutions – a wider analytical lens that has fruitfully been explored by global governance scholars in recent years. Work on ‘organizational ecosystems’Footnote 9 applies to international relations some of the insights of the 1980s ecological approach in the sociology of organizations, which took as its analytical unit the population of organizations (and not each organization as individual actors).Footnote 10 From this perspective, interaction between organizations occurs as they compete for scarce resources: the larger the population, the stronger the competition for those resources. As a result of such competition, the population adapts – mostly by reducing its size, eliminating the weakest organizations or, in the case of international organizations, also by expanding or reducing their areas of activity.Footnote 11 Moreover, while competition is the main form of interaction for organizational ecology, this approach does envisage cooperation when international organizations see benefits in suspending competition, while acting in the same niche,Footnote 12 with the intention of creating more resources or to achieve a wider goal that cannot be attained individually.Footnote 13
Adopting the mindset of a wider space of international organization interaction (or its ‘ecosystem’) is a step in the right direction, as it leaves behind images of fragmentation or regime collisionsFootnote 14 that risked obfuscating the fact that, more than a problem to be resolved, such interaction is a particular form through which legal meaning is createdFootnote 15 and power is exercised.Footnote 16 Despite its advantages, such an interactional mindset still provides very little insight into the actual characteristics of interaction. The regulatory space is not static: its form and texture shift constantly as international organizations interact. Even if we posit the existence of an ecosystem of interaction, we know very little about its variation in terms of size and density.
Let us begin by exploring the question of size. The demise of formal international organizations has been greatly exaggerated. While growth rates in the formation of intergovernmental international organizations did decrease by 20% in the first years of the twenty-first century, compared to the previous decade,Footnote 17 their growth rate from 2000 to 2022 has kept, at 58.92%, just below the rate of all other international institutions (including non-governmental organizations) in the same period (67.34%).Footnote 18 That is, while the growth peak of international organizations between 1980 and 1999 (315%)Footnote 19 was indeed remarkable, a lower rate of organizational growth does not mean, by any measure, that we live in a world with less international organizations – or even with a decreasing international organization density. On the contrary, if the current trend continues (and it is a modest trend), we might expect a doubling of international organizations every 50 years or so.
But that is just half of the story. Just as they are created, international organizations also die. Eilstrup-Sangiovanni analysed the population of defunct international organizations and found that, of 561 international organizations created between 1815 and 2005, 218 had ceased to exist by 2015 – that is, 39% of the overall rate of mortality.Footnote 20 Since 2000, though, the mortality rate has decreased, stabilizing around 7.6%.Footnote 21 Of such rate, almost half implied some kind of organizational continuation (be it succession or merger/absorption by some other international organization), while the other half involves the actual end of activities, under the form of termination or some form of desuetude.Footnote 22 Thus, while the space of international organization interaction is indeed growing, such growth is not only the result of new organizations, but also of the underlying dynamics of international organization mortality, which is not uniform at all: global international organizations tend to have a lower mortality rate than regional international organizations which, in turn, have varying mortality rates depending on the time period.Footnote 23
The second question is one of density. Often, the idea of proliferation of international organizations suggests an evenly dense texture of governance; that is, an even distribution of new organizations in the space of governance.Footnote 24 However, the opposite is the case. Recent quantitative work has explored overlap as the most intense form of interaction and finds that overlap among international organizations tends to cluster around certain issues and around certain scales of organization. Haftel and Lenz measured the overlap (understood as coincidence of policy competency and membership) among a sample of the 76 most authoritative international organizations developed by the Measure of International Authority project, which is a good indicator to assess the overall political relevance of international organizations.Footnote 25 The study found that 40 per cent of such international organizations do overlap in either mandate or membership, or both, with overlap roughly doubling from 1970 until today – a trend driven mostly by the expansion of competences, rather than by membership.Footnote 26 Most importantly, perhaps, such overlap tends to cluster around particular issues. Confirming what intuition might suggest, the issues in which most overlap seems to occur are financial stabilization, development and humanitarian aid, research policy and human rights, while the issues with less overlap are industrial policy, currency, competition policy, macroeconomic coordination and taxation and, finally (and perhaps unsurprisingly), energy.Footnote 27
Ultimately, then, the ecosystem of international organizations is composed of some densely populated islands of intense interaction, while other islands show low intensity interaction and, finally, other areas no interaction at all. These islands, in turn, shift size constantly – not by increasing or decreasing their membership but rather by expanding the mandate of already existing organizations.
