Ref. Ares(2021)3754897 - 08/06/2021
Ref. Ares(2022)442654 - 20/01/2022
LEGAL OPINION
WTO COMPATIBILITY OF A CARBON BORDER ADJUSTMENT
MECHANISM (CBAM) IN COMBINATION WITH FREE ALLOWANCES
UNDER THE EU EMISSIONS TRADING SYSTEM (ETS)
Prof. dr. Freya Baetens
University of Oslo (Norway) - University of Leiden (Netherlands)
11 May 2021
1
The present Legal Opinion was drafted upon invitation of CEMBUREAU, CEFIC, EUROFER,
Fertilizers Europe, and Eurometaux.
Disclaimer
The views expressed in this Legal Opinion are attributable only to the author in a personal,
independent and impartial capacity, and not to any institution which she is associated with, or
to the associations mentioned above.
2
1.
Question
1.
I have been asked to provide a legal opinion assessing the WTO compatibility of the EU-
proposed Carbon Border Adjustment Mechanism (CBAM) in combination with free allowances
under the Emissions Trading System (ETS). Specifically, I have been asked to analyse whether:
- A CBAM can be designed in a manner that is compatible with WTO law while;
- Maintaining free allowances during a transitional period until 2030 (based on the fact
that carbon leakage covered by the current ETS is the risk of relocation to third countries
whereas the CBAM would cover part of the carbon leakage risk not yet covered by EU
ETS which is the import of CO2 intensive products);
- Accepting that the CBAM will need to be designed to avoid so-called ‘double protection’
which can be done by taking into account the level of free allowances, and foreseeing in
an export rebate or discount based on the destination principle.
2. Conclusion
2.
CBAM measures, including export rebates, can be designed in a WTO-consistent manner so that
they do not breach the non-discrimination provisions (most-favoured-nation treatment under
GATT Article I and national treatment under GATT Article III), the tariff bindings (GATT Article
II) and the prohibition on quantitative import restrictions (GATT Article XI).
3.
Should a breach of WTO law nevertheless be found, the CBAM could be justified through
reliance on the GATT Article XX exceptions, in particular subparagraph (b) (‘measures
necessary to protect human, animal or plant life or health’) or (g) (‘measures relating to the
conservation of exhaustible natural resources if such measures are made effective in conjunction
with restrictions on domestic production or consumption’), provided that the requirements of
the chapeau of GATT Article XX are met (‘such measures are not applied in a manner which
would constitute a means of arbitrary or unjustifiable discrimination between countries where
the same conditions prevail, or a disguised restriction on international trade’).
4.
In accordance with the available environmental and economic impact data, the combination of the
existing free allowances under the ETS and the new CBAM (including export rebates) can
accommodate environmental interests because a well-calibrated CBAM can address risks that are
not covered under the existing ETS. WTO law does not
per se prohibit such co-existence of
measures, provided that an appropriate methodology is used to calculate the exact scope of CBAM
measures so as not to overlap with the coverage of ETS allowances.
5.
In light of the fact that the lack of such co-existence risks being highly damaging, and the fact that
there is no automatic legal prohibition on the combination of such measures, it would be overly
restrictive to rule out co-existence at this point. Should a WTO Member decide to initiate
proceedings against an EU combination of measures, the EU could mount a strong legal defence
as there is no precedent for a
per se prohibition. Even if a WTO panel were to find that the EU
measures had to be adjusted, this would have to be executed at the earliest several years into the
future, would not apply retro-actively and would still allow the EU a considerable margin for
further negotiations, as is inherent to the WTO system.
3
3. Background
6.
On 10 March 2021, the European Parliament adopted a Resolution towards a WTO-compatible
EU Carbon Border Adjustment Mechanism.1 In this Resolution, the European Parliament
expresses its concern regarding the adverse impacts of climate change which represent a direct
threat to human livelihoods and terrestrial and marine ecosystems.2 Both the EU and its
Member States are committed under the Paris Agreement to undertake action based on the
latest available scientific evidence with the objective of achieving climate neutrality by 2050 at
the latest.3 The European Parliament is of the opinion that the current design of the EU ETS, in
particular the provisions on carbon leakage, have not provided effective incentives for the
necessary decarbonisation of industry sectors. 4
7.
