AN ANALYSIS OF PRECAUTIONARY MEASURES IN THE WTO: THE CASE OF SCIENTIFIC (IN)SUFFICIENCY

AN ANALYSIS OF PRECAUTIONARY MEASURES IN THE WTO: THE CASE OF SCIENTIFIC (IN)SUFFICIENCY Author Eric Gilbers ANR 698427 Supervisor P. Delimatsis Sufficiency of scientific evidence is required for imposing an SPS measure (art 2.2 SPS). However, in the situation where there is no such sufficiency, states resort to taking precautionary measures. There are several complications in assessing this requirement. This thesis attempts to provide a clearer insight into the intricacies of the notion of scientific sufficiency in particularly with regards to precautionary measures. It will illustrate that this notion is opaque and both the WTO and EU judicial bodies seem to alternate between intrusive and less intrusive standards of Review. The appropriate solution would be an independent body such as an expert panel or science court.

AN ANALYSIS OF PRECAUTIONARY MEASURES IN THE WTO: THE CASE OF SCIENTIFIC (IN)SUFFICIENCY ERIC GILBERS 1

SUMMARY Sufficiency of scientific evidence is required for imposing an SPS measure (art 2.2 SPS). However, in the situation where there is no such sufficiency, states resort to taking precautionary measures. There are several complications in assessing this requirement. This current state of opaqueness requires an appropriate resolve because it affects the discretion of the state and respectively the WTO in assessing said measures. First, this thesis will expand on the notion of precaution. Thereafter, it will provide an insight in the ramifications of SPS measures themselves. Then, it will deal with the notion sufficiency and how the WTO legal system deals with it. In particular, the notion of standard of review will be evaluated. Overall, it will illustrate that this notion is opaque and both the WTO and EU judicial bodies seem to alternate between intrusive and less intrusive standards of review. The appropriate solution would be an independent body such as an expert panel or science court. 2

ACRONYMS AB: Appellate Body AG: Advocate General ECJ: European Court of Justice EFSA: European Food and Safety Authority EU: European Union GC: General Court of the European Union SPS: Sanitary and Phytosanitary Measures US: United States WTO: World Trade Organization 3

OUTLINE 1. INTRODUCTION... p5 2. SCIENCE AND PRECAUTION...p7 2.1 Precaution.......p7 2.2 Critique on precautionary measures..p8 2.3 Tension between law and science... p9 3. PRECAUTIONARY MEASURES IN THE WTO...... p12 3.1 Relevant Legislation....p12 3.2 Risk assessment 5.1 SPS...... p11 3.3 Sufficiency under 2.2/5.1......p17 3.4 Sufficiency under 5.7. Precaution...... p18 3.5 Scope of article 2.2/5.1 SPS versus 5.7 SPS... p19 4. SUFFICIENCY CASE LAW EXPLORED 4.1 Insufficiency of scientific evidence v uncertainty...p25 4.2. Changing evidence...p27 4.3 The case of differing opinions, Dueling Science and Minority Opinions explained......p29 5. STANDARD OF REVIEW AND INSUFFICIENCY.....p33 5.1 Standard of Review in WTO Law...p33 5.2 Procedural versus Substantive review...p37 6. STANDARD OF REVIEW IN EU LAW....p42 6.1 Standard of review of National Measures...p44 6.2 Standard of review of EU measures...p47 7. CONCLUSION...p49 4

Chapter I INTRODUCTION Suppose there is an outbreak of a new virus affecting food safety. A recent example of such a threat to food safety was the E.coli outbreak in France and Germany 1. In this case the EU placed a temporary ban on the import of all seeds and beans from Egypt. The European Food Safety Authority (EFSA) stated that seeds from Egypt could have been the source for food poisoning outbreaks (at least 48 people died in Germany). As a result of the EFSA findings, the EU banned the imports of seeds and beans from Egypt until 31 October 2011. Not only seeds and beans were sacrificed: "Given the possible severe health impact of exposure [...] it seems appropriate to consider all lots of fenugreek from the identified exporter as suspect, "the EFSA said in a report on its investigations. The EFSA added that the number of EU countries that had received parts of the suspected lots is much larger than previously known and added: "It cannot be excluded that other member states and third countries were supplied." 2 Naturally, many interests are at stake in such a situation. States, who are often inclined to act in a manner that is protective of their domestic market, can defend the health of their citizens by imposing a ban on this specific food. Regardless of whether the evidence is conclusive and even though the risk is not certain, such a ban can be put in place. In this particular example, this will have the consequence that farmers and vendors cannot sell their products to these countries and make enormous losses. Situations like these are not rare. The World Trade Organization (WTO), whose main aim it is to prevent barriers on trade, can play a pivotal role in resolving these cases. 1 The Telegraph. (2011, 6 5). Egyptian fenugreek seeds linked to E.coli banned by EU. 2 Ibid. 5

