1 Introduction

While greenhouse gas (GHG) emissions from international aviation and maritime transport are steadily on the rise, the ambition of multilateral responses was for a long time quite stagnant.Footnote 1 The slow pace of multilateral responses has long been a source of frustration for the European Union (EU), that over the past years has responded with consistent steps towards its own unilateral emission reduction policy.Footnote 2 Matters changed dramatically however, when both the International Civil Aviation Organization (ICAO) and International Maritime Organization (IMO) took considerable strides towards multilateral emission reduction responses. This came to a head in 2016, with the ICAO agreeing on a Carbon Offsetting and Reduction Scheme for International Aviation (CORSIA),Footnote 3 while the IMO adopted a Global Data Collection Scheme (DCS) for international shipping emissions.Footnote 4

Despite its nominal support for these agreements, the EU continues to express concern as to their ambition. Thus at a 2019 climate summit it noted that ‘the current global targets and measures envisaged’ by the IMO and ICAO ‘even if implemented in full, would fall short of the necessary emissions reductions’ for the EU’s goal of decarbonisation by 2050.Footnote 5 In view of this, the Union is consistently refusing to align its standards with those of the ICAO and IMO. Multilateral and unilateral measures are therefore starting to occupy the same regulatory space, with the possibility of further divergence in the future. Indeed, matters are already heating up. At the time of writing, as of 1 January 2021, international offset credits, including those ‘deemed to be eligible’ under CORSIA will not be recognised under the EU Emissions Trading Scheme (ETS).Footnote 6 Conversely, in a resolution of late 2019, the ICAO adopted an ‘exclusivity clause’, ‘determining’ the CORSIA to be ‘the only global market-based measure applying to CO2 emissions from international aviation’.Footnote 7

Against the backdrop of these policy developments, this article will analyse the relationships between these competing climate change responses under public international law. In light of the unique nature of the EU as a supranational actor, Sect. 2 will first set out the theoretical framework for the present analysis of the Union’s collective unilateral action. Section 3 then turns to the concrete elements of the EU’s ‘ultimatum strategy’, used here to describe the EU’s threats and use of unilateral acts as a means of catalysing and steering multilateral action. It will consider the emerging divergencies between multilateral and EU unilateral measures. Section 4 proceeds to analyse the extent to which the current multilateral standards may condition the EU’s unilateral regulatory competence. Notably, it does not conduct a full compatibility analysis of the EU measures with all of the applicable rules. Section 4.1 focuses on the debated relationship between the EU and the Chicago Convention on International Civil Aviation (Chicago Convention) and the CORSIA rules in Annex 16.Footnote 8 Section 4.2 then examines the extent to which the EU may have a duty to align with the IMO DCS in Annex VI of the International Convention for the Prevention of Pollution from Ships (MARPOL).Footnote 9

Supplementing these fields is the customary law of state jurisdiction, which governs state competence to regulate the conduct and consequences of an event.Footnote 10 Section 5 explores these more general jurisdictional rules on the apportioning of regulatory competence. Of particular relevance is the scholarly debate on jurisdictional ‘reasonableness’ and interest-balancing in light of the new position taken in the 2018 Restatement (Fourth) of US Foreign Relations Law.Footnote 11 It also considers the underlying normative question as to the desirable relationship between the ends and the means, particularly in light of the objective in the United Nations Framework Convention on Climate Change (UNFCCC) and Paris Agreement of preventing ‘dangerous’ climate change, by keeping global warming ‘well below’ 2 °C.Footnote 12

2 Theoretical Framework: EU (Collective) Unilateralism from an International Law Perspective

While the EU claims considerable autonomy to adopt unilateral standards, in reality it operates in what Wessel has termed a ‘global normative web’, where ‘many of its positions and decisions are closely connected to policies and decisions of other international bodies’.Footnote 13 The question of the EU’s relationship with the international legal order is then a highly contentious one, a detailed analysis of which goes beyond the scope of this article. At its core, as noted by Ziegler, much turns upon whether one conceives of the EU ‘as a creature of international law’ or rather ‘as sui generis because of its advanced constitutionalisation’.Footnote 14 This article takes a moderate international law perspective, conceiving of the EU as an international organisation, based on multilateral constituent treaties, and endowed with legal personality by virtue of the conferral of sovereignty from its Member States.Footnote 15 At the same time, one cannot ignore the far-reaching supranational competences of the EU institutions, and the highly-evolved and integrated nature of the EU legal order. As such, the relationships between the decisions of the various actors cannot be fully understood without some consideration of the EU law perspective on its position in the international legal order.Footnote 16

The present enquiry is further premised upon a neutral approach to ‘unilateral acts’, as neither lawful or unlawful per se.Footnote 17 These are defined as acts ‘formulated by a State with the intent of producing certain legal effects under international law’.Footnote 18 The core characteristic of a unilateral act is that it is taken without recourse to international procedure.Footnote 19 Such acts are an exercise of state sovereignty, and will only be illegal when there are restrictions in international law.Footnote 20 Notably, the EU is considered to exercise ‘collective unilateralism’, as it is a group of states acting together as a single legal person.Footnote 21 Despite this neutral conceptualisation, as noted by Sands, unilateralism remains a ‘term of art’, ‘because the territorial limits to the exercise of sovereign autonomy remain in a state of flux, and because the standards set by international law remain incomplete in many areas and ambiguous and open-textured in many others’.Footnote 22 As will be seen, this statement made in 2000 could not be more true twenty years later, in the context of the EU climate-protective measures.

