SECURING COMPATIBILITY OF CARBON BORDER ADJUSTMENTS WITH THE MULTILATERAL CLIMATE AND TRADE REGIMES

Abstract

The European Union (EU) is contemplating the adoption of a carbon border adjustment mechanism (CBAM), which would extend its domestic carbon price to emissions that are produced outside its borders but are embodied into its imports of carbon-intensive commodities. In doing so, the EU is testing the boundaries of permissible unilateral action at the interface of international climate and trade law. However, the question of whether the proposed CBAM is compatible with these two multilateral legal regimes is yet to be addressed in an integrated manner. This article seeks to fill this gap in the scholarship and makes two main arguments. First, the CBAM as presently designed does not respect the principle of Common but Differentiated Responsibilities and Respective Capabilities (CBDRRC) and needs to be adjusted through two forms of differential treatment: a full exemption for least-developed countries and Small Island Developing States and the use of CBAM-generated revenue to support decarbonisation efforts in other affected developing countries. Secondly, this CBDRRC-based differentiation should be permissible under WTO law on the grounds that it does not amount to discrimination between countries where the same conditions prevail.

I. INTRODUCTION

While carbon border adjustment measures (CBAMs) have been a popular topic in the scholarship over the past two decades, prospects for their actual implementation seemed fairly remote until recently since policymakers had generally dismissed such measures for being complex to administer and likely to trigger trade disputes and undermine multilateral climate change negotiations.¹ And yet, as the Intergovernmental Panel on Climate Change (IPCC) has recently observed, the previously largely academic debate on carbon border adjustments is shifting to real policymaking as an increasing number of jurisdictions are considering the introduction of CBAMs.² The European Union (EU) has taken the lead in this, with the Commission publishing a proposal for an EU regulation establishing a carbon border adjustment mechanism on 14 July 2021,³ about four months ahead of the global climate summit in Glasgow. If adopted by the European Parliament (EP) and Council,⁴ the proposed CBAM would make the EU the first jurisdiction worldwide to extend its domestic carbon price to emissions that are generated outside its borders but are ‘embodied’ into its imports of carbon-intensive commodities. Canada and the United States (US) are exploring similar measures.⁵ For their part, the BASIC countries (ie, Brazil, China, India and South Africa), as well as Russia, have expressed grave concerns over CBAMs, claiming that they go against World Trade Organisation (WTO) law and the principle of Common but Differentiated Responsibilities and Respective Capabilities (CBDRRC).⁶

International trade scholars have spilt much ink in assessing the WTO-compatibility of CBAMs, both before and after the EU's proposal was revealed.⁷ Conversely, the question of consistency with the CBDRRC principle has received less attention in the scholarship,⁸ presumably because it is seen as a secondary consideration in practical terms (ie, CBAMs can be legally challenged in the WTO dispute settlement system, but not really elsewhere), or even as an obstacle in ensuring WTO-compatibility. This article seeks to fill a gap in these academic discussions by considering for the first time both issues together, asking if it is in fact possible to design CBAMs in a manner that is consistent with both the CBDRRC principle and WTO law.

This integrated approach is necessary because CBAMs are, after all, trade-related climate measures that ought to be duly embedded into the existing multilateral legal frameworks, and yet ensuring this twin compatibility is not straightforward. On the one hand, the CBDRRC principle, as the most important and enduringly controversial principle of the United Nations (UN) climate regime, raises fundamental questions of equity and fair burden-sharing in global mitigation action.⁹ In particular, it requires consideration of whether certain countries should be granted differential treatment in the context of CBAMs in light of their differentiated responsibilities for causing climate change and respective capabilities to address it. On the other hand, it is not clear whether any such country differentiation would be permissible under WTO law. And this requires us to reflect upon how to foster ‘mutual supportiveness’¹⁰ between the multilateral climate and trade regimes.

Based on this premise, the article proceeds as follows. Section II begins with an overview of the main design features of the proposed EU CBAM. Using the EU's proposal as a case study is useful in terms of contextualising the subsequent analysis under international climate and trade law, but the core arguments made have broader implications for CBAMs being contemplated by other jurisdictions. Section III turns to assessing its compatibility with the UN climate regime, and in particular the CBDRRC principle. This involves dealing with the vexing question which forms of differential treatment the principle entails and for which countries. It is argued that this question cannot be answered on the basis of an abstract articulation of the CBDRRC principle, as in previous contributions,¹¹ but necessitates a careful analysis of how it has been operationalised ¹² in the mitigation provisions of climate change treaties—most notably the 2015 Paris Agreement (PA).¹³ It will be seen that the CBAM does not, contrary to what the European Commission maintains, respect the CBDRRC principle and needs to be adjusted to ensure that it does. However, this does not mean the non-application of the CBAM to all developing countries,¹⁴ as several scholars argued in the pre-Paris context.¹⁵ Rather, it calls for more nuanced forms of differentiation among developing countries.

Section IV explores whether this CBDRRC-adjusted CBAM is consistent with WTO law. It is argued that, while CBDRRC-based differential treatment is prima facie in conflict with the most-favoured-nation (MFN) treatment obligation of the General Agreement on Tariffs and Trade (GATT), it should be permitted under Article XX GATT and so be WTO-compatible. In making this argument, attention is drawn to an important aspect of the introductory clause (or chapeau) of Article XX GATT, which has received little attention in WTO jurisprudence thus far. By its express terms, the chapeau only prohibits discrimination between ‘countries where the same conditions prevail’. It is submitted that CBDRRC-grounded country differentiation does not amount to discrimination under GATT Article XX-chapeau because conditions in the countries involved are different in terms of their responsibilities/capabilities for climate conservation. Section V concludes.

II. OVERVIEW OF THE EU'S CBAM PROPOSAL

The Commission first announced its intention of proposing a CBAM in the European Green Deal of December 2019 and the proposed regulation forms part of its ‘Fit for 55 Package’ adopted in July 2021.¹⁶ This puts forward a set of legislative proposals with a view to meeting the targets enshrined in the European Climate Law—ie, a reduction in EU greenhouse gas (GHG) emissions by (at least) 55 per cent compared to 1990 levels by 2030, and the ultimate objective of ‘climate-neutrality’ (net-zero GHG emissions) by 2050.¹⁷ Among these legislative initiatives, the most closely linked to the CBAM is the revision of the EU's Emissions Trading System (ETS). This is a market-based mechanism for pricing carbon and reducing GHG emissions that presently operates in the European Economic Area (EEA) and covers emissions from energy-intensive power stations and industrial plants, as well as commercial flights between the 30 EEA countries.¹⁸

The CBAM is expected to enter into force on 1 January 2023 and would apply to products in five sectors (cement, iron and steel, aluminium, fertilisers and electricity) imported into the EU from all third countries, with the exception of Iceland, Liechtenstein, Norway (which are already part of the EU's ETS) and Switzerland (whose ETS is linked to the EU one).¹⁹ After a three-year transition period, from 1 January 2026,²⁰ importers of targeted products would have to: (i) apply for authorisation to import and set up a CBAM account with the competent authorities of EU Member States where they are established;²¹ (ii) submit a ‘CBAM Declaration’ by 31 May of each year with the total actual direct²² emissions embedded in their imports,²³ as verified by accredited verifiers;²⁴ and (iii) buy and surrender sufficient ‘CBAM certificates’ to cover these emissions.²⁵ The price of these CBAM certificates will mirror the weekly average price of emission allowances auctioned under the EU ETS,²⁶ thereby ensuring that imported and domestic products are subject to the same carbon price. However, unlike ETS emission permits, CBAM certificates cannot be traded.²⁷

