Unity or Fragmentation in the Deep Blue: Choices in Institutional Design for Marine Biological Diversity in Areas Beyond National Jurisdiction

Delegations are in the final stages of negotiating the proposed Agreement under the United Nations Convention on the Law of the Sea on the conservation and sustainable use of marine biological diversity of areas beyond national jurisdiction (BBNJ Agreement or Agreement). The Agreement will have tremendous scope. Geographically it covers all ocean areas beyond national jurisdiction, meaning approximately 60 percent of the earth’s surface. Substantively it deals with a range of complex topics necessary for the conservation and sustainable use of marine biological diversity of areas beyond national jurisdiction, including marine genetic resources, sharing of benefits, measures such as area-based management tools, including marine protected areas, environmental impact assessments and capacity-building and the transfer of marine technology. Existing scholarship primarily explores the substantive choices for the Agreement; little examines its proposed institutional structure. This article critically assesses the competing positions advanced during negotiations for the Agreement’s institutional structure – the ‘global’ and ‘regional’ positions – and reviews the middle, or ‘compromise’ position adopted by the draft text. It suggests that both global and regional actors will be necessary to conserve and sustainably use marine biological diversity of areas beyond national jurisdiction, and that some form of coordinating mechanism is required to allocate responsibility for particular tasks. Two principles are proposed for use in combination to provide a mechanism to help coordinate Agreement organs (global) and regional or sectoral bodies, namely, the principles of subsidiarity and cooperation. These principles are found in existing international and regional structures but are advanced here in dynamic forms, allowing for temporary or quasi-permanent allocation of competences, which can change or evolve over time. This position is also grounded in the international law of treaties and furthers dynamic views of regional and global ocean governance by offering practical coordinating principles that work with the existing Agreement text.


INTRODUCTION
Nearly two decades of planning and negotiations on an Agreement under the United Nations Convention on the Law of the Sea on the conservation and sustainable use of marine biological diversity of areas beyond national jurisdiction (Agreement) appear to be nearing a conclusion. The fourth intergovernmental conference (IGC) to negotiate the text of the Agreement has been postponed twice because of the COVID-19 pandemic but is likely to be convened before the end of 2022. The upcoming IGC may conclude the process.
The proposed Agreement (United Nations [UN], 2020a) deals with a novel problem -how to conserve and sustainably use marine biological resources existing in areas beyond national jurisdiction (ABNJ). ABNJ include marine areas falling outside of coastal states' exclusive economic zones, archipelagic waters and continental shelves, and are described in the law of the sea as the high seas (the water column) and Area (the deep seabed) (United Nations [UN], 2017, p. 4). ABNJ are estimated to cover about 60 per cent of the Earth's surface and water depths in ABNJ are on average more than 4 km in depth with a maximum depth of over 10 km; ABNJ form part of a single, interconnected world ocean (United Nations [UN], 2017, p. 1). ABNJ also demonstrate ecological connectivity and ocean circulation connectivity with coastal state waters (Popova et al., 2019).
Around 95% of the habitat occupied by life on Earth in all its forms is in ABNJ, yet much less than one millionth of the water column and seabed has been studied in detail (United Nations [UN], 2017, p. 1). International governance bodies collectively cover virtually all ABNJ, with some overlapping in geographic scope (PEW, 2016, p. 3). However very few bodies communicate or coordinate with one another, and many are concerned solely with a particular sector -e.g., a single category of fisheries -and are not competent to employ an ecosystem-based management approach in their coverage area. Only six organizations are capable of adopting binding ecosystem management measures in ABNJ: the Commission for the Conservation of Antarctic Marine Living Resources, the General Fisheries Commission for the Mediterranean, the Mediterranean Action Plan for the Barcelona Convention, the OSPAR Commission, the South East Atlantic Fisheries Organization and the South Pacific Regional Fisheries Management Organization (PEW, 2016, p. 6). However even these organizations are limited to particular sectors and their combined geographical reach does not cover all ABNJ. Moreover, as highlighted by the PEW Charitable Trust, '[a]lthough there are nearly 20 high seas governance organizations, none has a comprehensive cross-sectoral mandate with regulatory authority and a focus on conservation in areas beyond national jurisdiction' (PEW, 2016, p. 7).
Marine biodiversity in ABNJ is under threat from the cumulative effects of fishing, shipping and other sectors (International Union for Conservation of Nature [IUCN], 2019). Globally, marine fishery resources are in decline. Fish stocks within biologically sustainable levels decreased from 90% in 1974 to 65.8% in 2017, with 59.6% classified as being maximally sustainably fished stocks; the percentage of stocks fished at biologically unsustainable levels increased from 10% in 1974 to 34.2% in 2017 (Food and Agriculture Organization [FAO], 2020, p. 7). Deep sea living resources are especially vulnerable, with many species having low productivity and recovery rates that are long and not assured (Food and Agriculture Organization [FAO], 2020, p. 144). As briefly summarized by the IUCN, two-thirds of fish stocks in ABNJ are unsustainably fished, biodiversity is affected by ship generated noise, toxic spills, and marine debris, and deep seabed mining may destroy habitats. Climate change is aggravating existing stressors and causing mass movements of species (International Union for Conservation of Nature [IUCN], 2019, p. 2; Intergovernmental Panel on Climate Change [IPCC], 2021).
The complexity of this problem requires a multifaceted solution. The current version of the Agreement is divided into ten main parts: • general provisions (Part I, including definitions, principles, and scope), • sharing of marine genetic resources and their benefits (Part II), • area based management tools (Part III, including marine protected areas), • environmental impact assessments (Part IV), • capacity-building and transfer of marine technology (Part V), • institutional arrangements (Part VI), • financial resources and funding (Part VII), • implementation and compliance (Part VIII), • settlement of disputes (Part IX), • non-parties (Part X), • good faith and abuse of rights (Part XI), and • final provisions (Part XII).
The draft Agreement also contains two Annexes, one dealing with indicative criteria for the identification of areas requiring protection through the establishment of area-based management tools, including marine protected areas (Annex I), and the other dealing with types of capacity-building and transfer of marine technology (Annex II).
A whole range of fascinating issues arise under the Agreement, especially related to the complex topics of access and benefit sharing for marine genetic resources (MGRs). MGRs can be derived from a wide range of material and put to many different uses, some of which may be commercially lucrative. As noted in the United Nations Technical Abstract of the First Global Integrated Marine Assessment on the Conservation and Sustainable Use of Marine Biological Diversity of Areas Beyond National Jurisdiction: 'MGRs can be drawn from all levels of biota in the ocean, from bacteria to fish, and have potential importance to the economics and sustainability of many sectors, including the pharmaceutical industry (new medicines), cosmetics, the emerging nutraceutical industry, aquaculture (new high-value high nutrition, healthy foods) and biomedicine' (United Nations [UN], 2017, p. 17). Recent scholarship has analyzed the proposed MGR access and benefit sharing regime under the draft Agreement (Rogers et al., 2021), and a 'tiered' model for access and benefit sharing has been advanced (Humphries et al., 2020). 1 An issue which has been subject to less analysis, but that goes to the heart of the Agreement, is the question of the institutional structure necessary to implement its provisions. This article commences with a brief historical summary of the origins of the Agreement, and then examines competing visions underlying the institutional proposals by some of the negotiating participants. In the third section the article explores the 'compromise' position adopted by the current text and fleshes out its implications. The fourth section offers two principles as mechanisms to use in balancing global and regional governance mechanisms, namely, subsidiarity and cooperation, before turning to preliminary conclusions.