Evidently, as noted by the International Court of Justice (ICJ), international organisations ‘are subjects of international law and, as such, are bound by any obligations incumbent upon them under general rules of international law’.Footnote 23 However, determining which obligations are ‘incumbent’ upon the EU is not a straightforward exercise.Footnote 24 The EU is a party to the UNFCCC and the 1998 Kyoto Protocol, Article 2(2) of which designates the ICAO and IMO as the fora within which Annex I Parties to the UNFCCC ‘shall pursue’ the reduction of greenhouse gases from aviation and marine bunker fuels.Footnote 25 At the same time, international transport emissions were excluded from the scope of states’ mandatory reduction commitments, with the 2015 Paris Agreement remaining silent on this issue. Arguments have been made that Article 2(2) of the Kyoto Protocol provides for a ‘transfer of responsibilities’ to these specialised organisations, which—at least once exercised—exclude concurrent unilateral measures.Footnote 26 This contribution follows the opposing view that does not read the Kyoto Protocol as excluding parties’ parallel competence.Footnote 27 Such exclusivity does not appear supported by the text or by the broader context of the agreement which encourages individual reduction policies more generally.

Determining the relationship between the IMO, ICAO and EU measures thus requires further unpacking. As it is not a state, the Union is not a full member of the ICAO and the IMO. The EU is also not a formal party to the 1958 Chicago Convention, with its recent CORSIA amendment to Annex 16. It is further not a party to the MARPOL and its Annex VI, which was amended to contain the IMO DCS. As will be discussed, for its part, EU law does not particularly support systemic coherence, employing a very restrictive interpretation of when the Union may itself be bound by obligations incumbent upon all of its Member States.Footnote 28

What results is a ‘conflict’ of norms in the broader sense, defined here as ‘a situation where two rules or principles suggest different ways of dealing with a problem’.Footnote 29 There is a considerable group affected by the competing regulatory frameworks, as the EU measures seek to regulate conduct and circumstances beyond its territory. This gives rise to issues of state jurisdiction under customary international law. Notably, there is heated discussion as to whether customary international law binds international organisations generally, particularly given their limited—if at all—contribution to its formation.Footnote 30 This relates to the broader debate on whether being a subject of international law with separate legal personality entails that ‘general rules’—itself a contested category—are necessarily binding.Footnote 31 In this regard, Reinisch appealingly argues that ‘if the functionally limited personality of international organizations develops over time […] they are in fact subject to [custom] when they act in a way capable of infringing it’.Footnote 32 This aligns with ‘the legal premise that “subjects of international law” are usually “subject to international law”’.Footnote 33 For its part, the Court of Justice of the European Union (CJEU) has unequivocally recognised that custom binds the EU, and has indeed applied customary jurisdictional rules.Footnote 34 In light of the forgoing, and the fact that the customary rules are binding upon all of the EU Member States, custom is considered here as being very relevant to the delimitation of EU jurisdictional competences.

Before examining these issues in more detail, Sect. 3 will now consider the concrete points of tension arising from the EU’s ‘ultimatum strategy’, particularly since the conclusion of multilateral agreements in 2016.

3 The EU’s Persistent ‘Ultimatum Strategy’ in the Field of Aviation and Maritime Transport

The EU’s unilateral approach to aviation and maritime emissions has long been a subject of contention. Its 2002 Sixth Community Environment Action Programme (6th EAP) noted the ‘priority action’ of ‘identifying and undertaking specific actions to reduce greenhouse gas emissions from aviation if no such action is agreed within the International Civil Aviation Organisation by 2002’.Footnote 35 Along parallel lines, the Union was required to identify and undertake ‘specific actions’ for the reduction of emissions from marine shipping, ‘if no such action is agreed within the International Maritime Organisation by 2003’.Footnote 36 This was notably part of a two-stranded approach, the EU alternatively lobbying for the inclusion of transport emissions in the new agreement to be negotiated under UNFCCC.Footnote 37

Over the subsequent years, the EU steadily increased the pressure on the international community through threats and consecutive steps of unilateral action that would include operators beyond its territory.Footnote 38 This is referred to here as the EU’s ‘ultimatum strategy’. Sections 3.1 and 3.2 will briefly outline the relevant policy background and state of play for aviation and maritime emissions respectively. Despite the EU’s nominal prioritisation of multilateralism,Footnote 39 it is shown that the EU’s ‘ultimatum strategy’ has now shifted its objective from achieving the conclusion of multilateral agreements, to increasing the level of ambition and environmental integrity envisioned in their design. According to the EU, this is necessary in order for it to achieve its 40% reduction commitment under the Paris Agreement, to which ‘all sectors of the economy must contribute’.Footnote 40

3.1 The EU Emission Trading Scheme and the ICAO CORSIA

Turning first to aviation, the EU’s unilateral approach is well-known and quite controversial. After years of threats, the EU adopted the Aviation Directive, which extended the EU’s ETS to cover international aviation emissions for flights to and from EU territory as of 2012.Footnote 41 Push-back from industry led to the Air Transport Association of America (ATAA) case before the CJEU.Footnote 42 There the EU was accused, among other things, of violating the Chicago Convention and overstepping its jurisdiction. For its part, the CJEU did not find that the extended ETS violated international law.Footnote 43 However, threats of economic sanctions from other states pressured the EU to restrict the scope of the Aviation Directive to flights within the European Economic Area (EEA).Footnote 44 This was notably a temporary move, considered ‘desirable’, ‘in order to sustain the momentum’ of ICAO negotiations of a single global market-based measure for international aviation emissions.Footnote 45

In late 2016, the ICAO reached an agreement on an offsetting scheme in the form of the CORSIA.Footnote 46 Seeking to take into account parties’ ‘special circumstances and respective capabilities’, the CORSIA has a ‘phased’ and ‘routes-based’ approach, with base year 2019–2020.Footnote 47 Participation in the ‘pilot phase’ from 2021–2023, and the ‘first phase’ from 2024–2026, is voluntary.Footnote 48 It is only during the second phase from 2027–2035 that states whose share in the total international revenue tonnes per kilometre (RTK) is above a certain threshold are obliged to participate.Footnote 49 As noted by Erling, the ‘most distinct feature’ of such an offset scheme is that overall emissions can grow without limits, as long as emissions are compensated by ‘offset units’ coming from different emission reduction projects in other industries.Footnote 50 This differs fundamentally from the EU ETS, which sets a quantitative cap on emissions, allotting a set amount credits to be traded on the market.