According to the European Commission, the carbon price alignment sought by the CBAM is necessary to address the risk of carbon leakage as the EU increases its climate ambition in line with the 2030 and 2050 targets. Carbon leakage would occur if reduced carbon emissions within the EU are offset by increasing carbon emissions outside the Union, through the relocation of EU industries to countries with less stringent climate policies and/or increased EU imports of carbon-intensive products from such countries. This would not only diminish the effectiveness of the EU's mitigation efforts, but could also result in no net reduction (or even an increase) in carbon emissions at the global level.²⁸ Thus the overarching objective of the CBAM is to ‘prevent the risk of carbon leakage in order to fight climate change by reducing GHG emissions in the Union and globally’.²⁹ However, these environmental goals are closely intertwined with economic concerns about a ‘level playing field’ between EU and third-country producers in the absence of an internationally-agreed uniform carbon price.³⁰ In fact, carbon leakage has been described as a ‘lose-lose’ scenario: a loss of competitiveness for EU industries on global markets with no environmental gain, as emissions just migrate to other locations with no or lax climate regulation.³¹ To date, carbon leakage risks have been mitigated through the free allocation of emission allowances under the ETS to EU industries in energy-intensive and trade-exposed sectors.³² But this is considered to be problematic as it weakens the carbon price signal to EU industries compared to full auctioning and reduces the incentive to invest in low-carbon production.³³ As free allowances are expected to be gradually phased out under the revised ETS by 2035, the CBAM would be phased in as an alternative mechanism to address carbon leakage risks.³⁴

The CBAM also aims to push for more ambitious climate action in the EU's trading partners. At a general level, the EU-equivalent carbon price seeks to act as an economic incentive for foreign producers to introduce cleaner production technologies and lower emission levels.³⁵ More specifically, the CBAM seeks to encourage trading partners to adopt carbon pricing as the preferred mitigation policy option, and to closely follow the EU's own pricing system. This is most evident in the ‘EU-led carbon club’ aspect of the CBAM which, as noted above, currently exempts four countries from the application of the scheme. It is possible to add other countries to the list, subject to the same conditions: either by accepting the application of the EU's ETS; or by concluding an agreement with the EU fully linking their emissions trading systems with that of the EU and charging the same price as the EU for carbon emissions.³⁶ For products originating in non-exempted countries, an importer may seek a reduction in the number of required CBAM certificates to take account of any carbon price paid in the country of production.³⁷ This ‘CBAM discount’, however, only applies to explicit carbon policies (whether in the form of a carbon tax or under an emissions trading scheme) and ignores costs associated with other regulatory approaches to climate mitigation. At present, only a few of the top exporters in CBAM-covered sectors would be able to benefit from the CBAM discount, namely China, South Korea and the United Kingdom (UK).³⁸

To further contextualise the discussion of CBDRRC-compatibility that follows it is important to consider which of the EU's trading partners would be most exposed to the CBAM if adopted as currently designed. According to the Commission's Impact Assessment, based on a simple analysis of current trade flows, these would include Russia, Ukraine, and Turkey, ranking among the top-ten exporters for most CBAM-covered sectors, followed by some European countries (Belarus and UK). BASIC countries, which have voiced fierce opposition to the CBAM, also feature among the top exporters in specific sectors, notably: aluminium (China and India) and iron/steel (China, Brazil and India). Similarly, some North African countries (Algeria, Egypt and Morocco) are among the top-five exporters of cement and fertilisers.³⁹

Least-developed countries (LDCs)⁴⁰ and Small Island Developing States (SIDS)⁴¹ generally account for only a small share of EU imports of the targeted products. But the CBAM's impact on these countries is estimated to be considerable given the relative importance of such exports for their economies. For instance, in 2020, Mozambique was the sixth largest exporter of aluminium to the EU, accounting for nearly seven per cent of the country's gross domestic product (GDP). In that same year, Trinidad and Tobago was the fourth largest exporter of fertilisers to the EU, and while Senegal had a much smaller share, exports of fertilisers to the EU represented about five per cent of its GDP. The other nine LDCs affected by the CBAM are per sector: cement (Cambodia, Chad, Guinea, Haiti and Uganda), fertilisers (Afghanistan, Ethiopia and Madagascar), iron and steel (Myanmar, Niger and Sierra Leone).⁴² The Commission further acknowledges that compliance costs are likely to be higher in these countries when compared to other trading partners, given their lower capacity to both decarbonise production processes as well as to measure and verify the carbon intensity of products exported to the EU.⁴³ There are no signs in the CBAM proposal that special treatment will be granted to these countries, even if the EP had called for it in March 2021.⁴⁴

III. CARBON BORDER ADJUSTMENTS AND THE INTERNATIONAL CLIMATE REGIME

A. Unilateral Trade Measures under the UNFCCC

Before considering the compatibility of CBAMs with the CBDRRC principle, it seems appropriate to clarify whether the international climate regime contemplates the use of trade measures to address climate change in the first place. The key provision in this regard is Article 3.5 UNFCCC, which provides: ‘[m]easures taken to combat climate change, including unilateral ones, should not constitute a means of arbitrary or unjustifiable discrimination or a disguised restriction on international trade’.⁴⁵

It has been suggested that the effect of this provision is neutral, neither allowing nor forbidding the adoption of trade measures (including unilateral measures) to tackle climate change.⁴⁶ It is true that Article 3.5 UNFCCC cannot itself create rights or obligations since its function, as a principle, is to guide Parties’ actions in achieving the objectives of the Convention and implementing its provisions. However, it can be argued that by setting out explicit conditions on the use of trade-related climate measures, the UNFCCC implicitly recognises (or takes for granted) that Parties may resort to such measures, even unilaterally.⁴⁷ It certainly does not prohibit such measures outright, and accepts that their use may be permissible provided they do not amount to ‘arbitrary and unjustifiable discrimination’ or a ‘disguised restriction on international trade’. These terms are not defined in the UNFCCC and it is uncertain whether they should be interpreted in line with WTO jurisprudence concerning similar language in the chapeau of Article XX GATT. However, it is worth noting that the wording of these two provisions is not identical, as the chapeau of Article XX GATT has an additional element in that it only prohibits arbitrary and unjustifiable discrimination ‘between countries where the same conditions prevail’—and the significance of this will be explored in Section IV.

B. CBDRRC Principle: Content and Status

The CBDDRC principle is enshrined in Article 3.1 UNFCCC, which reads: ‘[t]he Parties should protect the climate system for the benefit of present and future generations of humankind, on the basis of equity and in accordance with their common but differentiated responsibilities and respective capabilities. Accordingly, the developed country Parties should take the lead in combating climate change and the adverse effects thereof.’⁴⁸ While this principle has underpinned global efforts to fight climate change from the very start,⁴⁹ its core content and legal status remain deeply contested.