Issues such as those mentioned above are dealt with under the Sanitary and Phytosanitary (SPS) Agreement. Under particular circumstances specified in the SPS agreement, States can make the exception and impose moratoria, much like the example above. Of paramount importance is article 2.2 which provides that there should be sufficient scientific evidence for this to be allowed. For many scholars, Members, and the WTO, there is much confusion as to what this sufficiency entails. Not only is it difficult to establish what this means under article 2.2, but even more so under the exception for sufficient evidence 5.7 i.e. the precautionary measure. When this sufficiency under 2.2 is not found, article 5.7 still offers a safe haven. Based on the precautionary principle, article 5.7 provides that WTO Members can impose provisional measures, provided that they meet the requirements set out therein. In short, within the realm of anti-protective regulation there is tension between science and law. This tension in itself is age-old and is observed in many legal disciplines, however still it seems not be resolved (chapter 2 will expand on this). This paper attempts to evaluate the concept of precautionary sufficiency within WTO jurisprudence and points out the lacunae within WTO legislation. The main question is whether this case law developed towards the better, and how its development relates. Main sources employed entail case-law and scholarly opinions. Chapter 1 discusses, briefly, the tension of science and law and the development of the notion of precaution. Chapter 2 examines the WTO law and jurisprudence vis-à-vis food safety i.e. SPS measures. Chapter 3 will deal with the concept of sufficiency. Finally, Chapter 4 provides a synthesis. 6

CHAPTER II SCIENCE AND PRECAUTION 2.1 PRECAUTION The notion of precaution is age-old, evidence traces it millennia back 3. Inter alia, Kaiser cites the Rio Declaration of 1992 while saying that [p]robably the most influential statement of the Precautionary Principle we find in principle 15 of [the Declaration] 45 : if there are threats of serious or irreversible damage, lack of full scientific certainty shall not be used as a reason for postponing cost-effective measures to prevent environmental degradation 6. Also the Wingspread Statement states: When an activity raises threats of harm to human health or the environment, precautionary measures should be taken even if some causeand-effect relationships are not fully established scientifically. 7. So mankind knows of this principle for centuries, if not millennia. Not nearly as long has it been such a popular concept amongst scholars as it is today. Literature about a diverse range of definitions is ample. It is illustrative that e.g. Graham 8 argues that there is no such thing as the precautionary principle since there are multiple interpretations and uses subject to debate. These interpretations are not within the scope of this paper but deserve mentioning nonetheless. Most notable example is the difference between the preventive principle and the precautionary principle. Both deal with risks and are ex ante, however the essence of the precautionary principle lies in the fact that it 3 Sadeleer. (2004). The Precautionary Principle and the Concept of Precaution. p2 4 UNEP. (1972, 16 6). United Nationds Development Programme. Retrieved 11 1, 2011, from Rio Declaration on Environment and Development: 5 Kaiser, M. (2004,) p112 6 David Kriebel, J. T. (2001, 3 29). 7 Wingspread agreement The Wingspread Consensus Statement on the Precautionary Principle Wingspread Conference on the Precautionary Principle. 8 John GrahamThe Perils of the Precautionary Principle: Lessons from the American and European Experience. p10 7

intends to assist the regulators in managing those situations where (identified or unidentified) uncertainties prevent any meaningful assessment of risk. This distinguishes it from prevention. On this, Trouwborst 9 differs with Sadeleer that the distinction between the preventive principle and the precautionary principle rests on a difference of degree in the understanding of risk. Prevention is based on certainties. Preventive measures are [...] intended to avert risks for which the causeand-effect relationship is already known 10. Conversely, precaution [..] comes into play when the probability of a suspected risk cannot be irrefutably demonstrated 11. Even though a risk is per definition not certain, it implies that there is a basis on knowledge of something that can be measured, but precaution deals with risks whose existence cannot be established because of scientific uncertainty. 12 The answer to the division is that dealing with a known risk would mean dealing with prevention and not precaution. Discussions about divisions such as these show the extent of ambiguity which surrounds the notion of precaution. For clarity s sake, this paper narrows down the definition of precaution. Based on inter alia the abovementioned literature, in principle the notion of precaution deals with a measure that prevents a potential harm or risk when there is no certain knowledge. The notion of precaution can be divided into three core components: 1) scientific uncertainty, 2) threat of harm and 3) precautionary action itself. In short, the principle provides guidance on how to react to a uncertain risk of harm. 9 Trouwborst, A. (2002). Prevention, Precaution, Logic and Law 10 Sadeleer, D. (2002). Environmental Principles: From Political Slogans to Legal Rules. p 74-75 11 Ibid. p118 12 This begs questions such as; when is a probability no longer an uncertainty? Is it not part of the dictionary definition of risk that it implies a gamble, a guess, and uncertainty? The difference of uncertainty and sufficiency will be addressed later. 8

Gruszczynski correctly points out that the differences of risk and potential risk are beyond semantic 13. This paper emphasizes the component that triggers a precautionary measure, the first component of precaution: scientific uncertainty or lack of science. In the words of this paper a precautionary measure is defined as follows: A measure aimed against a potential harm or risk which has to be taken, even if there is no sufficient evidence available for such harm or risk. 1 Sadeleer, D. (2002). Environmental Principles: From Political Slogans to Legal Rules.. Oxford: Oxford University Press. 1 Ibid. 1 This begs questions such as; when is a probability no longer an uncertainty? Is it not part of the dictionary definition of risk that it implies a gamble, a guess, and uncertainty? The difference of uncertainty and sufficiency will be addressed later. 1 Gruszczynski, L. (2010). Regulating Health and Environmental Risks under WTO Law, a Critical Analysis of the SPS Agreement. Oxford: Oxford Scholarschip Online p 1 Social Decision-Making under Scientific Controversy, Expertise, and the Precautionary Principle Olivier GODARD* p9 <http://halshs.archives-ouvertes.fr/docs/00/62/40/27/pdf/preccon3-97.pdf> 2.2 CRITIQUE ON PRECAUTIONARY MEASURES A precautionary note on precaution is necessary. Science can always be manipulated because it is often based on normative benchmarks. Olivier Godard argues that risk regulators might feel inclined to resort to precautionary measures because of a heightened sense of urgency or pressure within the society. This could open up the Pandora s box of temptations of manipulating science for the benefit of political and economic interests 14. In the current globalizing world, markets are becoming increasingly liberalized. Through the removal of market obstacles, states can more easily access other markets. Conventional economic theorist such as, inter 13 Gruszczynski, L. (2010). Regulating Health and Environmental Risks under WTO Law, a Critical Analysis of the SPS Agreement 14 Godard (1997) Social Decision-Making under Scientific Controversy, Expertise, and the Precautionary Principle 9