Responding to international developments, the EU adopted Regulation 2017/2392, ‘amending Directive 2003/87/EC to continue current limitations of scope for aviation activities and to prepare to implement a global market-based measure from 2021’.Footnote 51 While the title suggests an intention to align EU and ICAO measures, this has not proven to be entirely the case. Rather, the Commission is instructed to present a report considering ways to implement the ‘relevant ICAO instruments’ through a revision of the ETS.Footnote 52 This report ‘shall also examine the ambition and overall environmental integrity of the global market-based measure, including its general ambition in relation to targets under the Paris Agreement, the level of participation, its enforceability, transparency, [and] the penalties for non-compliance […]’.Footnote 53 Based on its assessment of these criteria, the Commission must provide a proposal to, ‘where appropriate’, ‘amend, delete, extend or replace’ the current derogations (Art. 28b(3) Aviation Directive). Importantly, this proposal shall be ‘consistent with the Union economy-wide greenhouse gas emission reduction commitment for 2030 with the aim of preserving the environmental integrity and effectiveness of Union climate action’.Footnote 54 EU policy expressly states that ‘in the absence of a new amendment, the EU ETS would revert back to its original full scope from 2024’.Footnote 55

In the meantime, some alignment is sought through amendments to the EU’s existing Regulation on monitoring and reporting (MRR) and Regulation on verification and accreditation of verifiers (AVR).Footnote 56 A further, more contentious measure is the Delegated Regulation ‘supplementing’ the Aviation Directive as regards the adopted ICAO MRV measures (Delegated Regulation regarding ICAO MRV measures).Footnote 57 This delegated act ‘complements’ the existing EU measures in respect of flights and operations that are not currently covered by the EU ETS Directive’.Footnote 58 The new Article 28c of the Aviation Directive requires that these measures be ‘based on the relevant instruments adopted in the ICAO’ and ‘avoid any distortion of competition’.Footnote 59 However, they must also be consistent with the EU’s own monitoring and reporting principles and its verification requirements.Footnote 60

Embedded in EU policy is thus a tension between alignment with ICAO standards and maintaining its desired level of environmental protection. At the time of writing, this is already leading to several divergencies between EU and ICAO measures. For example, the Implementing Regulation for the MRR only recognises two methods for the monitoring of fuel consumption, while the CORSIA SARPs recognise five.Footnote 61 This poses a restriction on the newly included group of operators based in EEA outermost regions or dependencies and territories of EU Member States.Footnote 62 Furthermore, the EU MRV requirements do not foresee in the simplified CORSIA ‘CO2 Estimation and Reporting Tool’ (CERT) for small scale operators.Footnote 63 The EU notably has its own Small Emitters Tool (SET), however this is only open to a more restricted group.Footnote 64 As regards verification, the EU measures require verifier accreditation by a ‘national accreditation body’, of which each EU Member State has one.Footnote 65 Under CORSIA, ‘verification bodies’ are defined more broadly as an ‘accredited independent third party’.Footnote 66 These bodies could be accredited by any CORSIA member, not just an EU Member State.Footnote 67

Furthermore, in terms of scope, the categories exempted from MRV requirements under the EU and ICAO measures do not match entirely.Footnote 68 For example ‘scientific research and testing’ is exempt under the EU measures but not the CORSIA. Interestingly, under the EU Delegated Regulation, covered aircraft operators are only ‘recommended’ but not obliged to verify and report emissions from flights between two third countries (Art. 2(3)). This contrasts with the CORSIA SARPs which for operators from participating countries are applicable to ‘all international flights’ on or after January 2019.Footnote 69 The EU’s rationale, however, is largely based on doubts as to the jurisdiction of the Commission to regulate these ‘extraterritorial’ flights.Footnote 70

The above comparison illustrates that the EU is already maintaining a stricter standard at the stage of MRV requirements. Responding to the emerging divergencies, in late 2019 ICAO Resolution 40–19 was adopted, providing that ‘the CORSIA is the only global market-based measure applying to CO2 emissions from international aviation.Footnote 71 The aim of this move is ‘to avoid a possible patchwork of duplicative State or regional MBMs [market-based measures]’.Footnote 72 Intriguingly, despite calls from both the European Parliament and the Commission, none of the EU Member States made a formal reservation to the exclusivity clause.Footnote 73 Taken together, interesting questions thus arise for the legal relationship between these overlapping measures, considered further in Sect. 4.

3.2 The EU Maritime MRV Scheme and the IMO DCS

Closely following the aviation sector, the EU’s ultimatum strategy has clearly manifested itself in the field of international shipping. By mid-2013 the Commission was clear that the ‘deadline has passed without sufficient international action’ on maritime emission reductions.Footnote 74 It therefore proposed a regulation introducing a monitoring, reporting and verification scheme for maritime transport as a ‘first step’ towards inclusion in the EU emission reduction target.Footnote 75 This was realised in Regulation 2015/757 (EU Maritime MRV), under which operators of ships above 5000 gross tonnes are obliged to monitor, by means of the fuel consumed, the amount of CO2 emitted on voyages to and from EEA ports.Footnote 76 Swiftly after, in 2016, the IMO Marine Environmental Protection Committee (MEPC) adopted its own mandatory global data collection scheme (DCS) for CO2 emissions as an amendment to Annex VI of MARPOL.Footnote 77 As of 1 January 2019, the IMO DCS obliges ships above 5,000 gross tonnes to monitor and report fuel consumption to their Administration (flag state).Footnote 78