PROGRESS OF BIODIVERSITY BEYOND NATIONAL JURISDICTION NEGOTIATIONS 2
Following successful conclusion of the 1982 UN Convention on the Law of the Sea (United Nations [UN], 1982) and two implementing agreements -the 1994 Agreement relating to Part XI of (United Nations [UN], 1994) and the 1995 Fish Stocks Agreement (United Nations [UN], 1995) -states turned their attention to gaps in the existing framework of international treaties and arrangements that regulated marine resources, including resources in ABNJ. These gaps are significant (Gjerde et al., 2008;Greiber and Druel, 2014;Warner, 2015, pp. 758-764;Friedman, 2019, p. 453). As highlighted by the PEW Charitable Trusts, the patchwork of international bodies and treaties managing ocean resources in ABNJ may have overlapping jurisdiction, but 'virtually no mechanisms exist to coordinate across geographic areas and sectors.' The piecemeal approach that results degrades the marine environment and its resources and makes use of management and conservation tools such as marine protected areas (MPAs) very challenging (PEW, 2016, p. 1). In addition the geopolitically defined rights and jurisdictions established under UNCLOS -dividing competences between distance-measured zones -cannot 'recognize or take into consideration the extent of marine ecosystems and the connectivity between biological habitats and species' (Popova et al., 2019, p. 98).
To address such concerns, and the urgent need to protect marine ecosystems (Goodwin, 2015), in 2004 the UN General Assembly established an Ad Hoc Open-ended Informal Working Group to study issues relating to the conservation and sustainable use of marine biological diversity beyond areas of national jurisdiction (United Nations General Assembly [UNGA], 2004). 1 Note, however, that caution has been urged in relation to the potential for monetary benefits from such a regime: Leary (2018Leary ( , 2019. 2 The following historical overview relies upon the summary provided in DOALOS (2017).
The Working Group met eleven times and in its final report recommended, inter alia, the establishment of a Preparatory Committee to develop a legally binding instrument for the conservation and sustainable use of marine biological diversity of ABNJ (DOALOS, 2017, pp. 2-7).
This Preparatory Committee was established by UN General Assembly Resolution 69/292 of 19 June 2015, which set out the elements to be covered in the Agreement. The UN General Assembly decided in paragraph 2 that 'negotiations shall address the topics identified in the package agreed in 2011, namely the conservation and sustainable use of marine biological diversity of areas beyond national jurisdiction, in particular, together and as a whole, MGRs, including questions on the sharing of benefits, measures such as area-based management tools, including marine protected areas, environmental impact assessments and capacity-building and the transfer of marine technology' (United Nations General Assembly [UNGA], 2015).
Other bodies in the UN, including the General Assembly, addressed issues related to the Agreement (for a summary, see DOALOS, 2017, pp. 8-9

TWO ALTERNATIVE VISIONS
To understand its proposed institutional structure, it is important to understand the Agreement's purpose and the balance which it must strike between competing interests. As noted by Payne, the Agreement must balance (1) the freedoms of the high seas, (2) the common heritage of humankind, and (3) the obligation of states to protect the marine environment (Payne, 2018, p. 120).
During the negotiations at the first, second and third IGCs different views about the nature and purpose of the Agreement and its place within the larger system of oceans governance architecture led to a range of positions regarding the institutional structure necessary to support its goals. The most distant positions regarding institutional structure may be divided into two opposing models, namely, the 'global' and 'regional' models. A third, 'hybrid' model, was also advocated, and is discussed below in relation to the current draft Agreement. 3 It should be noted, however, that some negotiating teams did not advocate positions falling squarely in either model. Some states accepted portions of both models. The Pacific Small Island Developing States, for example, advocated a strong, global-level decision making body while at the same time arguing for a strong regional component and a critical role for regional bodies in implementing the Agreement (Gjerde et al., 2019, p. 38).