With regard to the content, the first element of the principle refers to the ‘common responsibility’ of States to protect the environment and is generally understood as a recognition that climate change is a matter of ‘common concern’ of humankind, which requires the widest possible cooperation by all States.⁵⁰ However, the exact meaning of the remainder of the principle has generated much debate over the years. In particular, the term ‘differentiated’ indicates the need for differential treatment between countries, but the basis for such differentiation is ambiguous. On a plain reading of Article 3.1 UNFCCC, differentiation is to be determined on the basis of two factors—responsibility for causing climate change and capability to address it. Yet it does not specify how these should be measured and the extent to which they may be linked. Nonetheless, the expectation in that same provision that ‘developed countries should take the lead’ may be read as implying that responsibility was primarily conceived of in relation to varying historical contributions to climate change which, in turn, gives industrialised countries a greater capacity to tackle the problem.⁵¹ In other words, the UNFCCC sees responsibility and capability as intrinsically linked rather than as distinct factors, with enhanced capability being the direct result of industrialisation and historical responsibility for GHG emissions.⁵²

However, there is a growing consensus that the understanding of the CBDRRC principle has markedly evolved during the period from the UNFCCC to the Paris Agreement, both in its articulation and operationalisation of the substantive commitments of the agreement, and most notably those concerning climate change mitigation.⁵³ Article 2(2) PA reaffirms that the agreement ‘will be implemented to respect’ the CBDRRC principle, but this is qualified by this being ‘in light of different national circumstances’. Arguably, this qualifier of ‘national circumstances’ introduced in the Lima negotiations brings a dynamic and flexible element to interpreting both responsibilities and capabilities and broadens the parameters for differentiation between countries which is no longer premised on their historical contributions alone.⁵⁴ Thus, it allows for a more nuanced approach to differentiation, taking into account a wider array of criteria in assessing ‘differentiated responsibilities’ (ie, not only past but also current and projected future GHG emissions) and ‘respective capabilities’ (eg, financial and technical capabilities, human capacity and other factors).⁵⁵ At the same time, this amalgamation of country-specific responsibilities and capabilities makes determining which countries may be deemed to have a ‘high’ or ‘low’ responsibility/capability in the global fight against climate change increasingly complex.

There is no doubt that the EU itself falls within the high responsibility/capability category and should therefore assume a leadership role in global efforts to address climate change. But this does not help in answering the vexing question of how differentiation should be integrated into CBAMs and which ‘low’ responsibility/capability countries are entitled to it. In fact, this question cannot be answered on the basis of the general formulation of the CBDRRC principle in Article 3.1 UNFCCC and Article 2.2 PA. This is because legal principles are highly abstract, embodying general standards against which to evaluate governmental decision-making but leaving considerable room for interpretation.⁵⁶ Put differently, the CBDRRC principle is open-ended: while it may sway decision-makers in a particular direction, it does not itself specify particular actions.⁵⁷

This enduring ambiguity over what the principle entails and when it applies has spawned considerable debate over its legal status. These discussions have typically centred on the question of whether the concept of common but differentiated responsibilities (CBDR), which has been repeatedly endorsed in international environmental instruments since its first authoritative formulation in the 1992 Rio Declaration,⁵⁸ has attained the status of customary international law binding erga omnes. Most scholars have come to agree that CBDR does not constitute a customary principle and is thus devoid of general applicability outside the confines of the treaties in which it is found.⁵⁹ In the multilateral climate regime, CBDRRC is explicitly expressed as an overarching principle in the operational provisions of the UNFCCC (Article 3.1) and PA (Article 2.2)—two legally-binding treaties (inter partes) with near universal participation. The chapeau of Article 3 UNFCCC (‘shall be guided’) and Article 2.2 PA (‘will be implemented to respect’) create normative expectations that this principle will be taken into consideration,⁶⁰ and given proper weight,⁶¹ by the Parties in their ‘actions to achieve the objective’ of these agreements. The EU, as the relevant international actor for present purposes, does not contest this. Moreover, it sees the CBAM as an essential policy tool for meeting the objective of ‘a climate-neutral Union by 2050 in line with the Paris Agreement’ and expressly recognises that it ought to ‘respect the [CBDRRC] principle’.⁶²

Therefore, the key issue is not so much whether but how the CBDRRC principle should be duly taken into account when adopting CBAMs. With regard to the ‘how’ question, the principle as set out in Articles 3.1 UNFCCC and 2.2 PA does not offer an adequate benchmark. This is because, as seen above, its core meaning remains unsettled and it does not prescribe any specific form of differentiation. But it would be wrong to suggest that, as a result, the principle is of no legal (or practical) significance to CBAMs and can be fully disregarded. This would be tantamount to neglecting the central function which CBDRRC has played as a ‘framework’ or ‘structuring’ principle in the international climate architecture, providing the bedrock of the burden-sharing arrangements initially crafted in the UNFCCC with regard to climate change mitigation (as well as adaptation, financial assistance and technology transfer), and shaping their subsequent elaboration in the Kyoto Protocol and Paris Agreement.⁶³ As will be further examined below, the mitigation provisions in these agreements, which are directly relevant to CBAMs as carbon-reduction measures, implement in more precise terms the CBDRRC principle through a set of differentiated obligations among State Parties. It goes without saying that these substantive commitments are legally-binding and,⁶⁴ therefore, CBAMs ought not to (unilaterally) undercut the differentiation balance (multilaterally) established therein.⁶⁵ The next section considers whether the CBAM proposed by the EU respects the CBDRRC principle as operationalised in the mitigation pillar of the (currently applicable) Paris Agreement, and challenges the European Commission's position on this.

C. Does the Proposed EU CBAM Respect the CBDRRC Principle?

While the CBDRRC principle receives marginal attention in the Commission's proposal, it does put forward a number of arguments as to why the proposed CBAM is considered to be consistent with that principle. First, the Commission sees the CBAM as an ‘essential element’ of the EU's leadership role in global climate governance, ‘addressing the risks of carbon leakage as a result of the increased Union climate ambition’.⁶⁶ To the extent that the prospect of carbon leakage is genuine,⁶⁷ it is true that the CBDRRC principle places the EU in a dilemma. On the one hand, the EU is called upon to increase its own mitigation efforts (in casu, by phasing out ETS free allowances), but it is hard to see why and how the EU should do so if it would simply result in a shifting of carbon emissions abroad. This would not only compromise the effectiveness of the EU's mitigation policies, but may also jeopardise the achievement of the multilaterally-agreed temperature targets set out in the Paris Agreement.⁶⁸ On the other hand, the EU should respect the fact that not all countries are to contribute in equal measure to the common goal of mitigating climate change, in view of their differentiated responsibilities/capabilities.

