Not-so-voluntary labelling in the WTO tuna-dolphin dispute

9 November 2011

On 15 September, a WTO panel handed down a mixed ruling in a dispute brought by Mexico over the US “dolphin-safe” label for tuna products. The three panel members — confronted with the difficult question of whether requirements for a voluntary label can be a de facto mandatory regulation and as such disciplined by WTO law — handed a weak victory to Mexico by deciding that the measure, in fact, amounted to a mandatory technical regulation and was too trade restrictive. The panel members visibly struggled to find common ground on this critical point, however, with one member issuing a dissenting opinion.

While an appeal is expected, the decision is already considered a landmark ruling on the WTO’s Agreement on Technical Barriers to Trade (TBT). Given the increasing use of labels — be it for fair-trade coffee, organic vegetables or a product’s carbon-footprint — the ruling could have far-reaching consequences and joins a number of other TBT cases currently at the WTO.

The panel considered three essential TBT questions in particular: the mandatory vs voluntary nature of standards; what constitutes an international standard; and when a technical regulation can be considered more trade restrictive than necessary. Although Mexico had brought claims under both the TBT Agreement and the WTO’s General Agreement on Tariffs and Trade (GATT), the panel only considered the TBT arguments.  Regarding the TBT non-discrimination claims made by Mexico, the experts sided with the US in ruling that the label did not discriminate on the basis of nationality, and found it unnecessary to look at the parallel claims under GATT. Article XX GATT — virtually synonymous with the GATT’s/WTO’s  Tuna-Dolphin saga so far — was not addressed at all in the present dispute. Interestingly, the US never even raised a defence under GATT Article XX. This was risky, as they could not necessarily expect the panel to dismiss Mexico’s TBT non-discrimination claims and not rule on the GATT as a result.

The dispute, with roots dating back to the pre-WTO era, concerns a voluntary labelling scheme for tuna products caught in a dolphin-safe manner. The US 1990 Dolphin Protection Consumer Information Act (DPCI Act) prevents the use of the label for tuna from the Eastern Tropical Pacific (ETP) caught with so-called purse seine nets by encircling dolphins, even if independent observers can certify that no dolphins were killed or seriously injured during the specific catch. For tuna caught outside the ETP, on the other hand, it suffices for the captain of the vessel to certify that no dolphins were intentionally netted or encircled during the catch. Non-ETP tuna importers are not obliged to prove that no mortalities or serious injuries occurred.

Washington justifies the differential treatment with the unique relationship between tuna and dolphin schools in the ETP, where tuna swarms tend to follow dolphin groups beneath the surface. Fishermen strategically use this natural relationship by intentionally setting on dolphins in order to catch the tuna that swim below. Historically, many dolphins were killed as a result. Modern fishing practices, however, allow dolphins to be released alive when the tuna is captured.

Eligibility for the Department of Commerce-sanctioned label is no legal prerequisite for importation. However, in reality most of the processing and end-use market is closed for non-certified tuna as processors, wholesalers and retailers hedge against unlabelled products. Mexico, whose fleet predominantly uses purse-seine nets, says that this limitation de facto makes the label mandatory – and as such disciplined by the TBT Agreement’s stricter rules on governmental technical regulations, as opposed to the less stringent rules on voluntary standards.

Can something voluntary be de facto mandatory?

The use of “behind-the-border” measures affecting international trade is increasing. The TBT Agreement is meant to ensure that technical regulations and standards are not misused to unfairly discriminate between products and producers or to create unnecessary obstacles to international trade. According to the TBT Agreement, technical regulations and standards differ only in their degree of compulsiveness. A technical regulation is mandatory, while a standard is voluntary. Up until the current case, this distinction was considered to be rather straightforward. The dispute at hand, however, suggests otherwise.

The treatment of otherwise voluntary measures, which do not tolerate the parallel application of any alternative measures — or at least no measures that follow a different approach — is delicate. In the case of the Commerce dolphin-safe label, the US prohibits labels that deploy a “non-injury” rather than a “fishing method” approach; It prohibits labels that require that no dolphin was killed or seriously injured during the catch but permit the use of purse-seine nets. As US consumers largely refuse to purchase products that bear no label, they crowd out products that may have been caught in an arguably “dolphin-safe” manner, but by a fishing method outlawed by the only label-type available. Does this scenario make the regulation in effect mandatory?

