IV.75 Canada argued that the EC stated level of protection for the three natural hormones, "... i.e. ensure the EC consumers that there are no residues left over other than the naturally produced ones by the animals themselves", reflected a fundamental misunderstanding. Canada stated that this was not a level of protection, nor an MRL. The purported risk posed by residues of the natural hormones would occur regardless of whether the residues were produced endogenously or administered exogenously. Thus, the EC measures did not achieve a higher level of protection with respect to the natural hormones, as required by Article 3.3.

(g) Article 3.3 of the SPS Agreement

IV.76 The European Communities observed that the maintenance of a higher level of protection than that afforded by a Codex standard was explicitly allowed by Article 3.3 of the SPS Agreement. This Article provided an exception to the obligation to base SPS measures on international standards:

"Members may introduce or maintain sanitary or phytosanitary measures which result in a higher level of sanitary or phytosanitary protection than would be achieved by measures based on the relevant international standards, guidelines or recommendations, 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. Notwithstanding the above, all measures which result in a level of sanitary or phytosanitary protection different from that which would be achieved by measures based on international standards, guidelines or recommendations shall not be inconsistent with any other provision of this Agreement."

A footnote to this paragraph provided that:

"For the purposes of paragraph 3 of Article 3, there is a scientific justification if, on the basis of an examination and evaluation of available scientific information in conformity with the relevant provisions of this Agreement, a Member determines that the relevant international standards, guidelines or recommendations are not sufficient to achieve its appropriate level of sanitary or phytosanitary protection."

IV.77 Canada observed that the first sentence of Article 3.3 could be divided into two parts. The first phrase set out a precondition for the departure from the relevant international standards: the Member's measures must result in a higher level of sanitary or phytosanitary protection than would be achieved by measures based on the international (Codex) standards. Any other departure was not contemplated. The second phrase presented two alternative justifications for such a departure: first, scientific justification; second, as a consequence of the level of sanitary or phytosanitary protection a Member determined to be appropriate in accordance with the relevant provisions of paragraphs 1 through 8 of Article 5. The footnote to Article 3.3 provided that there was scientific justification if "... on the basis of an examination and evaluation of available scientific information in conformity with the relevant provisions of this Agreement, a Member determines that the relevant international standards, guidelines or recommendations are not sufficient to achieve its appropriate level of sanitary or phytosanitary protection". Article 3.3 also required that the measures not be inconsistent with any other provision of the SPS Agreement.

IV.78 Canada suggested that the two justifications (i.e., situations) were similar, and complementary, but were intended to deal with two different situations. The first justification was intended to deal with situations where an international standard was outdated and needed to be re-evaluated, and that standard did not provide the level of protection it was thought to have provided. Thus, "... an examination and evaluation of available scientific information in conformity with the relevant provisions of this Agreement ..." (inter alia, Articles 5.1 and 5.2) revealed that the international standard was "... not sufficient to achieve [the Member's] appropriate level of sanitary or phytosanitary protection." The second justification contemplated a situation when a Member wished to maintain a higher level of protection than that afforded by the international standard, where the scientific analysis underlying the standard was valid. However, the second justification required that the different level of protection be ".... in accordance with the relevant provisions of paragraphs 1 through 8 of Article 5." This would include the risk assessment requirements set out in Articles 5.1 and 5.2, as well as the consistency disciplines of Article 5.5.

IV.79 In Canada's view, the EC measures failed to meet these requirements for derogations from the obligation in Article 3.1 in four ways. First, with respect to the three natural hormones, the EC measures failed to provide a higher level of protection than would be achieved by measures based on the Codex standards. Since the levels of natural hormones in beef derived from untreated livestock varied widely, depending upon the sex, age and fertility cycle of an animal, the levels of these hormones in beef derived from hormone-treated livestock were well within the levels of natural variation. Since the European Communities did not regulate the exposure of consumers to higher levels of these hormones occurring in the meat of untreated animals, the EC measures failed to achieve any purported higher level of protection. Canada claimed that the naturally occurring hormones, whether endogenous or exogenous, were identical in chemical structure. The fact that these substances were administered exogenously had no bearing on whether or not they were carcinogenic. Thus the EC stated level of protection for the three natural hormones, "... i.e. ensure EC consumers that there are no residues left over other than the naturally produced ones by the animals themselves" was not a level of protection, nor even a MRL. The purported risk posed by residues of the natural hormones would occur regardless of whether the residues were produced endogenously, or administered exogenously.

