PROVOCATIVE CLIMATE PROTECTION: EU ‘EXTRATERRITORIAL’ REGULATION OF MARITIME EMISSIONS

Abstract

In 2015, frustrated by the slow pace of negotiations in the International Maritime Organisation, the EU issued Regulation 2015/757 on the monitoring, reporting, and verification of carbon dioxide emissions from maritime transport. Echoing the controversial Aviation Directive, the Regulation is intended to support a unilateral market-based measure, and includes emissions from outside EU territory. This raises the question whether, according to international law, the EU has jurisdiction to regulate such ‘extraterritorial’ circumstances. In exploring the appropriate jurisdictional bases, we argue that neither the Law of the Sea Convention, nor world trade law definitively decide this issue. We therefore devote more detailed attention to the customary international law of State jurisdiction supplementing these regimes. We seek to build on the existing analysis by examining climate change as a ‘common concern of mankind’. We argue that this emerging concept has distinct legal implications that can and should be accommodated within the interest-balancing exercise underlying the jurisdictional analysis.

I. INTRODUCTION

With the conclusion of the Paris Agreement in December 2015, the international community reinforced its resolve to mitigate the devastating effects of global warming on ecosystems and human welfare.¹ The Paris Agreement represents the first globally inclusive instrument to tackle climate change as a ‘common concern of mankind’.² However, despite the political success, there remains a considerable ‘ambition gap’ between countries’ voluntary pledges and the emission reductions required to prevent a dangerous rise in the global temperature.³ In particular, the Paris Agreement is silent on maritime emissions, which remain excluded from States’ mandatory commitments under the Kyoto Protocol.⁴ The 1997 Kyoto Protocol encourages Annex I parties to seek a multilateral solution ‘working through’ the International Maritime Organisation (IMO).⁵ However, almost two decades later, the IMO is yet to agree on a binding global instrument for the reduction of CO₂ emissions.

Impatient with the slow international negotiations, the EU has repeatedly threatened unilateral action, though to little avail.⁶ Finally, in 2015 the Union went ahead and adopted Regulation 2015/757 (the Regulation) setting out a monitoring, reporting, and verification (MRV) scheme for maritime emissions as the first step towards a market-based measure (MBM) incorporating these emissions into the EU's independent reduction commitment.⁷ MBMs use economic policy tools to influence the decisions of private operators, particularly through harmonizing the dual incentives of profit-making and pollution reduction.⁸ The maritime MRV serves as a prerequisite for such an MBM, providing the data necessary to determine operators’ emission reduction targets.⁹ Importantly, the scope of the Regulation includes the complete duration of voyages to and from EU ports, meaning that operators will be required to monitor and report CO₂ emitted outside EU territory. Should the chosen MBM maintain the same scope, then these ‘foreign’ emissions will likely entail further costs for operators.

There have already been considerable political objections from the shipping industry to the EU's decision to go it alone.¹⁰ In anticipation of increasing scrutiny in the run-up to implementation, this article explores the legal issues surrounding the EU's competence to unilaterally prescribe such measures.¹¹ This issue is not uncontroversial; the negative international reaction to a similar EU measure in relation to aviation emissions led it to pause implementation pending negotiations within the International Civil Aviation Organisation (ICAO).¹² However, the EU has not appeared poised to back down on either aviation or maritime emissions should multilateral negotiations fail to meet its standards.¹³ This raises the question whether, according to international law, the EU has jurisdiction to legislate Regulation 2015/757 and a further MBM with the same geographical scope. As the EU is yet to decide on a MBM, it is valuable to consider how the applicable rules of international law may affect the appropriate design of the projected measure. Importantly, as an international organization, the EU remains subject to international law, including the law of State jurisdiction that applies to the competences conferred upon it by its Member States.¹⁴

Three legal sources are of particular relevance here: the UN Convention on the Law of the Sea (LOSC),¹⁵ the law of the World Trade Organization (WTO) and the customary international law of jurisdiction. The LOSC conditions the MRV's regulation of international shipping and port State measures against foreign-flagged vessels, including its regulation of activities taking place beyond the EU's territory and maritime zones. The WTO is relevant insofar as the EU measures may unlawfully restrict international trade. Both of these regimes have been explored by scholars and research institutions, with differing results.¹⁶ Part of the divergence arises from the conflicting conceptions of ‘extraterritoriality’ that underlie the different analyses. This is ultimately a question of the customary law of State jurisdiction, which governs the geographical scope of the regulatory power of States and regional organizations.

This last regime deserves further attention, not least because of its important complementary role in relation to the two more specific fields. Indeed, the LOSC has been described as a ‘clarification’ and ‘manifestation’ of the customary international law of the sea, including the law of jurisdiction.¹⁷ The WTO Appellate Body (AB) has also held that WTO law ‘should not be read in clinical isolation from public international law’.¹⁸ The latter finding is of particular relevance for the question of whether there is an implicit jurisdictional limitation in the general exemptions of Article XX of the General Agreement on Tariffs and Trade (GATT).¹⁹ The customary law of jurisdiction's complementary role is essential here because, as will be argued, while the trade and maritime regimes do contain rules on State regulatory competence in their respective fields, they do not definitively answer the question of whether the EU has jurisdiction to regulate maritime emissions outside of its territory.

On a more doctrinal level, a focus on customary law is further merited because it is this regime that is primarily designed to target the fundamental regulatory issue at stake here, namely how far a State or regional organization may go when unilaterally protecting a shared interest. We seek to build on the existing discussion by looking in more detail at climate change as a particular type of shared interest, namely a ‘common concern of mankind’. We argue that this emerging concept has a distinct definition and legal implications, and that when States act to protect a common concern this affects the interest-balancing exercise underlying the jurisdictional analysis.²⁰

Section II explores the nature of climate change as a ‘common concern of mankind’ and analyses its legal implications for the obligations of States and regional organizations such as the EU. It then examines the controversial issue of ‘extraterritoriality’ in international law, a matter which often arises when States take action to respond to common concerns. Section III turns its focus to Regulation 2015/757, analysing the relevant ‘extraterritorial elements’ that trigger the need for jurisdictional justification. Drawing from the discussion in the literature, Section IV considers to what extent a basis can be found in the LOSC and WTO law, in particular the GATT, for the EU's competence to legislate the MRV and the projected MBMs. It concludes that, in principle, both regimes accommodate and, in fact, recognize particular regulatory competences to protect common (environmental) concerns. However, as mentioned, neither regime definitively decides the matter of the EU's jurisdiction, thus directing us to a more detailed consideration of customary international law. Section V therefore explores customary law's cornerstone test of a ‘substantial and genuine connection’ between the regulating State and the regulated subject matter. This flexible requirement offers room for development beyond jurisdiction's Westphalian roots, in a world where the common interest should also carry weight. In adapting to the realities of global interdependence we seek an interpretation of States’ jurisdictional rights that accommodates their obligations to respond to climate change as a common concern of mankind.

II. COMMON CONCERNS AND ‘EXTRATERRITORIALITY’

By 1988, climate protection had secured its place on the international agenda, with United Nations General Assembly (UNGA) Resolution 45/53 declaring climate change a ‘common concern of mankind’.²¹ While admittedly still an emerging term,²² we argue in favour of the common concern as a distinct concept with a unique definition and legal implications. Section IIA seeks to define the common concern, exploring its relationship with the concepts of the global commons, common heritage of mankind and obligations erga omnes. Following a similar logic to that underlying obligations erga omnes, it is submitted that the recognition of common concerns brings with it common obligations for States. Section IIB explores the sources of these obligations, focusing on their application in the field of international environment law. As is the case with the EU's maritime plans, measures taken in response to common concerns are likely to raise issues of extraterritoriality. Burdened with a decidedly negative reputation, ‘extraterritoriality’ remains a contested concept in international law. Completing the conceptual framework, Section IIC explores this issue in more detail, clarifying the terminology used here.