The issue is critical, as the Mexican tuna fleet invested heavily to comply with what it understood to be internationally-accepted standards. This occurred after the US and Mexico, with other states, concluded the Agreement on the International Dolphin Conservation Programme (AIDCP). The agreement established a “non injury” label, requiring the US to change its labelling practice to adopt the standard. Mexico’s modern fishing fleet complies with this standard. Today AIDCP is considered to be one of the most efficient and comprehensive programmes for animal conservation worldwide. Comprising a complex system for monitoring and verification, it has reduced observed dolphin mortality in the ETP by around 99 percent.

Nonetheless, consumer protests and a series of Federal Court rulings prevented the US Department of Commerce from changing its label. The objective of the now-retained fishing-method approach — focusing on the practice of setting on dolphins and on purse-seine nets — is to address also non-observed injuries and mortalities that could result from the stress that dolphins experience when being chased, or from dolphin calves being separated from their mothers.

This interplay of public and private actions undeniably severely affected Mexican tuna exports.

While the three panellists agreed that a document was mandatory when it prescribed or imposed something, they failed to find consensus on whether the DPCI Act was of such compulsory nature.

Regulation of information can be mandatory, majority finds

The US dolphin-safe label “regulates information in a binding or compulsory fashion,” the majority opinion of the panel report states. “It is not compulsory to meet these requirements but they do prescribe or impose the conditions under which a product may be labelled dolphin-safe.” One could certainly argue — and many have — that this is the very nature of a labelling scheme. It prescribes certain requirements that need to be met in order for a product to be eligible for the label.

The US dolphin-safe label reaches beyond this, the majority of the panel ruled. The case is about “regulating in a binding fashion the information that may be conveyed.” The label “prescribes certain requirements that must be complied with in order to make any claim relating to the manner in which the tuna was caught in relation to dolphins,” the two panellists found.

This is the crux of the matter. According to the panel, a regulation’s binding character may well concern the regulation of consumer information. If a measure regulates what sort of information may be given to consumers, thereby restricting producers’ ability to market their products fully, it may well qualify as a mandatory regulation. The panel noted: “the US Act prohibits [not only the use of the dolphin-safe label and similar labels] but the use of terms such as "porpoise" or ‘marine mammal’ or any statement relating to dolphins whether misleading or otherwise.” The regulation thus prohibits not only false and misleading information, but it restricts the kind of lawful information that may be made available to consumers. This negative compulsory regulation of information — the exclusion of any alternative labels, including the AIDCP dolphin safe label that Mexican fishermen comply with — makes the otherwise voluntary standard a mandatory regulation, the panel majority decided.

The emerging distinction between a restriction on the misuse of voluntary labelling and a restriction resulting in a quasi compulsory regulation is thin. This may have a bearing in situations where perceptions differ from science, for example when “organic” labels require products to be free of genetically modified organisms (GMOs) because this corresponds to what consumers expect, while scientists argue that the concept of organic production (without fertilizers and pesticides) is unrelated to the presence or absence of GMOs. Yet, if such de facto mandatory regulations are left unregulated, the TBT Agreement would run the risk of losing part of its “bite.” In the Tuna ruling, the panel sought to find a balance between these different concerns — for the moment, however, it remains at most a case-specific construct.

A dissenting opinion

The highly contentious nature of the discussion is also reflected in the dissenting opinion issued by one panel member who disagreed that the measure was mandatory. She/he stressed that “mandatory compliance” as a distinctive quality of technical regulations “relates more fundamentally to the fact that the measure at issue prescribes or imposes compliance with specific requirements to allow a product to be marketed, without allowing discretion to depart from them.” The expert found that the actions that made the label “necessary” (rather than mandatory) were private in nature – namely consumers‘ preferences – and could not be associated with the US government.

Dissenting opinions are no novelty in WTO law. But nowhere before have panellists disagreed on a turning point – that is, an element that impacts all that follows in the panel’s legal analysis. If one other member had agreed with the “dissenting opinion,” the panel would have had to reject all of Mexico’s TBT claims, which relied entirely on the label’s classification as a technical regulation. In an arguably risky move, Mexico had abstained from making possible claims under the TBT provisions relating to voluntary standards.