New, Global Model
The first, global model prioritizes universal values. Its advocates argue for the creation of a powerful new international organization. Labels used to describe such a model included 'global' or 'heavy' regimes (see e.g., Young and Friedman, 2018, p. 127), or 'Ocean Governance Authority' (see Freestone, 2018). Proponents of this approach seek to create a strong, permanent, robust international organization with organs, binding decisionmaking powers, delineated competences, financial security, and enforcement capabilities (see e.g., High Seas Alliance [HSA], 2018). Gjerde, Clark, and Harden-Davies describe the global model as entailing the creation of a new body which could actively and directly implement the Agreement including, at a minimum, a decision-making body, scientific/technical body, and secretariat. The Agreement institutions would take binding decisions and should be able to establish MPAs and EIAs in ABNJ (Gjerde et al., 2019, p. 37).
Such a robust structure would ensure the full and comprehensive implementation of the provisions of the Agreement, in a context where the fragile natures of the creatures and ecosystems in the areas beyond national jurisdiction are under threat and there is a need to prevent irreparable harm. Others advocating a full and robust new international organization include a compliance organ  see also Gjerde et al., 2018, p. 9; High Seas Alliance [HSA], 2020).
More broadly, some proponents have suggested that the legal status of ABNJ requires a different system of global governance. They argue that the concept of the 'common heritage of humankind' should be applied to ABNJ. This principle is found in Article 5 of the draft Agreement. In the Preamble to the United Nations Convention on the Law of the Sea (UNCLOS) the seabed, ocean floor, subsoil and resources are described as 'the common heritage of mankind, the exploration and exploitation of which shall be carried out for the benefit of mankind as a whole, irrespective of the geographical location of States' (United Nations [UN], 1982). Article 136 of UNCLOS, provides that 'The Area and its resources are the common heritage of mankind.' The fundamental nature of this principle is demonstrated by UNCLOS Article 311(6), which does not allow amendments to the principle and prohibits parties from joining agreements which derogate from it. 4 Applied to the current Agreement, those advocating a global institutional model have argued that the resources of ABNJ must be subject to the common heritage of humankind principle in the same manner as the Area, which is governed by the International Seabed Authority (ISA) (see e.g., Payne, 2019, pp. 206-210). Such views were expressed at the end of the 3rd IGC: 'During the closing session on Friday, 30 August, the African Union highlighted the common heritage of humankind as a cornerstone in the drawing up of the BBNJ agreement' (International Institute for Sustainable Development [IISD], 2019a, pp. 21-22). In the view of proponents, the regulation of ABNJ must be global/universal, not sectoral, and the common heritage of humankind should extend to the living resources of the Area (Freestone, 2016, p. 241). In a similar vein, others have suggested removing all references to high seas freedoms from the text (International Institute for Sustainable Development [IISD], 2019b, p. 15).
Interestingly, a variant of the global view which arose in negotiations advocates using an existing global oceans organization to fulfill all aspects of the Agreement. This would obviate the need to create a new body and should therefore save on costs.
The ISA has been proposed as a suitable organization, on the condition that its mandate be expanded to cover all of the topics included in the Agreement (see e.g., Mossop, 2015, p. 839). Proponents of this view suggest that the principle underpinning the status of the Area -the common heritage of humankind -is already fully appreciated by the ISA and should be extended from the deep seabed to the superjacent water column and its living resources. Currently the ISA does not have direct jurisdiction over the MGRs in ABNJ, since the definition of the Area's 'resources' excludes living resources (Lodge, 2015, p. 230). This is because the ISA's primary purpose was to regulate seabed mining, itself a sensitive topic. The ISA also does not have authority over marine scientific research in the Area (Stephens and Rothwell, 2015, p. 573). However, the ISA does have a duty to protect the marine environment from mining activities, including to adopt rules, regulations and procedures for 'the protection and conservation of the natural resources of the Area and the prevention of damage to the flora and fauna of the marine environment' (United Nations [UN], 1982, Art 145(b)).
Similarly, it is required to apply, alongside sponsoring states, best environmental practices and the precautionary approach to activities in the Area (International Tribunal for  Proponents argue that the ISA should be reshaped to become the regulatory body for all resources beyond national jurisdiction -living and non-living, in the subsoil, on the seabed and in the water column. The extension of ISA authority to the water column would be a matter of critical importance, both because of the lack of a coordinated regulatory framework governing it, and because of the potential high value of MGRs in the water column (see e.g., Mossop, 2015, p. 838). It should be noted in this vein that the ISA already has taken small steps in this direction. It has issued recommendations for contractors related to the environmental impact of their activities, including on environmental impact assessments (International Seabed Authority, Legal and Technical Commission [ISA], 2020), and has created a regional environmental management plan for the Clarion-Clipperton Zone (International Seabed Authority [ISA], 2012, 2021).
Nevertheless not all states may accept a strong Agreement role for the ISA. States which are not parties to UNCLOS (including the United States), and which therefore are not represented on the ISA organs, would be unlikely to accept the ISA serving as the institutional structure for the Agreement (Payne, 2019, p. 211). Questions have been raised about whether the ISA could be re-tooled to cover all of the matters addressed in the BBNJ Agreement. Concerns also have been expressed about the way it currently functions, including an alleged lack of transparency in the ISA's decision-making (Ardron, 2018).

Existing, Regional Model
The opposing model, falling at the other end of the spectrum, potentially avoids the creation of any new organs. Proponents suggest that existing regional and international organizations, including regional fisheries management organizations (RFMOs), can be used to adequately oversee and implement the Agreement. Labels used to describe such a model include 'regional' or 'light' regimes (Young and Friedman, 2018, p. 127). Gjerde, Clark, and Harden-Davies have described this model as requiring a minimal institutional structure since the Agreement would rely almost entirely upon regional and global institutions and sectoral bodies. Parties could set out broad guidance and perhaps meet to report on progress, but otherwise would continue with the status quo (Gjerde et al., 2019, pp. 36-37).
The underlying motivations for this position include the high costs of setting up a new, full-fledged organization, and concerns about such an organization's potential to intrude upon the competences of existing regional and sectoral bodies, as well as to infringe high seas freedoms, including the freedom to fish. 5 The question of the applicability of the Agreement to fish arose during the 3rd IGC negotiations on the topic of marine scientific research and MGRs. A distinction was made between fish as a commodity (subject to the high seas freedom to fish) and fish as a MGR (falling under the Agreement) (International Institute for Sustainable Development [IISD], 2019a, pp. 6-7 and 22). It is highly unlikely that a version of the Agreement which leaves implementation primarily to existing regional RFMOs and other sectoral organizations will gain significant support, especially given their difficulties in implementing MPAs in ABNJ (Freestone, 2018). Recent research demonstrates strong levels of interconnectivity between some areas within national jurisdiction and ABNJ, and this underlines the need for high levels of coordination beyond RFMOs and sectoral organizations (Popova et al., 2019), coordination which does not exist at present. 6 A 'regional' model was used by the UN Fish Stocks Agreement which, instead of creating new organs, imposed direct obligations on states and required use of regional organizations. These obligations included, under Article 8(1), the requirement that a party cooperate, either directly or through an appropriate subregional or regional fisheries management organization (RFMO) and, under Article 8(3), the duty to join an existing organization if that party was not already a member (United Nations [UN], 1995). Where no such organization exists, Article 8(5) requires parties to create one (United Nations [UN], 1995). Such a Fish Stocks Agreement framework would not likely satisfy the needs of the BBNJ Agreement. The Fish Stocks Agreement is limited in scope to straddling and highly migratory fish stocks (United Nations [UN], 1995, Art 2); it does not cover the complex and wide-ranging issues raised by the BBNJ Agreement. Moreover, the nearly two-decade long process leading to the upcoming IGC reveals widespread consensus that current RFMOs and sectoral arrangements are inadequate to support conservation and sustainable use of marine biological diversity, including in ABNJ (see e.g., Rayfuse, 2015).

Role for Regional Organizations
Nevertheless, RFMOs and other existing organizations remain important to the Agreement. This is recognized in the Agreement itself. Draft Article 6(1) expressly provides that 'States Parties shall cooperate under this Agreement for the conservation and sustainable use of marine biological diversity of areas beyond national jurisdiction, including through strengthening and enhancing cooperation with and among relevant legal instruments and frameworks and relevant global, regional, subregional, and sectoral bodies and members thereof in the achievement of the objective of this Agreement.' Cooperation through regional organizations is also recognized in Article 197 of UNCLOS: 'States shall cooperate on a global basis and, as appropriate, on a regional basis, directly or through competent international organizations, in formulating and elaborating international rules, standards and recommended practices and procedures consistent with this Convention, for the protection and preservation of the marine environment, taking into account characteristic regional features' (United Nations [UN], 1982).
The involvement of regional and sectoral bodies may prove crucial to the effective operation of the Agreement, if those bodies can be encouraged to cooperate and coordinate their activities (Gjerde et al., 2018). Regional and sectoral bodies also can help customize global norms to specific geographical areas (Gjerde and Yadav, 2021). Organizing such bodies will prove challenging. However, the example of the Sargasso Sea Commission shows that some progress can be made by relying upon several entities to fulfill different parts of an overall mandate (Freestone, 2016(Freestone, , 2021.