IV.80 Canada noted, furthermore, that the European Communities did not regulate the exposure of consumers to far higher levels of natural hormones occurring in a variety of foods. Second, there did not appear to be a scientific justification for a higher level of protection. An examination and evaluation of available scientific information in conformity with the relevant provisions of the SPS Agreement revealed that the six hormones in question did not present any harmful effects to the health of the consumer when used under the appropriate conditions as growth promoters in farm animals. Third, since the level of protection determined to be appropriate by the European Communities was not in accordance with the relevant provisions of paragraphs 1 through 8 of Article 5, the EC measures could not be a valid consequence of that level. Fourth, the EC measures were inconsistent with Article 5 of the SPS Agreement. Canada submitted, therefore, that the EC measures were contrary to Article 3.

IV.81 The European Communities argued that the term "appropriate level of protection" appeared in Article 3. The SPS Agreement gave the right to Members, when introducing or maintaining their measures, to choose between measures which resulted in a level which was higher than that which would result by measures based on the international standards (Article 3.3), if there was scientific justification, or a level of protection which the member determined to be appropriate. The European Communities argued that the text of the footnote to Article 3.3 confirmed the power, which the SPS Agreement left to Members, to make risk management decisions that reflected societal value choices distinct from the strict scientific process of risk assessment. As it had been said : "the choice of appropriate level of protection appears to be the unilateral prerogative of each WTO member State ..." ⁶⁰

IV.82 The European Communities observed that in both of the options provided in Article 3.3, which were available to a Member when taking a sanitary or phytosanitary measure, there must have existed "a potential risk for adverse effects". In other words, it was implicit that in order to need a level of protection there must have been some hazard against which a Member needed to protect. However, this only implied the identification of a hazard, not an assessment of the probability that it would cause damage. The SPS Agreement left Members free to define the level of probability they wanted to assume: this might range from zero to infinite; it also left them free to decide the type of measure they might choose to ensure that the level of protection considered by them to be appropriate was achieved. The approach adopted by the SPS Agreement was in conformity with previous GATT law and practice, and was a sensible approach for the purpose of establishing multilateral rules and disciplines to guide the development and progressive harmonisation in order to minimize the negative effects on trade from national sanitary and phytosanitary measures. The approach of the SPS Agreement was also in conformity with democratic regulatory procedures, where frequently a dichotomy was operated in the decision making process between: risk assessment and risk management. The first established strictly the scientific basis for regulatory action. The second (risk management) was the process by which the competent authority of a Member decided what action to take in the face on the assessment submitted to it by the scientists. Such action was based on factors such as public health and environmental protection, relevant legislation and legal precedent, application of social, economic and political values and consumer concerns. The risk management phase, therefore, in a democratic legislative system, expressly recognized the importance of societal value choices.

IV.83 The European Communities claimed that this seemed to be also the view of Canada. In its Statement on Implementation for the Agreement Establishing the World Trade Organization (31 December 1994, Canada Gazette, Part I, page 4888), it was stated:

"... Although it is not necessary for members to base their chosen levels of protection on risk assessment, they should work to minimize negative trade effects when choosing the level".

The above interpretation of the disciplines the SPS Agreement imposed on Members was apparently supported also by the United States, since when the USTR had presented the results of the Uruguay Round for approval to Congress it had stated :

"The SPS Agreement thus explicitly affirms the right of each government to choose its levels of protection, including a "zero risk" level if it so chooses. A government may establish its levels of protection by any means available under its law, including by referendum. In the end, the choice of the appropriate level of protection is a societal value judgement. The Agreement imposes no requirement to establish a scientific basis for the chosen level of protection because the choice is not a scientific judgement"⁶¹ (emphasis added).