alia, the contemporary school s Krugman 15 argue that it is through this liberalization that a more optimal situation for free trade can be achieved. Free trade allows for specialization and in turn, the global level of output increases. Next to the economies of scale it is because of specialization that countries and regions can allocate their resources in such a way that the country or region has a relative comparative advantage. In the end this will result in higher quality of products and services. However, along with liberalization of markets, national regulatory activism arose. Many states try to protect their domestic products by imposing barriers. According to Lukasz Gruszczynski, in his book Regulating Health and Environmental Risks under WTO law 16, this regulatory activism can be observed in particular in the case of risk regulation where governments, responding to the fears and demands of their domestic constituencies, adopted a wide range of regulatory measures aimed at the protection of the environment and human health and safety 17. Thus, confirming Godard that indeed- states are often influenced by ulterior motives to impose barriers. Precaution, both its definition and the motives for its application, is not a straightforward concept. The extent of academic and professional writing on this principle exemplifies this. It is rather peculiar that organizations such as the WTO and the EU employ such a vague concept. Since these bodies aim, inter alia, at optimal market access, allowing such an opaque principle as a tool for Member States does not make much sense. One of the raisons-d être for dispute settlement bodies is this dealing with this very aspect. However, acknowledging that regulatory 15 Krugman (1991) The move towards free market zones p13 16 Gruszczynski, L. (2010). p2 17 Gruszczynski. (2008 ). The SPS Agreement within the Framework of WTO Law. The Rough Guide to the Agreement s Applicability. 158 10

issues cannot simply rely on black letter law of treaties, it is for case law to establish a more detailed interpretation. A too diffuse manner of interpretation within case law clashes with one of the very basic legal concepts of legal certainty, this concept requires from law and case law to be consistent, something this paper will examine later 18. Overall, it seems that the most founded critique regarding WTO jurisprudence pertaining to precaution revolves around the possibility of covert protectionism. 2.3 TENSION BETWEEN LAW AND SCIENCE Tension between law and science is present in many different areas. Within WTO law this is manifested prominently in sufficiency. Notably, V.R. Walker 19 argues that science cannot be a neutral arbiter for triggering precautions. He goes past the myth of value-neutral science as a criticism. The notion that a decision cannot be purely scientific implies that when lawmakers define the factual predicate for taking precautions, non-scientific decisions are necessary. As a consequence, defining the factual base for taking precautions requires nonscientific decisions. Furthermore, when fact finders identify risks, or make a finding that the legal trigger for taking precautions has been satisfied, nonscientific decisions are also necessarily involved. According to Walker, lawmakers often utilize vague terms for the legal trigger and mask nonscientific decisions in language that sounds scientific. Also Walker finds that finding whether scientific uncertainties exist or not cannot be purely scientific. Thus, as a red line through this paper, the essence remains that scientific sufficiency should not be regarded as taking science as an absolute concept. 18 See conclusion 19 Walker, The Myth of Science as a Neutral Arbiter for Triggering Precautions, 26 B.C. Int'l & Comp. L. Rev. 197 (2003) p22 11

CHAPTER III PRECAUTIONARY MEASURES IN THE WTO 3.1 RELEVANT LEGISLATION Before going into the core issue, it first has to be established what exactly an SPS measure is. Annex 1 20 provides that SPS measures are measures designed to protect human, animal or plant life or health 21 from various risks. In effect, it balances the sovereign right of Member States to provide the appropriate level of health protection against the curtailing effects of protectionist measures resulting into barriers to trade. The Agreement while permitting governments to maintain appropriate Sanitary and Phytosanitary protection, reduces possible arbitrariness of decisions and encourages consistent decision-making 22. Member states can only apply a SPS measure to ensure the food safety and plant health. As this paper will demonstrate, this has to be based on a proper assessment of risks and should be based on sufficient scientific evidence. Another aim of the SPS agreement is to further harmonize SPS measures based on international standards, guidelines developed by relevant organizations. This implies that measures conforming to relevant international standards are presumed to be in compliance with the SPS Agreement s requirements. If members want to introduce or uphold measures more severe than provided for by relevant international 20 Annex 1 of the SPS Agreement 21 Annex A of the SPS Agreement 22 Food safety and quality: trade considerations OECD Publishing p 53 12

