Carbon labelling schemes at the WTO: Real or imagined conflict?

14 October 2015

Would carbon labelling initiatives be covered by the WTO’s Agreement on Technical Barriers to Trade?

It is broadly recognised, including in the WTO, that climate change is possibly the greatest sustainable development challenge presently facing the global community. The 1992 UN Framework Convention on Climate Change (UNFCCC), with its near-universal membership, represents the international response to this challenge with the ultimate objective of stabilising “greenhouse gas emissions concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system.” To this end, the Intergovernmental Panel on Climate Change (IPCC) has estimated that global greenhouse gas (GHG) emissions would need to be reduced to close to zero during the second half of this century in order to achieve the internationally-agreed target of limiting average temperature increase to below two degrees Celsius from pre-industrial times. And yet, the critical question remains which policies and specific measures are most appropriate and effective to mitigate climate change, a matter the UNFCCC largely leaves its parties to decide.

In this context, “carbon labelling” schemes for various products have emerged and proliferated since 2007 as one possible means to reduce GHG emissions, even though not as prominently as other market-based information instruments – such as energy-efficiency labelling schemes – or more direct forms of climate mitigation regulation – minimum energy-efficiency requirements – and support – renewable energy subsidies. [Ref 1] These, mainly private, carbon labelling initiatives are nevertheless seen as potentially problematic from a WTO law perspective, as evidenced by the discussions on the possible trade impacts of carbon labels that have taken placed in the WTO Committee on Trade and Environment (CTE). (WT/CTE/M/49) Of most significance among the various WTO legal texts is the Agreement on Technical Barriers to Trade (TBT), which deals specifically with labelling requirements, and whose core provisions have been decisively clarified by the Appellate Body in four recent disputes: US-Clove Cigarettes; US-COOL; US-Tuna II, and EC-Seal Products. This article seeks to appraise the key implications of the latest TBT case law for carbon labels. In doing so, it will challenge some common (mis)perceptions regarding the WTO-compatibility of measures based on so-called “non-product-related” processes and production methods (PPMs), while unfolding the central issues that remain to be addressed.

Main features of carbon labels

As a backdrop to our subsequent discussion, it seems first useful to briefly outline the main features of carbon labels. In terms of objectives, carbon labels are basically intended to provide consumers with a visible representation of the GHG emissions that can be attributed to a specific product, and thereby enable them to differentiate between high-carbon and low-carbon products in their purchasing decisions. In addition to this consumer information objective, carbon labels further seek to effect changes in consumer choice towards less emission-intensive consumption and production patterns, and thereby contribute to climate change mitigation. With regard to operating entities, carbon labels were originally introduced by UK supermarket chains Tesco and Marks&Spencer in 2007, and since then have proliferated mainly as private sector standards. [Ref 2] Nonetheless, a few carbon labelling schemes have been adopted by governments, or developed with significant governmental involvement. At present, the largest such scheme is the “Carbon Reduction Label,” operated by the UK Carbon Trust – a government-funded non-for-profit organisation – in association with the British Standards Institute and the Department for Environment, Food and Rural Affairs, followed by Japan’s “Carbon Footprint of Products System” and France’s carbon footprint pilot programme. Currently these are all voluntary schemes. [Ref 3]

It is also important to distinguish between two types of carbon labelling schemes; the so-called “food miles” used by commercial companies, which strictly focuses on GHG emissions released during a product’s transportation and is based on the single criterion of the distance travelled by air freighted products to reach the consumer; and the “product carbon footprint” (PCA) that instead seeks to account for the total amount of GHG emissions generated during the entire life cycle of a product from “cradle-to-grave,” through all stages of their production, processing, transportation, sale, consumer use and disposal or recycling. All governmental, and some private, carbon labelling schemes follow this life cycle analysis (LCA), but even so each employs a specific calculation methodology. As a result there is presently no generally accepted method for determining the carbon value of a product. (WT/CTE/M/49/Add.1)

Are carbon labels subject to TBT disciplines?