The European Commission is in the process of developing methodologies to ascertain a
product’s carbon and environmental footprint, and studying the traceability of products and
services in order to identify their impact. 5 Around 27% of global CO2 emissions from fuel
combustion relate to internationally traded goods and 90% of international goods transport is
carried out at sea.6 Ensuring effective and meaningful carbon pricing as part of a broader
regulatory environment can serve as an economic incentive to develop production methods
with a lower GHG footprint and spur investments in innovation and new technologies. Adopting
an effective CBAM plays an important role in this context, particularly as none of the nationally
determined contributions (NDCs) submitted in the framework of the Paris agreement would
seem to be sufficient to achieve ‘the objective of keeping the global temperature increase […] to
well below 2°C’. 7 While the EU’s domestic GHG emissions have been substantially reduced, the
GHG emissions embedded in imports to the EU have been constantly rising to the extent that
these net imports now represent more than 20% of the EU’s domestic CO2 emissions.
8.
The analysis below addresses the
legal aspects of a CBAM in light of WTO law (in particular the
General Agreement on Tariffs and Trade (GATT) and the Agreement on Subsidies and
Countervailing Measures (ASCM)), but the decision whether to adopt a CBAM and if so, which
form such CBAM will take, is as much, if not more, a
political decision. Ministers of different
European Member States have called for its establishment as a CBAM could be effective (better
than existing instruments), legitimate (in compliance with WTO law and the Paris Agreement)
and fair (transparent for all and coordinated with trading partners). A CBAM is also supported
by non-State actors such as the industry (because it levels the playing field – EU producers
currently pay a carbon cost while non-EU importers do not) and NGOs (because it counteracts
carbon leakage and may induce non-EU producers to cut emissions).
9.
But John Kerry, former US Secretary of State, now US President Joe Biden’s envoy on climate,
warned the EU that a carbon border tax adjustment should be a ‘last resort’,8 urging the EU to
avoid putting a new tariff in place before the COP26 in November 2021. Likely to draw criticism
1 European Parliament resolution of 10 March 2011 towards a WTO-compatible EU carbon border adjustment
mechanism (2020/2043(INI)).
2
Ibid., A.
3
Ibid., D.
4
Ibid., G.
5
Ibid., H-I.
6
Ibid., J.
7
Ibid., para. 1.
8 Leslie Hook, ‘John Kerry warns EU against carbon border tax’
, Financial Times (12 March 2021).
4
from the EU’s other major trading partners, is the change the European Parliament inserted into
the CBAM report in the form of a levy that the EU would impose on imports from jurisdictions
with lower climate change standards
.9
10.
Nevertheless, the EU’s CBAM plans and their potential to expand into a blueprint for a global
‘carbon club’ are gaining support worldwide, as noted in, for example, a recent Financial Times
(FT) article which addressed the risk of high-emission countries ‘gaming’ the mechanism by
channelling their renewable power to exporters while reserving dirty power for home
production, if the EU were to grant importers from these countries a reduced carbon tariff for
using green energy. As the FT noted: ‘[t]he main function of a CBAM is to serve as an incentive
for other countries to introduce their own carbon pricing, so that the mechanism does not apply
between them. The more countries in such a “carbon club”, the greater the cost of standing
outside it.’ 10 As such, a CBAM seems to have gained the necessary momentum to be established.
4. Analysis
4.1
Can a CBAM be designed in a manner that is compatible with WTO law?
11.
Potential WTO law issues due to CBAM measures affecting imports – The question is how
a CBAM can be designed so as to be compatible with WTO rules, in particular, the non-
discrimination provisions (most-favoured-nation treatment under GATT Article I and national
treatment under GATT Article III), the tariff bindings (GATT Article II) and the prohibition on
quantitative import restrictions (GATT Article XI).
12.
CBAM measures could be compliant with WTO law, depending on the design of the calculation
method. First, if designed as a tax, the measure should be determined on a fair and objective
basis that relates to the specific products being taxed and not to their national origin. In other
words, a CBAM tax would have to be construed either as an ‘internal tax or other internal charge
of any kind’ applied ‘directly or indirectly’ to EU products to be compliant with GATT Article
III:2. Alternatively, it has to be designed as a law, regulation or requirement affecting an EU
product’s ‘internal sale, offering for sale, purchase, transportation, distribution or use’ to be
conform to GATT Article III:4. Both types of measures are recognised as border adjustable, but
price measures, such as taxes, can be more easily adjusted at the point of sale or consumption,
while an internal regulation is only adjustable for imports by applying the same or an equivalent
regulation also on domestic products.
13.