A. Defining the Common Concern

Common concerns are essentially shared problems requiring a response in the form of collective action. As the term implies, the two core components are ‘commonness’ and an issue's nature as a ‘concern’. Turning first to ‘commonness’, it is submitted that concerns are ‘common’ when they are shared by all individuals, irrespective of their nationality.²³ As such, common concerns represent more than merely coinciding interests of individual States.²⁴ This is reflective of a shift in international law from a focus on individual State interests to those of the wider international community made up not only of States but of all human beings.²⁵ This was described by the ICTY in the Tadić case, as a development from a ‘State-sovereignty’ to a ‘human-being-oriented’ configuration of the international community.²⁶

Of course, such a definition of ‘common’ remains vulnerable to the criticism that not all individuals share the same values and interests, and therefore cannot be grouped into a collective ‘international community’ in which such interests are clearly delineated. While this may be the case, there are arguably some matters of such a fundamental nature that they can be considered ‘common’, the primary one being collective survival.²⁷ Adding a functional element, a concern is also common when ‘no single State can resolve the problems they pose, or receive all the benefits they provide’.²⁸

A common concern can be characterized as a type of common interest that takes the form of a ‘problem’,²⁹ targeting specific ‘processes’ or ‘protective action’.³⁰ This characteristic distinguishes common concerns from the linked concepts of the ‘common heritage of mankind’ and the ‘global commons’. Global commons are areas outside of the jurisdiction of any State, such as the high seas, outer space, and the atmosphere.³¹ The common heritage of mankind is a broader concept, referring to shared ownership or control over resources of the ‘common heritage’, including those within the territory of individual States.³² Common concerns are typically issues of environmental law, as ecosystems are not neatly delineated according to State territory.³³

At first glance, common concerns bear considerable resemblance to obligations erga omnes, which are obligations owed by all States to the international community as a whole.³⁴ Such obligations arise in relation to certain rights of such importance that all States have a ‘legal interest’ in their protection.³⁵ Similarly to the ICTY in Tadić, Crawford suggests that obligations erga omnes are owed to the ‘international community’ made up not only of States, but also international organizations and natural legal persons.³⁶ This view would further align the concept with common concerns. There is, however, a difference in emphasis. Erga omnes obligations are generally related to issues of State responsibility rather than State jurisdiction. They become relevant when one State's responsibility is invoked by another State that is not directly injured by a violation but has an interest in compliance.³⁷

Importantly, common concerns are linked to obligations erga omnes, as recognition of the former may well trigger the latter. This was explained in Belgium v Senegal, where the ICJ held that ‘the common interest implies that the obligations in question are owed by any State party to all the other States parties to the Convention’.³⁸ Arguably then, where a common concern, as a type of common interest, has been recognized in a treaty containing obligations aimed at its protection, then these are obligations erga omnes partes. It remains unclear, however, to what extent a common interest existing only in customary international law may give rise to general obligations erga omnes.³⁹

B. State Obligations to Respond to Common Concerns?

As the common concern is a newly evolving concept, its precise legal consequences are yet to crystallize in legal doctrine. Fundamentally, the notion that a common interest implies common obligations seems consistent with the rationale of erga omnes obligations provided in Belgium v Senegal. Based on the current state of the law, we argue that one can discern, at the very least, a duty to cooperate and to consider the interests of other legally equal States and their citizens when taking action affecting common concerns.⁴⁰ It is further submitted that, under international environmental law, States have a due-diligence obligation to anticipate, prevent and minimize the aggravation of common environmental concerns.⁴¹ The following paragraphs will explore key sources of these general obligations, and consider their more specific operationalization in relation to climate change.

While international law was traditionally founded on the individualistic, ‘negative’ rules of delimitation and abstention, the positive duty to cooperate now plays an increasingly central role.⁴² Article 1 of the UN Charter, as the ‘magna carta of cooperation’, notes the key UN aim ‘[t]o achieve international co-operation in solving international problems of an economic, social, cultural, or humanitarian character’.⁴³ Further support can be found in Principle 4 of the supplementary Declaration of Friendly Relations and Cooperation among States.⁴⁴ In fact, in the MOX Plant Case, the International Tribunal for the Law of the Sea (ITLOS) found the duty to cooperate to be a ‘fundamental principle in […] general international law’.⁴⁵ According to Perrez, this duty is inherent in sovereignty itself, which today is generally accepted to include certain responsibilities as well as State powers.⁴⁶ While a State's primary responsibility is to its citizens, the recognition of community interests has brought with it recognition of sovereign responsibilities towards the ‘higher’ international community.⁴⁷

States must also take into account the interests of the international community when taking action on a national level likely to affect others. This flows from of the basic norm of sovereign equality, according to which a State's sovereign rights are inherently limited by the equal sovereignty of other States.⁴⁸ As such, ‘[t]he existence of a number of sovereignties side by side places limits on the freedom of each State to act as if the others did not exist’.⁴⁹ From a more ‘humanity-oriented’ perspective, it could be argued that national regulators have sovereign ‘other-regarding’ obligations towards foreign individuals, where this may impact their right to self-determination.⁵⁰ The precise form of these obligations will depend on the applicable law.

International environmental law contains several due-diligence obligations which could be seen to operationalize a State duty to consider ‘others’ in relation to common environmental concerns. The first is the customary duty to ensure that activities conducted within a State's jurisdiction do not cause damage outside of their territories.⁵¹ Another relevant principle is that of precaution, which provides that in cases of ‘threats of serious or irreversible damage’, ‘lack of full scientific certainty’ shall not serve as a justification for failing to prevent environmental degradation.⁵² While the former has traditionally been applied to more isolated transboundary harm, it is now well accepted that the no harm principle covers more complex harms attributable to multiple States, a fact confirmed by its incorporation in the preamble of the UNFCCC.⁵³

In relation more specifically to climate change, the International Law Association (ILA) Draft Articles Relating to Climate Change provide that States have a due diligence obligation to ‘take all appropriate measures to anticipate, prevent or minimise the causes of climate change, especially through effective measures to reduce greenhouse gas emissions’.⁵⁴ Some authors suggest that this due-diligence threshold may go beyond the obligations contained in the Kyoto (and Paris) frameworks, as neither are currently sufficient to prevent a dangerous rise in the global temperature.⁵⁵ Indeed, the recently published ‘Oslo Principles’ go so far as to argue that States have a legal obligation to achieve set, per capita emission reductions.⁵⁶ This ‘permissible GHG quantum’ is calculated based on scientific data on the maximum amount of GHG that can ‘safely’ be emitted over a period of time.⁵⁷ Importantly, States’ respective obligations are to be adjusted according to the principle of common but differentiated responsibility (CBDR) so that more developed States shoulder a financial burden proportionate to their capabilities and historical contributions.⁵⁸ Notably, while the Oslo Principles purport to clarify existing legal obligations, they have not in themselves been incorporated into positive law.⁵⁹

The above analysis demonstrates considerable support for a sovereign duty to cooperate and consider the interests of other States and non-nationals in the context of common concerns. This extends to a due-diligence obligation to prevent and minimize their contribution to common environmental concerns, including the GHGs causing climate change. It would thus seem only consistent that States’ obligations to respect ‘others’ embedded in principles of public international and international environment law be reflected in the law of State jurisdiction. While this may seem intuitive, the relationship between States obligations and State jurisdictional rights to respond to common concerns is far from clear-cut. This is largely due to the fragmented and decentralized development of the law of jurisdiction, based on the Westphalian premise of independent States. With their high premium on sovereign independence, jurisdictional rules are designed to ensure reciprocal respect for territorial boundaries. Measures with ‘extraterritorial’ effects may threaten the freedom of other States and are suspected of jurisdictional overreach.⁶⁰ This is typical of measures targeting transboundary issues, particularly common environmental concerns, the EU maritime MRV being a key example.

Within the law of jurisdiction, ‘extraterritoriality’ remains a contested concept.⁶¹ In a neutral sense, it may be understood in relation to States or regional organizations, as ‘encompassing the area beyond its territory’.⁶² Views differ as to whether measures addressing extraterritorial conduct, such as emissions on the High Seas, are necessarily also ‘extraterritorial measures’ or an ‘exercise of extraterritorial jurisdiction’. Before proceeding, it is thus valuable to clarify our understanding of this issue within the broader context of customary international law.