3.3 A Legal Space of International Organization Interaction
Such a constantly shifting landscape is mostly understood by organizational ecology to be naturally occurring. Seemingly adopting the factual premises of their ecosystem metaphor, interaction among international organizations is often understood to occur in a space of nature, in which organizations interact as naturally occurring organisms in a biological ecosystem. In this reading, law is crucially absent. The size of populations and their density are functions of spontaneous organizational dynamics, and the demand for regulation only appears at a later moment (if at all), when the malfunction of interaction (for example, as incoherence or forum shopping) triggers demands for some kind of top-down ordering principles (such as hierarchy) to the otherwise organic process of interaction.Footnote 28
However, the ecosystem of international organization interaction does not occur in a legal vacuum – it is neither pre-legal, nor independent from the law.Footnote 29 On the contrary, it is thoroughly constituted by international law, both in its size and in its density. International lawyers have developed a vocabulary for the wider sphere of governance in which international organizations interact. The image providing more traction here is that of space, which has been explored in different contexts. First, as I have argued elsewhere, it is possible to read the interaction among international organizations as part of the ‘global regulatory space’Footnote 30 – a notion originally suggested in a domestic setting as a reaction against the narrow reading of regulatory processes in terms of conflict between public authority and private interests.Footnote 31 Transposing such a notion to global governance, interactions can be seen as part of an emergent ‘global administrative space’; that is, ‘a space, distinct from the space of inter-state relations governed by international law and the domestic regulatory space governed by domestic administrative law, although encompassing elements of each’.Footnote 32 Following a similar line, but in a different context, the non-hierarchical interaction among regimes has been interpreted as a ‘legal space’ in Europe. Such European legal space ‘coalesces what sixty years of Europeanization of domestic public law have achieved, and articulates the specific relationship between the various legal orders as one of mutual structural dependency’.Footnote 33 In Latin America, in turn, interactions between the Inter-American Court of Human Rights, domestic courts and other international courts (such as the International Criminal Court) have also been read as reflecting an Inter-American legal space that contributes to the dynamic legal developments that have characterized that region in the last decade.Footnote 34
In all these iterations, the image of a legal space is a step in the right direction. It adopts as its starting point the wider unit of international organization interaction, thus providing a useful alternative to limitations of legal hierarchical order.Footnote 35 Moreover, in highlighting its legal nature, the image of the legal space provides a counternarrative to the notion that interaction among international organizations occurs organically, instead emphasizing that the ecosystem is legally constituted, and is thus the outcome of the agency of people in positions of power.
With that in mind, let us go back to the size and density of the organizational ‘ecosystem’. As a manifestation of a legal space, the size of the space ‘international organization interaction’ is a product of legal reasoning. As Sabine Müller-Mall has explained, a ‘legal space’ can be read as being composed by acts of normative perception, which are necessarily relational: the perception of a particular legal utterance only makes sense in relation to other perceived legal utterances. A particular legal space is thus defined by that constant process of relational observations.Footnote 36 The same can be said of international organizations: the limit of the space in which international organizations interact is defined by the relation that the interpreter makes of a particular point of the internal law of that organization (a point in space) with her interpretation of the law of a different organization, and then with the law of another organization and so on. The space defined by that line linking these different legal utterances in reference to different organizations is the space of international organization interaction – the outer limits of the ‘ecosystem’, as it were.