Second, the burden of the measure for each of the different sectors covered by the ETS should
not be ‘in excess of’ that applied, on average, to the sector of like domestic products, pursuant
to the first sentence of GATT Article III:2. Alternatively, in accordance with the second sentence
of GATT Article III:2, the products could be argued to fall under the looser category of ‘directly
or competitive’ or substitutable products (DCSP), rather than the stricter category of ‘like’
products. In that case, an argument needs to be made that consumers would make a distinction
between environment-friendly and -unfriendly produced goods, as the other parameters remain
the same (physical characteristics, end uses, tariff classification) and would lead to the strict ‘in
excess’ test applying to like products. As a result, even if the EU were to treat imported products
9 European Parliament resolution of 10 March 2011 towards a WTO-compatible EU carbon border adjustment
mechanism (2020/2043(INI)) para. 16 (emphasis added).
10 Martin Sandbu, ‘Time is ripe for EU to start a carbon club’,
Financial Times (30 March 2021).
5
differently than domestic products, the difference is not ‘dissimilar’11 and/or the intent is not
such ‘so as to afford protection’ to domestic products (SATP-standard).
14.
In other words, once an adjustable measure in compliance with GATT Articles II and III has
been adopted, the adjustment cannot impose a heavier burden on imported products as
compared to like domestic products.12 Such likeness is most frequently determined on the basis
of a competitive relationship in the market. As an illustration, EU cement would probably be
‘like’ US cement, irrespective of their carbon footprint which could create a
de facto discrimination (which could nevertheless be justified – see below). Such finding could be
averted, if the average carbon price levied on EU cement were to be imposed on US cement and
individual US importers were to be charged a lower price if they could demonstrate that they
emitted less. In such cases, however, the overall group of US cement importers might be treated
more favourably than the overall EU cement producers. Alternatively,
de facto discrimination
could be avoided if the CBAM measure would focus on inputs (for example, the energy used in
the production of cement) rather than end products. This would remove the risk of EU
producers to be treated less favourably.
15.
Potential WTO law issues due to CBAM measures affecting exports – A conditional export
rebate could also be part of the CBAM, to be applied to exports to countries that have not
implemented a regulatory framework aiming at GHG emissions reduction that can be
considered equivalent to the EU ETS. Export exemptions could be challenged as prohibited
subsidies in violation of ASCM Article 3.1(a) (financial contributions by a government or any
public body, such as the EU, in the form of government revenue that is otherwise due but is
forgone or not collected).13 However, export rebates should not be deemed subsidies if they are
designed and applied in accordance with the stipulations in the footnote to ASCM Article 1: ‘the
exemption of an exported product from duties or taxes borne by the like product when destined
for domestic consumption, or the remission of such duties or taxes in amounts not in excess of
those which have accrued, shall not be deemed to be a subsidy.’ Furthermore, the ASCM
confirms that rebate on exports, being a form of indirect taxes, are border adjustable.14
16.
Destination principle – The destination principle is part of a two-sided coin: on the one side,
it enables products destined for export to be relieved of some/all taxes charged in the exporting
State and, on the other side, it allows the importing State to impose a tax on the product, which
is applied to similar domestic products. Generally, it seems such taxes are compliant with WTO
law.
11 ‘[D]issimilar taxation of even some imported products as compared to directly competitive or substitutable
domestic products is inconsistent with the provisions of the second sentence of Article III:2’ (WTO, Appellate
Body Report,
Japan – Alcoholic Beverages II, p. 27): as a result, ‘balancing more favourable treatment of some
imported products against less favourable treatment of other imported products’ is in breach of WTO rules
(WTO, Appellate Body Report,
Canada – Periodicals, p. 29).
12 Joost Pauwelyn and David Kleimann, ‘Trade Related Aspects of a Carbon Border Adjustment Mechanism. A Legal
Assessment’,
Briefing requested by the INTA Committee, EP/EXPO/INTA/FWC/2019-01/Lot5/1/C/02 (2020) p. 9.
13 As defined under ASCM Article 1.1(a)(1).
14 ASCM footnote 58 defines ‘indirect taxes’ as ‘sales, excise, turnover, value added, franchise, stamp, transfer,
inventory and equipment taxes, border taxes and all taxes other than direct taxes and import charges’; for these,
border adjustment is permitted in line with ASCM Annex I(g) and (h), as opposed to ‘direct taxes’: ‘taxes on wages,
profits, interests, rents, royalties, and all other forms of income, and taxes on the ownership of real property’; for
these, border adjustment prohibited pursuant to ASCM Annex I(e). Joost Pauwelyn and David Kleimann, ‘Trade
Related Aspects of a Carbon Border Adjustment Mechanism. A Legal Assessment’,
Briefing requested by the INTA
Committee, EP/EXPO/INTA/FWC/2019-01/Lot5/1/C/02 (2020) p. 8.