standards, they must ensure that they are supported by a scientific justification 23. Members are at liberty to pick their own standards but they have certain limits. Article 3.3 allows Members to adopt more stringent SPS measures if there is a scientific justification, or as a consequence of the level of sanitary or phytosanitary protection a Member determines to be appropriate in accordance with the relevant provisions of paragraphs 1 through 8 of Article 5. 24 Even though, there has been similar debate on the use of risk assessment, article 3.3 is not within the scope of this thesis because this thesis focuses on article 2.2 in relation to the precautionary measures. Of particular importance, and central to this thesis, is article 2.2 of the SPS Agreement mandates WTO Members to base SPS measures on scientific principles and prohibits their maintenance without sufficient scientific evidence. Article 2.2 reads that WTO members shall ensure that any SPS measure is based on scientific principles and is not maintained without sufficient scientific evidence, except as provided for in paragraph 7 of Article 5 25. Following EC-Hormones article 2.2 has to be read together with article 5.1. In the Hormones case, the Appellate Body underlined the importance of science and risk assessment as a safety pin on the use of the SPS by stating that; The requirements of a risk assessment under Article 5.1, as well as of sufficient scientific evidence under Article 2.2, are essential for the maintenance of the delicate and carefully negotiated balance in the SPS Agreement between the shared, but sometimes competing, 23 Article 3.3 24 Ibid. 25 A. Text of Article 2 Article 2: Basic Rights and Obligations 13

interests of promoting international trade and of protecting the life and health of human beings 26. Article 5 provides for the risk assessment. Article 5.1 provides that Member States must ensure that their measures are; based on an assessment, as appropriate to the circumstances, of the risks to human, animal or plant life or health, taking into account the risk assessment techniques developed by the relevant international organizations. 27 Article 5.2 requires the risk assessment to take into account various factors, including available scientific evidence. Furthermore, article 5.3 allows, although subject to several conditions, the right for Member States to set their own Appropriate Level of Protection or ALOP. These conditions are inter alia, following article 2.3, that; these measures are not applied in a manner which would constitute a disguised restriction on international trade 28, article 5.4, to take into account the objective of minimizing negative trade effects and article 5.6 to be not more traderestrictive than required to achieve their appropriate level of sanitary or phytosanitary protection, taking into account technical and economic feasibility 29. 3.2 RISK ASSESSMENT 5.1 SPS Hitherto, all cases before the WTO dispute settlement system which were under the SPS agreement have been found to be in violation of article 5.1. Article 5.1 indicates that SPS measures have to be based on a risk assessment; 26 Appellate Body Report: European Communities EC Measures Concerning Meat and Meat Products, Para. 177 27 5.1 SPS 28 Article 2.3 29 Article 5.6 SPS 14

Members shall ensure that their sanitary or phytosanitary measures are based on an assessment, as appropriate to the circumstances, of the risks to human, animal or plant life or health, taking into account risk assessment techniques developed by the relevant international organizations. 30 A risk assessment is defined in Annex A.4 of the SPS Agreement as follows: The evaluation of the likelihood of entry, establishment or spread of a pest or disease within the territory of an importing Member according to the sanitary or phytosanitary measures which might be applied, and of the associated potential biological and economic consequences; or the evaluation of the potential for adverse effects on human or animal health arising from the presence of additives, contaminants, toxins or disease-causing organisms in food, beverages or feedstuffs. 31 Article 5.1 provides for two different types of risk assessment. Both with different requirements, the type of risk assessment that has to be employed depend on the objective of the SPS measure. The SPS agreement refers to two types of risk assessment: pest/disease risks or quarantine risk assessment and food borne risk assessment. The first type of risk assessment, which must be conducted where the risk at issue is from pests or diseases, contains three requirements. The other type of risk assessment: the assessment of food-borne risk is conceptualized as a two-step analysis 32 : 30 Article 5.1 SPS 31 SPS Agreement Annex A, para. 4 32 Ibid. 15

Table I: risk assessment 33 Step Pests or diseases Food-borne risk 1 The risk assessor needs to identify two sets of data: First step: two sets of data need to be identified: the diseases [or pests] whose entry, establishment or spread Members wants to prevent within its territory as well as potential biological and economic consequences associated with the entry, establishment or spread of these disease [or pests] 2 Second step: requires an assessment of likelihood with respect to each set of data. 3 Third step: during this phase risk assessors are expected to evaluate the likelihood of entry, establishment or spread of an identified disease or pest according to the SPS measures that might be applied. First, the identification of adverse effects to human or animal health and life arising from the presence of certain substances such as additives and toxins in food, feedstuffs, and beverages. This step can be further divided into 2 distinctive phases. 1. Identification of additives, contaminant, toxins, or diseasecausing organism in particular food, beverages, or feedstuffs, and 2.Identification of any possible adverse effects on human or animal health. If such adverse effects are found, the second step of the analysis requires the evaluation of the potential of occurrence of these effects 33 Table based on Gruszczynski (2006) The Role of Science in Risk Regulation under the SPS Agreement p14 16

This paper will not go into the details of these procedures, yet it is important to keep in mind that the WTO makes a distinction between procedural approaches in the objective of SPS measures. Why is there such risk assessment? Jacqueline Peel claims that the very fact of having to take a risk assessment into account and respond to its findings could have the salutary effect of forcing national regulators to articulate objectives, to assess means, and to rationalize results, a substantial improvement for the regulatory processes of many nations. 34. In essence, risk assessment is established to prevent Members to abuse this regulatory measure. However, Paul Slovic s take on this provides caution: Risk assessment is inherently subjective and represents a blending of science and judgment with important psychological, social, cultural, and political factors. In addition, [ ] social and democratic institutions, remarkable as they are in many respects, breed distrust in the risk arena. 35 In short, Slovic claims that whoever controls the definition of risk controls the rational solution to the problem at hand. So, even though, risk assessment is used to maintain a certain rational approach towards regulatory measures, it should be borne in mind that the establishment of the risk assessment itself is inherently subjective since it is -to an extent- a construct. 3.3 SUFFICIENCY UNDER 2.2/5.1 In Japan-Agricultural Products II, the Appellate Body came up with a test to establish sufficiency. Japan had imposed a prohibition on the imports of practical agricultural products These practical agricultural products 34 Peel, J. (2004). Jean Monnet Working Paper 02/04 Risk Regulation Under the WTO SPS Agreement: Science as an International Normative Yardstick? Retrieved 8 8, 2011, from Jean Monnet Working Paper. p51 35 Slovic. (1991). Trust, Emotion, Sex, Politics, and Science: Surveying the Risk-Assessment Battlefield. p63 17