As with civil aviation, the IMO and EU measures do not mirror each other.Footnote 79 Article 22(3) of the EU Maritime MRV is said to have ‘anticipated’ this situation, instructing the Commission to ‘review this Regulation’ and, ‘if appropriate, propose amendments […] in order to ensure alignment with that international agreement’.Footnote 80 However, the Commission’s Proposal for a Regulation amending the Maritime MRV to ‘take appropriate account of global data collection system’, already rejects complete alignment.Footnote 81 According to this proposal, ‘the objective is not to replace the EU MRV Regulation by the IMO DCS’.Footnote 82 Rather, the ‘main challenge’ is ‘to find appropriate ways to help market actors implement both schemes while streamlining and reducing administrative burden as possible, and while preserving the objectives of the EU MRV Regulation’.Footnote 83 Interestingly, these objectives appear to be framed largely in terms of the contentious differences between the data collection measures.Footnote 84

To note some key examples, the EU Maritime MRV requires data on the actual load carried or CO2 emitted, while the IMO DCS focuses only on the fuel consumed during a ship’s voyage.Footnote 85 The EU Maritime MRV further includes domestic CO2 emissions—i.e. emissions from voyages within EU states, while the IMO DCS only covers fuel consumed on international voyages.Footnote 86 The 2019 Impact Assessment accompanying the proposed regulation does not support further alignment, considering that this would undercut the required data for the design of ‘coherent and cost-effective climate, energy and environmental policies’.Footnote 87 In addition, the EU Maritime MRV requires the appointment of independent accredited verifiers.Footnote 88 This differs from the IMO DCS, where verification of the monitoring and reporting activities is to be carried out by a ship’s Administration according to national procedures. A ‘departure’ from third party accreditation was considered unacceptable for the EU, potentially harming the ‘reliability’ and ‘accuracy’ of the system.Footnote 89

Another issue is that of transparency. Under the IMO scheme, the Secretary General of the IMO will produce an annual summary of the aggregated anonymised data to be presented to the MEPC.Footnote 90 This differs from the EU Maritime MRV which requires the publication of information ‘with the aim of informing the public and allowing for an assessment of the CO2 emissions and the energy efficiency of maritime transport per size, type of ships, [and] activity’.Footnote 91 This EU strategy has a potential naming and shaming effect, and is intended to ‘incentivise the uptake’ of energy efficiency measures by operators.Footnote 92 The 2019 Impact Assessment considered that following the IMO’s lower transparency standards would reduce the capacity for the EU Maritime MRV to address market failures and achieve ‘positive environmental impacts’.Footnote 93

Perhaps most ominously, the EU has been clear that concrete reduction measures must now follow to ensure that the sector ‘contributes its fair share’ to the objective of the Paris Agreement on keeping climate change ‘well below 2 °C’.Footnote 94 To this end, Directive 2018/410 instructs the Commission to keep IMO progress ‘under regular review’.Footnote 95 Holding the ultimatum in place, it further notes that ‘[a]ction from the IMO or the Union should start from 2023’.Footnote 96 In 2018, the IMO did agree on an Initial Reduction Strategy, with a minimum goal of 50% GHG reductions from shipping by 2050, compared to 2008.Footnote 97 While supportive of this strategy, the European Parliament’s resolution at the UNFCCC Conference of the Parties (COP) 25 simultaneously ‘urges the Commission to propose, as soon as possible, additional EU actions […] such as the inclusion of the maritime sector in the ETS’.Footnote 98 Looking ahead, this thus remains a very contentious policy point.

4 Competing Climate Change Responses: Legal Relationships Between the EU, ICAO and IMO Rules

The previous section has illustrated a shift in the focus of the EU’s ‘ultimatum strategy’ from the conclusion of multilateral agreements to their subsequent design, and even implementation by participating states. As policy regarding transport emissions develops on multiple fronts, we are left with concurrent measures containing different monitoring, reporting and verification standards. The longer-term scenario is potentially more concerning, with acutely diverging reduction policies for aviation and maritime emissions. As discussed in Sect. 2, the point of departure for this analysis is that unilateral measures are permitted unless there are international rules providing otherwise. However, as will be seen, it is not so straightforward to determine which rules of international law apply to the EU here, and what they precisely entail. Sections 4.1 and 4.2 will now consider the relationship between the EU measures and those of the ICAO and IMO respectively.

4.1 The Relationship Between the EU and the ICAO Measures on Aviation Emissions

The CORSIA was adopted within the broader framework of the Chicago Convention on International Civil Aviation. This treaty constituted the ICAO, Article 37 endowing the organisation competence to adopt standards and recommended practices (SARPs). Of particular relevance in the present context is ICAO Resolution A39-3, which decides to ‘implement’ the CORSIA (Art. 5), and sets out the general design framework. Further operationalisation can be found in Annex 16, Volume IV of the Chicago Convention, which contains the evolving CORSIA SARPs.Footnote 99

At the outset, as noted by Martinez Romera in her comprehensive study, neither the Convention’s Annexes nor ICAO resolutions are strictly binding on ICAO members, as these were not originally an ‘integral part’ of the Convention.Footnote 100 The ICAO’s adoption of the ‘exclusivity clause’ in Resolution A40-19 would then seem to have more of a political than a legal weight for the EU, particularly as the latter is not an ICAO member. Nonetheless, the European Parliament was ‘deeply concerned’, ‘urging’ Member States to file reservations, ‘so as to preserve the Union’s legislative autonomy with regard to measures intended to reduce GHG emissions from the aviation sector’.Footnote 101 From this statement, it is unclear whether the Parliament considers its legislative competence actually limited, or whether the use of the word ‘autonomy’ is intended to refer to the de facto pressures that will arise if its Member States face conflicting ICAO standards. The latter seems more likely, given the EU’s position on its relationship with the Chicago Convention, discussed later in this section.

Turning to the Chicago Convention, one element that does bind the parties is the procedure in situations of non-compliance with SARPs. This can be found in Article 38 which provides that:

any State which finds it impracticable to comply in all respects with any such international standard or procedure [as adopted by the ICAO according to Article 37], or to bring its own regulations or practices into full accord with any international standard or procedure after amendment of the latter, or which deems it necessary to adopt regulations or practices differing in any particular respect from those established by an international standard, shall give immediate notification to the International Civil Aviation Organization of the differences […] [emphasis added].