C. ‘Extraterritoriality’ in the Law of State Jurisdiction

Extraterritorial jurisdiction has been defined broadly as ‘referring to the exercise of sovereign power or authority by a State outside of its territory’.⁶³ Here, a distinction must be drawn between prescriptive jurisdiction, which pertains to a State's competence to ‘to adopt legislation providing norms of conduct which govern persons, property or conduct’,⁶⁴ and enforcement jurisdiction, which describes a State's competence to ensure compliance with a State's laws.⁶⁵ While enforcement jurisdiction is strictly territorial, there are several recognized bases in international law for States to prescribe measures that extend beyond their territory.⁶⁶ There remains, however, considerable disagreement as to precisely when such a jurisdictional assertion should be characterized as ‘extraterritorial’, and the implications this has for a measure's legality.

A more traditional view approaches extraterritoriality through the lens of the classical bases of jurisdiction. To start with, while States can clearly regulate domestic conduct occurring entirely within their territory, in some circumstances they may also assert jurisdiction when a constituent element of an activity is only initiated or completed within their territory (subjective and objective territoriality respectively).⁶⁷ It is also generally accepted that States may legislate with regard to their nationals abroad (nationality principle) and to acts which threaten essential State interests (protective principle). In unique cases, States may also regulate ‘certain conduct committed by foreigners against foreigners, outside its territory and not implicating that State's essential interests’.⁶⁸ These are generally crimes of such a grave nature that all States have jurisdiction (universality principle). On this view, the customary bases of nationality, protection and universality are seen as ‘extraterritorial’ bases or principles, supporting jurisdictional assertions. Notably, these classical jurisdictional principles, while well-recognized in legal theory, are far from clear-cut and consistently applied in practice.⁶⁹ Today it is therefore generally accepted that what is required for a valid assertion of prescriptive jurisdiction is a ‘sufficient nexus’ or ‘substantial and genuine connection’ between the legislating State and the regulated subject matter at issue.⁷⁰ The classical principles can provide, either together or individually, evidence of such a link. The ‘substantial and genuine connection’ requirement will be discussed further below.

Another view is that measures taking into account foreign conduct are not ‘extraterritorial’ when there is a ‘territorial link’ or connection, even where this is very weak. Joanne Scott for example, argues that such measures are instances of ‘territorial extension’ of State or EU law, and can still be based on the territoriality principle.⁷¹ The controversial Air Transport Association of America (ATAA) case of the Court of Justice of the EU (CJEU) appears to support this conclusion.⁷² There, the Court was asked to determine whether the EU Aviation Directive (2008/101/EC) which expanded the EU Emissions Trading Scheme (ETS) to include aviation emissions was ‘extraterritorial’ and thus in violation of international law.⁷³ The territorial scope of the Aviation Directive is comparable to that of the present maritime MRV, taking into account carbon emitted throughout the entire duration of the journey, including above the High Seas and over the territory of other States. The Court did not consider this to give rise to extraterritoriality, holding that as the aircraft landing and departing from EU aerodromes were ‘physically in the territory’ of one of the Member States, they were subject to the ‘unlimited jurisdiction’ of that Member State and the Union.⁷⁴

Contrastingly, from yet another perspective, a measure is to be qualified as extraterritorial precisely because it addresses conduct outside of the territory of the forum State.⁷⁵ This approach differs on two key points from the findings in the ATAA case. Firstly, it draws a clearer distinction between prescriptive and enforcement jurisdiction, both of which require a valid basis under international law. Measures which condition the import or export of goods and services on conduct and circumstances abroad are concerned with extraterritorial conditions.⁷⁶ The legislation itself requires a basis under international law, the absence of which can constitute an international wrong.⁷⁷ The fact that such measures are only imposed at the border is a question of enforcement, which requires its own valid basis in international law, as well as a valid act of prescriptive jurisdiction. Secondly, measures which do take into account extraterritorial conduct or circumstances are not automatically considered illegal. This will depend on whether a State can demonstrate a ‘substantial and genuine connection’ with help from other bases of jurisdiction. This last approach will be taken here, however, to avoid confusion, this article uses the phrase measures with an ‘extraterritorial element’.⁷⁸

This raises the threshold question of when a measure contains a sufficient degree of external influence to constitute a relevant ‘extraterritorial element’, triggering questions of jurisdiction. This is especially important with regard to outward-looking market-based-measures, which do not, strictly speaking, force foreign actors to comply, but rather provide a (strong) incentive for them to do so.⁷⁹ While a full exploration of this discussion goes beyond the scope of this article, it is submitted, along the lines proposed by Bartels in the context of trade law, that a measure has a relevant extraterritorial element when its application is ‘objectively defined’ by foreign circumstances.⁸⁰ Thus, even where a measure has been framed around territorial conduct or presence (ie ‘territorialized’), such as presence in an aerodrome, the actual requirements to be met at that point relate to extraterritorial events. Notably, this ‘objective operation’ threshold does not encompass all external effects. It requires that the legislation contemplate and condition, either directly or indirectly, specific extraterritorial events. This is consistent with the sovereignty-oriented frame, as economic pressure can be just as compelling as physical pressure in shaping the behaviour of foreign actors. In fact, economic pressure, when applied over time, can lead to even greater, more structural changes in the activities of foreign actors. It is irrelevant in this regard, whether the financial pressure takes the form of a sanction or neutral trade restriction.⁸¹

Section II has sought to clarify the concepts of ‘common concerns’ and ‘extraterritoriality’ in international law. These issues are related, as measures taken in response to common concerns tend to contain extraterritorial elements to bolster their effectiveness. Doctrinally however, there remains a conflict of paradigms between community-oriented common concerns managing interdependence, and the State-centric law of jurisdiction promoting independence. For States this may lead to a problematic gap between their obligations to respond to common concerns and their jurisdictional rights to do so. Still, these two frameworks need not be incompatible. It will be shown that the jurisdictional rules applicable here can and do accommodate measures protecting the climate as a common concern. Turning to the case at hand, Section III will first explore the ‘extraterritorial elements’ of Regulation 2015/757 giving rise to the need for jurisdictional justification.

III. PROVOCATIVE CLIMATE PROTECTION: THE ‘EXTRATERRITORIAL ELEMENTS’ OF THE EU'S MARITIME MEASURES

The EU has made no secret of its plans to go it alone. In 2009, frustrated by the slow pace of international negotiations, it notified the international community that should the IMO fail to reach an international agreement by 2011, it would proceed to include maritime emissions in its own community target.⁸² By this time, the IMO had made some progress, amending MARPOL Annex VI to include the binding Ship Energy Efficiency Management Plan (SEEMP) and Energy Efficiency Design Index (EEDI) for new ships.⁸³ However, the EU's environmental impact assessment predicted that even with these measures, EU-related emissions would increase more than 50 per cent by 2050 compared to 1990-levels.⁸⁴ What was needed was an instrument to cut emissions for the international maritime transport sector as a whole.⁸⁵ The present maritime monitoring, reporting, and verification regulation (MRV) signals the first step, providing the data necessary for the operation of a market-based-measure (MBM) for maritime emissions.⁸⁶ Notably, the MRV itself is expected to achieve up to two per cent emission reductions as a freestanding measure.⁸⁷ Regulation 2015/757 and the planned MBMs will now be examined in more detail.

A. Regulation 2015/757

Regulation 2015/757 applies to all ships above 5000 gross tonnage involved in commercial transportation, ‘regardless of their flag’.⁸⁸ ‘Companies’ responsible for the ships’ operation are required to monitor the CO₂ emitted during voyages to, between, and from ports under the jurisdiction of an EU Member State as of 2018.⁸⁹ This must be done using approved technologies in accordance with one the of four standard monitoring methods listed in Annex I.⁹⁰ Companies must provide the Commission with a ‘monitoring plan’ by 2017, detailing how they will meet these requirements (Article 13). As of 2019, companies must then provide emissions reports to the Commission and the concerned flag State authorities, documenting all required information for each ship during the previous reporting period (Article 11(1)). ‘Verifiers’, appointed by national accreditation authorities, must approve both the monitoring plans and resulting reports before they can be sent to the Commission.⁹¹

Ships are obliged to carry a valid ‘document of compliance’ evidencing that they have met these conditions when arriving at, within, or departing from a port under EU jurisdiction.⁹² Each EU Member State is further instructed to take ‘all the measures necessary to ensure compliance of ships flying its flag’ (Article 19(1)). Article 19(2) requires Member States to ensure that ‘any inspection of a ship in a port under its jurisdiction’ is done in accordance with Directive 2009/16/EC on Port State Control Procedures. Importantly, Article 20(1) requires Member States to set up and impose an ‘effective, proportionate and dissuasive’ penalty system for ships that fail to comply. As noted by Churchill, while one would expect flag States to impose sanctions, this provision does not preclude, and in fact appears to require, port State action, also in relation to ships of non-EU nationality.⁹³ Member States of the port of entry may further issue an expulsion order for ships failing to comply for two or more consecutive reporting periods (Article 20(2)).