It has been the subject of much debate whether the TBT Agreement applies to carbon labels mainly because these are inherently based on non-product-related or unincorporated process and production methods. [Ref 4] In other words, the amount of GHG emissions generated during a product’s transportation or/and production does not affect in any evident manner the physical characteristics of the final product. The application of the TBT Agreement to measures specifying non-product-related PPMs was one of the most knotty controversies during its negotiation process, resulting in rather ambiguous provisions (Annex 1) that have not yet been fully clarified by the WTO dispute settlement organs. [Ref 5] Nonetheless, the text of Annex 1 TBT Agreement omits the term “related” when referring to labelling requirements, and thus an argument can be made that the product-related/non-product-related distinction is largely irrelevant in determining the agreement’s coverage of labelling schemes. Further, this proposition finds support in the recent TBT rulings, which seem to indicate that all labelling requirements are covered by the TBT Agreement irrespective of the PPM being addressed by the label. [Ref 6] The most obvious example is the US dolphin-safe labelling scheme at issue in US-Tuna II, which was scrutinised against TBT disciplines even though it was clearly based on non-product-related PPMs, namely fishing methods for catching tuna. The discussions in the CTE Committee, moreover, reveal that the PPM issue is actually not the crux of disagreement among WTO Members over the application of the TBT Agreement to carbon labelling schemes. It is, rather, the fact that these schemes are predominantly private sector initiatives. (WT/CTE/M/49, paras. 7-20)

While the TBT Agreement is quite unique in explicitly addressing voluntary standards adopted by “non-governmental” bodies, the extent that it is applicable to private sector standards is a complex and unsettled question, which will among other things depend on what is meant by a “recognised body.” [Ref 7] However, what does seem clear is that carbon labelling schemes are subject to TBT disciplines if adopted and applied by governmental bodies of WTO Members, or otherwise with significant governmental involvement. [Ref 8] Which TBT disciplines apply to such governmental carbon labels will, in turn, depend on whether they qualify as a mandatory technical regulation (Article 2 TBT) or a voluntary standard (Article 4 and Annex 3 TBT), but these are broadly similar for the core non-discrimination and necessity obligations discussed next.

Would carbon labels be compatible with TBT disciplines?

The TBT Agreement explicitly recognises that WTO Members have a right to adopt technical regulations and standards necessary to fulfil certain legitimate policy objectives, which certainly encompass consumer information and climate change mitigation carbon labelling schemes (TBT Agreement, Preamble, sixth recital; AB Report in US-COOL, paras. 445 and 453). However, as reaffirmed by the Appellate Body in US-Clove Cigarettes, this right to regulate is not unbound but subject to the condition that such measures are not used as a means of “arbitrary or unjustifiable discrimination” or otherwise create an “unnecessary obstacle” to international trade. (AB Report in US-Clove Cigarettes, paras. 92-95) A pertinent question is, therefore, whether carbon labels can be designed and applied in a manner that meets these requirements.

A means of arbitrary or unjustifiable discrimination?

As is well-known, non-discrimination is one of the basic principles in WTO law, formulated in Article 2.1 for technical regulations and Annex 3.D for standards of the TBT Agreement, and requiring WTO Members to accord “treatment no less favourable” to imported products than that accorded to “like products” of domestic origin – national treatment obligation – or originating in any other country – most-favoured-nation (MFN) obligation. It is often claimed that carbon labelling schemes are in tension with these non-discrimination obligations because they are likely to have a detrimental impact on the competitive opportunities for imported high-carbon products from one country vis-à-vis low-carbon products of domestic origin or/and originating from other countries. [Ref 9] This perception, however, needs to be refined. To begin with, whether or not high-carbon and low-carbon products are “like products” can only be assessed on a case-by-case basis, depending on “the nature and extent of [their] competitive relationship” in a given market. (AB Report in US-Clove Cigarettes, para. 120)

Likeness between products in WTO law has been traditionally determined on the basis of four criteria – the products’ physical characteristics, their end-uses, consumers’ tastes and habits, and tariff classification – which do not include PPMs per se. However, there is no reason why unincorporated PPMs cannot be considered through these likeness criteria, as these PPMs can affect consumer preferences and demand. Thus, there is a possibility that the competitive relationship between high-carbon and low-carbon products in a given market is too weak, so these are deemed “unlike” in that particular market. However, it seems improbable that this type of situation will often arise in practice because if high-carbon and low-carbon products are not, or only weakly, competing on the market, why introducing a carbon labelling scheme in the first place?

But even if high-carbon and low-carbon products are found to be “like”, this does not necessarily mean that they cannot be treated differently, or that any detrimental impact on the competitive opportunities for high-carbon products is prohibited. For origin-neutral measures, detrimental impact is only prohibited in cases where it does not stem exclusively from a legitimate regulatory distinction. (AB Report in US-Clove Cigarettes, paras. 173-175 and 182)

To better illustrate this subtle point, in US-Tuna II, the Appellate Body did not condemn the US dolphin-safe labelling scheme justbecause it differentiated between tuna products on the basis of how the tuna had been caught, an unincorporated PPM. It did so because this regulatory distinction causing detrimental impact on Mexican tuna products was not “calibrated to the risks to dolphins arising from different fishing methods in different areas of the ocean.” (AB Report in US-Tuna II, para. 297) In other words, labelling schemes based on unincorporated PPMs are not per se inconsistent with the TBT non-discrimination obligations. But they need to be designed and applied in an even-handed manner so that differences in the treatment of like products are fully explained or justified by a legitimate policy objective.