Limitations of the Not Undermining Clause?
Others have suggested that even if RFMOs are not granted a strong, formal role in the BBNJ Agreement -for example, if they are deemed to be excluded from Agreement control by the phrase 'not undermining' in Article 4(3) 7 -the Agreement organs still may wish to enter into separate MOUs with RFMOs. An example of such an MOU arrangement is that existing between OSPAR and NEAFC, which has allowed the creation of MPAs in the North-East Atlantic (Haas et al., 2021, p. 229). MOUs could strengthen 'the ecosystem-based approach for fisheries management, or [provide] a common and consistent framework for states to follow' (Haas et al., 2021, p. 229). Such a model, as suggested by De Lucia, would envisage the BBNJ Agreement organs as helping to coordinate regional and sectoral bodies to ensure holistic ecosystem-based coverage (De Lucia, 2019, p. 7). Thus, the emphasis under an MOU arrangement would be on structured consultation, cooperation and coordination, rather than a hierarchical, top-down order (Friedman, 2019, pp. 454-455). The BBNJ Conference of the Parties (COP) could take binding decisions, but these would only bind parties, not regional RFMOs or other bodies. Nevertheless, parties could implement these decisions with respect to their own nationals (flag ships) or work through regional or sectoral bodies.
The Agreement organs also could work directly with regional or sectoral bodies. Since the Agreement contains the 'not undermining' principle, the COP and other Agreement organs would need to consult and cooperate with RFMOs and other bodies in order to fulfill the goals of BBNJ decisions (De Lucia, 2019;Friedman, 2019, pp. 454-455;Haas et al., 2021, p. 229).
Interestingly, as pointed out by Freestone, there would be no conflict between the Agreement's COP and an RFMO if the latter was unable to support a COP decision, for example, in relation to an MPA. Such a situation could arise where only some of the members of the RFMO are parties to the Agreement. In such a case the COP could designate an area as an MPA. The parties to the Agreement, many of whom also will be members of the RFMO, would be obliged to respect it. They could respect the MPA through their own actions and, perhaps, by advocating for the adoption of compatible policies in the RFMO. The MPA could be implemented under the aegis of the Agreement organs (if they have competence to enforce it), under the RFMO (again, if it has competence), or directly (between state parties to the Agreement) (cf Gjerde et al., 2019, p. 30). Parties to the Agreement would not be required to fish in the MPA, since no RFMO 'obliges its parties to fish in certain areas' (Freestone, 2018, p. 130). Interestingly, this issue would gradually disappear as more members of the RFMO became parties to the Agreement; when all became parties, then the RFMO itself could be used to help members comply with their Agreement obligations.

Pacta Sunt Servanda
So long as the Agreement creates a binding obligation, then parties are required to fulfill it in good faith. This rule is fundamental under the law of treaties and is codified in Article 26 of the Vienna Convention on the Law of Treaties (United Nations [UN], 1969). Article 26 provides that '[e]very treaty in force is binding upon the parties to it and must be performed by them in good faith.' Article 26 is supported by the rule that parties cannot invoke the inadequacy of their own laws and procedures to escape liability for breach of a treaty obligation. This is codified in Article 27 which provides that '[a] party may not invoke the provisions of its internal law as justification for its failure to perform a treaty.' Applied to the present context, once the Agreement enters into force parties are obliged to uphold its binding norms and binding decisions in good faith. Thus, for example, if the COP or another organ creates a legally binding MPA, parties will be required to implement it through their own actions. They would not be allowed to take an inconsistent action before a regional RFMO without being in breach of the Agreement.
The binding nature of the Agreement and the decisions of its organs will be necessary, but not sufficient, to ensure its effectiveness. The Agreement organs and RFMOs/sectoral bodies must work together, with regional organizations being able to aid in the implementation of Agreement decisions. This is why a positive duty to cooperate in the text of the Agreement, such as that found in Article 6(1), is so important.
A duty to cooperate through international organizations to conserve marine biological diversity is found in Article 5 of the Convention on Biological Diversity: 'Each Contracting Party shall, as far as possible and as appropriate, cooperate with other Contracting Parties, directly or, where appropriate, through competent international organizations, in respect of areas beyond national jurisdiction and on other matters of mutual interest, for the conservation and sustainable use of biological diversity' (United Nations [UN], 1992a).
Cooperation, however, does not mean delegation. Simple delegation of functions to sectoral bodies would not only reproduce the inefficiencies and incoherence of the current system; it also may give rise to the potential for interference in Agreement-related measures by powerful states and other actors. As noted by De Santo et al. (2019, p. 4), some states have a vested interest in using existing sectoral bodies for the BBNJ Agreement project because of the benefits they derive from the current system, or the extensive powers they exercise in existing RFMOs.
As a result, mutual support and cooperation are necessary between Agreement organs and regional bodies, not simple delegation.

THE DRAFT BIODIVERSITY BEYOND NATIONAL JURISDICTION AGREEMENT 'COMPROMISE'
The 'compromise' position adopted by the current draft of the Agreement is similar to that found in recent multilateral environmental agreements (MEAs), namely, an institutional structure comprising a conference of the parties, a scientific and technical body, a clearing-house mechanism, and a secretariat. The Agreement's institutional position falls in the middle of the spectrum between the global and regional models discussed earlier. It could be described as a 'hybrid' model, one which moves beyond the status quo but falls short of a centralized and powerful global body (Gjerde et al., 2019, pp. 37-38). Nevertheless, since the word 'hybrid' has been used in many different ways by negotiators of the BBNJ Agreement, it would be inaccurate to label 'hybrid' as a 'model.' Rather, the term describes the range of positions falling in between the two extremes of the 'global' and 'regional' models (Clark, 2020, p. 3).
Part VI of the Agreement sets out its institutional arrangements, including the creation of four organs: a Conference of the Parties (COP) (Article 48), a Scientific and Technical Body (STB) (Article 49), a Secretariat (Article 50), and a Clearing-house Mechanism (CHM) (Article 51) (United Nations [UN], 2020a).
These four articles are riddled with square brackets and hence several issues remain unresolved. 8 In addition, a number of textual proposals have been made seeking to modify the draft Agreement, as submitted by delegations and compiled and grouped article-by-article by the President of the IGC (United Nations [UN], 2020b). The proposals concerning the institutional section of the Agreement in the latter document generally tend to suggest text for deletion, rather than propose new, alternative drafts.
Interestingly, the different options set out in the draft Agreement's square bracketed text have the potential to move the Agreement closer to either end of the global-regional spectrum. In other words, the assumptions underpinning, and consequences of adoption of, some of the square bracketed text could produce ripple effects across other provisions of the Agreement. These could push the Agreement along the spectrum, depending upon whether the bracketed text is, or is not, accepted.