6
17.
As a consequence, an export rebate/discount based on the destination principle seems to be
possible under WTO law. The question remains whether it is reasonable to expect the
Commission to design the CBAM as described. There are two potential reasons: first, it would
be WTO law compliant. Second, the CBAM could increase taxes for exports into States with
no/lower carbon prices to meet CBAM’s/the Green Deal’s objective. Hence, it takes into account
the carbon prices in the importing States – it would, therefore, only make sense to take into
account (positive and sufficient) carbon prices in other States if there is a tax in the importing
State similar to the CBAM. The destination principle would, hence, not only take away the risk
of double taxation but also be in line with the objectives of the Green Deal.
18.
Symmetrical or asymmetrical approach – Most experts seem to agree that a CBAM tax would
be WTO-conform, if it charges the imports in conformity with GATT Articles II:2 (a) and III:2
and rebates the exports in conformity with GATT Ad Article XVI.15 An asymmetrical system,
whereby only import taxes but no export rebates are provided, is also possible but somewhat
controversial. 16 In addition, such an asymmetrical system could result in double taxation if one
WTO Member were to adopt an asymmetric approach while another applied a symmetric one
– but this is not illegal (although policy-wise it may be preferable). Yet, export rebates could
potentially delay climate efforts within the EU (in case some EU producers would choose to
avoid paying the cost of carbon by exporting carbon-intensive products), unless their design
provides an incentive for EU producers to improve the emission intensity of their products.
19.
Justification in case of breach – Should a breach of WTO law nevertheless be found, for
example because of the calculation method of the import tax, the CBAM could be justified
through reliance on the GATT Article XX exceptions, in particular subparagraph (b) (‘measures
necessary to protect human, animal or plant life or health’) or (g) (‘measures relating to the
conservation of exhaustible natural resources if such measures are made effective in conjunction
with restrictions on domestic production or consumption’). Under subparagraph (b), it could,
for example, be argued that a CBAM protects humans, animals and plants from the negative
effects of climate change, although the application of the necessity test might raise some
evidentiary issues. Under subparagraph (g), only a relationship between the goal and the
measure would need to be established: as climate change arguably affects all natural resources,
the burden of proof might be easier to meet under this subparagraph. In both cases, a sufficient
nexus with the EU territory can be established.17 The important thing to keep in mind is that an
economic justification would not be covered by GATT Article XX, and would endanger the
15 Anna Dias, Stéphanie Seeuws, Agnieszka Nosowicz, ‘EU Border Carbon Adjustment and the WTO: Hand in Hand
Towards Tackling Climate Change’, 15
Global Trade and Customs Journal 1 (2020) pp. 15–23; Matthew Porterfield,
‘Border Adjustments for Carbon Taxes, PPMs, and the WTO’, 41
U. Pa. J. Int'l L. (2019) p. 1; Ulrike Will,
Climate
Border Adjustments and WTO Law: Extending the EU Emissions Trade System to Imported Goods and Services,
(Brill 2019); Joel P. Trachtman, ‘WTO Law Constraints on Border Tax Adjustment and Tax Credit Mechanisms to
Reduce the Competitive Effects of Carbon Taxes’,
Resources for the Future Discussion Paper 16-03 (2016); Gabrielle
Marceau, ‘The Interface between the Trade Rules and Climate Change Actions’, in: Deok-Young Park (ed.),
Legal
Issues on Climate Change and International Trade Law (Springer 2016) pp. 3-39; Joost Pauwelyn, ‘Carbon Leakage
Measures and Border Tax Adjustments under WTO Law’, in: Geert Van Calster and Denise Prévost (eds.),
Research Handbook on Environment, Health and the WTO (Edward Elgar 2013) p. 448.
16 Reinhard Quick, ‘Carbon Border Adjustment -
A dissenting view on its alleged GATT-compatibility’,
ZEuS 4
(2020) p. 555.
17 Barbara Cooreman,
Global environmental protection through trade: a systematic approach to extraterritoriality
(EE 2017).
7
environmental defence of CBAM.18 The key question in connection with GATT Article XX(b)
may be whether less trade restrictive alternative measures could achieve the same objective.
20.