In the Commission's view, this can be ensured without any form of differentiation in the CBAM since it has been ‘designed in such a manner that it does not directly depend on the overall level of ambition of a country’.⁶⁹ This is partly correct, if emission reduction targets are considered to be the most obvious quantitative indicator of each Party's overall level of climate ambition under the Paris Agreement. In this regard, the EU's emission reduction target is economy-wide,⁷⁰ whereas its CBAM would only apply to selected sectors and only insofar as the foreign products are exported to the EU.⁷¹ Hence, it is inaccurate to claim that the CBAM would harmonise emission reduction targets between the EU and the exporting countries through the back door.⁷² But this should not mask the fact that, for particular sectors or products, the CBAM does seek to incentivise a reduction in the carbon emissions of exporting countries, even if not to an equal extent as the EU. Indeed, the Commission's Impact Assessment estimates ‘a 1.0% emissions reduction in the EU and a 0.4% in the rest of the world in CBAM sectors by 2030’,⁷³ albeit that it does not specify the impact on individual third countries.⁷⁴ So, the question is whether this emission-reduction effect of CBAMs upsets the balance of differentiated obligations on mitigation achieved in the multilateral climate regime.

It certainly appears to be in tension with the CBDRRC principle as operationalised by the UNFCCC and its Kyoto Protocol.⁷⁵ Based on an understanding of the principle that relied almost exclusively on historical and then (in 1997) current responsibility, the Kyoto Protocol adopted a stringent ‘binary’ or ‘bifurcated’ approach to mitigation differentiation. In essence, it established legally-binding, quantified and economy-wide emission reduction targets for each developed country (listed in Annex I UNFCCC),⁷⁶ while fully exempting developing countries (that is, non-Annex I countries) from emission reduction obligations.⁷⁷ As per the UNFCCC preamble, this was to recognise that ‘per capita emissions in developing countries are still relatively low’, back in 1992, and that ‘the share of global emissions originating in developing countries will grow to meet their social and development needs’.⁷⁸ In this normative context, the adoption of CBAMs would have clearly eroded the Kyoto burden-sharing arrangements by encouraging developing countries to reduce (rather than increase) carbon emissions. As such, an exemption for developing countries was objectively justifiable to preserve their unrestricted right to use the remaining atmospheric space (or carbon budget) for their socio-economic development.

However, it is far from obvious that a blanket CBAM-exemption for developing countries (or non-Annex I countries) can be similarly reconciled with the CBDRRC principle as operationalised in the Paris Agreement, which no longer allocates an exclusive responsibility for reducing GHG emissions to developed countries (or Annex I countries). Instead, the Paris Agreement embraces a bounded self-differentiation approach to mitigation commitments,⁷⁹ shifting away from the strict annex-based division of responsibilities/capabilities that underpinned the Kyoto Protocol.⁸⁰ It imposes a binding obligation of conduct on each Party to ‘prepare, communicate and maintain nationally determined contributions (NDCs) that it intends to achieve’, coupled with a good-faith expectation that domestic measures will be implemented to achieve the objectives of such NDCs.⁸¹ Each Party chooses its own contribution, thus leading to a spectrum of self-differentiated mitigation targets and implementation measures. But this self-differentiation is disciplined by strong normative expectations—albeit not mandatory obligations—on ‘progression’ and the ‘highest possible ambition’ for each successive NDC, ‘in light of different national circumstances’. While each Party is left to determine how its NDC reflects the ‘highest possible ambition’, it is clear that all Parties are expected to undertake more ambitious mitigation action over time.⁸²

Within this global mitigation trajectory, the Paris Agreement still assigns a leadership role to developed countries by requiring them to undertake economy-wide absolute reduction targets, whilst ‘developing countries should continue enhancing their mitigation efforts, and are encouraged to move over time towards economy-wide emission reduction or limitation targets in the light of different national circumstances.’⁸³ Such an expectation has been met in practice, with several developing countries committing to economy-wide reduction or limitation targets, including the BASIC countries.⁸⁴ This direction of travel towards more ambitious climate action by all Parties reflects the nature of climate change as a ‘common concern’, requiring effective participation by all major contributors.⁸⁵ At present, and departing from the assumption underlying the Kyoto burden-sharing arrangements, non-Annex I countries include some of the world's largest carbon emitters even in per capita terms.⁸⁶ For this reason, in order to achieve the temperature goals established in the Paris Agreement,⁸⁷ all Parties are committed to contribute to the global peaking of GHG emissions ‘as soon as possible’ and to undertake rapid reductions thereafter, so as to achieve a balance of GHG emissions by sources and removals by sinks by 2050.⁸⁸ In this regard, the latest IPCC Sixth Assessment Report warns that the aggregate emission reductions implied by NDCs to 2030 are far behind what is necessary to meet the 1.5°C, or even 2°C, temperature goals and that ‘limiting warming to either level implies accelerated mitigation actions at all scales’.⁸⁹ This applies in particular to G20 countries (including some developing countries affected by the EU's CBAM), since they are collectively responsible for more than three quarters of global GHG emissions.⁹⁰

It is hard to see how the Paris legal framework, which sets firm expectations on all Parties to ratchet up mitigation action and move towards low-emission production strategies,⁹¹ can still justify exempting all non-Annex I countries from CBAMs based on CBDRRC grounds. Put simply, things have moved on from the Kyoto scenario in which CBAMs would be out of step with the CBDRRC principle because developing countries should not be expected to reduce GHG emissions at all.

There is, however, an important qualification to this argument and this concerns the specific situation of LDCs and SIDS, and which is acknowledged in the mitigation and other provisions of the Paris Agreement.⁹² Of most relevance is Article 4.6 PA, which provides that such countries ‘may prepare and communicate strategies, plans and actions for low greenhouse gas emissions development reflecting their special circumstances’.⁹³ This provision follows those applicable to all Parties, and its permissive character is clearly significant. It differentiates LDCs/SIDS from all other Parties (including other developing countries), meaning that these countries can—but are under no obligation to—undertake emission-reduction action. This special treatment of LDCs/SIDs reflects the fact that they bear the least historical and current responsibility for the climate emergency (presently accounting for only seven per cent of global GHG emissions) and have the least capacity to adapt to new climate conditions,⁹⁴ thus being the most vulnerable countries to the adverse effects of climate change.⁹⁵ Insofar as the EU's CBAM would incentivise (or arguably, pressure) producers in LDCs/SIDS to lower carbon emissions, it is not in line with the CBRRDC principle as operationalised in Article 4.6 PA. Therefore, this provision is a solid basis for justifying an LDCs/SIDS exemption from the CBAM, as has been called for by the UN Conference on Trade and Development (UNCTAD) and other commentators.⁹⁶

The Commission, however, has dismissed the possibility of an LDC/SIDS exemption on grounds that it would run counter to the overarching objective of the CBAM by encouraging a growth of emissions in these countries and that it would be counterproductive by potentially locking them into high-carbon development paths.⁹⁷ Although the first concern is an overstatement, given that the risk of carbon leakage associated with LDCs/SIDS has been estimated to be negligible,⁹⁸ the second argument is not entirely misplaced. However, as seen above, the Paris differentiation balance does give LDCs/SIDS full discretion (‘may’) as to whether they embrace the decarbonisation of energy-intensive industries in light of their special circumstances—and ultimately, this national choice ought to be respected by the EU when designing the CBAM in the spirit of the CBDRRC principle.