Conference of the Parties (Article 48)
The first organ of the Agreement,

Decision-Making
In terms of voting procedures for the COP, two possibilities arose during the negotiations. The first is that the COP must take all decisions by consensus, with no other voting option being contemplated. If this position was to succeed then, in the absence of consensus, a proposed decision would fail. The second is that the COP must attempt to achieve consensus, but if this proves impossible it is empowered to decide in another manner (majority voting).
Both views were expressed by negotiating states during the second IGC (International Institute for Sustainable Development [IISD], 2019b, p. 7). Strong support for the latter position is found in High Seas Alliance's (HSA) proposals on institutional arrangements and cross cutting issues. In its 2018 Briefing Paper the HSA advocates for majority decision-making as a fallback for failure to achieve consensus, with dissenting members not being able to block a decision (High Seas Alliance [HSA], 2018, p. 2). In its 2020 recommendations the HSA modified its position to that of a two thirds majority vote, without referring to any preliminary attempt at consensus (High Seas Alliance [HSA], 2020, p. 2).
In response to the possibility of the lack of a voting procedure being established in the Agreement for the COP 9 , the IUCN has pointed out that 'it is highly unusual not to have a decision making process set out in the treaty' (International Union for Conservation of Nature [IUCN], 2020, p. 71). They provide examples of treaties where decision-making processes are spelled out. For example, Article 16 of the Convention on the Conservation and Management of High Seas Fishery Resources in the South Pacific Ocean expressly provides for both consensus and majority decision-making systems (South Pacific Regional Fisheries Management Organisation [SPRFMO], 2009). Another model which includes consensus and majority voting is found in the first UNCLOS implementing agreement, the Agreement relating to Part XI of UNCLOS (United Nations [UN], 1994). Section 3, subsection 2, provides for consensus as the general rule for decision-making for both the Assembly and Council; subsections 3 and 5, however, allow for decisions by either simple majority or a two-thirds majority. 10 At the 3rd IGC general support was shown for 'clearly outlining the possible decision-making functions of the COP, ' although divergent views remained about 'whether decision making should be only by consensus, or a fallback mechanism should be contemplated in cases where consensus cannot be reached' (International Institute for Sustainable Development [IISD], 2019a, p. 10).
The text under negotiation in Article 48(3bis) of the Agreement does not preclude the majority voting fallback position; nor, however, does it definitively provide for it: '[3bis. As a general rule, the decisions of the Conference of the Parties shall be taken by consensus. If all efforts to reach consensus have been exhausted, the procedure established in the rules of procedure adopted by the Conference shall apply.]' This wording leaves it up to the COP to negotiate an alternative voting format in its rules of procedure. Note that since the only specified rule in the text is that of consensus, if Article 48(3bis) is adopted in its current form then the COP will be required to adopt further voting rules by consensus. This may prove challenging.

Functions of Conference of the Parties
In terms of functions of the COP, its critical role is that of monitoring and ensuring implementation of the Agreement. Several unbracketed subsections of Article 48 suggest potential agreement on text, including those on the COP's ability to adopt decisions and recommendations, to exchange information, to establish subsidiary bodies, to adopt a budget, and to undertake other

Coordination With Other Bodies
In terms of the COP's function in promoting cooperation and coordination with other bodies, there remains some disagreement. Article 48(4)(c) mandates the COP to '[p]romote cooperation and coordination with and among relevant legal instruments and frameworks and relevant global, regional, subregional and sectoral bodies, with a view to promoting coherence among efforts toward, and the harmonization of relevant policies and measures for, the conservation and sustainable use of marine biological diversity of areas beyond national jurisdiction. . ..' However, the bracketed text that immediately follows reveals concern by some states about whether the COP should itself establish processes, or invite others to establish processes, for cooperation. Nevertheless, even if both sets of bracketed text are excluded, the quoted text is significant in that it sees the COP as being required to play an active role in coordinating measures related to marine biological diversity in ABNJ, by promoting coherence and harmonization of such measures.
The fragmented and patchwork nature of global ocean governance in ABNJ (see Mahon et al., 2015), makes this role of the COP very important. Depending upon the strength of Agreement organs, the function of promoting cooperation and coordination 'among' relevant legal instruments and frameworks will be critical. In this vein, Mahon et al. highlight literature that suggests that cooperation and coordination can be enhanced by 'systematic promotion of inter-institutional learning' (Mahon et al., 2015, p. 27). The need to coordinate the national, regional and global ocean governance systems to protect marine biological diversity in ABNJ is highlighted by the same authors in their conclusion that '[t]he entire set of governance arrangements for ABNJ and areas within national jurisdiction may be best approached as a single global ocean governance structure' (Mahon et al., 2015, p. 64).

Monitoring and Compliance
A final key role proposed for the COP remained unresolved at the end of the 3rd IGC, namely, its function in supporting monitoring and compliance. Bracketed Article 53(3) suggests that the COP could 'address cases of non-compliance' with the Agreement. However one state is reported to have requested the deletion of this latter phrase, another state opposed the compliance mechanism altogether; others expressed support for such a mechanism (International Institute for Sustainable Development [IISD], 2019a, p. 19). As will be suggested below, there is need for a strong monitoring and compliance mechanism in the final version of the Agreement, whether undertaken by the COP or a subsidiary body, especially if the Agreement is to be implemented in part by a range of RFMOs and other bodies. A compliance organ would help to ensure accountability regardless of the level at which an Agreement action was taken -global, regional or national -and enable polycentric decision-making (see e.g., Gjerde and Yadav, 2021).