The question then arises whether the EU, while not a formal party, would be subject to Article 38 of the Chicago Convention, and on what basis this would occur. One option is through functional succession, according to which an organisation when acquiring competence from its Member States, succeeds to the relevant obligations already incumbent upon all of its Member States. There is, however, much debate on the meaning and validity of this doctrine, with differing views in both EU and public international law. From an EU law perspective, in International Fruit Company, the CJEU in principle recognised the possibility for functional succession, accepting that the EU had succeeded to its Member States’ obligations under the 1947 General Agreement on Tariffs and Trade.Footnote 102 However, it set the bar for functional succession very high, and in the ATAA case, explicitly rejected such succession in relation to the Chicago Convention.Footnote 103 According to the CJEU, ‘in order for the European Union to be capable of being bound, it must have assumed, and thus had transferred to it, all the powers previously exercised by the Member States that fall within the convention in question’.Footnote 104 Here, EU Member States were found to have retained rights falling within the scope of the Chicago Convention.Footnote 105

This strict approach to the functional succession doctrine has been criticised in the literature as being ‘too simplistic’,Footnote 106 and one which ‘makes little sense’.Footnote 107 According to Eeckhout, the doctrine itself rests on the notion that ‘a transfer of powers from the Member States to the EU also entails a transfer of the Member States’ international obligations’.Footnote 108 The CJEU’s narrow interpretation undermines the EU’s likeliness to adhere to important international treaties,Footnote 109 and generally exacerbates legal fragmentation.

This critique is related to a debate also avid under public international law, on the validity of the functional succession doctrine as such. The discussion is not only relevant for succession to treaties as a whole, but in relation to discrete international obligations. On one approach, international organisations, having a ‘separate legal identity’, are not bound by their members’ obligations, as this ‘would make a mockery of their independent legal existence’.Footnote 110 Accepting this, the CJEU could then be seen to have created its own sui generis approach, allowing the applicability of international agreements in its legal order for its own functional reasons, rather than as a matter of international law.Footnote 111 An opposing view supported here, focuses on the functional nature of international organisations’ competence.Footnote 112 This competence is derived from the sovereignty of its Member States, who cannot transfer more power than they have.Footnote 113 As noted by Schermers and Blokker, ‘an organization formed by states will be bound by the obligations to which the individual states were committed when they transferred powers to the organization’.Footnote 114 Contrary to the view of the CJEU, this should not be made dependent on whether there are other competences in a broader international agreement that have not been conferred to the international organisation.

As the 1944 Chicago Convention predates the 1958 EC Treaties, it could then well be argued that the EU must exercise said competence under the same conditions as those originally in place upon its Member States. This is notably different to the existing EU law provision in Article 351 of the Treaty on the Functioning of the European Union (TFEU), that rights and obligations with third countries made before 1 January 1958 or before the date of acceding states’ accession ‘shall not be affected by the provisions of the Treaties’. Article 351 TFEU has namely been interpreted as an obligation to allow Member States to adhere to their agreements, not to bind the EU itself to these obligations.Footnote 115 A further detailed legality assessment of the unilateral ETS with the Chicago Convention goes beyond the scope of this article.Footnote 116 This is a debate that could certainly reignite should the EU decide to revert the ETS to its original scope.

Taken together, while the EU may arguably be obliged to notify ICAO of differences, it appears free to consider it ‘necessary’ to pursue—through its Member States—higher environmental integrity standards. This latter point was indeed relied upon by the EU in establishing its position within the ICAO in respect of the first edition of the CORSIA SARPs.Footnote 117 In practice however, this is politically tenuous, and was criticised by industry as being ‘extremely damaging’ as ‘it creates the perception that it is acceptable for ICAO Member States to depart at their discretion’ from ICAO agreements.Footnote 118

4.2 The Relationship Between EU and IMO Measures on Maritime Emissions

The IMO Data Collection Scheme for maritime emissions was adopted by resolution as an amendment to Annex VI of the MARPOL Convention.Footnote 119 Contrary to the Annexes to the Chicago Convention, MARPOL Annex VI is binding upon parties.Footnote 120 For its part, the CJEU has repeatedly stressed that the EU ‘is not a contracting party to the Marpol 73/78 Convention, including Annex VI, and is not bound by it’.Footnote 121 From an international law perspective, one may ask similar questions regarding functional succession. Here, however, the situation is somewhat different.

To start with both the MARPOL and Annex VI are younger agreements, the latter having been added by the 1997 Protocol.Footnote 122 Still, following the arguments made above, it would seem consistent that, to the extent that the EU ‘progressively assumed powers previously exercised by the Member States’ at the time of the 1997 Protocol, these should be exercised in line with the obligations therein.Footnote 123 Yet there remains another hurdle, as two EU Member States, namely Austria and Hungary, were not party to Annex VI upon its amendment.Footnote 124 This presents a doctrinal problem, as accepting functional succession would effectively impose obligations on non-contracting states without their consent, contrary to the pacta tertii principle expressed in Article 34 of the Vienna Convention on the Law of Treaties.

Fundamentally, the IMO DCS is itself an amendment to Annex VI made by another international organisation, the IMO, subsequent to the EU’s acquisition and exercise of its own competence. It is then questionable whether such a later amendment would ever fall within the scope of functional succession, the rationale of which relies on the obligations existing prior to conferral of power. Accepting this, the IMO DCS must be characterised as a concurrent or competing decision, for which clear-cut rules of priority are difficult to determine.Footnote 125