Evidently, the requirements of the chapeau of GATT Article XX would need to be met (‘such
measures are not applied in a manner which would constitute a means of arbitrary or
unjustifiable discrimination between countries where the same conditions prevail, or a
disguised restriction on international trade’). If the calculation of the border adjustment is based
on the carbon host in the producers’ country, it is arguably not arbitrary. Differentiation
between developed and developing (particularly least developed) countries could be justified in
light of historical emission rates but it risks opening up a new area of controversy. It is important
to treat all developed countries the same but there is no agreement on any CO2 measurement
method,19 so the EU may be faulted by relying only on its own measurement system. The
WTO/GATT
Gasoline and
Shrimp/Turtle cases demonstrated that one needs to make allowance
for different systems in different countries. For example, what if Egypt or the US were to put in
place solar panel fields in the desert, but without adding place restrictions for industrial
production? In other words, ensuring compliance with the chapeau might turn into a real
difficulty.
21.
Remaining issues – Yet, CBAM measures are not a simple variant of border tax adjustments: it
ought to be clarified whether a CBAM is a domestic measure applied at the border or a border
measure. Different interpretations may ‘blur the distinction between “indirect” and “direct”
taxes and potentially open the floodgate for adjusting also “direct” taxes at the border. How can
one continue to reject the adjustability of a producer tax, such as a corporate tax, if one considers
a carbon tax on producers to be an adjustable “indirect” tax?’ 20 Moreover, a CBAM might raise
questions as to the nature of ‘typical environmental command and control regulations’:
currently they are considered producer regulations (GATT Ad Article III), but what if someone
would ‘define as product regulation any domestic legislation which affects to some extent the
internal sale of products’?21 In such cases, WTO Members could arguably justify border
adjustments for all types of environmental command and control regulations. The main risk
seems to be that CBAM measures would have a punitive or coercive effect, which would be
difficult to justify.
22.
Possible solution – Additional safeguards could be put in place to ensure WTO-conformity.
Legally the least problematic, but politically difficult, solution would be a CO2 consumption tax
on domestically consumed products, which could be accompanied by an equivalent tax on
imports imposed at the time of distribution or sale. Similar to a VAT, a typical ’indirect’ tax on
products would be adjustable at the border. For any measures, the calculation method of the
carbon price (in comparison to the treatment of domestic producers) would seem key.
18 E.g., James Bacchus, What trade and environment negotiations agenda for the WTO? Peterson Institute for
International
Economics,
Trade
Winds
series
(17
March
2021)
available
at
https://www.youtube.com/watch?v=MB193wdZZgU.
19
Ibid.
20 Reinhard Quick, ‘Carbon Border Adjustment -
A dissenting view on its alleged GATT-compatibility’, ZEuS 4
(2020) p. 570.
21
Ibid.
8
4.2
Can a CBAM maintain free allowances during a transitional period until 2030
and foresee in an export rebate or discount based on the destination
principle, if designed to avoid so-called “double protection”?
23.
This question is first and foremost not a legal question, but one of environmental and economic
impact assessment. Once such assessment has allowed to conclude that the absence of co-
existence of CBAM measures and free allowances does risk to have a significant impact (4.2.1),
the legal question is whether such absence is legally required (4.2.2).
24.
For the sake of clarity, the meaning of term ‘co-existence’ below implies the situation in which
CBAM measures would only be imposed above the benchmarks used to assess EU producers’
emissions. In other words, EU producers would maintain their current level of free allowances
up to the benchmarks and would be subject to the applicable CBAM measures for emissions
above those benchmarks. Equally, foreign producers would not be subject to CBAM measures
if their emissions fall within the benchmarks, only insofar as they are exceeding the benchmarks.
In this manner, all producers would be subject to the same standards.
4.2.1
Assessing the environmental and economic impact of the co-existence of
CBAM measures and free allowances under the ETS
25.
Net increase of global emissions – EU products are on average less carbon intensive but more
costly than foreign products, ‘an asymmetry that is set to expand as the European Green Deal is
operationalised’, likely to result in a loss of market share in foreign markets, which turn will
create ‘a net increase of global emissions’ as foreign consumers turn to more carbon intensive
and less expensive non-EU products.22 Such risk could be addressed by maintaining a level of
free allocation for both domestically sold and exported European products.
26.
Position of EU producers in third markets – In the absence of a rebate for exports, the CBAM
with full auctioning risks jeopardising the position of EU producers in third markets. Producers’
interests match environmental integrity goals since loss of production would increase carbon
leakage. Removing existing carbon leakage measures would expose EU producers to the full
compliance of carbon costs, hence reducing their financial ability to compete internationally
and, more particularly, to invest in innovative low carbon technologies when such costs cannot
be recovered from the market.
27.