Scientific and Technical Body (Article 49)
Some disagreement emerged in earlier negotiations on whether the Scientific and Technical Body (STB) should be ad hoc or permanent (International Institute for Sustainable Development  [IISD], 2019b, p. 13), and one state is reported to have opposed the establishment of the STB (International Institute for Sustainable Development [IISD], 2019a, p. 12). The current unbracketed text in Article 49(1), stating that a 'Scientific and Technical Body is hereby established' , offers hope that these issues have been resolved.
There appears to be some agreement that the STB shall be composed of experts, and that it can call upon other scientists or experts as required [Article 49(2)- (3)]. Disagreement remains about certain types of expertise, including expertise in relevant traditional knowledge of indigenous peoples and local communities, and whether additional expertise may be sought from existing arrangements and bodies.
In terms of functions, there is agreement that the STB shall provide scientific and technical advice to the COP and perform other functions as required by the COP [Article 49(4)(a) and (m)]. The importance of science-based decision-making by the COP is clear (cf. Johnston, 2019).
The other potential functions of the STB, which are numerous, are bracketed. These functions would allow the STB to play a strong role in helping parties to implement the various parts of the Agreement in areas ranging from monitoring MGRs, to making recommendations on ABMTs (including standard setting and review), to providing guidelines on, making recommendations to the COP on, and reviewing EIAs, to identifying technology and know-how, to advising on the development and transfer of marine technology, to assessing the effectiveness of measures for capacity-building and transfer of marine technology (including collaboration with regional and subregional committees, and by elaborating programs), and to establishing subsidiary bodies [Article 49(4)(b)-(l)]. In assessing the best available scientific evidence, the STB may rely upon the traditional knowledge of indigenous peoples and local communities (Mulalap et al., 2020). This knowledge is expressly referred to in numerous provisions (albeit mainly in brackets) of the Agreement: Articles 5(i), 10bis, 16(1), 17(4)(c), 18(2)(c), 21(4), 31(2), 32(1), 34(2), 35(3), 46(1)(b), 49(2), 51(4)(b), and 52(5)(e).
If most of the bracketed functions are adopted this would transform the STB from being merely an advisory body to a body tasked with implementing different aspects of the Agreement. These functions could have an impact on the relationship of the STB to other organs, and to the parties. For example, in negotiations regarding the STB's role, different views were expressed about whether it should be responsible for evaluating environmental impact assessments and deciding whether a proposed activity should proceed, or whether such matters should be left to states parties (International Institute for Sustainable Development [IISD], 2019b, pp. 9-10). The STB could play a valuable role in relation to ABMTs, particularly if its recommendations are transparent and science-based, not political. Also, the effect of a recommendation in relation to an ABMT could be weak or strong, depending upon the model chosen. One model, adopted by the International Seabed Authority (ISA), creates a default position under which the executive organ (the Council in the case of the ISA), must approve the recommendation by the Legal and Technical Commission for approval of a plan of work, unless a two-thirds majority votes against it (United Nations [UN], 1994, s 3(11)). 11 If such a position were to be adopted in the Agreement even a recommendation of the STB could have significant impact.
In addition, the role of standard setting in relation to ABMTs could bolster the effectiveness of the Agreement. If, for example, the STB were to adopt a data-driven (Visalli et al., 2020), dynamic and adaptive approach to ABMTs, one that ensured that they could react to changes in ocean conditions and movement of marine biological resources over time (both spatial and temporal), then ABMTs could more effectively deal with long term issues such as climate change (Ortuño Crespo et al., 2020). The inability to create broad, effective ABMTs, including MPAs, is a critical weakness of existing regional and sectoral organizations and this is where the STB, and the Agreement in general, could play an important role (see e.g., Clark, 2020, pp. 4-5).
Interestingly, Article 49 does not specify the decision-making rules of the STB. The opening clause in Article 49(4) -'[u]nder the authority and guidance of the Conference of the Parties'suggests that the rules of procedure of the STB, including voting rules, may be elaborated by the COP. It would be preferable to set out the STB's voting rules directly in the Agreement. Decisionmaking rules for similar bodies can be found in other agreements. For example, Section 3(13) of the Agreement relating to Part XI of UNCLOS provides that '

Secretariat (Article 50)
During negotiations at the 2nd IGC, different options were suggested for the Agreement Secretariat. These included creating a new independent secretariat or requesting the UN Division for Ocean Affairs and the Law of the Sea or the International Seabed Authority to perform its functions (International Institute for Sustainable Development [IISD], 2019b, pp. 13-14). These options remained on the table at the 3rd IGC (International Institute for Sustainable Development [IISD], 2019a, p. 17).
There appears to be some support for a permanent Secretariat (High Seas Alliance [HSA], 2018, p. 3). Such a body could be crafted to the needs of the Agreement, would be independent, and would not be constrained by pre-existing institutional practices and biases. A new secretariat might prove to be more expensive, however, since it could require the creation of an entire infrastructure from scratch.
Use of an existing body as a secretariat would likely be less expensive (depending upon the fees charged and the expenses related to the location of the host body) but may be constraining. A host arrangement might require the Agreement Secretariat to fit within existing institutional rules, processes and even, perhaps, value systems/paradigms of the host. Examples exist of secretariats being hosted by other bodies. The secretariat for the Convention on Biological Diversity and the secretariat of the Vienna Convention on the Protection of the Ozone Layer (taken together with its Montreal Protocol) are both hosted by the UN Environmental Programme (UNEP) (see e.g., Churchill and Ulfstein, 2000, p. 627;Kishore, 2011Kishore, , pp. 1051Kishore, -1082. The UNEP also assists with the more general coordination of a number of treaties (United Nations Environment Programme [UNEP], 2021).
In terms of functions, there appears to be agreement that the Secretariat should provide administrative assistance and support, service meetings, disseminate information, and perform such other functions as determined by the COP [Article 50 (2) As can be appreciated, this list of functions would enable the clearing-house mechanism to serve as a central hub for the Agreement. MGRs and access and benefit sharing (including capacity building) will likely be fundamental to the Agreement (Harden-Davies and Rogers et al., 2021), as will the related data management systems (Rabone et al., 2019), and EIAs (Gjerde et al., 2018). Interestingly, research collaboration may also require capacity building (Tolochko and Vadrot, 2021).
Article 51(4) sets out the CHM's role in facilitating cooperation and information sharing, including by: (a) matching capacity-building needs to available support, (b) promoting linkages with existing clearing-house mechanisms and other databases, repositories and gene banks '[including experts in traditional knowledge]' , (c) linking to private and NGO platforms for information exchange, (d) building on existing clearing-house mechanisms, (e) facilitating enhanced transparency, including by 'providing baseline data and information' , (f) facilitating international collaboration and cooperation, including scientific and technical forms [Article 51(a)-(f)].
Article 51(5), also bracketed, requires the CHM to recognize the special circumstances of small island developing states (and archipelagic developing states), and to facilitate their access to, and use of, the mechanism. The CHM is required to provide specific programs for those states.
In terms of administration of the CHM, several states expressed a preference for it to be managed by the Agreement's Secretariat; others preferred its management by the Intergovernmental Oceanographic Commission of the United Nations Educational, Scientific and Cultural Organization (IOC-UNESCO) Secretariat (International Institute for Sustainable Development [IISD], 2019a, p. 18). The IOC-UNESCO is already developing a new digital ocean ecosystem, the IOC Ocean Data and Information System (ODIS) (IOC-UNESCO, 2021;IODE, 2021).
The square bracketed text in Article 51(6) of the Agreement preserves both options and specifies that the IOC-UNESCO Secretariat should act in association with the ISA and IMO, and 'be informed by the Intergovernmental Oceanographic Commission Criteria and Guidelines on the Transfer of Marine Technology.' This second option specifically envisages the CHM working closely with other international bodies.
Such cooperative arrangements have been recommended by some as a way of making the clearing-house mechanism's information sharing and distribution functions nimbler and more accessible.
The final bracketed provision related to the clearing-house mechanism requires 'due regard' to be given to confidential information under the Agreement [Article 51 (7)].

