News and Analysis • Volume 10 • Number 7 • November 2006

A Preliminary Analysis of the WTO Biotech Ruling

The long-awaited report of the WTO panel examining the US-Argentina-Canada challenge of the EU’s application of its biotech regulations was finally released on 29 September. The most immediate question for other WTO Members will be how the more than 2000 page ruling will affect their policy and regulatory decisions on biotechnology.

The Findings in a Nutshell

The complainants challenged the EU on three accounts, namely (i) the EU’s alleged general moratorium on biotech approvals, (ii) its failure to approve a number of specific biotech products (referred to as ‘product-specific measures’), and (iii) national-level bans in several EU member states on the marketing and import of specific biotech products that had already been approved at the EU-wide level.

Broadly speaking, the panel sided with the US, Argentina and Canada on all three counts. It concluded that the general and product-specific moratoria had led to an ‘undue delay’ in the completion of the EU’s approval procedures for biotech products, thus breaching Brussels’ obligations under the WTO Agreement on the Application of Sanitary and Phytosanitary Measures (SPS). The panel also rejected the EU’s defence of the national-level bans as precautionary measures and called on the EU to bring the measures in conformity with WTO rules.

Only the Application of EU Approval Procedures Assessed

The panel explicitly stated that it did not examine whether the EU’s approval procedures for biotech products were in line with its WTO commitments. Instead, the panel focused on European regulatory authorities’ application of those procedures. As a result, the ruling’s most immediate implications are for situations where approval procedures are in place, but might be applied inconsistently with WTO rules. As usual, however, the devil lies in the detail. Understanding some of the systemic implications of the dispute requires a closer look at the specific findings.

Are the Moratoria SPS Measures?

The panel concluded that the alleged general and product-specific moratoria did not constitute SPS measures. Specifically, it argued that while requirements and procedures may constitute SPS measures (such as the EU’s approval procedures), the application of such requirements and procedures would not meet the definition.

As a result of this general finding, the panel rejected virtually all claims by the complainants, including allegations that the moratoria were not based on scientific evidence or a risk assessment since such requirements only apply to SPS measures. Instead, the panel focused its case exclusively on the conclusion that the moratoria had led to ‘undue delay’ in the application of the approval procedures in violation of Article 8 and Annex C(1)(a) of the SPS Agreement.

SPS, TBT or Both?

The panel assessed whether the approval procedures were SPS measures or – as claimed by the EU – whether they also partly fell under the remit of the Agreement on Technical Barriers to Trade (TBT). This distinction is important since it will determine under which set of rules a measure will be assessed.

As a general finding, the panel concluded that a single measure (with multiple objectives) could fall under one agreement to the extent that it is applied to one of that agreement’s purposes, and under another when applied for other purposes. After examining the EU’s approval procedures in great detail, the panel concluded that the approval procedures set out in Directives 90/220 and 2001/18 constituted an SPS measure. The EU’s novel foods Regulation 258/97, on the other hand, was seen as falling under several agreements since it aimed to achieve multiple objectives, only one of which was thought to fall within the scope of the SPS Agreement.1

The practical and systemic implications of the general finding remain somewhat unclear since the panel only referred to the two directives, but not the novel foods regulation, when assessing consistency with the TBT Agreement. Given that it had already found that the procedures in the two Directives constituted an SPS measure, the panel – citing judicial economy – saw no need to examine Canada’s complaints under the TBT Agreement.

The panel took a rather broad approach to interpreting the scope of the SPS Agreement. In one notable conclusion, it rejected the EU’s argument that any risks to the environment other than those related to the life and health of animals or plants (such as adverse effects on soil or water organisms or on biogeochemical processes of an ecosystem) should fall outside the scope of the SPS Agreement. By taking such a broad approach, the panel has considerably raised the bar for any country wishing to justify its approval procedures (at least partly) as TBT measures.

When is a Delay ‘Undue’?

According to the panel, the obligation to ‘undertake and complete’ approval procedures without undue delay meant that “once an application ha[d] been received, approval procedures must be started and then carried out from beginning to end.” The panel referred to the dictionary meaning of ‘undue’ to say that approval procedures should be undertaken and completed with no ‘unjustifiable loss of time’. However, rather than attempting to define reasons which would render a given delay ‘undue’, the panel stressed that any determination of whether a particular approval procedure had been delayed unduly would need to be done on a case-by-case basis ‘taking into account relevant facts and circumstances’, thus making it difficult to draw more general, systemic conclusions for possible future cases.