Yet, even in the absence of a formal hierarchy, there are arguments to be made that the EU must respect the obligations of its Member States when exercising union competences. As such exercise is an extension of its Member States’ sovereignty, it seems logical that the EU must not place the former in a stranglehold with their other obligations. There is notably some recognition for this notion in the duty of sincere cooperation enshrined in Article 4(3) of the Treaty on European Union (TEU). In the case Manzi and Compagnia Naviera Orchestra, the question was raised whether the EU had violated Article 4(3)TEU and the pacta sunt servanda principle, ‘on the ground that that provision of the [relevant] Directive may lead to an infringement of [MARPOL] Annex VI and thereby require Member States party to the 1997 Protocol to infringe their obligations with regard to the other Contracting Parties thereto’.Footnote 126 The CJEU however, declined to answer this question, finding that as the Union is not a party it had no competence to review compatibility with MARPOL Annex VI.Footnote 127 According to the Court, this ‘may not be circumvented by relying on the alleged infringement of the principle of cooperation in good faith laid down in the first subparagraph of Article 4(3) TEU’.Footnote 128 Still, it was recognised that ‘although the European Union is not bound by an international agreement, the fact that all its Member States are contracting parties to it is liable to have consequences for the interpretation of European Union law’, in particular relevant secondary law.Footnote 129 From an international law perspective, it is questionable whether this construction gives sufficient respect to Member States’ other obligations. Such respect arguably extends beyond consistently interpreting concurrent rules to the creation of these rules themselves.

Drawing together these considerations, the EU would not seem bound by the IMO DCS, even though this amends MARPOL Annex VI, which has binding force. That being said, the EU arguably has some duty not to use its derived sovereign power to force its Member States to violate their other obligations. In the present context, there is not (yet) a strict ‘conflict’ of norms, as Member States can enact both sets of standards and operators can in principle comply with both.Footnote 130 Still, there remains a conflict in the broader sense defined above, given the differing views on how to respond to the same problem. The existence of these dual regulatory burdens triggers important questions on the balance of regulatory competences of equal sovereign states. This is an issue for the law of state jurisdiction which will now be considered further in Sect. 5.

5 Customary Rules on Regulatory Competence: The Law of State Jurisdiction

Supplementing the specific rules on international transport emissions, the customary international law of state jurisdiction contains general rules on state competence to regulate the conduct and consequences of an event.Footnote 131 Typically, these rules come into play when states seek to regulate conduct or circumstances occurring beyond their territory, as it is then that concurrent jurisdictional claims arise.Footnote 132 The present case is no exception. For aviation, the current EU ETS covers flights between EEA aerodromes, while the EU Delegated Regulation ‘supplementing’ the Aviation Directive extends key MRV requirements to outermost regions, or dependencies and territories.Footnote 133 Flights between states are considered ‘international’ under the CORSIA SARPs and are therefore also covered by the CORSIA.Footnote 134 A potential expansion of the ETS to all flights to and from the EEA or EU would obviously exacerbate the dual regulatory burden. As regards maritime emissions, the EU Maritime MRV applies to ships on voyages to and from all EU ports.Footnote 135 This covers conduct on voyages of large vessels flying all flags before and after they are in EU territory, and clearly anticipates the possibility for the EU to design a further MBM with the same geographical scope.Footnote 136

Jurisdictional assertions that reach beyond territorial boundaries raise questions of ‘extraterritoriality’, a contested topic in legal discourse.Footnote 137 In assessing the applicable jurisdictional rules, Sect. 5.1 will first consider the possible jurisdictional basis for the EU’s unilateral emission reduction measures. Section 5.2 then turns to the more normative question of interest-balancing, both doctrinally within the law of state jurisdiction, and more concretely in the context of international transport emissions.

5.1 The Jurisdictional Basis for the EU Emission Reduction Measures for International Transport

Unilateral measures seeking to regulate extraterritorial conduct require a valid jurisdictional basis. A key voice in the debate on when this may be the case has been the CJEU itself. In the ATAA case, the CJEU assessed the permissibility of the EU Aviation Directive in its full international scope according to customary jurisdictional rules. There it found that the EU had ‘unlimited jurisdiction’ based on the territorial presence of the aircraft in EU aerodromes.Footnote 138 However, as this author has argued elsewhere, this interpretation does not appear consistent with the theory and rationale of the law of state jurisdiction.Footnote 139 The fact that aircraft are present on one end of the journey is relevant for the legality of the enforcement measures. Enforcement jurisdiction ‘to ensure compliance with its laws’ is namely strictly territorial.Footnote 140 It is not however a sufficient basis for prescriptive jurisdiction, which pertains to ‘the authority of a state to make its law applicable to particular persons or circumstances’.Footnote 141 As will now be discussed, the reason for this has to do with the material protection of sovereign equality.

Essentially, the operation of the EU’s measures is de facto dependent on conduct or circumstances abroad. Such measures give rise to a ‘potential interference’ with the regulatory autonomy, and thus the sovereignty, of other states.Footnote 142 Notably, the interests of private actors are considered here as an extension of the regulatory choices of different states. Thus, where a home state has chosen not to burden its producers with certain environmental regulations, foreign state measures imposing costs on these actors raise issues of jurisdiction. Such ‘potential interference’ needs justification based on more than partial territorial presence, as this would otherwise disregard a measure’s very real impact on activities beyond the regulator’s territory. Recognition for the effects of one jurisdictional assertion on the regulatory space left for others, aligns with the function of jurisdiction rules, namely to assign and apportion regulatory claims.Footnote 143 Measures whose operation is de facto dependent on foreign conduct or circumstances are therefore characterised here as having an ‘extraterritorial element’, making them measures of interest under the law of state jurisdiction.

There are several bases which may support assertions of prescriptive jurisdiction with an ‘extraterritorial element’ under customary law. In addition to territory, these are nationality, effects, protection, and universality.Footnote 144 These bases may notably be applied individually or together, the relevant legal threshold being whether a regulator can demonstrate a ‘substantial connection’ to the subject matter in question.Footnote 145 To start with, the nationality principle provides a basis of jurisdiction over nationals—including ships flying a flag of, or aircraft registered to—a particular state.Footnote 146 In addition, according to the effects-doctrine, a state may regulate ‘conduct outside its territory that has or is intended to have substantial effect within its territory’.Footnote 147 Related to this is the protection principle, granting jurisdiction over conduct that threatens vital national interests.Footnote 148 These last two principles are based on states’ sovereign rights to respond to threats and protect their territory. Finally, the universality principle provides a basis for jurisdiction over conduct that threatens fundamental values of the international community, in the absence of any specific link to the regulating state.Footnote 149 This principle has evolved in the field of criminal law, with state practice to date largely focusing on grave crimes.