The space of international organization interaction is thus legally constituted, performative (in the sense that it mobilizes meaning to have an effect) and is always dynamic: the interpreter chooses the particular extension of the legal space with regards to the particular problematic that inspires her approach to interaction, which allows for the meaningful location of a particular set of legal utterances. Ultimately (and this is perhaps not clear enough in prior work on the legal space, with its focus on non-hierarchy) the space of interaction is not ‘out there’ – it is necessarily performed through interpretations linking legal utterances made in reference to the internal law of two or more international organizations.Footnote 37
A similar thing can be said of the varying density that characterizes the space of international organization interaction. Instead of a naturally occurring process of organic clustering, density of governance is a function of choices expressed through law, which influence how the power of an international organization is clustered around certain issues and not others. A productive way to start thinking about the density of certain clusters of inter-actual governance is to consider them in reference to two different variables: hierarchy between international organizations and mandate differentiation (for example, differentiation in terms of membership, geographical focus or subject-matter competence). Both these variables are constituted by international law, and both have an impact on governance density, as we now turn to see.Footnote 38
To begin with, the densest clusters of governance might emerge from international legal norms that create hierarchy among organizations but establish no clearly differentiated mandates. In that case, international organizations will need to continuously be in contact with each other to define their space of action, both vertically (in the sense of hierarchy) and horizontally (in the sense of mandate differentiation). That is the case of the peace-keeping and security fields, where the combination of Articles 53 and 103 of the UN Charter gives the UN supremacy over regional organizations with which it might collaborate, but very little is said as to the distribution of competences among them, thus triggering myriad legal problems in terms of responsibility.Footnote 39 Lack of differentiated mandates might lead to competition between organizations, which can be tackled by formal legal arrangements among organizations facilitated, in turn, by hierarchy – as has been, in fact, the case with the involvement of regional organizations in peace-keeping and security.Footnote 40
A medium level of density can be expected from legal arrangements creating hierarchical relations and establishing differentiated mandates, which require vertical interaction among international organizations of different hierarchical levels, but little horizontal contact. Despite its apparent lack of hierarchy, that is in practice the system created by Article XXIV of GATT, which regulates the interaction between the WTO and regional economic regimes.Footnote 41 Here, we might expect (at least in principle) more cooperation between organizations, expressed in formal institution arrangements that follow hierarchy – such as the notification procedure established in the GATT. A similar level of intensity can be expected from norms that establish non-hierarchical interactions between organizations with non-differentiated mandates, as they must interact horizontally to define their sphere of governance but have no need to horizontally interact with other international organizations. That is the case of the protection and use of plant genetic resourcesFootnote 42 and most of the issues studied by the organizational ecology literature, which focuses on populations occurring in precisely such contexts.Footnote 43 In this pattern of interaction we might expect to see more competition, as international organizations reach for the same pool of resources, without a framework of hierarchy.
And, finally, the least dense cluster of governance might emerge when the law establishes no hierarchy and clearly differentiated mandates, as here international organizations have little need to establish links with each other, if at all. In that case, very light interaction is to be expected. That would be the case, for example, of the interaction between the IMF and the WTO on issues of monetary policy and trade – in particular, with regard to the duty on behalf of the WTO’s Committee of Balance of Payment Restrictions to consult the IMF on the balance of payments situation of a WTO Member that, facing sharp disequilibrium in its balance of payments, wants to apply import restrictions.Footnote 44 Such low density of governance might create incentives for non-binding cooperation, as organizations are not competing for the same resources and might, in fact, adopt a shared goal that boosts each of them in their own area of competence.Footnote 45 Of course, in the extreme, total lack of hierarchy and crystal clear mandate differentiation would imply no interaction and, thus, no governance density at all.
3.4 International Organization Interaction and the Dual Role of International Institutional Law
The law of international organizations has a limited vocabulary to account for international organization interaction. Part of the problem, as Klabbers has pointed out, is that the underlying rationale organizing the law of international institutions (‘functionalism’) provides an account of the relation between members states and the organization, but remains silent on the engagement of the organization with the wider world (for example, non-member states, private parties and, important for our case, other international organizations).Footnote 46 While this observation is accurate,Footnote 47 it is not the whole story. The problem is not (only) functionalism. As with any theory, functionalism has proven to be useful for one thing (providing a rationale for the regulation of the principal–agent logic) but not for others. To make sense of the role of law with regards to relations among international organizations, we might need to start looking elsewhere.