Digressive allocation and cross-sectoral correction – Free allocation is digressive and
subject to the cross sectoral correction factor (CSCF) when the ETS cap is too strict. The
benchmarks recently adopted for the period 2021-2025 in Phase IV of EU ETS23 impose very
stringent requirements on EU operators; the way in which the benchmark setting is defined
means that only the top 5% EU producers will receive their allowances for free, provided that
there is no application of the CSCF. While the exact date is open to debate, it is likely that there
is a point where the EU ETS will run out of free allocation.
22 Andrei Marcu, Michael Mehling and Aaron Cosbey, ‘CBAM for the EU: A Policy Proposal’,
ERCST Roundtable on
Climate Change and Sustainable Transition (2021) p. 10.
23 Commission Implementing Regulation (EU) 2021/447 of 12 March 2021 determining revised benchmark values for
free allocation of emission allowances for the period from 2021 to 2025 pursuant to Article 10a(2) of Directive
2003/87/EC of the European and of the Council, O.J. 2021, L 87, p. 29.
9
28.
Impact mitigation for downstream sectors – A well-designed complementary CBAM/ETS
could mitigate the impact on downstream sectors since the carbon leakage measures should
limit the cumulative costs for downstream sectors.
29.
Legitimate (temporary) expectations – Industrial sectors subject to carbon leakage argue
that from an investment perspective, the next 9 years have been mapped out. Therefore,
introducing a CBAM and taking away the ETS allowances risks going against the investment
plans to which companies have already committed themselves until 2030. The industry’s request
that the CBAM’s ‘implementation should not lead to abrupt modifications of existing provisions
in order to secure legal certainty for long term investment decisions’,24 would seem reasonable
in this respect. In any case, benchmarks are continuously tightened, so the industry has to buy
an ever-increasing portion of their emissions above the benchmark.
30.
In sum – In accordance with the available environmental and economic impact data, the
combination of the existing free allowances under the ETS and the new CBAM can accommodate
environmental interests because a well-calibrated CBAM can address risks that are not covered
under the existing ETS.
4.2.2
Assessing the EU’s legal obligations with regard to the co-existence of CBAM
measures and free allowances under the ETS
31.
No explicit prohibition – There is no clear WTO rule, or any case law precedent, that would
seem to prohibit the combination of measures whereby CBAM measures would co-exist with
free allowances under the ETS. Without a proper economic analysis, it is not possible to proceed
to a correct compatibility analysis nor is it legally justified to pre-emptively conclude that the
combination of free allowances and a CBAM are
per se incompatible with WTO law. The fact
that such co-existence may not be the preferred political option should not impact the legal
feasibility and the framing of conditions under which WTO law can be complied with.
32.
Current ETS provides only partial coverage – The notion of carbon leakage comprises the
situation where either the production is transferred from the EU to other countries with lower
ambition for emission reduction,
or where EU products are replaced by more carbon-intensive
imports.25 Article 10(a) of the EU ETS Directive 2003/8726 tackles the first part of that carbon
leakage notion in that it foresees in a 100% free allocation for sectors that are determined to be
under a significant risk of carbon leakage; the 100% is applied on the basis of a benchmark set
at the level of the average of the 10% best performers. Existing carbon leakage measures (ETS
phase four) are therefore seen as partial, since they are based on strict performance benchmarks
set at the level of the average best 10% installations. The argument is that these benchmarks
have deliberately been set high because the EU wants to create incentives for less efficient
24 See e.g., EUROFER,
EU Green Deal on Steel: EU Climate Leadership Requires Waterproof Carbon Leakage Measures
(2 March 2020): https://www.eurofer.eu/publications/position-papers/eu-climate-leadership-requires-airtight-
carbon-leakage-measures/.
25 See defined as such in point 5 of the Communication from the Commission of 21st September 2020, Guidelines on
certain state aid measures in the context of the system for greenhouse gas emission allowance trading post 2021,
COM(2020) 6400final.
26 Directive 2003/87 establishing a system for greenhouse as emissions allowance trading within the Union, as last
amended by Directive (EU) 2018/410 of the European Parliament and of the Council of 14 March 2018, O.J. 2018,
L 76, p. 3.
10
producers to reduce their emissions so that they, too, would receive full allowances.27 In other
words, further carbon leakage measures benefitting less efficient installations would arguably
remove their incentive to become more efficient. The counterargument to this is that even with
a complementary CBAM, EU installations still maintain an incentive to improve their
performance and achieve the benchmarks, since they would save on their ETS compliance costs
and have a stronger position against imports.
33.