originating inter alia from the US, were apples, cherries, peaches (including nectarines), walnuts, apricots, pears, plums and quince. The argument put forward by Japan was that these products could be potential hosts of a pest i.e. codling moth 36. In response the US requested consultations with Japan. The United States argued that Japan violated, inter alia, Articles 2 and 5 of the SPS agreement. According to the US, Japan prohibited the importation of each variety of a product because the Japanese government required quarantine treatment until it was tested, despite finding that the treatment was effective for other varieties of a similar product. The Panel found that Japan acted inconsistently with Articles 2.2 and 5.6 of the SPS Agreement, and Annex B and, consequently, Article 7 of the SPS Agreement. Upon appeal by Japan, the Appellate Body upheld the finding that Japan s varietal testing was inconsistent with the SPS agreement. According to the Appellate Body, the ordinary meaning of: Sufficient is of a quantity, extent, or scope adequate to a certain purpose or object. From this, we can conclude that sufficiency is a relational concept. Sufficiency requires the existence of a sufficient or adequate relationship between two elements, in casu, between the SPS measure and the scientific evidence. [ ] The context of the word sufficient or, more generally, the phrase maintained without sufficient scientific evidence in Article 2.2, includes Article 5.1 as well as Articles 3.3 and 5.7 of the SPS Agreement. 37 36 WT/DS76/R panel report Japan Agricultural products 2..A. 1 37 Appellate Body Report on: Japan Agricultural Products II para. 73 18

So, the Appellate Body, this time, acknowledged that the Panel s finding regarding article 5.1 is relevant in the interpretation of article 2.2. This implies that there exists an assumed rational relationship between the measure and the risk assessment. So, similar to EC-Hormones 38 determined that article 2.2 and 5.1 should always be read together. it is 3.4 SUFFICIENCY UNDER 5.7. PRECAUTION As mentioned, Japan argued, in the case Japan Agricultural Products II 39, that the meaning of sufficiency in article 5.7 must be the same is in Article 2.2. This would mean that when there is insufficient scientific evidence, Article 5.7 applies. That is: in cases where there is no rational relationship between the available evidence and the SPS measure. 3.5 SCOPE OF ARTICLE 2.2/5.1 SPS VERSUS 5.7 SPS It is important to examine the scope of article 2.2/5.1 and 5.7, because it concerns most notably the notion of sufficiency, which is also the central issue of this thesis, and an important factor of the previously established definition of precaution. Thus far, it seems clear-cut: If the particular evidence is found to be sufficient, article 2.2 applies together with the risk assessment provided for within article 5.1. SPS measures can be invoked if they are based on scientific principles and sufficient scientific evidence and take account of the factors provided by article 5. Additionally, article 5.7 reads the exception to article 2.2, which is as follows; In cases where relevant scientific evidence is insufficient, a Member may provisionally adopt sanitary or phytosanitary measures on the basis of available pertinent information, including that from the relevant international organizations as well as from 38 Hormones para. 180, also Japan - Agricultural products II para. 82 39 Japan Agricultural Products II para. 12 19

sanitary or phytosanitary measures applied by other Members. In such circumstances, Members shall seek to obtain the additional information necessary for a more objective assessment of risk and review the sanitary or phytosanitary measure accordingly within a reasonable period of time. 40 From this follows that when the evidence is found to be insufficient article 5.7 applies. Article 5.7 functions as a safe haven form the requirement in article 2.2. Thus, it is a precautionary measure. However, the relation between 2.2/5.1 and 5.7 seems to be more intricate than 5.7 merely being a safe haven in the case of insufficiency of scientific evidence i.e. when the SPS measure falls outside of 2.2. In EC Approval and Marketing of Biotech Products, the Panel addressed the relation between 2.2/5.1 and 5.7. This case concerned a complaint the US launched regarding a moratorium applied by the EC on the approval of biotech products. This led to a restriction on imports of US agricultural and food products. Again the US argued that the measure was not consistent with, inter alia articles 2 and 5 of the SPS agreement. The Panel 41 found that article 5.7 is applicable in every case where relevant scientific evidence is found to be insufficient. It looked at the structure of the text and found that art. 5.7 provide the conditionality rule. So, if insufficiency of scientific evidence is found, then a provisional measure might be taken. This was contrary to the approach of the EC in this case, which argued that the distinguishing feature was the provisionallity of the measure 42 itself. This provisionallity refers to whether an SPS measure has provisional or a permanent character. This 40 Article 5.7 SPS 41 Panel Report: EC Biotech Products, para. 7.2939 42 Ibid., para 7.2939 ; in this context the panel also noted that if the provisional adoption of an SPS measure had been intended as a condition for the applicability of Article 5.7, the first sentence of Article 5.7 would, in our view, have opened with a different phrase, such as In cases where a Member provisionally adopts an SPS measure ( Ibid., para 7.2940 ). 20