A second argument advanced by the Commission concerning the CBAM's compatibility with the CBDRRC principle is that ‘it does not directly depend … on the policy choices made by a country’.⁹⁹ This statement is simply incorrect given that, as has been seen, the CBAM takes into consideration ‘explicit’ carbon pricing policies in trading partners, through either an exemption for those countries joining the EU-led carbon club or a CBAM discount for any carbon price paid in the country of production.¹⁰⁰ By contrast, it fails to account for the ‘implicit’ carbon prices associated with non-price regulatory approaches to reducing carbon emissions (eg, energy efficiency or emission performance standards) adopted in trading partners, whose exports of targeted products would be subject to the full ETS-determined carbon price. In any event, the key point is that the proposed CBAM would effectively equalise the costs (rather than overall levels) of lowering GHG emissions between EU and foreign producers through the imposition of an EU-equivalent carbon price on the products concerned. In other words, it is one thing to expect developing countries to reduce carbon emissions (which is Paris-compatible), and another to expect them to bear carbon costs equivalent to the EU.

This undercuts the substantial flexibility which the Paris Agreement gives concerning choice of means to pursue decarbonisation, since Parties are free to decide for themsleves whether or not to adopt carbon pricing instruments when implementing their NDCs. Most importantly, heterogeneity in carbon prices and other mitigation policies across jurisdictions remains acceptable as as long as this reflects each Party's ‘highest level of ambition’ in light of its national circumstances.¹⁰¹ Ultimately, this is a matter of national determination,¹⁰² as the key oversight mechanism established under the Paris Agreement (the so-called ‘Global Stocktake’) is only authorised to assess collective progress towards meeting the global warming targets, thus insulating individual Parties from any assessment of the the adequacy of their mitigation action under NDCs.¹⁰³

Even if the Commission's position that a uniform carbon price is necessary to enable the EU to increase its own climate ambition and avoid carbon leakage is accepted,¹⁰⁴ the EU should support developing countries in meeting the burden of complying with the CBAM for their exports of targeted products, both in terms of administrative requirements and increasing carbon prices.¹⁰⁵ The Commission itself acknowledges that ‘in the absence of compensating mechanisms, LDCs could argue that the introduction of a CBAM will be a disproportionate burden for them and […] conflict with the [CBDRRC] principle’.¹⁰⁶ This is a pertinent observation, but there is no obvious reason why it should only apply to LDCs. It can be argued that the CBAM would also impose a disproportionate burden on SIDS and other developing countries, insofar as it is designed to equalise carbon prices to an EU-equivalent level for the products concerned.

In this regard, Article 4.5 PA clearly requires (‘shall’) developed countries to provide financial support to assist mitigation by developing countries, ‘recognizing that enhanced support for developing country Parties will allow for higher ambition in their actions’.¹⁰⁷ The Paris differentiation balance thus reflects a critical understanding of the relationship between greater overall ambition, with developing countries assuming responsibility for lowering carbon emissions, on the one hand, and increased financial resources from developed countries to support such mitigation, on the other.¹⁰⁸

From this perspective, the Commission's CBAM proposal is deeply disappointing. It contains only a preambular provision which declares that the EU ‘stands ready to work with low and middle-income countries towards the decarbonisation of their manufacturing industries [and] should support less developed countries with the necessary technical assistance in order to facilitate their adaptation to the new [CBAM] obligations’,¹⁰⁹ without any firm commitment on financial assistance. Furthermore, the proposed regulation is strikingly silent on the use of revenue generated through the sale of CBAM certificates, which is estimated to exceed EUR 2.1 billion by 2030.¹¹⁰ The Explanatory Memorandum indicates that the intention is to allocate most of these additional resources to the Union budget, including to finance its COVID-recovery instrument ‘Next Generation EU’.¹¹¹ The suggestion that revenue earned from pricing emissions produced in developing countries would be used to subsidise a ‘greener’ recovery plan in the EU fundamentally subverts the Paris burden-sharing arrangements and ought to be seriously reconsidered. This revenue should be recycled back to the developing countries concerned to support their own decarbonisation programmes.¹¹²

The EP has gone some way in this direction by proposing a new article on the use of CBAM-generated revenue. It provides for increased EU funding to support climate mitigation in LDCs through the Neighbourhood, Development and International Cooperation (NDIC) Instrument,¹¹³ by an amount that ‘should correspond at least to the level of revenues generated by the sale of CBAM certificates’.¹¹⁴ However, the EP fails to explain why such EU support should be limited to LDCs and not extended to other developing countries. For the reasons previously mentioned, this would not respect the Paris differentiation balance as reflected in Article 4.5 PA.

To sum up, contrary to what the Commission claims, the EU's CBAM as presently envisaged is not compatible with the CBDRRC principle as reflected in the mitigation provisions of the Paris Agreement. To be brought into line with this principle, two adjustments need to be made: an LDC/SIDS exemption (based on Article 4.6 PA) and a revenue-recycling provision applicable to imports originating in developing countries (based on Article 4.5 PA).

IV. CBDRRC-COMPATIBLE CARBON BORDER ADJUSTMENTS AND WTO LAW

A. CBDRRC-Based Differentiation and MFN Treatment Obligation

Before considering the WTO-compatibility of the EU's CBAM, it is first necessary to determine how this measure should be characterised under WTO law, in order to determine the applicable GATT obligations. Scholars have discussed extensively whether CBAMs should be viewed as a ‘border’ or ‘internal’ measure, but this is not the place to contribute to this debate, nor to offer a comprehensive analysis of whether the EU's CBAM is consistent with the WTO.¹¹⁵ Rather, this article is concerned with whether the two forms of country differentiation that are necessary to bring the CBAM into line with the CBDRRC principle, as discussed in the previous section, are permissible under WTO law. For this purpose, the key provision is Article I GATT concerning the MFN obligation, which is a pillar of the multilateral trading system¹¹⁶ and applies to both border and internal measures affecting international trade in goods—and hence to the CBAM, no matter how it is characterised. Conversely, it is less likely that the provision of climate finance to developing countries through the NDIC Instrument would fall within the scope of Article I GATT, even if part of this financial support comes from CBAM-generated revenue.¹¹⁷

Essentially, Article I GATT prohibits discrimination between WTO members by requiring that any trade advantage that is accorded to any product from any country (or destined for any country) is also accorded immediately and unconditionally to like (or competitive) products from or destined for, all WTO members. There is no doubt that inserting an LDC/SIDS exemption into the EU's CBAM would be in direct tension with this core GATT obligation because it would offer a trade advantage (ie, an exemption from the obligation to purchase CBAM certificates/pay an EU-equivalent carbon price) to products originating from LDCs/SIDS (say, aluminium from Mozambique) that is not accorded immediately and unconditionally to like products originating in other WTO members (say, aluminium from Russia or China). In fact, it would be a clear case of de jure discrimination in contravention of Article I GATT. This is unsurprising, given the fundamental purpose of the MFN obligation is to preserve the equality of competitive opportunities for like products imported from all WTO members,¹¹⁸ regardless of the different conditions that may prevail in these countries. Hence, for the purpose of the MFN clause, the only basis for comparing WTO members and determining whether they should receive the same trade advantage is the ‘likeness’ of their products—in this case, whether they export CBAM-covered goods that are in a competitive relationship in the EU market.¹¹⁹ In this respect, it is very different from the non-discrimination obligation in the chapeau of Article XX GATT which, as discussed below, requires equal treatment of countries in like conditions.¹²⁰

But the same reasoning applies to the EEA/Switzerland exemption, which already exists in the EU's CBAM.¹²¹ Furthermore, most commentators have argued that, even leaving aside origin-based exemptions, the measure is likely to be in breach of Articles I (and possibly III) GATT on grounds of de facto discrimination.¹²² Hence, with or without the LDC/SIDS exemption, the EU's CBAM is most certainly in need of justification under the GATT general exceptions clause, to which the next section turns.

