The EU in search of a WTO-compatible Carbon Border Adjustment Mechanism

Cecilia Bellora, Lionel Fontagné 26 March 2022



The Paris Agreement initiated a cooperative approach to address climate change, but the movement is running out of steam – hence, the call for an international carbon price floor differentiated by income level (Parry et al. 2021). However, even if such a solution were considered, international differences in carbon prices would still imply carbon leakage through the emitting European industries moving part of their production to countries with less strict climate policies, and imports substituting for European production. The lower demand for fossil fuels in Europe depresses the price of these fuels, which leads to higher consumption and higher greenhouse gas emissions by non-constrained countries, adding indirect leakages.

The European Commission proposed a ‘Fit for 55’ package in July 2021, which includes a Carbon Border Adjustment Mechanism (CBAM) that aims to avoid carbon leakage (Tanabe 2022). On 15 March 2022, the European Council agreed on the mechanism’s overall design. In a recent paper (Bellora and Fontagné 2022), we address three questions raised by the CBAM: 

1. To what extent will it really reduce carbon leakage induced by EU climate policy? 

2. In a world of global value chains (Baldwin and Freeman 2021), will it restore a level playing field for EU producers paying for the carbon they emit?

3. Is it designed to minimise the likelihood of WTO panels or even the prospect of retaliation by trading partners?

Currently, the allocation of free allowances to the sectors (i) covered by the European Emission Trading Scheme (ETS) and (ii) exposed to trade (Boehringer et al. 2012) addresses direct leakage. The ‘Fit for 55’ package envisages reducing direct leakage with the CBAM. However, the mechanism will reduce indirect leakage only if it incentivises non-European countries to curb their emissions more significantly – hence, the German proposal to set up a club of countries that share the objective of reducing their emissions (Bundesministerium der Finanzen 2021).

Offsetting carbon at the European border

The CBAM complements the central tool of EU climate policy – the Emissions Trading Scheme (EU ETS). The Scheme, set up in 2005, covers 40% of EU emissions generated by EU-based firms in certain sectors. It sets a cap on these emissions to reduce them by 61% by 2030 , compared to 2005 levels. The CBAM will add a carbon price on imported products whose production-related emissions have not been taxed (or not at the same level as in the EU) by the exporter’s country.

More specifically, the mechanism envisaged by the Commission combines:

  1. the purchase of allowances by importers on a specific market, price taker with respect to the EU ETS 
  2. a taxation base equal to the emissions of the exporter, eventually inclusive of indirect emissions associated with the energy mix of the electricity consumed in production processes
  3. a compensation net of the carbon price paid by the exporter in its own country
  4. the phaseout of free allowances over ten years
  5. the absorption in the European budget of the resources generated by the CBAM.

The European Council’s 22 March agreement on the mechanism keeps the discussion open on the thorny issues of ending free allowances – which is required to be WTO compatible and called for to make climate policy efficient – and compensating exporters for losses in competitiveness.

The Carbon Border Adjustment Mechanism substituting for free allowances

Let us tentatively put numbers on the environmental and economic effects of such a compensation scheme. Our modelling features a reference path for the world economy till 2040. Nationally determined contributions of all countries are carefully examined, as well as the European Emission Trading Scheme and free allowances. Our general-equilibrium modelling traces production displacements across sectors and countries and, consequently, accounts for carbon leakages. We explicitly take into account global value chains, as the pricing of emissions related to intermediate consumption affects downstream competitiveness. We use the Mirage-VA model calibrated using the GTAP 10.1 MRIO database.

For clarity, we assume that the CBAM covers the same industries as the EU Emissions Trading Scheme. This goes well beyond the proposed regulation, which is restricted to cement, aluminium, fertilisers, electric energy production, iron and steel. Our assumption is consistent with the Commission’s longer-term objective and makes it easier to identify the main mechanisms at play. It also avoids exposing the EU to legal challenges at the WTO for ‘cherry-picking’ the industries that will continue to receive free allowances. Our results relate the long-term impact of the CBAM extended to all industries in the Emissions Trading Scheme.

We consider here two scenarios. In scenario 1, the CBAM is designed to offset the direct emissions of the  EU ETS sectors. Reference emissions are the European average. Free allowances are phased out over ten years as the CBAM comes into force. There is differential treatment for the least developed countries to facilitate WTO acceptance of the new European regulation and to align with the European Parliament’s recommendation.

Scenario 2 replicates scenario 1 but uses the emissions of the exporter country as the reference for the adjustment. It is more ambitious in offsetting emissions and incentivising non-participating countries. The drawback is the (moderate) risk of being challenged at the WTO, as compensation would differ across exporters. The administrative burden of collecting information on foreign emissions is also a potential source of challenge. This scenario most mimics the one envisaged for the EU regulation.