should be not relevant for the applicability of article 5.7. Otherwise, establishing provisionality of measure would on itself be enough to keep a SPS measure out of 2.2/5.1. As a consequence it would lower the threshold for the requirement of insufficiency to enter the safe haven that 5.7 provides. The Panel, furthermore, distinguished between applicability and consistency. Applicability deals with the triggering of article 5.7 and consistency deals with the respective requirements of article 2.2/5.1 or 5.7 43. So, a measure can be found to be applicable but it can still be inconsistent with particular requirements, thus violating it. The Appellate Body, in Japan Agricultural Products II, set out four cumulative requirements for a measure to be provisionally adopted following article 5.7. 44 A measure should be 1) imposed where relevant scientific information is insufficient, and 2) adopted on the basis of available pertinent information. A Member can only maintain this measure if the Member 3)) seek[s] to obtain the additional information necessary for a more objective assessment of risk and 4) review[s] the [..] measure accordingly within a reasonable period of time. This case will be addressed hereafter. The Panel s view in the Biotech case -initiallysuggests that if a consistency requirement is not met, there is a violation of article 5.7 but the measure is not inapplicable. This applicable versus consistency division seems to have no avail. The Panel noted that a Member State can benefit from this aforementioned right if it meets the requirements of Article 5.7. It stated: if a challenged SPS measure was adopted and is maintained consistently with the four cumulative requirements of Article 5.7, the situation is as provided for in paragraph 7 of Article 5 (Article 2.2) and the obligation in 43 Ibid., para. 7.2942. 44 Appellate Body report: Japan Agricultural Products II, para. 89 21

Article 2.2 not to maintain SPS measures without sufficient scientific evidence is not applicable to the challenged measure 45. So, despite the previous distinction, whenever a measure is found to be violating requirements from 5.7 because the circumstances are not as provided for in Article 5.7, article 5.7 becomes inapplicable and the SPS measure falls within the scope of article 2.2. Only after all four requirements are met, a SPS measure does not fall within the scope of article 2.2. Japan, in Japan Agricultural Products II, tried to find a way out of this peculiar situation. The Japanese Member State argued that the SPS measure in question was in conformity with the first requirement of article 5.7, and therefore fell outside the scope of article 2.2. The Appellate Body struck this submission down. [E]xcept as provided for in article 5.7 46 did not refer to merely the first sentence of said article, but instead to article 5.7 as whole 47. These four requirements are clearly cumulative in nature and are equally important for the purpose of determining consistency with this provision. Whenever one of these four requirements is not met, the measure at issue is inconsistent with Article 5.7. 48, thereafter the Panel established that the requirements of the second sentence of art. 5.7 were not met and therefore the Panel did not need to look at the requirements encapsulated in the first sentence 49. The consequence of this situation is that the Panel is send back to article 2.2/5.1, despite having established insufficiency before. For Gruszczynski, who critiques this approach, the more appropriate approach would be to regard the two different disciplines, 2.2/5.1 versus 5.7 as distinctive paths of compliance. He suggests an exclusive reading. When an SPS measure is found to be inconstant with a relevant provision, 45 Ibid., para. 7.2974 46 Panel Report: EC Approval and Marketing of Biotech Products, para. 7.2975. 47 Appellate Body Report : Japan Agricultural Products II, para. 90. 48 Japan Agricultural Products II, para. 89 49 Ibid. para. 91 22

it will not be send to the second regime. Thus, when scientific evidence is insufficient, article 2.2 and 5.1 are not violated because several of the requirements of article 5.7 are no met. Such an approach seems to be more logical compared to the current standard of the case law 50. He cites the US-Continued Suspension case Article 2.2 excludes from its scope of application situations in which the relevant scientific evidence is insufficient. In such situation, the applicable provision is Article 5.7 of the SPS Agreement 51 to evidence his view. Andrew Lang also finds that in all cases where scientific evidence is found to be insufficient for conducting a risk assessment there should be the inferred right to apply article 5.7 52. According to Lang, the other three requirements should be regarded as substantive requirements, instead of a condition, which are triggered by this right. He adds: at the level of principle: there is something objectionable in principle about a situation in which the right to take protective measures can become legally unavailable, even where a clearly legitimate reason exists for such measures 53. This makes sense. Not only because of the aforementioned arguments but also because the three other requirements rather seem to be parameters for the first requirement therefore not -as postulated Japan Agricultural Products II- equally important. In short 2.2 and 5.7 are mutually exclusive as obligation, provided that all four conditions are met. Gruszczynski maintains that the scope of this exclusion is not clear. He states: There is no case law which would explain the relationship between Article 5.7 and other provisions of the SPS Agreement. 54. 50 Gruszczynski L., 2010, p184-185 51 Appellate Body Report: US Continued Suspension, para. 674. 52 Lang. (2008) Provisional Measures under Article 5.7 of the WTO s Agreement on Sanitary and Phytosanitary Measures: Some Critisism of the Jurisprudence So Far P630 53 Ibid. 54 Gruszczynski L., 2010 p186 23