On the face of it, the Regulation only attaches sanctions to a failure to produce a certificate of compliance when in EU territory. Indeed, Marten argues that such an MRV concerns the territorial provision of information, which is not ‘stamped’ with the location where it came into being.⁹⁴ This distinction may not, however, be a reliable way of signalling suspect, ‘extraterritorial’ measures. Provisions formally framed as territorial requests for information, may, de facto, operate as extraterritorial conduct requirements.⁹⁵ For example, in the context of the present MRV, companies will only be able to prove compliance if they have abided by EU information collection methods when abroad. This constitutes a relevant extraterritorial element, triggering the need for the EU to demonstrate a ‘substantial and genuine connection’ to the regulated subject matter.

B. Possible Market-Based Measures

As discussed, the MRV is only the first step towards a further MBM aimed at harmonizing the objectives of profit-making and pollution reduction.⁹⁶ At the time of writing, the EU is yet to decide on the design of such an MBM but in its 2013 Communication (COM (2013) 479) it announced its preferred options. The first is a maritime ETS (METS) similar to that envisioned for aviation emissions.⁹⁷ Under this cap-and-trade scheme, operators are allotted a set amount of carbon credits based on specific emission reduction targets. Less energy-efficient operators whose emissions exceed their allotted targets are then forced to buy credits from other companies with a surplus. A maritime ETS with the same scope as the MRV would include in its calculations CO₂ emitted outside EU territory. As the measure imposes a cumulative cost on foreign emissions, it contains a clear extraterritorial element.

A second possible MBM is a ‘target-based compensation fund’ which would set the same target as a METS for an entire fleet, to be monitored by a ‘sector-wide entity’.⁹⁸ This entity would have an obligatory contractual relationship with the fleets, who would have to pay excess emissions charges and a membership fee into the fund, to be invested in energy efficiency.⁹⁹ Where the specific targets take into account the total distance travelled to or from EU ports, this would present similar jurisdictional concerns as a maritime ETS.¹⁰⁰ As was the case with aviation emissions, it is likely that the Union will meet greater resistance when it comes to implementing the further stages of its plan. In order to stand up to political criticism, it is important that the design of the final MBM be compliant with international law. The following section will therefore consider the more specific legality issues under the LOSC and relevant world trade law.

IV. PERMISSIBILITY OF THE MRV AND MARKET-BASED MEASURES UNDER THE LAW OF THE SEA CONVENTION AND WORLD TRADE LAW

The EU's well-known maritime plans have already sparked discussion amongst scholars and research institutions.¹⁰¹ There, the primary focus has been on the permissibility of a possible maritime ETS under the LOSC and WTO law, and views have been divided.¹⁰² This section seeks to demonstrate that not only a maritime ETS, but also the present MRV and the possible target-based compensation fund are, in principle, permissible under both the maritime and trade regimes. However, it will also be shown that neither regime definitively answers the question of whether the EU has exceeded the limits of its jurisdiction. For this, we are directed towards customary international law, an area that has received less structured attention to date, and will thus be the focus of Section V.

A. Permissibility under the Law of the Sea Convention

As alluded to above, there is some disagreement as to whether the Law of the Sea Convention (LOSC) permits the EU's measures, particularly a maritime ETS. Those against emphasize the tensions with the freedom of navigation. They consider an METS to be an exercise of ‘extraterritorial’ rather than port State jurisdiction, requiring an explicit, alternative basis under the law of the sea.¹⁰³ The argument then goes that there is no such explicit basis and, moreover, the EU measures would conflict with Article 92 LOSC, which enshrines the principle of exclusive flag State jurisdiction over ships on the High Seas. An METS would violate this principle, as the EU Member States are regulating activities on the high seas of vessels not flying their flag.¹⁰⁴ For this reason, the EU measures would also violate Article 89 LOSC which provides that no State may purport to subject the High Seas to its sovereignty. These authors may also point to Article 218 LOSC, which provides that port States may not institute proceedings pertaining to discharges occurring outside of their territorial waters or EEZ other than those ‘in violation of applicable rules and standards’, and thus arguably implies a contrario, that States or the EU are not competent to regulate other conduct on the high seas, let alone conduct that is not in violation of international standards.¹⁰⁵

This article supports the alternative view that requiring an explicit basis in the LOSC does not sufficiently consider the residual jurisdiction of States and regional organizations to set conditions for entry into their ports, which are, after all, part of their territory.¹⁰⁶ Under international law ships have no general right of access to foreign ports.¹⁰⁷ Article 25(2) LOSC further provides that the coastal State has the right to take the necessary steps with respect to ships proceeding to internal waters to prevent any breach of the conditions to which the admission of those ships is subject.

Moreover, UNCLOS implicitly acknowledges port States’ competence to regulate environmental matters, without providing a territorial limit. Article 211(3) LOSC recognizes the competence of coastal States (which port States also qualify as) to ‘establish particular requirements for the prevention, reduction and control of pollution of the marine environment’.¹⁰⁸ Maritime CO₂ emissions can be considered ‘pollution’, as climate change and air pollution harm marine life and human health.¹⁰⁹

In assessing port State jurisdiction to prescribe such measures, a distinction can be drawn between ‘static’, construction, design, equipment and manning (CDEM) standards and measures regulating operational pollution.¹¹⁰ Both the MRV and envisioned MBMs implicitly require vessels to be fitted with the equipment necessary for monitoring CO₂ emissions, and thus contain CDEM standards.¹¹¹ The requirements regarding the monitoring itself, as well as the reporting and attainment of verification and compliance documentation pertain to foreign conduct. While several IMO Conventions, in particular MARPOL, set CDEM standards for pollution prevention,¹¹² Article 211(3) does not specify whether port State pollution measures are limited to those international standards.¹¹³ Furthermore, the IMO Conventions themselves do not contain relevant prohibitions on higher unilateral standards.¹¹⁴ No explicit territorial limits can be found in the LOSC as regards pollution reduction measures regulating vessel operation prior to port entry.¹¹⁵

Notably, it has been argued that unlike operational measures, CDEM standards are in fact territorial, as they are not primarily aimed at regulating vessels outside a State's territory.¹¹⁶ However, as vessels will have to comply with construction standards before arrival at port, there remains an ‘extraterritorial element’ to be considered in light of customary international law.

By not geographically limiting the marine environment that States can protect, the LOSC accommodates State measures taken to remedy environmental degradation. Indeed, Article 211(2) LOSC supplements this by requiring States to ‘adopt laws and regulations for the prevention, reduction and control of pollution of the marine environment from vessels flying their flag’. These laws must have ‘at least’ the ‘same effect as generally accepted international standards’.

Furthermore, the principle of exclusive flag State jurisdiction does not preclude port entry conditions. While still a point of discussion, the systemic logic and drafting history of the LOSC arguably indicate that Article 92 LOSC is concerned with enforcement rather than prescriptive jurisdiction for activities on the High Seas.¹¹⁷ For its part, the EU MRV instructs Member States to ensure compliance only of vessels flying their flag (Article 19(1)), and companies are only required to submit emission reports to the ‘flag States concerned’ (Article 11(1)). This is without prejudice to the port State competence to require a document of compliance upon entry (Article 19(2)), and sanction non-compliance (Article 20(1)(3)).

Arguably therefore, the LOSC allows States to legislate port entry conditions on prevention reduction and control of polluting activities, including activities contributing to climate change, which occur outside of their territories. These conditions remain subject to the reasonableness requirements of proportionality, non-discrimination (Article 227 LOSC), and non-abuse of rights (Article 300 LOSC), discussed in Section VB.