CBAM would provide additional risk coverage – The existing carbon leakage measures, such
as free allowances under the EU ETS, cover the risk of relocation of EU industry to third
countries, whereas the CBAM would cover a part of the carbon leakage risk that is not yet
covered by the EU ETS: the import of CO2 intensive products. A well-designed CBAM should
eventually be able to address both risks: if a producer relocates to a third country with lower
carbon prices, it will have to pay the CBAM import tax if it wants to have access to the EU
market. Therefore, the CBAM should be aimed to prevent the risk of relocation to third
countries. As stated above
, in principle, the CBAM could charge the imports with the tax in
conformity with GATT Articles II:2 (a) and III:2 and rebate the exports in conformity with GATT
Ad Article XVI (although the latter is not necessary from a legal perspective). Alternatively, the
CBAM could be structured as a carbon tax on domestically consumed products, imposed at the
time of distribution or sale. Similar to a VAT, a typical ’indirect’ tax on products would be
adjustable at the border. The latter alternative (
i.e., a carbon tax on domestic goods) could
ultimately make the ETS scheme unnecessary, which would then be an argument for free
allowances under the CBAM. Some might argue that a system of ‘free allowances’ would seem
to introduce an element of discrimination favouring domestic products. This criticism can be
addressed by demonstrating, first, that the CBAM level would take into account the level of free
allowances, and second, that free allowances contribute to an
environmental goal,
e.g., surpassing the minimum requirements of CBAM.
34.
Methodology is key – If ETS were characterised as a tax on a product, the situation is rather
straightforward: a tax on a product is more easily applied than a methodology to calculate the
actual financial burden of a measure. If the CBAM were to fall under GATT Article III:2 second
sentence, then a WTO-complainant would have to prove the measure is imposed ‘so as to
protect’ domestic products (SATP-standard, as explained above). If the free allowances for the
EU domestic industry (and other circumstances related to ETS, such as carry over of allowances
to another year) are reflected in the CBAM for imported products, then there is equality in
opportunities and the complainant will not be able to meet the SATP standard.
35.
Since ETS may not be characterized as a tax on a product (because it is arguably difficult to call
legislation which refers to environmental performance of installations a fiscal measure),28 the
Commission will have to design a methodology to calculate the financial burden for each
product produced, bearing in mind that installations have different energy sources (coal,
nuclear, gas, renewable energy). The ‘no less favourable’ treatment standard (GATT Article III:4)
27 Gabriel Felbermayr and Sonja Peterson, ‘Economic assessment of Carbon Leakage and Carbon Border
Adjustment’,
Briefing requested by the INTA Committee, EP/EXPO/INTA/FWC/2019-01/Lot5/1/C/02 (2020) p. 13.
28 Case C-366/10,
Air Transport Association of America and Others v. Secretary of State for Energy and Climate
Change, Judgment of the Court (21 December 2011) ECLI:EU:C:2011:864.
11
requires effective equality of opportunities for imported products to compete with domestically
produced products.
36.
It has been suggested that CBAM should mirror ETS as much as possible. It has also been argued
that the pool of certificates to be created for imports would be another than the pool of
certificates existing under ETS. This could be problematic if the domestic system is more
beneficial (
e.g., from a purely administrative point of view) than the system created under
CBAM. Since the financial burden for each producer is different under ETS, a calculation
methodology could, for example, be built on the basis of a benchmark, namely the average
amount of certificates which are required for a ton of the specific product.
37.
Within this methodology, one has to position free allowances as another aspect of establishing
the financial burden of ETS on a given product. This average amount of certificates would have
to be surrendered by the importer unless it could demonstrate that the country of production,
or more specifically the installation, is more CO2 friendly than the EU’s average. The question
is what measures will be taken into account when allowing for these adjustments: production-
related CO2 measures or country-related CO2 measures.? Under GATT Article III, which
concerns products, one could reasonably argue that only production-related CO2 measures
would have to be considered. Since each industry is different, the methodology will have to be
established for each industry and for each product.
38.
One could apply both free allowances and CBAMs as long as it can be ensured that domestic
production is not privileged over imports. The Commission would therefore have to publish a
calculation which clearly demonstrates that the CBAM imposes a cost on importers comparable
to the average cost of ETS for a specific domestic industry sector (including the free allowances
it receives, and other privileges such as the ability to carry over allowances to the next year).
-
If the CBAM is fiscal, then GATT Article II:2 (a) and GATT Article III:2 first or second
sentence have to be applied. If the ‘tax’ is applied in accordance with GATT Article III:2,
it can be adjusted at the border, imposed on imports and rebated for exports (the EU
can freely decide whether to rebate the tax on export).29 If the free allowances are
correctly accounted for within the CBAM calculation, there is no subsidy issue. Of
course, those fiscal CBAM measures would still have to comply with GATT Article III:2.