The depths of science before the panel

Based on the majority opinion, the three panellists engaged in discussion on whether the measure – now identified as a technical regulation – was more trade restrictive than necessary to achieve its legitimate objectives.

Article 2.2 of the TBT Agreement provides that technical regulations shall not be adopted with the effect of creating an unnecessary obstacle to international trade. That is, they should not be more trade restrictive than necessary to achieve a legitimate objective - including the prevention of deceptive practices, or the protection of human health or safety, animal or plant life or health, or the environment. The panel considered the DPCI Act to pursue two of these legitimate objectives: the prevention of deceptive practices by ensuring that consumers are not misled or deceived about whether tuna was caught in a manner that adversely affects dolphins; and protecting animal health and life by discouraging the use of fishing techniques that adversely affect dolphins. According to the panel, these two issues are fundamentally complementary. “The measure seeks to provide a disincentive for certain behaviour, based on consumer choices.”

The panel did not appear overly concerned with the potentially far-reaching consequence of the approach, namely that members in the future may well feel empowered to use their consumers’ buying power more freely to influence the production methods in other members. As one observer pointedly put it: “PPMs (the controversial process and production methods) are here.”

Importantly, the panel found that “a measure that aims at the protection of animal life or health need not be directed exclusively to endangered or depleted species or populations, to be legitimate. Article 2.2 refers to ‘animal life or health’ in general terms, and does not require that such protection be tied to a broader conservation objective.” Though mentioned only in passing, this finding stirs up an ongoing discussion among animal welfare experts over whether animal life or health, or resource conservation, respectively, extend to animal welfare as such. Some experts have even argued that animal welfare is a fundamentally distinct concept not addressed by WTO law, or only as part of the concept of public morals.

The panel rejected Mexico’s argument that the legitimacy of the US’ objectives was impaired by the fact that alternative fishing methods effectively promoted by the US may actually lead to significant negative effects on other marine animals. Claims of policy incoherence are foreseen in some situations by the SPS Agreement, but do not figure explicitly in the TBT Agreement or elsewhere. Citing long-standing Appellate Body jurisprudence - which stresses the autonomy of members to define their policies and to determine the level of protection pursued - the panel found that it was the US’ right to focus on dolphins exclusively and to aim at the prevention of non-observed mortalities and injuries.

Although recognising the US objectives as legitimate, the panel nonetheless sided with Mexico when stating that the degree of regulation was more trade restrictive than necessary.

The panellists reached this conclusion after an 80-page discussion of the current label’s effectiveness and the question of whether serious injuries and mortalities among dolphins occur outside the ETP. The panel’s chosen starting point for this discussion was the notion that a measure may be considered more trade restrictive than necessary when a less trade restrictive alternative exists.

The panel engaged primarily in a discussion on effectiveness, focusing on animal welfare, conservation and consumer protection, rather than on trade impacts. The resulting in-depth analysis of scientific evidence is rather unusual for a WTO panel outside of SPS disputes. In SPS cases, special rules on the role of science provide more clarity and panels routinely rely on the use of outside scientific expertise.

The panel clearly struggled with the task. The experts even acknowledged that they did not possess the level of expertise required to make a final determination on some of the issues. In light of this, it might have been wise for the panel to seek the support of scientific experts, as allowed by the TBT Agreement.

Eventually, the panellists concluded that the current Commerce label was indeed more trade restrictive than necessary in the sense that it limited the information available to consumers – which in turn also limits the label’s effectiveness for discouraging the use of adverse fishing practices. The panel found fault with the label because it did not sufficiently regulate dolphin bycatch outside the ETP, even though dolphins are harmed outside the ETP by fishing techniques other than purse-seine nets. “Where tuna is caught outside the ETP, it would be eligible for the US official label, even if dolphins have in fact been caught or seriously injured during the trip,” the panellists criticised.