These diverging views illustrate the conflict of interests between the freedom of navigation and unilateral pollution reduction by port States. The LOSC sets out a general framework balancing these interests, with exclusive competences in territorial waters, and greater freedoms on the High Seas. However, precisely at the point of contact where foreign vessels enter territorial waters, the Convention says little, leaving open the question of the territorial limits of port State jurisdiction. Here, customary international law has an important role to play in clarifying the limits of States’ regulatory competence.¹¹⁸

B. Permissibility under World Trade Law

The EU measures must also pass the hurdles posed by the WTO Agreements governing barriers to trade.¹¹⁹ The MRV as well as both MBMs would raise tensions under the General Agreement on Tariffs and Trade (GATT) as they may unjustifiably erect obstacles to the international trade in goods.¹²⁰ They may also face issues under the General Agreement on Trade in Services (GATS), where States have made commitments regarding access to and use of port facilities and maritime transport services.¹²¹ It has further been submitted that a METS could constitute a ‘technical regulation’ under the Technical Barriers to Trade (TBT) Agreement.¹²² To date, however, it remains unclear whether non product-related process and production methods (npr-PPMs) such as those in an ETS, ‘lay down product characteristics or their related processes or production methods’ (Annex 1.1), falling within the scope of the TBT Agreement.¹²³ As the processes in the MRV and MBMs pertain to the vessels’ operation irrespective of their goods, it seems unlikely that they have a ‘sufficient nexus’ with the products on board as required by the Appellate Body (AB) in EC–Seals. ¹²⁴ The following section therefore does not address TBT, but limits its analysis to the GATT and GATS, which would otherwise continue to apply.

Looking in more detail at the possible violations, a distinction must be drawn between the target-based compensation fund as a fiscal measure¹²⁵ and a maritime ETS (METS) which is arguably a non-fiscal measure.¹²⁶ The compensation fund imposes costs on both EU and foreign fleets through membership fees and excess emission payments, the proceeds of which go to the EU.¹²⁷ These costs are reflected in the price of the transported products and related services, affecting the conditions of competition. As such, the measure could be characterized as an internal charge on all products that have travelled by ship, triggered by the internal sale of the goods (Article III:2 GATT).¹²⁸ Presumably, the membership fee will be the same for all EU and non-EU fleets and therefore in compliance with the national treatment (NT) requirement in Article III:2 GATT, and the ‘most favoured nation’ (MFN) requirement in Article I:1 GATT. However, assuming the fleets’ emission reduction targets do not take into account their distance from the EU, the payments for excess carbon emissions would likely lead to problems. This is because companies transporting ‘like’ products from further away will be faced with higher compliance costs than those from closer-by, particularly EU Member States, in violation of Article III:2 GATT.¹²⁹ This differentiation in compliance cost according to port of origin therefore also violates Article I:1 GATT's MFN principle, failing to accord the same ‘advantage’ to like products of all States.¹³⁰

An ETS, on the other hand, can arguably be characterized as a regulation rather than a tax. This is because operators receive something of value (carbon credits) for the price they have to pay, the cost of which is set by the market rather than the regulator.¹³¹ Again, the ETS would require operators from further away to surrender more credits than those closer by. Where a METS is construed as a border measure triggered by virtue of importation,¹³² this would likely violate the prohibition of quantitative restrictions in Article XI GATT. The higher cost for imports travelling from further away has the effect of altering the conditions of competition for like products from different origins.¹³³ Alternatively, an ETS could be construed as an internal regulation, as it also applies to intra-EU voyages, and is triggered by the internal sale of the imported goods.¹³⁴ A METS is then not immediately prohibited, but has to comply with the NT and MFN requirements in Articles III:4 and I:1 GATT respectively.¹³⁵ However, for the same reasons as with the compensation fund, an ETS differentiating compliance costs according to origin would likely violate the above-mentioned provisions.¹³⁶ The MRV itself could also be characterized as an internal regulation in the sense of Article III:4 GATT. As the standardized monitoring and reporting procedures do not differentiate compliance costs according to a product's origin, the measure appears to comply with Articles III:4 and I:1 GATT.

The EU instruments also constitute ‘measures affecting trade in services’ according to Article I:1 GATS.¹³⁷ Again, a compensation fund and ETS differentiating amongst ‘like’ services according to distance travelled from the EU would likely modify the conditions of competition in favour of EU-flagged vessels, violating the national treatment requirement in Article XVII GATS.¹³⁸ They would also fail to accord ‘treatment no less favourable’ (lowest possible compliance cost) to all like services and service suppliers, in violation of the MFN requirement in Article II GATS.¹³⁹

Nevertheless, the WTO agreements contain well-recognized ‘windows’ for the consideration of non-trade interests, even when these have extraterritorial effects.¹⁴⁰ These can be found in Articles XX GATT and XIV GATS, which contain exemptions for presumed violations which seek to serve certain public interest objectives. The following analysis focuses on GATT Article XX, as it has received the most interpretation from the WTO Dispute Settlement Body, its text and rationale running largely parallel with that of the GATS.¹⁴¹ Section III.1 therefore examines the possibility for exemption offered by the environmental policy objectives in paragraph (b) (the ‘protection of human, plant and animal life and health’), and paragraph (g) (the ‘conservation of exhaustible resources’) as conditioned by the requirements of the chapeau.¹⁴²

1. Article XX(b) GATT: ‘necessary for the protection of human, plant and animal life or health’

To provisionally qualify under Article XX(b), the maritime MRV and possible MBMs must pursue the objective of protecting human plant and animal life or health, and be ‘necessary’ for its achievement. The EU’s well-documented policy objectives make clear that the measures are intended to reduce international maritime emissions for the benefit of the global climate and air quality.¹⁴³ Based on IPCCC findings, the EU considers climate change mitigation instrumental to the protection of human and animal life and health.¹⁴⁴ As regards more narrowly to air quality, in US–Gasoline (AB) the AB found that reducing air pollution is a policy objective aimed at protecting life or health.¹⁴⁵

The next issue is whether the measures are ‘necessary’, in the sense that they make a ‘material contribution’ to the achievement of their objective in the absence of an alternative, less trade restrictive measure.¹⁴⁶ In Korea–Beef, the AB found that the ‘relative importance’ of the common interests or values that the measure purportedly serves may be considered.¹⁴⁷ The more vital or important the common interests, the easier it would be to accept the measure as ‘necessary’. Furthermore, in Brazil–Tyres it was recognized that ‘certain complex public health or environmental problems may be tackled only with a comprehensive policy comprising a multiplicity of interacting measures. […] Moreover, the results obtained from certain actions—for instance, measures adopted in order to attenuate global warming and climate change, […] —can only be evaluated with the benefit of time.’¹⁴⁸

According to the EU's impact assessment, both the MRV and planned MBMs can provide considerable GHG reductions compared to a business as usual scenario, evidencing a ‘material contribution’ to climate change mitigation.¹⁴⁹ As mentioned, the MRV alone is expected to provide up to two per cent in annual GHG emission reductions and €1.2 billion net industry savings on fuel bills.¹⁵⁰ Furthermore, a METS would be capable of achieving up to 21 per cent reduction in GHG emissions compared to the baseline scenario of only the existing instruments.¹⁵¹ For a target-based compensation fund, the predicted reduction is 16 per cent.¹⁵²

Notably, the issue of carbon leakage poses a threat to the effectiveness of climate-related regulations, and it has been argued that lower carbon prices outside the EU may lead businesses to relocate.¹⁵³ Yet avoiding carbon leakage is in fact a key reason for the expansive territorial scope of the EU's policy. If only intra-EU voyages were covered, vessels from outside of the EU would have a competitive advantage, which would be reflected in consumer prices. This would increase the consumption of cheaper, more CO₂ intensive products and services, defeating the climate protection objective of the EU measure.¹⁵⁴ There is a risk, however, that imposing costs on foreign producers may favour EU actors in a way that allows them to increase their own production of CO₂. This was discussed in relation to border carbon adjustment taxes by Colares and Rode, who allude to the danger that EU lobbying efforts aimed at levelling the playing may result in the economically suboptimal imposition of tariffs or other trade restrictions.¹⁵⁵ While this could certainly decrease the measures’ ‘material contribution’ to climate change mitigation, these negative effects could be avoided by careful design of the tariff and scope of the envisioned MBMs.¹⁵⁶

There must also be no less trade-restrictive alternative reasonably available to the proposed MBMs.¹⁵⁷ In EC–Asbestos, the AB held that WTO members cannot reasonably be expected to adopt an alternative measure that would not allow them to achieve their desired level of protection.¹⁵⁸ In 2009, the IMO recognized that technical and operational measures alone ‘would not be sufficient to satisfactorily reduce the amount of GHG emissions’ and an ‘overwhelming majority’ considered an MBM as a necessary supplement.¹⁵⁹ Furthermore, an ETS is not by nature overly trade restrictive, as it has been recognized in Article 17 of the Kyoto Protocol as a legitimate means of limiting emissions.¹⁶⁰ This evidences against the availability of a less trade-restrictive alternative.