As discussed in Section 3.3, interaction is an intensely legal phenomenon: both the outer limits of the space of interaction and its inner dynamics are constituted through law. Obfuscating the legal nature of interaction is, precisely, the main shortcoming of organizational ecology. However, international organization interaction is not only a legal phenomenon. When international organizations interact, such interaction is more than just their underlying legal regimes interacting. International organizations develop organizational autonomy, which is based on law, but goes beyond the relative normativity of their respective constituent document or internal law. It is crucial to focus also on the organization of the law and not just the law of the organization. Providing a legal account of international organization interaction requires, thus, to take seriously the difference between the law of the organization and the organizations itself – without going to the extreme of reifying the organization and considering as existing ‘outside the law’, as organizational ecology does. As Samantha Besson has rightly put it, international organizations are ruled and rule by international law.Footnote 48 The limit of functionalism in international organizations law is that it provides a theory for the first dynamic but not for the second, which is the space where interaction occurs.
Perhaps the best way to start going beyond such limitations is to consider international organizations as semi-open entities – that is, to see international organizations as fully structured organizations that are, nonetheless, sensitive to external pressures. Notably, both international organizations law and most work in organizational ecology consider each organization as a closed unit, not able to adapt.Footnote 49 Only recently have ecological studies started to account for processes of organizational openness and adaptation.Footnote 50 However, organizational theory has provided a vocabulary to speak of just such phenomena, at least since the 1960s, as a problem of ‘decoupling’.Footnote 51 Classical organization research took a strict functional view of the formal purposes of an organization, in which formal goals were aligned (‘coupled’) with their practices. However, critical organizational studies soon began to realize that organizations were open to their environment. And, in that open context, organizational practices were decoupled of formally stated goals – not as a problem of incoherence or lack of compliance with internal regulation, but rather as a way in which the organizations adapted to external pressures, while leaving formal goals unchanged. Decoupling of policy and practice started being read thus not as a problem, but as one of several organizational strategies of adaptation and resilience.
As with any organization, it seems reasonable to expect some level of functional decoupling in international organizations, though there seems to be little research on the matter.Footnote 52 For the purpose of international organizations law, though, the decoupling framework, with its underlying concept of a semi-open organization, is a useful starting point to explore the varying roles of law in international organization interaction – beyond functionalism’s narrow focus. As mentioned in Section 3.3, international organizations are constituted by law, but are more than legal regimes: they are autonomous organizations that govern through law. Law plays, therefore, two different roles in this context: first, it constitutes international organizations as autonomous actors and, second, it assigns competences and distributes rights and obligations for these organizations to operate.
In its first role, international law serves as the background norm for international organization interaction. This is, for lack of a better word, a ‘deep’ process of organizational constitution. This is obviously true for notions of statehood and state consent, through which international law constitutes the building blocks of the international ‘reality’ that create international organizations. More directly related to international organization interaction, international law also creates the doctrinal vocabulary that allows the very existence of international organizations to be taken for granted. In this regard, the background idea of autonomy is key.Footnote 53 Through the vocabulary of international law, the ontological existence of international organizations becomes possible, as does the idea of international organizations as autonomous actors, and not, say, organs of a state, parts of decentralized networks or private actors.
To be sure, functionalism seems not to work as a theory for this kind of proto-law of international organizations, as there is no specific function being assigned. Instead, international organizations law seems to play a role here more akin to that of Robert L. Hale which recognized for property rights: the rules that define the rights of property owners are more than recognition of rights to individual actors, but in fact distribute bargaining power between competing interests and, thus, power in society.Footnote 54 Similarly, the notion of autonomy distributes bargaining power between the organization, member states and other actors: states get a strong bargaining position, as their consent is required to create the organization but then the idea of autonomy gives the organization an enhanced position, as it is able to increasingly adopt decisions without such consent. Third parties, of course, are given no bargaining power at all.