CHAPTER IV SUFFICIENCY CASE LAW EXPLORED Hitherto, this paper has established that a crucial requirement for the applicability article of article 5.7 is scientific sufficiency. The primary of the four cumulative requirements for adopting a provisional measure is that relevant scientific evidence is insufficient. This triggers the application of Article 5.7. Hence, it is of paramount importance to determine in what circumstances this criterion will be met. It should be noted that the notion of sufficiency is a very complex and multifaceted aspect of the SPS measure. This part attempts to clarify this notion of sufficiency within this context. The US disagreed with Japan in the case Japan Agricultural Products II, The United States argued that sufficiency in article 5.7 pertains to sufficiency of the evidence available to allow for the conduct of a risk assessment. Hence, when a Member state desires to adopt said provisional measure, there should be a lack of information to do an objective risk assessment. This is different from a situation where sufficient evidence to conduct a correct risk assessment under 2.2 is established but the risk assessment under 5.1, does not support the measure, still violating article 2.2. The argument of the US for this is rather sensible: if article 5.7 is applied, as opposed to article 2.2, on the mere fact that risk assessment does not support the measure, it would defeat the purpose of the obligation of article 5.1 55. The use of article 5.7 is limited to cases where the available scientific evidence does not allow for a risk assessment, instead of providing for a loophole contained in a risk 55 Japan Agricultural Products II, para. 27 24

assessment. This contention was not settled by the Panel and the Appellate Body, for reasons of judicial economy. First the Appellate Body explicitly stated that a link exists between the first sentence of Article 5.7 and the obligation to perform a risk assessment under Article 5.1. 56. With this link established, it follows that relevant scientific evidence will be insufficient within the meaning of 5.7 if the body of available scientific evidence does not allow, in quantitative or qualitative terms, the performance of an adequate assessment of risks as required under Article 5.1, and as defined under Annex A(4) of the SPS Agreement 57. 4.1 INSUFFICIENCY OF SCIENTIFIC EVIDENCE V UNCERTAINTY Insufficient and uncertain: two words which appear to amount to the same point. Initially one would be able to argue insufficient evidence is uncertain evidence, since logically- certain evidence is in any way sufficient. Merriam-Webster defines sufficiency as 1 a : enough to meet the needs of a situation or a proposed end b : being a sufficient condition [..] and uncertainty as a : not known beyond doubt : dubious b : not having certain knowledge : doubtful c : not clearly identified or defined. However, WTO case law departs from this literal reading and provides that insufficiency and uncertainty are not the same. The Appellate Body in Japan Apples 58 refused to acknowledge these concepts as interchangeable 59. In Japan Apples, Japan had applied certain Phytosanitary requirements in order to protect against the risk of fireblight from the importation of apples. The US challenged this measure. 56 Japan Apples Appellate Body Report, id. Para. 179. 57 Ibid. Para. 179. 58 1344 Appellate Body Report, Japan Apples, para. 184. 59 Ibid. Japan - Apples para 184 25

This was the first time were the Panel and Appelate Body dealt with the first requirement of 5.7. The trigger for this applicability was not uncertainty per se 60. The AB in Australia - Salmon found that existence of unknown and uncertain elements does not justify a departure from the requirements of Articles 5.1, 5.2 and 5.3 61, so finding uncertainty does not - trigger the application of art. 5.7 62 (i.e. fulfills the function of the first requirement). The Suspension Panel complicates this by stating that the existence of scientific uncertainty does not automatically amount to insufficiency of relevant scientific evidence. Gruszczynski argues that this could mean that the Panel sees some types of uncertainty as amounting to insufficiency 63. For example, the Panels in US Continued Suspension and Canada Continued Suspension held that the fact that a number of aspects of a given scientific issue remain uncertain may not prevent the performance of a risk assessment 64. There could be an overlap between insufficiency and uncertainty. So, some types though not all- of uncertainty could trigger the requirement of 5.7. A. In particular, the Panels excluded from the determination of insufficiency under Article 5.7 theoretical uncertainty. This term refers to the uncertainty that always remains because science can never provide absolute certainty about the safety of the substance. The Panels here referred to the Appellate Body s finding in EC Hormones that theoretical uncertainty should not be assessed under Article 5.1 of the SPS Agreement 65. 60 Denise Prévost What Role for the Precautionary Principle in WTO Law after Japan-Apples? p 13 61 Appellate Body Report, Australia Salmon, para 130. 62 Appellate Body Report, Japan Apples, para 184. 63 Gruszczynski L., 2010 p180 64 Panel Report, US Continued Suspension, para.7.628; Panel Report, Canada Continued Suspension, para. 7.606. 65 Appellate Body Report, EC Hormones, para. 186. 26

B. Uncertainty arising from divergent views on a scientific issue, and uncertainties that may be present in the conclusions of a risk assessment despite the fact that there was sufficient evidence to perform the risk assessment. The Panels looked at the finding of the Panel in EC Approval and Marketing of Biotech Products. The Panel in Biotech found that "uncertainties present in a risk assessment may be taken into account by a Member when determining its SPS measure." 66 Such uncertainty, according to the Panels, does not mean that previously sufficient evidence has become insufficient nor should it ipso facto justify the applicability of Article 5.7 of the SPS Agreement. 67 4.2 CHANGING EVIDENCE Science changes. It does so most of the time and this state of flux characterizes our civilization. People such as Galileo Galilei came forth with new scientific information, completely the opposite of conventional scientific assumptions and political/religious preferences. Now, seeing that science, and the sufficiency thereof, forms is of such a value to SPS measures, it is worth examining how the WTO regards this issue. The compliance panel in Japan Apples 68 was faced with the question whether new scientific studies, presented by Japan, supporting Japan s revised measure against the risk (of fire blight) could be qualified as minority opinion 69. The panel experts did not agree with the conclusions of the new studies in question: they regarded the studies as 66 Panel Reports, EC Approval and Marketing of Biotech Products, para. 7.1525. Panel Report, US Continued Suspension, paras 7.631-7.634; Panel Report, Canada Continued Suspension, paras 7.609 7.612. 67 Panel Report, US Continued Suspension, para. 7.628; Panel Report, Canada Continued Suspension, para. 7.606. 68 Japan Apples (Article 21.5 Canada) 69 Panel Report, Japan Apples (Article 21.5 Canada), para. 8.146. 27