The third element of necessity recognized in Korea–Beef, namely the ‘relative importance’ of the objective, is particularly relevant here.¹⁶¹ As observed by Marceau and Trachtman, Korea–Beef broadened the necessity test to weigh the actual importance of the national (or EU) regulatory values against the trade values.¹⁶² This accommodates common concern protection by effectively lowering the ‘necessity’ threshold in Article XX(b). Applied to the present case, there can be no doubt that the mitigation of climate change is a ‘vital’ common interest meriting protection.¹⁶³ Brazil–Tyres further indicates leeway in the time permitted to judge the effectiveness of the EU's mitigation measures.¹⁶⁴

2. Article XX(g) GATT: ‘relating to the conservation of exhaustible natural resources’

The EU maritime measures may also be exempted under Article XX(g), which captures measures aimed at the conservation of exhaustible natural resources, as long as they ‘relate to’ that objective and are made effective in conjunction with domestic restrictions.¹⁶⁵ A leading case here is US–Shrimp, which concerned a US import ban on shrimp and shrimp products from countries that had not used certain turtle-friendly fishing nets.¹⁶⁶ There, the AB held that ‘the term “exhaustible natural resources” must be read […] in light of contemporary concerns of the community of nations about the protection and conservation of the environment.’¹⁶⁷ Considering the sustainable development objective embodied in the WTO Agreement, it went on to find that ‘the generic term ‘‘natural resources’’ in Article XX(g) is not ‘‘static’’ in its content or reference but rather ‘‘by definition, evolutionary’’’.¹⁶⁸ Here too, Article XX is thus interpreted flexibly to allow States to respond to emerging scientific evidence on common environmental concerns. Such an approach is not only for the benefit of a single State's regulatory autonomy, it also facilitates the protection of ‘higher’ shared interests.

In addition, the fact that a resource may be ‘renewable’ does not mean that it is not ‘exhaustible’.¹⁶⁹ This was confirmed in US–Gasoline (Panel) where the Panel found clean air to be an exhaustible natural resource as it can be depleted by means of pollution from fuel combustion.¹⁷⁰ It has been argued that by analogy, a ‘stable atmosphere’ is an exhaustible natural resource, as the necessary composition of gases is also disrupted by CO₂ emissions from fuel combustion.¹⁷¹ Furthermore, the EU's above-mentioned air quality objective indicates that the envisioned measures would also fall under a narrower reading of the provision.

Under Article XX(g), a measure must also ‘relate to’ its objective. This pertains to the ‘relationship between the general structure and design of the measure’ and ‘the policy goal it purports to serve’.¹⁷² What is required is ‘a close and genuine relationship of the ends and means’, where the rules are not merely ‘incidentally or inadvertently’ aimed at conservation.¹⁷³ As discussed, the MRV and MBMs would be primarily directed at reducing CO₂ emissions in order to mitigate climate change, and are capable of achieving significant reductions. The arguments supporting necessity are sufficient to meet the ‘related-to’ requirement here.

The above analysis demonstrates that the EU METS and planned MBMs could be provisionally justified under Article XX(b) and (g) GATT, both of which accommodate the protection of common concerns. However, the actual application of these measures must meet the stringent requirements of the chapeau, which will now be discussed in more detail.

3. Article XX GATT chapeau

The chapeau of Article XX GATT contains additional obstacles for the application of prescriptive measures provisionally justified under the provision's sub-paragraphs. Such measures must not constitute ‘arbitrary or unjustifiable discrimination’ or a ‘disguised restriction on international trade’. According to the AB in US–Shrimp, ‘the chapeau of Article XX is, in fact, but one expression of the principle of good faith’, prohibiting the abuse of rights and requiring States to exercise their rights ‘reasonably’.¹⁷⁴

To avoid ‘arbitrary or unjustifiable discrimination’, States must make serious best-efforts to reach a multilateral solution, before resorting to unilateral action.¹⁷⁵ In US–Shrimp (Article 21.5) the AB held that while a multilateral approach would be ‘preferred’, the actual conclusion of an agreement is not required.¹⁷⁶ In the present case, EU Member States have been lobbying for the regulation of maritime emissions in the IMO since 2009, submitting various ETS proposals.¹⁷⁷ By the MEPC68 meeting in May 2015, these repeated efforts had still achieved no results, the Committee deciding once again not to set a quantifiable reduction target for maritime emissions.¹⁷⁸ Even after issuing the MRV, the EU continues to stress its preference for a multilateral solution, noting that ‘[t]he likely timeline for the Regulation's adoption leaves ample opportunity for the IMO to make progress before the EU rules come into force’.¹⁷⁹

In US–Shrimp, the AB further found that a measure will not constitute arbitrary or unjustifiable discrimination in violation of the chapeau of Article XX if it is sufficiently flexible so as not to effectively require ‘all other exporting Members […] to adopt essentially the same policy’.¹⁸⁰ The measure must also ‘allow for comparably effective third country measures’, taking into account ‘the specific conditions prevailing in its territory’.¹⁸¹ With regard to flexibility, the MRV recognizes four different types of recording methods, allowing operators to choose a system best fitted to, or already present on their ships. The EU plans to accommodate third country conditions through its ‘staged approach’, in which it first implements the MRV before ‘pricing the emissions at a later stage’.¹⁸² This ‘facilitates the making of significant progress at international level’, allowing the final market-based measure to be better suited to comparable third country measures.¹⁸³ It remains to be seen whether the EU will include recognition for comparably effective measures in its chosen MBM. If it chooses an ETS, it may follow the approach taken in the Aviation Directive, which allowed for the recognition of third country policies where they adopt ‘measures that have an environmental effect at least equivalent to that of this Directive’.¹⁸⁴

Notably, in the field of climate change, consideration for third country conditions must also take into account the CBDR principle, which requires States to divide the cost of mitigation according to their respective capabilities.¹⁸⁵ Applied here, this arguably means that where a developing State has, for capacity reasons, chosen to implement less stringent climate protection standards, its measures should not be automatically rejected by the EU. The approach in the Aviation Directive that only recognizes measures with ‘equivalent’ environmental effect, may thus not take due consideration of the circumstances in the exporting country.

In Brazil–Retreaded Tyres it was further found that any discriminatory effects of the measure must be ‘explained by a rationale that bears … [a] relationship to the objective of a measure’.¹⁸⁶ Here, the higher costs allotted to operators further away is related to the fact that they travel further and therefore emit more. Finally, it has been suggested that the reasonableness requirement also obliges the EU to take into account and deduct any double taxation should the emissions already be subject to another regime.¹⁸⁷ This is something that the EU will have to consider when further designing its chosen MBM.