The fact that autonomy operates as a background norm, and not as a principle of the institutional ‘foreground’ norms, is the reason why international institutional law has such difficulty with locating the internal law of the organization (that derives from its autonomy) within the larger landscape of general international law: is the former part of public international law? Or is it akin to the domestic law of a state? To be sure, international organizations tend to defend the autonomy of their own legal system, as in the case of the European Union,Footnote 55 or the Andean Community.Footnote 56 In contrast, international adjudicators, such as investment tribunals, have been adamant that the internal law of international organizations is not autonomous but part of general international law.Footnote 57 When facing that very same question in order to define the relevant law applicable to responsibility emerging from breaches of internal law, the International Law Commission found it difficult to give a clear answer.Footnote 58 But this is understandable – it really could not, because the ultimate criteria to define the relevant law (that of autonomy) operates as a background norm.
The second role of law in international organization interactions derives from that legally constituted idea of autonomy and the distribution of bargaining power it entails. International law creates the ‘foreground’ institutional law; that is, it provides the vocabulary for states to create particular organizations and provides the doctrinal tools to define how that particular entity will act in the world – for example, by establishing rules applicable to delegated powers, legal personality, funding, internal structure and internal decision-making powers, among others.
Such a delegation of powers around a particular formal function, though, is only the beginning. As we have seen, international organizations are semi-open entities that need to adapt to their environment (even if their law fails to recognize such a wider environment). In that process, it is likely that their actual practice is decoupled from the formal goals and powers as established, for example, through its constituent document. This process of decoupling does not register in a narrowly functionalist reading of the organization, but it still takes place, through legal means: organ competences are expanded, implied powers are found, international legal personalities are presumed.Footnote 59 Overall, the organization takes advantage of the flexibilities provided by the open-ended texture of international law to adapt to its environment by shifting the shape of its practices, while keeping its overall formal function untouched. Again, this is not a problem of non-compliance with the internal law of the organization but rather a process of adaptation that is performed through legal interpretation.
Interaction among international organizations occurs in this space. Part of the stimuli that enters the semi-open organization is the friction with other international organizations, which is also dealt with through law. Once again, explicit delegation of powers is an important starting point: some international organizations have explicit provisions concerning powers to engage in external relations, as in the case of ASEAN, which has explicit treaty-making powers,Footnote 60 or the East African Community, which has the explicit power to ‘foster co-operative arrangements with other regional and international organisations whose activities have a bearing on the objectives of the Community’.Footnote 61 More often than not, though, the friction generated by interactions with other international organizations needs to be internally managed by referring to the flexibility offered by the internal law of the organization, for example, by finding the implicit power to engage in external relations, as in the case of the European Union.Footnote 62
Focusing on the organization as created by law, but existing beyond it, allows us to see law operate at both these levels (constitutive and operational). International institutional law provides not only the background norm of autonomy but also both the formal function that defines the organization and the indeterminate vocabulary and flexible mechanisms to perform the decoupling of practice from form, which facilitates the organization’s resilience and ultimate survival.
3.5 Conclusion
Interaction is the default status of international organizations. It might trigger inconsistencies, undermine state obligations under international law, hinder accountability, and create composite forms of inter-interinstitutional governance that may disproportionately affect states in the Global South. It seems only reasonable that such a common occurrence would have been long considered as part and parcel of mainstream international institutional law. That it has not reveals much about our current thinking about the relation between international organizations and international institutional law as different but intimately linked manifestations of power. International institutional law provides the background vocabulary for the very existence of international organizations as autonomous entities and distributes bargaining power along the different actors involved in their creation. Besides playing this background role, international organizations law also provides a large portfolio of legal doctrines that facilitates the decoupling of form and practice in international organizations – a safety valve that gives international organizations resilience, allowing them to adapt their internal functioning to external pressures, including the paramount pressure of interaction among international organizations.