The panel accepted the US’ contention that the AIDCP label, on the other hand, does not capture un-observed dolphin injuries and mortalities, thereby misleading consumers as to the level of dolphin protection. However, the panel also found that the Commerce label misleads consumers, as it suggests that no dolphins have actually been killed or seriously injured while, in fact, it can guarantee this only for the ETP. Both labels were thus found to be equally insufficient to achieve the US’ objectives, albeit in different ways. On that basis, the panel agreed with Mexico that it would be less trade restrictive to allow both labels to apply in parallel. It “would contribute to informing consumers about the precise dolphin-safe characteristics of the various techniques to harvest tuna,” the panel concluded its analysis. While the finding is somewhat Solomonic, it seems open to debate. Can two “wrongs” – neither label properly achieves the US’ objectives, the panel found – really make one “right”? The idea that this would be achieved through the cumulative informative effect of the two insufficient labels applying alongside each other certainly provides food for thought.

A possible consequence of the reasoning is rather bittersweet for Mexico. Instead of implementing the panel’s suggestion, the US could comply with the ruling by making its requirements stricter so as to encompass fisheries outside the ETP with similar vigour as those within the ETP.

What qualifies as an international standard?

Mexico further claimed that the AIDCP label should be recognised as a relevant international standard and as such – by means of Article 2.4 of the TBT Agreement – function as a benchmark for the US’ labelling practice, which on this basis should be ruled illegal. The panel eventually dismissed this argument, as it considered the AIDCP standard alone to be unsuitable to achieve the US’ chosen level of protection. It did, however, make some statements that could have far-reaching systemic consequences. In the absence of a definition of “international standard” in the TBT Agreement itself, the panel - as suggested by the Agreement - turned to guidelines by the International Standardisation Organisation (ISO guidelines) that define the term. On that basis, the panel established that an international standard had to be, inter alia, adopted by an international standardising or standards organisation.

The panel held that the AIDCP members collectively acted as a standardising body, but only after finding that the associated institution established through the agreement had a constitution, its own administration and (despite its primarily regional character) an open membership. As such, the panel severely limited the scope of what qualifies as an international standard. Following that argument, each international standard would have to be adopted by a standing body with an administration, even where countries decided to establish the standard otherwise, for example by explicit treaty. Whether this was intended by the drafters of the TBT Agreement when they referred to the ISO guidelines remains uncertain, but it is clear that in times of international roundtables and technical commissions, this narrow understanding of an international standard could be troublesome.

Questionable judicial economy

The dispute is likely to proceed to the appellate stage. Not only did the US loose the case, but the panel’s findings on quasi mandatory regulations could potentially have far-reaching consequences for the future application of the TBT Agreement, something the US or even Mexico might want to see clarified by the Appellate Body.

In that case, the Appellate Body might be confronted with a tricky constellation due to the panel’s exercise of judicial economy – that is, the neglect of legal arguments that it considered unnecessary to solve the dispute. Referring to its TBT findings on the non-discriminatory nature of the label, the panel decided not to examine Mexico’s discrimination claims under GATT (despite Mexico’s somewhat passionate insistence). While the provisions are indeed very similar, they do vary slightly, as TBT Article 2.1 focuses on discrimination “in respect of technical regulations,” while the GATT provisions are much broader. This could become problematic at the appeals stage, as the Appellate Body must rely on findings made by the panel in respect to facts.

A welcoming approach to amici curiae

Another procedural peculiarity was the panel’s treatment of amicus briefs, i.e. non-solicited submissions from non-parties. In perhaps the friendliest treatment accorded so far to an amicus submission, the panel not only generally accepted a submission received from an animal rights group and a university (as most panels and the Appellate Body now routinely do, only to dismiss them as not relevant). It explicitly noted that it had in fact considered information from the brief to be relevant, and had brought salient points to the attention of the parties for comments. Future amici may hope that this will find imitators.

Marie Wilke is ICTSD’s International Trade Law Programme Officer. Hannes Schloemann is a Director of WTI Advisors Ltd. and a partner at MSBH Bernzen Sonntag Rechtsanwälte.

This article is published under
26 November 2010
The world’s fisheries and fisheries trade will be affected by climate change in numerous ways. Foreseeing the changes and taking adequate adaptation and mitigation measures can help developing...
26 November 2010
In the lead-up to the UNFCCC’s Sixteenth Conference of the Parties (COP16), the UN collaborative initiative on Reducing Emissions from Deforestation and Forest Degradation, or “REDD” programme, has...