4. An implied jurisdictional limitation in Article XX GATT?

The analysis above indicates that the EU maritime measures could provisionally be justified within the WTO framework. However one key issue remains to be considered, namely whether Article XX GATT contains an implied jurisdictional limit that would exclude from its scope measures with an ‘extraterritorial element’ like those at issue here. Such measures have traditionally been frowned upon under world trade law, although over time the WTO has gradually taken a more permissive stance towards measures serving a recognized public interest. Thus in US–Tuna (1994), the Panel found that the GATT could, in principle, apply to policies ‘relating to things or actions outside the territory of the regulating State’.¹⁸⁸ However, while the Panel found that Article XX(g) could apply to conservation measures for resources outside a State's jurisdiction,¹⁸⁹ it held that where the measure coerced other States to change their policies within their own jurisdictions, it could never be ‘necessary’ within the meaning of Article XX(b).¹⁹⁰

This finding was reversed in US–Shrimp (1998), where the AB held that ‘requiring from exporting countries compliance with […] certain policies’ does not render a measure ‘a priori incapable of justification under Article XX’.¹⁹¹ ‘Such an interpretation renders most, if not all, of the specific exceptions of Article XX, inutile.’¹⁹² It did, however, require a ‘sufficient nexus’ between regulating State and the objective pursued by the measure, in order for that State to claim jurisdiction.¹⁹³ There the US needed to prove its connection with migratory sea turtles in order to be eligible for exemption under Article XX(g). The AB found that although the sea turtle species were ‘highly migratory animals’, the fact that they were known to occur within US territorial waters provided a ‘sufficient nexus’ between the endangered population and the US. Notably, however, this finding was limited to the specific circumstances of the case and the AB explicitly declined ‘to pass upon the question of whether there is an implied jurisdictional limitation in Article XX(g)’.¹⁹⁴

To date the question of whether there is an implied jurisdictional limit in Article XX remains unclear.¹⁹⁵ In line with Article 3.2 DSU, we are left with Articles 31 and 32 of the Vienna Convention on the Law of Treaties (VCLT) for interpretative assistance. Neither the object and purpose (Article 31(1) VCLT) nor the travaux préparatoires (Article 32 VCLT) of the GATT provide much clarity on the issue.¹⁹⁶ It is argued that, based on Article 31(3)(c) VCLT, other ‘relevant rules of international law’ applicable between the parties should be relied on here. Neither the UNFCCC nor the Kyoto and Paris agreements provide much clarity on the territorial limits of unilateral emission reduction policies. However, as the question pertains to jurisdiction, the most ‘relevant’ rules are those of customary international law of State jurisdiction.¹⁹⁷ Unlike MEAs, these rules apply between all WTO members and can also be applied to the non-environmental exemptions of Article XX.¹⁹⁸ As with the LOSC we are once again directed to customary international law to complete the jurisdictional analysis. Section V examines this in more detail, exploring the customary bases and limitations for State jurisdictional rights to respond to common concerns.

V. PROTECTING COMMON CONCERNS UNDER THE CUSTOMARY INTERNATIONAL LAW OF STATE JURISDICTION

We have seen that there is compelling legal support for the argument that States (and regional organizations) have a duty to respond to common concerns through cooperation and ‘other-regardingness’ (Section IIB). The latter more specifically includes States’ due-diligence obligation to anticipate, prevent and minimize their contribution to common environmental concerns. The above analysis has further shown that legislative measures aimed at doing so are at least accommodated, and arguably supported, by both the LOSC and world trade law (Section IV). Still, a jurisdictional analysis of both regimes is not complete without taking into account customary international law. From a doctrinal perspective, it is this regime that is fundamentally tasked with appropriately delimiting sovereign powers, taking into account both the right to independence and the realities of interdependence. Section VA explores how this balance could be struck in the relatively unchartered territory of common concerns. The overarching jurisdictional limitations are considered in Section VB.

A. Common Concerns and the ‘Substantial and Genuine Connection’ Requirement

From a jurisdictional perspective, measures with an ‘extraterritorial element’ are suspect, as they risk infringing upon the freedom of other equally sovereign States. As we have seen, it is now generally accepted that such measures can be justified where a State or regional organization can demonstrate a ‘substantial and genuine connection’ to the regulated subject matter.¹⁹⁹ As one of the founding fathers of this theory, FA Mann defined this requirement as a ‘meaningful connection’, to be determined by a ‘weighing of legally relevant elements’ to ‘justify, or make it reasonable for, a State to exercise legislative jurisdiction’.²⁰⁰ According to Mann, ‘merely’ an economic, political, commercial, or social interest is not sufficient to support the connection.²⁰¹ Today, this emphasis on the ‘legal’ nature of the connection seems falsely abstract. As argued by Bartels, where State practice and opinio juris demonstrate that certain interests merit the exercise of jurisdiction, then a customary rule develops to support this.²⁰² Essentially, this test for a lawful exercise of jurisdiction entails a balance of interests to determine, in a specific case, whether one State's interest can legitimize limiting the legislative freedom of another State.²⁰³

How then must one go about demonstrating such a ‘substantial and genuine connection’? Traditionally this is done using the five prescriptive bases of territory, nationality, protection, effects and universality discussed in Section IIC.²⁰⁴ However, as noted by Kamminga, the traditional bases ‘do not provide a coherent and straightforward model’ to authoritatively determine the legality of an exercise of prescriptive jurisdiction.²⁰⁵ While well recognized in theory, the fragmented State practice provides limited support in the relatively new field of common environmental concerns. For this reason, while the following section channels the analysis through the traditional bases, it seeks to combine their Westphalian origins with the realities of common environmental concerns.

As we have seen in the ATAA case, on one view, physical presence, such as that of vessels in EU ports, provides ‘unlimited jurisdiction’ based on the territoriality principle.²⁰⁶ However, as already noted, this appears to blur the conceptual distinction between prescriptive and enforcement jurisdiction (Section IIC). It also effectively allows all import and export measures an unlimited territorial scope. This approach thus permits powerful States to exercise unbridled market power, threatening the sovereign independence its proponents claim to protect. Finally, the international reaction to the CJEU's use of this reasoning illustrates that this is not a politically tenable interpretation of the substantial connection requirement. We therefore argue that States must do more than simply demonstrate a formal territorial connection to justify market entry conditions that take into account conduct abroad. While presence may be a sufficient trigger to justify enforcement, the threshold for demonstrating a legitimate interest in legislating is higher.

The nationality principle is of little help here, as the EU measures also cover non-EU flagged vessels. The protective principle, allowing for regulation of activities threatening a vital national interest, could conceivably cover action causing grave environmental harm.²⁰⁷ However, this is difficult to apply to situations with such diffuse causality as climate change, and the threshold for what constitutes a ‘threat’ is furthermore a high one.²⁰⁸

We therefore turn to effects-doctrine, which provides a basis for jurisdiction over conduct by foreign nationals occurring outside of its territory where it has a ‘substantial effect’ within its territory.²⁰⁹ The validity and scope of the effects-doctrine is controversial, however. While some view it as an extension of the objective territoriality principle, it is well accepted in the United States as an independent basis of jurisdiction,²¹⁰ and can be found in section 403(2) of the Third Restatement of the Foreign Relations Law of the United States (US Third Restatement).²¹¹

Accepting that effects could provide a ground for jurisdiction, we arrive at the threshold issues of the requisite causal link and the scale of the effects. Turning first to causality, some argue that the effects-doctrine should be interpreted strictly, and focus, like the traditional principle of harm prevention, on delineated damage with a direct causal link to another actor. According to this view, the EU must be able to identify ‘a specific environmental outcome’ in its territory caused by GHG emissions from ships.²¹² However, this focus on specific, direct harm does not fully comprehend the complex processes involved in climate change. The fact that cause (maritime and emissions) and effect (climate change) cannot be isolated to one specific environmental outcome does not mean that there is no causality.²¹³

A broader take on the effects-doctrine recognizes the indirect effects felt globally using a process-based understanding of anthropogenic interference with the climate system. As discussed, CO₂ emissions cause an imbalance in the atmospheric concentration of greenhouse gases that leads to climate change. In turn, climate change leads to the destruction of ecosystems, extreme weather and other environmental harm occurring in all States, including the EU.²¹⁴ Exploring this issue in the field of State responsibility, Peel considers that GHG emitted by a State could constitute the ‘proximate cause’ of environmental injury (effects), where it makes a ‘meaningful or material contribution to the risk of climate change harm’.²¹⁵ On one view, each cumulative tonne of GHG makes a ‘material contribution’ to the risk. Yet in the context of responsibility this would resemble a strict liability test, ‘at odds’ with the primary due-diligence obligations in the field of climate change. An alternative approach based on the precautionary principle is thus preferred.²¹⁶ Notably, in the context of jurisdiction, causality serves to demonstrate an interest in legislating to mitigate harm, rather than the liability of another State for causing harm. This supports rather than conflicts with States’ due-diligence obligations. The ‘material contribution’ threshold is then better framed in terms of degree, where the activity regulated must either immediately or over time make an objectively provable contribution to the harm. Scientific and economic evidence is particularly helpful in this respect.²¹⁷