In the realities of our complex, globalised world, the jurisdictional bases cannot always be applied in a clear-cut manner. Today, it is therefore generally-accepted that to validly assert prescriptive jurisdiction, states must demonstrate a ‘genuine’ or ‘substantial connection’ to the regulated subject matter.Footnote 150 This may be made up of one or more of the abovementioned bases. Yet the application of the ‘substantial connection’ requirement in the relatively newer context of climate change remains quite unclear. Some support can be found in the literature for such expansive EU measures. Considering the issue more generally, Ringbom for example notes that the principles of jurisdiction ‘seem to leave some room for port States to apply requirements which relate to activities beyond their maritime zones’.Footnote 151 Scott considers such measures an exercise of ‘territorial extension’, which have received measured acceptance from both the CJEU or the Appellate Body of the World Trade Organisation.Footnote 152

A detailed analysis of the operationalisation of substantial connection requirement goes beyond the scope of this article, and has been the subject of extensive debate elsewhere.Footnote 153 In short, it is submitted by this author that a state may have a ‘substantial connection’, when it can demonstrate, based on consistent scientific evidence, that the subject matter regulated measurably contributes to an increase in the risk of reasonably foreseeable grave harm.Footnote 154 This aligns with the preventative rationale of the effects and protection bases, while also infusing elements of precaution from international environmental law. Such a claim is arguably reinforced when the measure aims to protect a community interest of ‘common concern’, such as climate change.Footnote 155 In the view of this author, the EU can demonstrate such a substantial connection to the international maritime and aviation transport activities, which contribute a substantial proportion of total global emissions cumulatively resulting in anthropogenic climate change.

5.2 Consideration of Other Interests as a Condition on Unilateral Jurisdiction?

The classical principles analysed above all focus on the interests of the regulating actor. Clearly, there is a myriad of other interests also engaged when the EU refuses to align with multilateral measures regulating emissions from international transport. This gives rise to tensions with the principle of sovereign equality. As authoritatively explained by Meessen, in a jurisdictional context this principle entails the right of states ‘to pursue the regulatory goals it has decided to adopt’.Footnote 156 Notably, tensions with this principle may arise even in the absence of competing legislation, as the choice not to regulate also falls within the regulatory autonomy afforded by sovereign equality.

In the case of a competing multilateral measure the problem is clearly exacerbated, as it is not one sovereign interest against the other, but one against many. In practice however, it is a rather artificial exercise to seek to balance competing claims based on the number of sovereign states supporting a measure. Indeed, in all multilateral settings real power imbalances will inevitably shape the final agreements reached, meaning that an international agreement is not the product of perfect sovereign equality.Footnote 157 For the EU, while its legislation formally represents the interests of all of its Member States, its law-making intuitions, in particular the Commission, have far-reaching autonomy. Their decisions are then by no means a negotiated compromise reached by its Member States.

The question remains therefore, whether and how states must consider the legitimate interests of others when exercising jurisdiction with an extraterritorial element. In the literature, this is referred to as a matter of ‘reasonableness’ or ‘interest-balancing’, the terminology largely stemming from the 1987 Restatement (Third) of US Foreign Relations Law (US Third Restatement).Footnote 158 For a long time, this was one of the most comprehensive and authoritative documents dealing explicitly with jurisdictional interest-balancing beyond the field of private international law. Of particular importance is §403 which precludes states from exercising jurisdiction, despite the existence of a valid basis, when this would be ‘unreasonable’. Importantly this ‘rule of reason’ was considered by the drafters to reflect customary international law.Footnote 159 According to §403(2) the unreasonableness of an act was to be determined ‘by evaluating all relevant factors’ including:

(c) the character of the activity to be regulated, the importance of regulation to the regulating state, the extent to which other states regulate such activities, and the degree to which the desirability of such regulation is generally accepted.

(d) the existence of justified expectations that might be protected or hurt by the regulation;

[…]

(g) the extent to which another state may have an interest in regulating the activity; and

(h) the likelihood of conflict with regulation by another state.

In terms of concrete application, this very open interest-balancing test may point in several directions at once. Indeed, the test was subsequently rejected by US courts as being unworkable in practice, particularly when applied by domestic courts.Footnote 160 Published in 2018, the US Fourth Restatement confirms a lack of state practice and opinio juris to support §403’s rule of reason as a requirement of custom.Footnote 161 In fact, according to the Fourth Restatement, the only element of ‘reasonableness’ required by international law is that of a ‘genuine connection’ to the regulated subject matter.Footnote 162 Beyond that, the Reporters’ Notes consider that ‘states often seek to reduce conflicts of prescriptive jurisdiction through various rules of domestic law that are often motivated by international comity but are not required by international law’.Footnote 163 We thus see a shift in the characterisation of reasonableness from one of custom to one of voluntary comity.

In the view of this author, while the emphasis on the requirements for the formation of custom is understandable, the characterisation of jurisdictional self-restraint as purely voluntary goes too far in the opposite direction. Following this approach would namely mean that international law contains no binding protections for the sovereign interests of states whose regulatory autonomy is affected by prima facie valid jurisdictional assertions. This again does not seem to give sufficient attention to the principle of sovereign equality. It also risks overlooking the related principles of non-intervention and non-interference, which serve to realise sovereign equality.Footnote 164 While the precise operationalization of these principles is unclear, they certainly apply to the exercise of legislative jurisdiction.

It is argued that, rather than manifesting as defined limitations, these principles serve to inform national comity doctrines such as ‘reasonableness in interpretation’ and the ‘presumption against extraterritoriality’.Footnote 165 The blending of law and comity is further catalysed by the widely-accepted doctrine of consistent interpretation with international law.Footnote 166 As such, state practice demonstrating jurisdiction self-restraint is necessarily a reflection of both custom and comity. Applied to the present case, it seems that these principles of sovereign equality, non-interference and non-intervention support a duty of the EU to align its measures with those of the ICAO and IMO.