reflecting laboratory conditions as opposed to real world conditions 70. Emphasizing that the latter was important to demonstrate the Panel did not agree with Japan s standpoint. Thus, new scientific studies are not to be regarded as a minority opinion 71.. The panel in EC Biotech suggested that an EC Member State would never be able to impose provisional measures under Article 5.7 of the SPS Agreement due to the existing EC-level risk assessment. 72 I.e. since a risk assessment has already been undertaken, sufficient scientific evidence (Articles 2.2 and 5.1) exists and will always exist. 73. 74 This interpretation is problematic. Because this would mean that new evidence demonstrating a greater degree of risk than that which was previously understood could never trigger a provisional measure under Article 5.7 of the SPS Agreement. 75 This interpretation would not only mean that member states have to anticipate and work with evolving health risks, but on a more fundamental level, it ignores the fact that science is not necessarily a set-in-stone type of process. It could be very well the case that that previously unknown or unavailable evidence could have a significant impact on conventional or accepted risk perceptions. So logically, sufficient scientific evidence can always become insufficient in light of the new evidence/research. Luckily, the Panel in US/Canada Continued Suspension recognized that:.. there could be situations where existing scientific evidence can be put in question by new studies and information. There could 70 Ibid., para. 8.140-8.144. 71 Panel Report, EC Approval and Marketing of Biotech Products, par. 7.3062. 72 EC Biotech Panel Report, supra note 7, para. 7.3260. 73 Caroline Henckels, GMOs in the WTO: A Critique of the Panel s Legal Reasoning in EC Biotech, 7 MELB. J. INT L L. 295 (2006). 74 Japan Apples Appellate Body Report, supra note 37, para.180-182. But see the panel decision in US/Canada Continued Suspension, Panel Report, US/Canada Continued Suspension of Obligations in the EC Hormones Dispute, para, 7.597-7.599, WT/DS31/R (circulated on Mar. 31, 2008) stating that scientific evidence which was previously deemed sufficient could subsequently become insufficient. 75 See Japan Apples Appellate Body Report, supra note 37, para.180-182. 28

even be situations where evidence which supported a risk assessment is unsettled by new studies which do not constitute sufficient relevant scientific evidence as such to support a risk assessment but are sufficient to make the existing, previously relevant scientific evidence insufficient 76. Bearing this in mind, the Appellate Body in US/Canada Continued Suspension finally reversed the position taken by the panel in EC Biotech. 77 It found that there situations could occur where the new scientific developments themselves do not permit the performance of a new risk assessment that is sufficiently objective 78 fall within the scope of Article 5.7. This new approach, which essentially requires that a previously conducted risk assessment must be reviewed in light of new evidence, means that new evidence could be used either to perform another risk assessment 79 or simply to render insufficient scientific evidence which was previously considered sufficient. Though it should be noted that [t]he possibility of conducting further research or of analyzing additional information, by itself, should not mean that the relevant scientific evidence is or becomes insufficient 80. So, the difficult question for the Appellate Body was not in holding that sufficient scientific evidence can be rendered insufficient through new evidence (a proposition which the US and Canada dispute seemed to accept: Both the United States and Canada accept that evidence which at some point in time was sufficient to perform a risk assessment could become insufficient at a later point in time. 81 ), but rather under what 76 Panel Report, US Continued Suspension, para. 7.620; Panel Report, Canada Continued Suspension, para. 7.598 77 US Continued Suspension Appellate Body Report, supra note 10, para. 701 78 Ibid. 79 This in accordance with Annex A(4) of the SPS Agreement, i.e. when it does not allow, in quantitative or qualitative terms, the performance of an adequate assessment of risks 80 US Continued Suspension / Canada Continued Suspension, para. 702 81 Ibid. para. 700 29

circumstances could this shift occur. The Appellate Body started its analysis by recognizing that the mere supplementation of existing scientific knowledge is not enough to render information insufficient, before determining that the panel s critical mass standard for determining insufficiency under Article 5.7 requiring a critical mass of new evidence that calls into question the fundamental precepts of previous knowledge and evidence so as to make relevant, previously sufficient, evidence now insufficient 82 as far too difficult and inflexible. 83 Instead, the Appellate Body determined that Members should be permitted to take a provisional measure where new evidence from a qualified and respected source casts doubts as to whether the previously existing body of scientific evidence still permits a sufficiently objective assessment of risk. 84 The AB stated that a WTO Member should be permitted to take a provisional measure where new evidence from a qualified and respected source puts into question the relationship between the pre-existing body of scientific evidence and the conclusions regarding the risks 85 In other words, existing scientific evidence can become insufficient if new evidence casts into doubt the soundness of previous conclusions. 86 In short, this (re-) evaluation of the sufficiency is more practical than the previous approach and leans towards a more flexible approach. It acknowledges that sciences does change continuously, on the other hand it does not render measures particularly transparent because in certain field scientific knowledge can change radically. 82 US Continued Suspension Appellate Body Report, supra note 10, para. 704 83 Ibid. para. 699-702. 84 Ibid. para. 703. 85 Hormones 2008 para. 703 86 Andrew T.F Lang, Provisional Measures Under Article 5.7 of the WTO s Agreement on Sanitary and Phytosanitary Measures: Some Criticisms of the Jurisprudence So Far, 42 J. WORLD TRADE 1085, 1096-1097 (2008) 30