Reduced leakage but competitiveness losses

Since the goal of the CBAM is to curb leakages, reduction of leakages is the first metric of its efficiency. We consider cumulated emissions (and leakages) over 2021–2040. Here, leakages are increases in the greenhouse gas emissions of third countries caused by the European climate policy. 

Implementing the European nationally determined contributions without an instrument to reduce leakages would generate cumulated leakages amounting to 18.5 Gt of CO2-equivalent. Allowances freely allocated to the industries exposed to leakage reduce these leakages by a third. Substituting the CBAM for free allowances, regardless of its design, further reduces leakage by up to two-thirds in scenario 2, compared to the situation without any instrument to deal with leakages. Scenario 2 is more efficient than scenario 1, but the difference is of second order (1 Gt). Put differently, leakages offset 62% of the reduction in EU greenhouse gas emissions achieved by the European nationally determined contributions in the absence of free allowances or the CBAM, 41% with free allowances, and 22% with CBAM in scenario 2. The CBAM does the job.

However, achieving this environmental goal has an economic cost. Comparing the counterfactual with the reference scenario, European imports of  EU ETS products decrease, as expected, but so do EU exports. Exports of intermediate goods decline because Emissions Trading Scheme producers now must buy their allowances and therefore lose competitiveness on third markets. European exporters of final goods use  EU ETS products as intermediates that are now more expensive, whether they are imported or produced in Europe, which induces a loss of competitiveness on the European market as well as on third markets. Here is the thorny issue of compensation for exporters that the March 15 agreement left out.

Lastly, the expected impacts of the CBAM on third countries deserve attention. First, the EU – a large country – enjoys a positive terms-of-trade effect when introducing the CBAM (+0.7% in scenario 1 and 0.9% in scenario 2). This is consistent with the theoretical prediction that the taxing country is extracting rent from the exporters (Balistreri et al. 2019) and may pose difficulties for the CBAM to be accepted by WTO members.

Another possibly contentious issue is the impact of the CBAM on the bilateral exports of the main European trading partners of carbon-intensive products. Figure 1 shows the impact on EU bilateral exports to (in red) and imports from (in blue) selected countries, with a darker shade for final products. The absolute variation in billions of US dollars is on the vertical axis, while relative changes in percentages are shown on the bars. For the sake of brevity, we report only results for scenario 2. 

Figure 1 Impact of the Carbon Border Adjustment Mechanism on EU27 bilateral trade, in 2040 (scenario 2)

Interestingly, we see trade imbalances favourable to Canada, Japan, and the US. The UK and European Free Trade Association are an extreme case: they benefit from low-carbon compensation and hence increase their exports, while this accumulates with reduced competitiveness of EU exporters (this reduction is also large in other markets). On the other extreme, the CBAM deeply affects India, with a -26% drop in its exports of intermediates to the EU.

A last contentious issue with the WTO could be the absence of clear earmarking of the large revenues generated by the  CBAM (up to +51% in EU tariff revenues in 2040 in scenario 2) in favour of environmental policies.

Coming back to our initial three questions, the  CBAM significantly curbs European carbon leakages but at a cost, and not only for the sectors covered by the mechanism and the EU Emissions Trading Scheme. In parallel, WTO compatibility necessitates (at least) the end of free allowances, whose modalities remain to be agreed upon among member states and with the Parliament.


Baldwin, R, and R Freeman (2021), “Risks and global supply chains: What we know and what we need to know”, NBER Working Paper 29444.

Balistreri, E J, D T Kafine and H Yonezawa (2019), “Optimal environmental border adjustments under the general agreement on tariffs and trade”, Environmental and Resource Economics 74(3).

Bellora, C, and L Fontagné (2022), “EU in search of a WTO-compatible Carbon Border Adjustment Mechanism”, CEPII Working Paper No. 2022-01.

Boehringer, C, J C Carbone and T F Rutherford (2012), “Unilateral climate policy design: Efficiency and equity implications of alternative instruments to reduce carbon leakage”, Energy Economics 34.

Bundesministerium der Finanzen (2021), “Steps towards an alliance for climate, competitiveness and industry – building blocks of a cooperative and open climate club”, August.

Parry, I, S Black and J Roaf (2021), “Proposal for an international carbon price floor among large emitters”, Staff Climate Note No 2021/001, International Monetary Fund.

Tanabe, Y (2022), “Japan should lead the global effort to decarbonise: Reflections on the EU’s Carbon Border Adjustment Mechanism”,, 23 January.



Topics:  Environment EU policies

Tags:  Carbon Border Adjustment Mechanism, carbon leakage, emission trading scheme, EU, greenhouse gas emissions, paris agreement

Economist, CEPII

Professor of Economics in the Paris School of Economics, Université Paris I Panthéon Sorbonne


CEPR Policy Research