Applying this to carbon labels, food miles appear most problematic as they differentiate between high-carbon and low-carbon products on the basis of a single criterion of how far the product has travelled. This mileage-based criterion has an in-built bias against imports of long-distance food products and the de facto discrimination cannot be fully justified by the objectives of informing consumers or mitigating climate change. Quite the contrary, studies have shown that food miles do not generally provide a reliable indication of the carbon value and climate change impact of food products, as most GHG emissions are generated during the production phases – about 80 percent – rather than during transportation. [Ref 10] This could lead to the arbitrary and misleading result that a product is presented to consumers as “low-carbon” while more GHG emissions were, in fact, generated during its entire life cycle when compared to a locally produced like product. From this angle, carbon footprint labels are expectedly more even-handed given the broader range of factors included in the life cycle analysis, but concerns have been equally raised regarding the selectivity of the accounting criteria – for example, by ignoring GHG emissions from domestic transportation or associated with capital plant – and the lack of impartiality in its application – for instance, by relying on secondary data or default values, particularly for developing countries where accurate data may be more difficult to obtain. [Ref 11]

Unnecessary obstacle to trade?

In addition to the non-discrimination obligations just seen, the TBT Agreement sets forth an independent requirement that technical regulations (Article 2.2 TBT Agreement) and standards (Annex 3.E TBT Agreement) are not prepared, adopted, or applied with a view to or with the effect of creating unnecessary obstacles to international trade. The full potential of this necessity requirement in challenging the WTO-consistency of product regulations and standards, even non-discriminatory, is still to be unveiled by the Australia-Tobacco Plain Packaging dispute presently before several WTO panels.

Nevertheless, the Appellate Body clarified in US-Tuna II that this necessity test involves a weighing and balancing process of a number of factors, namely: (i) the degree of contribution made by the measure to the legitimate objective at issue; (ii) the trade-restrictiveness of the measure; and in most cases (iii) whether a less trade-restrictive alternative measure is reasonably available that would make an equivalent contribution to the relevant legitimate objective, taking into account the risks non-fulfilment would create. (AB Report in US-Tuna II, paras. 318-322)

The second and third of these elements are arguably not so problematic for carbon labelling schemes, given that labels are generally viewed as one of the least trade-restrictive instruments available to pursue environmental protection objectives, when compared to other forms of government regulation. [Ref 12]

It may, however, be more difficult to meet the first condition and establish that carbon labels make at least some contribution to climate change mitigation. (AB Report in US-Tuna II, para. 317; AB Report in US-COOL, para. 476) This is because there appears to be little evidence that consumers are actually responding to carbon labels, with their impact on purchasing decisions and thus on GHG emissions reduction still unclear. [Ref 13] It is also noteworthy that carbon labelling does not figure among the key mitigation policy proposals contained in the UNFCCC negotiating text for a post-2020 climate change agreement. If carbon labels are not being perceived as effective or necessary from a climate change perspective, it would logically be more onerous to demonstrate they are nonetheless a necessary obstacle to international trade before WTO adjudicators.

Real or imagined tension?

This article has sought to debunk a number of contentions concerning the implications of the TBT Agreement for carbon labels. First, the fact that carbon labelling schemes are based on non-product-related PPMs does not per se exclude these measures from scrutiny under the TBT Agreement. The threshold question for applying the TBT Agreement to carbon labels is, instead, whether they are developed and implemented by governmental bodies of WTO Members, or otherwise with significant government involvement. Given the presently limited governmental practice in relation to carbon labels, and their apparent low-profile in the ongoing UNFCCC negotiations, it is quite possible that any perceived tensions with TBT disciplines are more imagined than real. A separate question is whether, and to what extent, WTO Members should also be held responsible under WTO law for private sector carbon labels, but this is an entirely different matter from the often unhelpful incorporated or unincorporated PPM dichotomy. [Ref 14]

But even if carbon labelling schemes were to be adopted by more WTO Members in the future, it is a misperception that these schemes are typically inconsistent with TBT non-discrimination disciplines solely because they distinguish between products on the basis of non-product-related PPM criteria. A conclusive legal assessment cannot be made in the abstract, as it will depend on the specific design and application of individual carbon labels, and in particular whether the carbon calculation method employed is even-handed or impartial in light of the climate change mitigation objective. That being said, however, the most critical question remains whether carbon labels are an effective tool for addressing climate change in the first place. However, this is not a concern primarily from a WTO law perspective – the necessity requirement – but from a climate change law and policy standpoint. Put differently, ifthe potential of carbon labels in reducing GHG emissions is uncertain or marginal,why should governments bother introducing them?