GLOBAL AND REGIONAL COORDINATION
How, if at all, has the draft Agreement resolved the tensions between the global and regional models discussed earlier? Will the 'not undermining' principle contained in Article 4(3) constrain or strengthen the Agreement?
Some scholars have argued that the 'not undermining' principle should be restricted to not undermining the effectiveness of measures adopted by RFMOs and sectoral bodies (see e.g., Clark, 2020, p. 4). This approach is taken in the 1995 Fish Stocks Agreement (United Nations [UN], 1995), and if similar wording could be adopted in the BBNJ Agreement this would be welcomed.
However, the current wording in Article 4 (3) is more open ended: '[t]his Agreement shall be interpreted and applied in a manner that [respects the competences of and] does not undermine relevant legal instruments and frameworks and relevant global, regional, subregional and sectoral bodies.' If this wording is retained, a mechanism to allocate competences between Agreement organs and regional bodies will be necessary. Perhaps an approach which relies upon a combination of the principles of subsidiarity and cooperation could be used.

Subsidiarity
The principle of subsidiarity may have originated in Roman Catholic law or German constitutional law (Hartley, 2014, p. 123), and has been developed and refined in European Union law (see generally Hartley, 2014, pp. 122-128). The principle is meant to help decide at which level an action should be undertaken -the regional or the local level. Applied to the Agreement context, it could be used to resolve questions about whether an action should be taken by Agreement organs (alone), a range of RFMOs, a single RFMO, or parties (alone or in groups). In other words, using the language of our two models, it would help us determine whether something should be done at the global, regional, or other (i.e., local) level. The principle of cooperation, as discussed below, would enable something to be done simultaneously at multiple levels.
Although the complex legal regime of subsidiarity used by the European Union (EU) cannot -and should not -be transposed to the Agreement, the principle of subsidiarity can be applied. A brief explanation may help put in context the ways in which the principle could be used to allocate (and reallocate) competences between global and regional entities under the Agreement. 12 Subsidiarity under the EU legal system is intertwined with two other principles: conferral and proportionality. As succinctly stated in Article 5(1) of the Treaty on European Union (TEU) '[t]he limits of Union competences are governed by the principle of conferral. The use of Union competences is governed by the principles of subsidiarity and proportionality' (European Union [EU], 2007).
The principle of conferral mandates that EU institutions act only within the competences conferred upon them by the treaties or Community Law (their limits). Competences not conferred upon the EU remain with member states [TEU, Art 4(1)]. The principle of subsidiarity, therefore, only arises where a competence can be used by the EU and does not fall solely within the prerogative of national authorities.
It should be noted in this regard that under the system of EU law once powers are conferred upon EU organs then it is difficult to return them to national authorities. It is possible, via a new treaty or amendment to the existing treaty regime, but generally powers once transferred remain with the EU organs. This part of the EU model -semi-permanent accumulation of powers by one body -is not recommended for adoption for the BBNJ Agreement. The Agreement in its current form does not have such an expansionist design. Rather, it is meant to help achieve a result, the conservation and sustainable use of marine biological diversity of areas beyond national jurisdiction. That result can be achieved by a range of actors, including global or regional bodies, or both acting together, and thus there is no priority in favor of global or regional control. Subsidiarity as proposed here is not unidirectional. Contrary to some suggestions (see e.g., Adewumi, 2021, pp. 2 and 14), as a principle it does not push toward regional or global competence.
The principle of subsidiarity assists in EU law where the EU has the competence to act. The question is whether the EU should act or instead leave the matter for member states. The principle of subsidiarity helps answer this question. Article 5(3) of the TEU sets out a triple test (identified with inserted letters [A]- [C] in the quotation below), as follows: "(3) Under the principle of subsidiarity, [A] in areas which do not fall within its exclusive competence, the Union shall act [B] only if and in so far as the objectives of the proposed action cannot be sufficiently achieved by the Member States, either at central level or at regional and local level, but can rather, [C] by reason of the scale or effects of the proposed action, be better achieved at Union level".
The procedure for subsidiarity is set out in the 'Protocol on the application of the principles of subsidiarity and proportionality' , which is appended to the TEU (see further Hartley, 2014, pp. 124-126). This procedure allows for widespread consultation with national bodies before action is taken at the EU level. Article 5(3) of the TEU demonstrates that so long as the matter can fall within both EU and national competences and is not solely within the domain of the EU (EU exclusive competence, [A]), then the principle of subsidiarity can be used to allow actions to be undertaken at the most local level at which they would treaties, and that proposed under the Agreement, are completely different. As a result, the EU model is not being proposed here; in fact, a very different one is suggested. be effective. In contrast, actions can only be undertaken at the regional level if they cannot be accomplished at the local level.
To provide an example under the BBNJ Agreement, if a proposed measure can be achieved entirely within national legal systems, then it should be implemented nationally; if it imposes an obligation that requires regional action, then it should be implemented regionally; the same logic extends to global obligations. The criteria used to decide whether an issue would be more effectively resolved at a particular level (global, regional, national, or a combination) would be the same criteria employed to assess the measure. In other words, if an MPA is required for an area, which entity or entities could ensure its successful implementation?