Risk and Precaution The panel found that the national-level bans were not based on an adequate risk assessment in violation of Article 5.1 of the SPS Agreement. It also ruled that the bans could not be justified as precautionary measures under Article 5.7 since sufficient evidence had been available to carry out a risk assessment at the EU-level. The panel also pointed to the fact that the EU’s regulatory bodies did not revise the conclusions of their original EU-level assessment of the safety of approved GMO products in light of the ‘new scientific information’ provided by those member states that had imposed national-level restrictions.

Importantly, the panel reasoned that Members’ right to adopt provisional SPS measures in situations where relevant scientific evidence was insufficient (under Article 5.7) was a ‘qualified right rather than an exception’. In the panel’s view, this implied that when a complainant alleges that a country’s SPS measure is not based on a risk assessment (in violation of Article 5.1 of the SPS Agreement) and the defendant has invoked Article 5.7, the burden is on the complainant to establish inconsistency with both provisions.

Rather than requesting the national bans to be lifted, the panel called for the measures to be brought in line with the SPS Agreement. It thereby left the door open for EU member states to maintain national level bans if they are able to provide an assessment showing a level of risk that warrants stricter measures than those adopted at the EU-level.

The Question of Likeness

While the US had argued its case entirely under the SPS Agreement, Canada and Argentina had also invoked the TBT Agreement and the General Agreement on Tariffs and Trade (GATT) in their challenge of the product-specific measures. They alleged, inter alia, that the measures had led to discrimination against imported biotech products (‘less favourable treatment’) vis-àvis domestic non-biotech counterparts (based on the assumption that the process of genetic modification does not render a product different as such). The panel effectively skirted this question, explicitly stating that it did not think it necessary to address it.

Some of the panel’s specific arguments are nevertheless interesting. In response to Argentina’s claim of a GATT violation, the panel felt that Argentina had failed to show that the alleged ‘less favourable treatment’ of imported biotech products had been a result of the foreign origin of the products, which would have violated the national treatment principle. Indeed, the panel noted that Argentina did not allege that imported biotech products had been treated differently from domestic biotech products.

The question of ‘likeness’ could become important if biotech approval procedures themselves were to be challenged, given that some argue that genetically engineered products are not different from other products ‘as such’ and should therefore not be subject to different regulations. By emphasising the need to show that the foreign origin of a product was the motivation for the alleged discrimination – rather than, for instance, the “perceived difference between biotech products and non-biotech products in terms of their safety” (para. 7.2514) – the panel appears to have raised the hurdle for complaining parties in a future dispute to show that different rules for biotech and non-biotech products amount to discrimination between ‘like’ products.

Relationship with Other Bodies of International Law

At the request of the EU, the panel assessed the applicability and relevance of the Cartagena Protocol on Biosafety and the Convention on Biological Diversity (CBD). In its analysis, the panel examined the Vienna Convention’s provision that when interpreting any treaty, other relevant rules of international law “applicable in the relations between the parties” should be taken into account (Article 31c). Given that not all parties in the WTO dispute are also parties to the Cartagena Protocol and the CBD, the panel felt that it did not have an obligation to take into account their provisions. Nevertheless, the panel noted that it had the option of doing so, but concluded that the provisions cited by the EU in its defence were not relevant in this case.

Since he measures at issue were not taken to implement the Cartagena Protocol as such, the ruling does not provide guidance on the question whether measures to implement provisions of a multilateral environmental agreement might be presumed to be WTO compatible.

Concluding Remarks

It is important to stress that the panel’s findings and their implications are very specific to the EU context. Nevertheless, a number of arguments and implications are noteworthy. The panel has taken a rather broad approach to the scope of the SPS Agreement, making it virtually impossible to avoid scrutiny of challenged biotech regulations under the SPS Agreement. Furthermore, the ruling has provided little guidance on when a delay should be seen as ‘undue’, leaving this question to be decided on a case-by-case basis.

At the same time, the ruling might have facilitated the defence of an SPS measure as precautionary (under Article 5.7) by placing the burden on complainants, who must prove not only that a measure is inconsistent with the SPS Agreement’s risk assessment requirements, but also that it does not qualify as a justified precautionary measure. In addition, its conclusions regarding the discrimination between ‘like’ products appear to have made it more difficult to challenge measures specifically targeted at biotech products (such as approval procedures) as discriminatory vis-à-vis measures for non-biotech products.

Heike Baumüller is Programme Manager – Environment and Natural Resources at ICTSD. Yvonne Apea, formerly Programme Co-ordinator – Dispute Settlement at ICTSD, is now Conference and Project Manager at the Economist Intelligence Unit. The views expressed in this article are those of the authors and not those of the institutions.