In this sense, this author shares the view that there is perhaps too much anxiety over the potential WTO-illegality of yet currently non-existent or non-prominent climate change measures. [Ref 15] A more constructive “trade and climate change” debate should, instead, focus on identifying first which measures are proving most effective from a climate change mitigation viewpoint and then whether WTO law imposes duly or unduly any constraints on such measures. Tackling climate change is simply too urgent to waste time and resources on abstract conflict scenarios with WTO law.

Gracia Marin Duran. Lecturer in International Economic Law, School of Law, University of Edinburgh

[Ref 1] Tamiotti, Ludivine et al., Trade and Climate Change – WTO-UNEP Report. Geneva: World Trade Organisation and United Nations Environment Programme, 2009 (part IV).

[Ref 2] Baddeley, Shane, Cheng, Peter and Wolfe, Robert. “Trade Policy Implications of Carbon Labels on Food.” The Estey Center Journal of International Law and Policy 13.1 (2012): 59-93, pp. 65-67.

[Ref 3] Borg-Michaleff, Cheryl. “Product Carbon Footprinting: Calculation and Communication Standards in the Making.” Carbon and Climate Law Review 4.2 (2010): 178-189, pp. 179-180.

[Ref 4] Appleton, Arthur. “Private Climate Change Standards and Labelling Schemes under the WTO Agreement on Technical Barriers to Trade.” International Trade Regulation and the Mitigation of Climate Change. eds. Thomas Cottier, Olga Nartova and Sadeq Z. Bigdeli. Cambridge: Cambridge UP, 2009, pp. 137-141.

[Ref 5]  Marín Durán, Gracia. “NTBs and the WTO Agreement on Technical Barriers to Trade: the Case of PPM-Based Measures Following US – Tuna II and EC – Seal Products.” European Yearbook of International Economic Law (2015):  87-136, pp. 94-110.

[Ref 6] Kudryavtsev, Arkady. “The TBT Agreement in Context.” Research Handbook on the WTO and Technical Barriers to Trade. eds. Tracey Epps and Michael J. Trebilcock. Cheltenham: Edward Elgar Publishing, 2013, p. 46.

 [Ref 7]  Kudryavtsev, Arkady. Private Sector Standards as Technical Barriers to Trade. Oisterwijk: Wolf Legal Publishers, 2015.

[Ref 8]   Norpoth, Johannes. “Mysteries of the TBT Agreement Resolved? Lessons to Learn for Climate Policies and Developing Country Exporters from Recent TBT Disputes.” Journal of World Trade 47.3 (2013): 575-600, p. 581.

[Ref 9] Appleton, Arthur. “Private Climate Change Standards and Labelling Schemes under the WTO Agreement on Technical Barriers to Trade.” International Trade Regulation and the Mitigation of Climate Change. eds. Thomas Cottier, Olga Nartova and Sadeq Z. Bigdeli. Cambridge: Cambridge UP, 2009, p. 141.

[Ref 10]   Ibid., pp. 134-135.

[Ref 11]  Bolwig, Simon and Gibbon, Peter. Global Forum on Trade and Climate Change – Counting Carbon in the Marketplace. Paris: Organisation for Economic Cooperation and Development, 2009, pp. 4 and 12-14.

[Ref 12]  Ankersmit, Laurens J. and Lawrence, Jessica C. “The Future of Environmental Labelling: US – Tuna II and the Scope of the TBT.” Legal Issues of Economic Integration 39.1 (2012): 127-147, pp. 128-129 and 133-134.

[Ref 13] Bolwig, Simon and Gibbon, Peter. Global Forum on Trade and Climate Change – Counting Carbon in the Marketplace. Paris: Organisation for Economic Cooperation and Development, 2009, pp. 4 and 17.

[Ref 14] Charnovitz, Steve. “The Law of Environmental PPMs in the WTO: Debunking the Myth of Illegality” Yale Journal of International Law 27 (2002): 59-110.

[Ref 15]   Kulovesi, Kati. “Real or Imagined Controversies? A Climate Law Perspective on the Growing Links Between the International Trade and Climate Change Regimes” Trade, Law and Development 6:1 (2014): 55-92.

This article is published under
14 October 2015
How to track fish from sea to fork to ensure positive environmental outcomes? In the latest chapter of what is generally agreed to be the longest running international trade and environment dispute,...
14 October 2015
Do the recent WTO rulings related to product labelling provide enough flexibility for Members to achieve various non-trade policy objectives? WTO Members and consumers have largely ignored the...