Cooperation
The principle of cooperation, or the obligation to cooperate, is found in numerous sections of the Agreement, namely, in Articles 2, 6,12,14,15,20,23,27,28,43,48,51,52,53 . The duty to cooperate for the conservation and sustainable use of biological diversity in ABNJ, either directly or through competent international organizations, is expressly provided in Article 5 of the Convention on Biological Diversity (United Nations [UN], 1992a). The principle of cooperation is found in several ocean-related treaties (Houghton, 2014, pp. 123-124).
The principle of cooperation also is embraced by EU Law. The first paragraph of Article 4(3) of the TEU provides that '[p]ursuant to the principle of sincere cooperation, the Union and the Member States shall, in full mutual respect, assist each other in carrying out tasks which flow from the Treaties.' The obligation is one of sincere cooperation and applies to the relations of the Union and member states. The subsequent paragraph also requires member states to ensure fulfillment of treaty and institutional obligations. The word 'ensure' is meant to make the obligation one of effect.
Applied to the Agreement, if the principle of subsidiarity is combined with that of cooperation, a mutually reinforcing framework for implementation is created. It works at three overlapping levels. Firstly, Agreement organs are required to act where they are better placed to do so, and national or regional bodies cannot do so, or do not exist. These organs might be vested with final competence over an area, or only interim/preliminary competence, exercisable until RFMOs or other sectoral bodies obtain capacity to do so. Secondly, RFMOs of which parties to the Agreement are members (1) would be unable to act contrary to Agreement obligations (without causing a breach by their members who are parties to the Agreement), and (2) may be required to act in support of Agreement obligations if their members are already bound to do so under other treaties, including UNCLOS (see e.g., International Tribunal for the Law of the Sea [ITLOS], 2015, para 218). Thirdly, individual state parties would be required to act to implement Agreement obligations where they can assist. A role for individual states under the Agreement has been proposed under the adjacency principle (Dunn et al., 2017), and individual state action would be required in cases where there are high levels of ecological connectivity between ABNJ and areas within national jurisdiction (Popova et al., 2019).
In other words, a combination of the principles of subsidiarity and cooperation can be used to foster vertical and horizontal coordination in support of the Agreement. Mahon et al. have suggested using a combination of vertical and horizontal arrangements to understand ocean governance by employing a 'single global ocean governance structure as comprising 'globalregional issue-based networks' and 'regional clusters' [to provide] a framework that may help to improve understanding of the very complex, disordered and fragmented set of arrangements for the ocean' (Mahon et al., 2015, p. 64). Gjerde and Yadav have suggested a polycentric model of ocean governance (Gjerde and Yadav, 2021). The current article contributes to this discussion by proposing the use of subsidiarity and cooperation as dual coordinating principles.
This approach may appear novel but is in fact implicit in the binding nature of treaty obligations. As noted earlier, parties are required to fulfill binding treaty obligations in good faith. To the extent that the Agreement creates binding obligations, parties must pursue actions at all available levels to fulfill themlocally, in their national legal systems; regionally, through an RFMO or sectoral body; and internationally, in cooperation with other parties, RFMOs and Agreement organs. The obligation to cooperate becomes one of result. As noted by the ITLOS, the duty to cooperate for the conservation and management of shared resources under UNCLOS applies to 'each and every State Party' and may be implemented individually or through subregional, regional and global organizations; cooperation also may be with both member and non-member states of regional organizations (International Tribunal for the Law of the Sea [ITLOS], 2015, paras 207(i) and 215). Moreover, the obligation to cooperate is a due diligence obligation which must be undertaken in good faith and be meaningful (International Tribunal for the Law of the Sea [ITLOS], 2015, para 210). The Agreement should be interpreted in light of these ITLOS decisions in order to ensure that the principle of cooperation is both robust and legally enforceable.

CONCLUSION
Negotiations on the final shape of the BBNJ Agreement are at a critical juncture and many issues remain to be resolved. For the upcoming IGC to be successful, it will be important for negotiators to understand the implications of the different textual choices in the draft Agreement, and the forces that are pulling states in different directions.
The present article explores the two polar positions put forward for the institutional structure of the Agreementthe 'global' and 'regional' models. The current draft text rests somewhere in the middle of these extremes, but the shape of the Agreement institutions will vary significantly depending upon which set of bracketed provisions is, or is not, adopted.
In addition to identifying preferred textual choices and other ways of strengthening the Agreement, this article suggests a way of moving beyond the 'not undermining' impasse, namely, by invoking a combination of the principles of subsidiarity and cooperation as a way of allocating responsibilities along the global-regional-national spectrum. Subsidiarity ensures the continued relevance of existing regional and sectoral bodies: if they can effectively implement a decision of the COP, they should do so. However, where a decision cannot be effectively implemented by these bodies, or nationally, then the Agreement organs should do so, either as a permanent arrangement, or as a temporary/interim arrangement. In this sense the principle of subsidiarity is not static. As regional and global bodies evolve, competences may shift from one to the other; new regional bodies may emerge to fill gaps, allowing collective, regional action. In addition, the principle of cooperation should be applied to ensure that all relevant actors play their part, including the Agreement organs, regional and sectoral bodies, and parties.
Some have suggested a simpler 'fall back' position to global structures in cases where regional and sectoral bodies will be inadequate (for example, in relation to the creation of MPAs in ABNJ) (e.g., Clark, 2020, p. 5). In such circumstances, the Agreement organs themselves would need to step in to implement or even create the relevant measure (e.g., MPA).
The model proposed here, by adding cooperation, moves beyond the 'fall back' position and rejects the necessity of an 'either/or' approach of exclusive competence. Competence should be shared, with all relevant institutions and states working collectively to ensure effective implementation of Agreement measures. A very wide spectrum of possible cooperative arrangements could be used to achieve the results required by the Agreement, including actions by single states or groups of states, actions by single regional or sectoral bodies or groups of them, collaborative arrangements between regional/sectoral bodies and Agreement organs, or management (perhaps temporary) by Agreement organs.
This approach is in line with the ways in which treaties are implemented in other areas of international life. A state which is a party to a treaty has an obligation as to result. It does not matter whether the state faces challenges because of inadequate domestic legislation or other similar factors -the state is required to uphold the treaty in good faith. To this extent, I have suggested that if the Agreement creates clear and binding norms, both in its text and through decisions of its organs, then its parties will be required to use all means at their disposal to implement them. These include national actions, regional actions, and global actions.
This point is critical. It demonstrates the weakness of an either/or approach to global or regional solutions. A combination may be required, with regional entities supporting the global Agreement organs. For such a system to be effective, a binding Agreement and the capacity to make binding decisions (i.e., by the COP or another organ) are needed. However, there is one caveat. The further the Agreement's institutional model drifts from the 'global' end of the spectrum, the more important it will be for the Agreement to have effective implementation, monitoring, and compliance mechanisms. These mechanisms can be guided by the international jurisprudence related to responsibility, cooperation and due diligence (see e.g., International Tribunal for the Law of the Sea [ITLOS], 2015). The more actors that are involved in implementing the Agreement, the more important it will be to monitor them to ensure coherence and consistency. As a result, an additional institution -an implementation and compliance committee -is recommended for the Agreement (see High Seas Alliance [HSA], 2020).
Difficult, substantive issues remain to be resolved in the upcoming negotiations, including questions regarding MGRs and the sharing of benefits, area-based management techniques, marine protected areas, environmental impact assessments, capacity building and transfer of marine technology, Agreement financing, implementation and compliance, and dispute settlement. Considerable attention will be paid to resolving questions in each of these areas to make the BBNJ Agreement as comprehensive, rigorous, and effective as possible. To be successful, we also must ensure that the Agreement is supported by a strong institutional framework, one that specifically allows for global and regional action, in concert.

AUTHOR CONTRIBUTIONS
DB conducted the research and wrote the manuscript.