The next threshold issue for effects-based jurisdiction is the scale of the effects themselves. It seems a logical starting point that the greater the effects, the greater a State's claim to a genuine connection. When considered from the perspective of a single State this is hardly controversial, and indeed appears reflected in the requirement of ‘direct and substantial effects’ well rooted in both US and EU antitrust law.²¹⁸ While there remains little legal clarity on what this means in practice,²¹⁹ one could safely assume that the irreversible environmental harm caused by climate change would meet this threshold. This arguably applies more generally to all common concerns, which by definition have been recognized by the international community to be grave enough to require action. Accepting this reasoning, based on the territorial effects alone, the EU arguably has a sufficient nexus to regulate global maritime emissions and enforce this regulation in its ports.²²⁰

However, the analysis of effects-based jurisdiction for common concerns does not end here. It is further submitted that the gravity of the effects should not only be measured in terms of local intensity, but also in terms of total territorial scope. Thus, where a State is legislating to mitigate harmful effects felt globally, this should lend weight to its claim to a ‘substantial and genuine connection’ to the subject matter. This is a way of representing the interests of the broader international community using the State-centric tools of the law of jurisdiction.²²¹ Indeed, from both a practical and normative perspective it is necessary to consider the legitimate interests of other effected actors, rather than artificially limiting the analysis to the acting State.²²² Support for this can be found in the section 403(2) of the US Third Restatement, which provides that States must consider, inter alia, the importance of the regulation to the international political, legal, or economic system and the regulation's consistency with the traditions of the international system. Fundamentally, this approach accommodates the jurisdictional rights necessary for States to fulfil their obligations to respond to common concerns, improving the coherence between different legal regimes.

The consideration of community interests as a prescriptive basis is not alien to the law of jurisdiction. Indeed, it is the very rationale underlying the universality principle, which provides for jurisdiction with respect to certain grave crimes under international law, without any link or nexus to the legislating State.²²³ There, the intrinsic gravity of the crime gives all States an interest in legislating.²²⁴

It could be argued that this interpretation of the effects-doctrine is problematic, as it constitutes a carte blanche for all States to prescribe far-reaching measures in response to common concerns. This risk should not, however, be overstated. Firstly, while global effects certainly bolster a State's claim to a reasonable interest, they are not necessarily determinative. Other factors such as the strength of a formal ‘territorial connection’ and the gravity of the territorial harm will also play a role. Causality also poses an important limitation, as the subject matter regulated must make a real contribution to the local and global harm. The measure in question must also be capable of mitigating the harm caused. In the present case, maritime emissions materially contribute to climate change, and the MRV and envisioned MBMs are capable of reducing them. Finally, the concrete exercise of jurisdiction must also comply with the jurisdictional ‘reasonableness’ limitations discussed in Section VB below.

From a more positivist perspective, explicit consideration of these jurisdictional questions will help to increase clarity on the way in which international law treats acts aimed at protecting common concerns. This could occur within a variety of fora including the WTO, International Tribunal for the Law of the Sea (ITLOS), the CJEU, and even the ICJ.²²⁵ In this way, unilateral action, such as that of the EU, serves to further the development of international law.²²⁶

B. Jurisdictional Limitations

The concrete exercise of jurisdiction must also meet the reasonableness criteria of good faith, proportionality and non-discrimination.²²⁷ It is submitted that in the context of common concerns, the good faith requirement entails that a State has made best-effort attempts to reach a multilateral solution. This is in line with the duty to cooperate in response to common concerns discussed above in Section II. It is also reflected world trade law in the chapeau of Article XX GATT, and in Articles 118 and 300 LOSC.²²⁸ As discussed, the EU has met this condition.

The proportionality principle directs us to consider the jurisdictional assertion's level of intrusiveness. Firstly, the design of the MRV and planned MBMs focusses on reducing the greatest amount of emissions at the lowest cost to operators, the Commission noting that ‘[s]hips above 5000 GT account for around 55 per cent of the number of ships calling into Union ports and represent around 90 per cent of the related emissions’.²²⁹ With regards to the EU MRV, it should be noted that the burden posed on operators is largely administrative, as most ships have already been fitted with the monitoring technology, as well as the possibility of offsets by fuel savings. The MBMs are obviously somewhat more intrusive, with additional costs arising from the obligatory fund contributions, purchase of additional carbon credits, and investment in new technology by operators. The latter, which the measures are intended to incentivize, would however also provide the company with fuel-saving offsets. The investment costs are further limited as much of the technology already exists, the MBMs being aimed at removing the market barriers to their actual use.²³⁰ As such, the Commission's impact assessment predicts a possible negative total cost in the mid-to-long term.²³¹

Under the proportionality principle, not only the impact on private economic operators but also on other States must be taken into account. Thus, a measure must not unduly limit the legislative freedom of other States to apply their own comparable measures.²³² This is also visible in the discussion on unjustifiable discrimination of the chapeau of Article XX GATT (see Section IVB). In the context of climate change, the proportionality requirement also reinforces the duty to act in accordance with the CBDR principle.²³³ This means that the final MBM will require a mechanism for recognition and integration of comparable third State measures, with consideration of the differing capacities of the participating States.

As regards the non-discrimination requirement, we have seen that the MBMs may pose a heavier burden on ships travelling from further away. However, as the EU's measures address all ship operators irrespective of their flag, and all voyages irrespective of the port of origin, it is difficult to maintain that they discriminate between operators.²³⁴ In any event, the differentiation based on origin is directly related to the objective of reducing emissions, and its absence would greatly diminish the measures’ effectiveness. We therefore conclude that the EU's measures, if applied proportionately, are permitted by the customary law of jurisdiction.

VI. CONCLUSION

By taking provocative climate action the EU has proved willing to extend its environmental legislation beyond its own borders. The recently adopted Regulation 2015/757 is a key example, incorporating ‘extraterritorial’ emissions, including those of foreign operators. The Regulation is intended as a precursor to a market-based measure (MBM), echoing the earlier adopted—though temporarily doomed—Aviation Directive. In international law, unilateral measures with such ‘extraterritorial elements’ are jurisdictionally suspect, as they may be seen to trample on the territorial prerogatives of other sovereigns. This raises particular tensions when such measures are taken in response to internationally recognized ‘common concerns’ like climate change. In such situations, we are confronted with a conflict of paradigms between the interdependence-focussed common concern obligations and the independence-focussed jurisdictional rights.

Our analysis has demonstrated, however, that these two paradigms need not be incompatible. The Law of the Sea Convention (LOSC) does not pose clear-cut territorial limits to port and coastal States’ (and the EU's) regulatory competences. At the same time, the law of the World Trade Organization (WTO) allows States and the EU to justify non-neutral trade restrictions furthering global environmental aims, as long as some territorial connection can be established. Ultimately, however, the LOSC and WTO law remain overly vague regarding the connection which international law requires between the regulating State (or the EU) and the subject matter at issue. We have therefore devoted special attention to the customary law of state jurisdiction supplementing these regimes. We have posited that, while the law of jurisdiction demands a substantial connection for an assertion to be presumptively lawful, such a connection should be construed in light of common concerns, which by their very nature have diffuse territorial effects in all States. It is the combination of climate change's indirect effects on the territory of EU, and the nature of the climate change problématique as a common concern, which serves as the legal basis for the EU's climate change unilateralism, more specifically for the MRV and the envisaged MBM.

Outstanding issues for the institutional design of unilateral measures pertain to their application to developing countries in light of the principle of common but differentiated responsibilities, as well as, somewhat relatedly, to the global welfare maximization potential of unilateral measures. This calls for vigilance in determining the precise design of climate-protective unilateral trade measures. However, given that the best option of adequate multilateral measures has not materialized, carefully crafted unilateral measures remain a defensible second-best option for addressing global climate change.