-
If the CBAM is regulatory, the same problem arises under GATT Article III:4. The EU’s
calculation method determining how many certificates have to be surrendered at the
border has to take all domestic privileges into account, in particular free allocation.
Under GATT Article III:4 the rebate on exports could raise questions under the Subsidies
Code. If the free allowances for the domestic industry are taken into account when
calculating the CBAM for imported products, however, this is likely to neutralize the
complaint under the ASCM that the domestic industry receives a ‘benefit’.
39.
If the EU overcomes the GATT Article III hurdle and assuming there is no violation, it still has
to overcome GATT Article I-based objections, as described above. If it applies different
treatments for different countries, it risks committing a violation which it would have to justify
under GATT Article XX. Under the chapeau of GATT Article XX, the issue of methodology
resurges. Here the EU would have to prove that it has considered all CO2 related measures taken
29 GATT Ad Article XVI and footnote 1 Subsidies Code.
12
by the exporting country when calculating the CBAM. In sum, the situation is legally clear but
methodologically more complex – but not impossible.
40.
WTO litigation – Any CBAM (whether co-existing with free allowances under the ETS or not)
could give rise to trade tensions and could be challenged at the WTO. In light of the fact that
many countries are currently developing their own CBAM, such WTO challenge would seem
less likely as WTO Members would risk undermining their own mechanisms. What does seem
likely is that the next decade will see a considerable amount of ‘fine-tuning’ of different CBAMs
so as to arrive at a ‘carbon club’ with largely the same or similar CBAMs, if only to make the
calculation methods of carbon costs comparable. Nevertheless, even if the EU CBAM would be
challenged at the WTO, this would not entail immediate drastic changes.
-
First, WTO Members have to attempt to reach a mutually agreed solution through
consultations – a large number of cases are settled in this stage.
-
Second, if a WTO panel is constituted, the EU could mount a strong legal defence as there
is no clear rule on, or precedent for, a
per se prohibition of either a CBAM or a combination
of measures such as CBAM and free allowances under ETS.
-
Third, even if a WTO panel were to find that the EU measures had to be adjusted,
depending on whether the complaining member is a member of the Multi-Party Interim
Appeal Arbitration arrangement,30 or when the WTO Appellate Body would again be able
to review appeals, the dispute might be subjected to appeal proceedings.
-
Fourth, even if a WTO report recommending adjustment of EU measures would stand, the
EU would have a ‘reasonable period of time’ (in practice frequently several years) to
implement this report. During this time, the EU would have a considerable margin for
further negotiations, as is inherent to the WTO system. It is exactly this ‘nature and
structure’ of the WTO covered agreements that led the Court of Justice of the EU to
deny direct effect to WTO law in the EU legal order on the ground that ‘to require the
judicial organs to refrain from applying the rules of domestic law which are inconsistent
with the WTO agreements would have the consequence of depriving the legislative or
executive organs of the contracting parties of the possibility […] of entering into
negotiated arrangements even on a temporary basis’.31
-
Fifth, WTO reports at most recommend an adjustment of measures for the future. Even if
the co-existence of a CBAM and free allowances were to be found WTO-inconsistent,
the adjustments required to make the EU measures WTO–compliant would not apply
retro-actively so no ‘benefits’ would need to be refunded.
41.
In sum – In light of the fact that the lack of such co-existence risks being highly damaging and
that there is no automatic legal prohibition on the combination of such measures, ruling out co-
existence would be a form of ‘pre-emptive self-limitation’ that is neither legally mandatory nor
necessary. WTO law does not
per se prohibit such co-existence of measures, provided that an
appropriate methodology is used to calculate the exact scope of CBAM measures so as not to
30 JOB/DSB/1, Multi-Party Interim Appeal Arbitration Arrangement Pursuant to Article 25 of the DSU (30 April
2020), available at https://trade.ec.europa.eu/doclib/docs/2020/april/tradoc_158731.pdf.
31 Case C-149/96,
Portuguese Republic v Council of the European Union, Judgment of the Court (23 November 1999)
ECLI:EU:C:1999:574.
13
overlap with the coverage of ETS allowances. The Commission should develop its system bearing
clear WTO rules and existing precedents into account, in the knowledge that it has a strong legal
position in potential WTO dispute settlement proceedings and, in any case, in further negotiations
with third countries.
14