B. Is CBDRRC-Based Differentiation Permissible under Article XX GATT?

1. Provisional justification under Article XX(g) GATT

As is well settled in WTO case law, Article XX GATT lays down a conditional exception for a measure that is inconsistent with the substantive obligations established in the GATT, provided that: (i) it is provisionally justified under one of the policy grounds listed in paragraphs (a) to (j); and (ii) it meets the requirements of the chapeau.¹²³ With regard to the first step, Article XX(g) GATT offers the most promising approach given that it is most often invoked to justify trade-related environmental measures.¹²⁴ This provision requires that the measure at issue be ‘related to the conservation of exhaustible natural resources’ and is ‘made effective in conjunction with restrictions on domestic production and consumption’. The CBAM is likely to meet this latter requirement,¹²⁵ given that it operates together with the EU's ETS and other restrictions on domestic carbon-intensive production.¹²⁶

As to the first condition, it is highly likely that a stable climate (ie, a global atmosphere with a safe level of GHG concentrations)¹²⁷ can be considered to be an ‘exhaustible natural resource’ within the meaning of Article XX(g) GATT, based on the dynamic and flexible interpretation of this term in WTO jurisprudence. Notably, in US – Gasoline, clean air was found to be an exhaustible natural resource and, in US—Shrimp the Appellate Body (AB) emphasised the importance of interpreting this concept in an evolutionary manner, ‘in light of the contemporary concerns of the community of nations about the protection and conservation of the environment’, with reference to relevant international instruments.¹²⁸ Following the wealth of scientific evidence that has emerged over recent decades on the urgency of tackling the climate crisis, and its political endorsement within UNFCCC processes and other multilateral fora, it seems difficult to deny that a stable climate falls within the purview of Article XX(g) GATT. While the question of whether this provision is subject to an implicit jurisdictional limitation remains unsettled,¹²⁹ this is unlikely to arise in the case of CBAMs. This is because ‘the change in the Earth's climate and its adverse effects’ is explicitly recognised as a ‘common concern of humankind’ in the UNFCCC,¹³⁰ with which all WTO members can claim to have a ‘sufficient nexus’.¹³¹

Therefore, the key issue under Article XX(g) GATT is the extent to which the EU's CBAM is ‘closely and genuinely related to’¹³² the conservation of a global atmosphere with low GHG density. This requirement is not very demanding and, unlike the necessity test under other subparagraphs of Article XX GATT, does not require the CBAM to be scrutinised for the extent to which it contributes to the climate conservation objective, nor to be balanced against less trade-restrictive alternative measures.¹³³ It just needs to be shown that the CBAM is not ‘disproportionally wide in reach and scope’ in relation to the climate conservation objective, and that the means-to-end relationship is a ‘close and real one’.¹³⁴

This may raise questions about how real the risk of carbon leakage, which the CBAM aims to address, actually is in the first place. In this regard, the Commission's Impact Assessment acknowledges that ‘[t]he evidence of the existence of carbon leakage is not always conclusive or suggests that it is difficult to isolate carbon leakage as a single factor in [firms’] relocation decisions’.¹³⁵ In fact, it is mainly ex-ante theoretical analyses that point to there being a substantial risk of carbon leakage in the absence of protection mechanisms (such as ETS free allowances), particularly for emission-intensive and trade-exposed sectors, whereas ex-post studies often find that carbon leakage is occurring at a much lower rate. This matter has also received considerable attention in the recent IPCC Sixth Assessment Report, which confirms the need for further research on the ‘existence and extent of carbon leakage’ and on the implications of CBAMs and other instruments designed to address it.¹³⁶

While shedding further light on these issues is certainly important, it does not have a direct bearing on whether a CBDRRC-adjusted CBAM can be WTO-compatible.¹³⁷ Thus, on the assumption that the CBAM can meet the thresholds of Article XX(g) GATT, the next section turns to the chapeau, which is of most relevance for appraising the permissibility of CBDRRC-based differentiation under WTO law.

2. CBDRRC-based differentiation and Article XX-chapeau

a) Discrimination between countries in ‘the same conditions’

The introductory clause of Article XX GATT sets out a number of horizontal requirements for measures provisionally justified under one of its paragraphs. By its express terms, the chapeau only prohibits discrimination between countries ‘where the same conditions prevail’ that is ‘arbitrary or unjustifiable’.¹³⁸ As a matter of logic, the first step of the analysis should be to determine whether the differential treatment at issue occurs between countries in the same conditions, and only where this is the case does the question of whether the resulting discrimination is arbitrary and unjustifiable arise.¹³⁹ While the AB confirmed this reading of the chapeau in EC – Seal Products,¹⁴⁰ WTO judicial bodies have often skipped over the first step and focused on the arbitrary/unjustifiable discrimination test. In doing so, they have simply equated any kind of differentiation between countries with discrimination under the chapeau, without enquiring whether the countries concerned were in fact in the same conditions. This approach cannot be right, because it contradicts the very text of Article XX GATT and overlooks the fact that treating countries in dissimilar conditions differently is not discriminatory under the chapeau and hence permissible without need for further justification. In such cases of different conditions, the arbitrary/unjustifiable discrimination limb of the chapeau would not be applicable.¹⁴¹ For this reason, a principled approach is necessary in order to decide the threshold issue of which conditions are relevant to the comparison of countries under the chapeau and when they can be considered to be the same or sufficiently different.¹⁴²

These questions remain largely unexplored in WTO jurisprudence and, to some extent, also in the scholarship.¹⁴³ In US – Gasoline (1996) ¹⁴⁴ the AB simply assumed that the prevailing conditions were the same while in US – Shrimp it assumed that these may instead be different,¹⁴⁵ but without explaining in either case the rationale behind such assumptions. The latter concerned an import prohibition on shrimp products from non-certified countries because they had not used a certain fishing net prescribed by the US (ie, approved Turtle Excluder Devices (TEDs)) when catching shrimp. The US successfully argued that this GATT Article XI-inconsistent import ban was provisionally justified under Article XX(g) GATT as it related to the conservation of sea turtles, all species of which were listed as threatened with extinction in Appendix I of the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES).¹⁴⁶ Under the chapeau, however, the AB found the measure discriminated unjustifiably because it made market access conditional on the adoption of ‘essentially the same regulatory programme’ by exporting countries as that applied to US shrimp trawl vessels, based on a ‘single, rigid and unbending’ standard (ie, use of TEDs), without allowing for ‘any enquiry into the appropriateness of the regulatory programme for the conditions prevailing in those countries’.¹⁴⁷

This statement could be open to various interpretations but¹⁴⁸ in the subsequent compliance proceedings US – Shrimp (2001) the AB clarified that the US was only required to recognise third-country regulatory programmes that are ‘comparable in effectiveness’—in casu, TED-comparable measures for the conservation of the endangered sea turtles. This provides sufficient latitude to exporting countries to design regulatory programmes that are suitable for their specific conditions whilst achieving the level of environmental protection sought by the importing country.¹⁴⁹ Therefore, this standard does not demand that the regulating WTO member compromise the achievement of its environmental objectives in order to accommodate prevailing conditions in different countries.¹⁵⁰ It allows flexibility in terms of the means to achieve the relevant environmental outcome (in casu, no killing or harming of endangered sea turtles).