Importantly however, the forgoing discussion is premised on the assumption that the EU and multilateral measures are real alternatives to each other. In reality this is clearly not the case. For shipping, in the EU’s view, the IMO DCS lacks the transparency, reliability and completeness of information that its Maritime MRV offers. Based on its impact assessment, the EU considers its level of protection necessary to achieve its projected 2% reduction in GHG emissions compared to business as usual standards.Footnote 167 For aviation, it is clear that a carbon-offsetting scheme simply does not provide the same level of protection as a cap-and-trade scheme which puts a ceiling on emissions.Footnote 168 The latter was considered by the ICAO, but somewhat unsurprisingly, given the pressure from the aviation industry, ultimately was not chosen.Footnote 169 Indeed, the International Coalition for Sustainable Aviation went so far as to argue that replacement of the ETS with the CORSIA would, given the latter’s environmental weakness, ‘constitute a breach of Europe’s obligations under the Paris Agreement’.Footnote 170

This raises questions as to how the duty to respect regulatory autonomy relates to the obligations in the multilateral climate agreements. It is clear that the EU is bound by the UNFCCC and the Paris Agreement to which it is a party.Footnote 171 Yet as discussed, silently continuing the approach of Article 2(2) of the Kyoto Protocol, the Paris Agreement does not explicitly include international aviation and maritime transport emissions. There are compelling arguments to be made that the UNFCCC’s recognition of climate change as a ‘common concern of humankind’ and its overarching objective of preventing ‘dangerous’ climate change imply a (collective) obligation for states to adopt sufficiently ambitious reduction targets.Footnote 172 This would hold irrespective of the voluntary element in the setting of National Determined Contributions (NDCs). Arguably then, the aviation and maritime transport sectors, as sizeable contributors to global warming, would need to make a meaningful contribution to the required mitigation targets. This appears reflected in the EU’s position that ‘all sectors of the economy must contribute’ to achieving its 40% reduction commitment under the Paris Agreement.Footnote 173

For their part, it is doubtful whether the IMO and ICAO consider that the climate change regime contains a binding goal for them. The IMO has stressed that ‘the Paris Agreement does not include international shipping’.Footnote 174 Its Initial Reduction Strategy notes quite vaguely that it is ‘aimed at enhancing IMO’s contribution to global efforts by addressing GHG emissions from international shipping’ including ‘the Paris Agreement and its goals’.Footnote 175 In similarly ambiguous wording, the ICAO notes that ‘work is being undertaken to explore a long-term global aspirational goal for international aviation in light of the 2 °C and 1.5 °C temperature goals of the Paris Agreement’.Footnote 176 Resolution A40-18 explicitly denies ‘any attribution of specific obligations to individual States’, in the achievement of ‘a collective medium-term global aspirational goal of keeping the global net carbon emissions from international aviation from 2020 at the same level’.Footnote 177 This issue is sure to gain attention in the future.

The lack of defined ambition of the ICAO and IMO aggravates the interest-balancing dilemma in the law of state jurisdiction. While the principle of sovereign equality requires states to respect each other’s regulatory space, it is questionable whether autonomy necessarily prevails when it hampers the achievement of the near-universally accepted objective of the climate change regime. This is to say nothing of the far broader ethical implications that formalistic respect for sovereign equality may have for the recognition of the interests of future generations and of vulnerable states with less bargaining power. We are thus left with important normative dilemmas for the further crystallisation of international climate change rights and obligations.

6 Conclusion

With the ‘emission gap’ still gaping, there is an urgent need for regulatory responses to the increasing emissions from international transport. While the emergence of multilateral rules is to be welcomed, doubts remain as to the adequacy of their level of ambition. As a global actor with economic clout, the EU is in a unique position to unilaterally exercise pressure on ICAO and IMO policy. Questions arise, however, as to the extent to which the development of these competing multilateral responses to climate change may limit the EU’s regulatory freedom. This issue is steeped in a myriad of disagreements, which hinge on varying conceptions of the relationship between the EU and the international legal order. The present contribution has sought to navigate these debates from the perspective of international law.

In the context of aviation, it could be argued that the EU has functionally succeeded to certain obligations under the Chicago Convention, including Article 38 on non-compliance with ICAO SARPs. Still, as the CORSIA is not binding and may be deviated from if ‘necessary’, the EU retains a degree of regulatory freedom. The ICAO’s 2019 ‘exclusivity clause’ would not seem to alter this relationship, though it may impose de facto limitations on EU regulatory autonomy. For the regulation of shipping emissions under MARPOL Annex VI, functional succession does not seem plausible. While binding on parties, Annex VI is a younger agreement which was amended to include the multilateral scheme after the EU had been conferred regulatory competence in the same field. Despite the lack of hard limits, it is nonetheless arguable that the EU has an indirect duty to respect the obligations of its Member States, derived from the functional nature of its competences. While the EU principle of sincere cooperation supports some alignment through consistent interpretation, under international law this should arguably extend to policy design.

In light of the looming regulatory burdens, more general questions of state jurisdiction have also been considered. It is argued that the EU may well be able to demonstrate a valid jurisdictional basis for its expansive measures. This must, however, be exercised with sufficient respect for the sovereign equality of other states. Operationalising concrete jurisdictional limitations remains, however, complex. Even assuming that sovereign equality and non-interference support a duty for the EU to respect the multilateral responses from the ICAO and the IMO, this may well undermine the UNFCCC and Paris Agreement aim of preventing ‘dangerous’ climate change. Between these competing norms and interests, the EU then remains both a hero and a hegemon. More fundamental questions remain for the design of the international legal system as a whole. As the legal relationships continue to evolve, it is to be hoped that necessity will be the ‘mother of invention’.Footnote 178