As an effects-based equivalence requirement, this could be used to challenge the EU-led carbon club exemption currently found in the CBAM, which could be seen as the EU using its market power to compel other WTO members into adopting essentially the same carbon pricing policies. Even if there are important differences between the two cases, including in terms of the coercive effect of the measures at issue,¹⁵¹ it could still be argued that the EU should consider non-price regulatory instruments that may be comparable in environmental effectiveness to its ETS as a basis for exempting third countries from the CBAM. In this scenario, the CBAM would not serve its environmental purpose since the risk of carbon leakage would hardly materialise in countries whose emission-reduction policies are as stringent as those of the EU.

Recognising equivalence of non-price climate regulations would further support the bottom-up approach to implementation under the Paris Agreement discussed earlier, which gives Parties ample flexibility with respect to the adoption of mitigation measures to meet their NDCs.¹⁵² However, it also raises the complex question of how to compare different emission-reduction policies and who should ultimately determine their effects-based equivalence. As the IPCC has recently noted, ‘comparing the stringency of [mitigation] policies over time or across jurisdictions is very challenging and there is no single widely accepted metric or methodology’.¹⁵³ In fact, this comparability assessment is likely to be more complex in the climate change context than in the first US – Shrimp case, where the possibility of comparing sea turtle conservation programmes was envisaged in the challenged US scheme but inflexibly applied.¹⁵⁴ But the possibility cannot be foreclosed and, as suggested by the EP, the CBAM proposal should be amended so that the EU can actively engage with trading partners (eg, as part of an ‘international carbon club’ or other platform) on comparability assessments of price and non-price regulatory approaches.¹⁵⁵

That being said, the regulatory flexibility standard as applied in the second US – Shrimp case is not very helpful to CBDRRC-based differentiation in the CBAM for two reasons. First, the two underlying multilateral environmental agreements (MEAs) significantly differ in relation to the CBDRRC principle. The CITES treaty at issue in the US – Shrimp cases does not incorporate this principle and imposes the same obligations on all Parties to protect endangered species (including all recognised species of sea turtles), with no differentiation in light of varying responsibilities/capabilities.¹⁵⁶ By contrast, as we have seen, the Paris Agreement does not expect some countries, notably LDCs and SIDS, to make comparable efforts in reducing carbon emissions and it entitles other developing countries to be supported in their mitigation action. Thus differential treatment of the countries concerned is demanded by the design of the CBAM, and is not simply an evaluation of policy equivalence in reducing GHG emissions. Secondly, in the US – Shrimp cases the AB did not specify which of the conditions prevailing in the US and complainant countries were considered to be different.¹⁵⁷

In EC – Seal Products, the AB offered some more guidance on how to identify the conditions that are relevant to the comparison of countries under the chapeau. Drawing from dictionary definitions, it first stated that the term ‘conditions’ has a number of meanings and encompasses a number of circumstances facing a country.¹⁵⁸ This does not help much and it needs to be circumscribed. Otherwise, the vast array of differences in socio-economic, environmental and other circumstances across WTO members could open the floodgates to claims that conditions are not same and that the arbitrary/unjustifiable discrimination limb of the chapeau is not applicable. The AB somewhat delimited the range of prevailing conditions that are relevant by referring to ‘the subparagraphs of Article XX, and in particular [that] under which the measure has been provisionally justified, [as providing] pertinent context’.¹⁵⁹ In other words, the relevant conditions for the purpose of the chapeau are those that ‘relate to the particular policy objective of the measure under the applicable subparagraph of Article XX’ and, as such, will vary on a case-by-case basis. If the respondent considers such conditions are not ‘relevantly the same’ between the countries concerned in a particular case, it has the burden of proving that claim.¹⁶⁰

This approach makes good sense, as there would be no point in comparing countries on the basis of factors that have nothing to do with the policy objective for which the challenged trade measure was instituted.¹⁶¹ In the CBAM context, similarities or differences in conditions between the countries involved should therefore be directly related to the goal of conserving a stable climate, and it is irrelevant whether such countries differ in terms of other factors (eg, moral or religious values). It seems clear that countries’ respective responsibilities/capabilities for climate change mitigation are related to the CBAM's conservation objective, and it would be odd to suggest otherwise when the CBDRRC principle has underpinned the UNFCCC regime from its very foundation. The AB did not predetermine all conditions relevant to the chapeau discrimination analysis as this would be unfeasible, and indeed undesirable, given the broad range of policy objectives (and hence, potentially relevant conditions) reflected in the subparagraphs of Article XX GATT.¹⁶² Accordingly, it did not exclude the possibility that differentiated responsibilities/capabilities towards global environmental challenges could be a relevant condition for comparing countries under the chapeau, provided a relationship to the measure's objective can be shown. Contextual support for this is found in the preamble of the WTO Agreement,¹⁶³ which provides that members should seek ‘to protect and preserve the environment … in a manner consistent with their respective needs and concerns at different levels of economic development’.

But even if differentiated responsibilities/capabilities for climate conservation can be accepted as a relevant condition in the case of CBAMs, the assessment of whether these are sufficiently similar or different across countries needs to be based on objective criteria, rather than merely be asserted by the regulating WTO member. In fact, the AB took a similar position in the EC – Tariff Preferences case, which concerned the distinction between discrimination and lawful differentiation in the granting of tariff preferences under the GATT Enabling Clause.¹⁶⁴ It held that differential treatment was permissible insofar as it responds to the special needs of particular developing countries whose existence is based on an ‘objective standard’, including ‘broad-based recognition set out in the WTO Agreement or in international instruments adopted by international organisations’.¹⁶⁵ It is also noteworthy that, in US – Shrimp, the AB showed a willingness to use MEAs as factual evidence in the application of the chapeau requirements.¹⁶⁶

Applying this approach to the EU's CBAM, it is evident that the CBDRRC principle as given effect in the Paris mitigation provisions is the most broadly-recognised basis upon which to establish whether the countries involved are similarly-situated in terms of their responsibilities/capabilities for climate conservation. A concern here is that relying on the Paris Agreement could be unworkable as it may open the way to claims that such conditions are never the same, and hence that any form of differentiation for whichever country in the CBAM is permissible under the chapeau. As we have seen, it is true that the Paris differentiation model embraces the idea that each country's responsibility/capability is unique and different from every other country. But such claims can be restrained because what matters under the chapeau is the similarity of conditions, and not whether conditions are identical between any two countries since this would render the test largely ineffective in practice.¹⁶⁷

From this standpoint, Article 4.6 PA gives the EU a firm basis on which to argue that responsibility/capability conditions in LDCs and SIDS are relevantly different when compared to other countries affected by the CBAM, and hence that the LDC/SIDS exemption is permissible under the chapeau. LDCs/SIDS are the only group of countries not expected to undertake emission reduction commitments under the Paris Agreement, and no other country can claim to be similarly-situated in this regard. As to differentiation in favour of other developing countries in the form of EU financial assistance, this may be challenged under the chapeau even if not previously found to be inconsistent with the MFN obligation.¹⁶⁸ If it is, a similar line of defence would be available to the EU by relying on Article 4.5 PA, which only entitles developing countries to be financially supported in their mitigation efforts, and it would be unfounded to claim that developed countries affected by the CBAM are in this same condition.

To recap, introducing the two forms of differentiation in a CBDRRC-adjusted CBAM as recommended in Section III.C above should be permissible under WTO law, as these do not amount to discrimination between countries that are similarly-situated in terms of their responsibilities/capabilities for climate conservation. Importantly, this assessment of similarity should be based on the CBDRRC principle as operationalised in the Paris mitigation provisions, rather than upon generic references to that principle or to the principle of special and differential (S&D) treatment under WTO law.¹⁶⁹ These principles do not serve the same systemic purpose and, as such, follow distinct differentiation markers and country categories.¹⁷⁰ In particular, pursuant to the CBDRRC principle, LDCs and SIDS are granted the same differential treatment in the Paris mitigation provisions, while the two country categories are not placed on an equal footing under the WTO principle of S&D treatment.¹⁷¹ This is not a minor given that only 7 out of 38 countries falling in the SIDS category are also recognised as LDCs by the UN.¹⁷²

b) Arbitrary and unjustifiable discrimination

To further support the argument made in the previous section, it is useful to consider what would happen if it was not followed and it is simply assumed that prevailing conditions are the same for all countries affected by the EU's proposed CBAM. In this scenario, CBDRRC-based differentiation is likely to be condemned as unjustifiable discrimination under the chapeau, creating an unnecessary friction between the multilateral climate and trade regimes. This stems from the so-called rational connection (or regulatory rationality) standard, which was formulated in Brazil – Retreaded Tyres and reaffirmed in subsequent cases ‘as one of the most important factors’ in the assessment of unjustifiable discrimination,¹⁷³ and which focuses ‘on the cause of the discrimination, or the rationale put forward for its existence’¹⁷⁴ by the regulating WTO member. More specifically, discrimination between countries in same condition would be unjustifiable ‘when the reasons given for this discrimination bear no rational connection to the objective falling within the purview of a paragraph of Article XX, or would go against that objective … to however small degree’.¹⁷⁵

In Brazil – Retreaded Tyres,¹⁷⁶ the AB appeared to elevate the question of whether the reasons for the discrimination can be reconciled with the objective of the measure into some sort of litmus test: that is, any degree of contradiction will be in itself dispositive for a finding of unjustifiable discrimination under GATT Article XX-chapeau.¹⁷⁷ It displayed more caution in EC – Seal Products, where the discrimination at issue resulted from an exception from a sales ban on seal-containing products derived from hunts traditionally conducted by Inuit and other indigenous communities.¹⁷⁸ But even in this case, the AB remained ambivalent as to whether it was willing to accept that discrimination under the chapeau can be justified by legitimate reasons (in casu, protection of indigenous communities) other than the primary objective of the measure (in casu, seal welfare).¹⁷⁹ A unidimensional test in which one single purpose is the metric for justifying discrimination under the chapeau is excessively rigid. It renders particularly difficult the justification of discrimination that is caused by an exception, whereas real-life policymaking often needs to strike a balance between competing legitimate objectives.¹⁸⁰ This is because the rationale for an exception will not only differ from, but may often contradict, the objective underlying the general rule.¹⁸¹ As has been argued elsewhere, this rigidity proved problematic in EC – Seal Products, and to a lesser extent in Brazil – Retreated Tyres.¹⁸²

Applying the rational connection standard to CBDRRC-based differentiation in the CBAM, it is clear that the LDC/SIDS exemption is particularly vulnerable to challenge. Its rationale (ie, respecting the CBDRRC principle) is not related to the measure's overarching objective of preventing carbon leakage in order to combat global warming. Instead, it compromises the achievement of that objective by promoting the growth of carbon-intensive production in these countries,¹⁸³ even if only to a small degree in quantitative terms. Given the nature of climate change as a problem resulting from the overall accumulation of GHG emissions in the atmosphere, emissions by any State, regardless of location, contributes to the problem and affects all other States.¹⁸⁴ As has been seen, granting special treatment to LDCs/SIDS in Article 4.6 PA is not motivated by the goal of mitigating climate change per se, but rather by fairness considerations stemming from the CBDRRC principle.¹⁸⁵ However, this apparent tension between Paris-based differentiation and WTO law can—and should be—easily avoided through a proper application of the first (‘same conditions’) limb of Article XX-chapeau, as previously argued.

V. CONCLUSIONS

Reflecting its long-standing ambition to play a leadership role in the global battle against climate change, the EU may soon become the first jurisdiction in the world to price emissions embodied in its imports of carbon-intensive commodities. In doing so, it is showing courage and experimenting with climate policies that are yet to be tried elsewhere and which may have the potential of incentivising urgently needed global action to fight climate change. However, the EU is also testing the boundaries of permissible unilateral action at the intersection of pre-existing multilateral legal regimes. Before the proposed CBAM sees the light of the day, it ought to be made compatible not only with WTO law, but also with the UNFCCC framework, without which the pursuit of global climate targets would ultimately be in vain.

In this article, it was argued that the EU's CBAM as currently designed does not respect the CBDRRC principle as operationalised in the Paris mitigation provisions. To be brought in line with this principle, it needs to be adjusted through two forms of differential treatment: a full LDC/SIDS exemption (based on Article 4.6 PA) and a revenue-recycling provision to support decarbonisation in other affected developing countries (based on Article 4.5 PA). It was also argued that such CBDRRC-grounded differentiation is permissible under WTO law. This differential treatment should not be considered discriminatory within the meaning of the chapeau of Article XX GATT since the countries concerned are not similarly-situated in terms of their responsibilities/capabilities for climate conservation. Admittedly, the proposition that countries’ differentiated responsibilities/capabilities are a relevant condition for the purpose of the chapeau's discrimination analysis is yet to be tested in WTO dispute settlement. But it is critically important to avoid a potential and undesirable clash between WTO law and the UNFCCC regime. Indeed, it has broader significance for a mutually supportive relationship between WTO law and other MEAs which equally demand CBDRRC-based country differentiation in the protection of global natural resources, such as the Convention on Biological Diversity.¹⁸⁶