The European Communities wondered if Canada agreed with the above statement from the US Statement of Administrative Action.⁶²

IV.84 In response to the arguments of Canada that the EC measures did not meet the requirements for derogation from the obligation to base the measures on Codex's recommendations, the European Communities responded that it was not possible to limit exposure of its consumers to residues from the hormones occurring naturally in the meat of untreated animals. The European Communities did not know how one could regulate the levels of residues from naturally occurring hormones other than banning all meat and foods from human consumption, but assumed this was not what Canada was suggesting. This argument of Canada also disregarded the fact that hormones naturally occurring in animals and other foods had formed part of the human diet and had entered the metabolism of the human body throughout the course of human evolution. The naturally occurring hormones could not be compared with exogenously administered carcinogenic substances given to animals for growth promotion. As Dr. Liehr had said in his written opinion submitted to the Panel: "as the amount of hormone or metabolite necessary for tumour induction is not known, the amount of exogenous hormone or metabolite necessary for tumour induction in addition to unknown amounts of endogenous hormone or metabolite have not yet been determined".

IV.85 The European Communities further contended that this argument of Canada also amounted to dictating the type of measures the European Communities should put in place in order to achieve the level of protection it alone might decide. But the SPS Agreement imposed no requirement to establish a scientific basis for the chosen level of protection because the choice was not a scientific judgement. Canada's submission, therefore, in fact attacked not only the level of protection chosen by the European Communities, but also its measures, by insisting that residues of hormones above naturally present levels did not pose any risk to health and, therefore, did not warrant the application of any measures to control them other than MRLs recommended by Codex Alimentarius. If the Panel were to accept this line of argument, it would in the future be open to any country to oppose any health measure which was based on the precautionary principle. There were very few examples of health hazards where all scientists agreed on the degree of risk, and countries might be allowed to make judgments as to what degree of risk they were willing to accept. But the SPS Agreement recognized the fact that scientific certainty was rare and many scientific determinations required judgments between differing scientific views. The SPS Agreement preserved the ability of governments to make such judgments.

IV.86 The European Communities stated that Canada's argument amounted to offering a solution to safeguard against the potential risk to human health from the use of these hormones for growth promotion which the European Communities had considered carefully but rightly rejected, because it did not meet its appropriate level of protection. The European Communities also noted that the MGA was authorized for growth promotion only in Canada and the United States, but apparently nowhere else. The Codex Commission had never examined scientifically this hormone and had never recommended any standard. In the view of the European Communities, Canada in essence argued that the European Communities should accept meat treated with MGA and check only for residue limits as they were set by Canada. The European Communities submitted that this argument would fly in the face of the SPS agreement, which explicitly allowed the European Communities to adopt a level of protection which it deemed appropriate (no residue at all). Therefore, the European Communities could choose to adopt a measure which was no more trade restrictive than required to achieve its appropriate level of protection, taking into account technical and economic feasibility, i.e. a prohibition of use of MGA for animal growth promotion.

IV.87 The European Communities also observed that the remaining three reasons invoked by Canada in its submission were unsubstantiated, because there was a growing part of "available" scientific evidence which argued that these hormones were dangerous to human and animal health and the European Communities was basing its measures on this part of the available scientific evidence, as the SPS Agreement clearly allowed it to do. In its view, there were usually several ways of dealing with any given hazard, and the SPS Agreement did not require Members to change the type of measure they chose. For example, one of the many ways of tackling pathogenic organism in food was irradiation, and some Members allowed it to be used for certain foods. Codex had adopted standards for irradiation of foods but this did not mean that every Member was now obliged to allow irradiation of all foods. Similarly, if a Member had chosen a level of protection against a contaminant on the basis of an MRL, this did not mean that it should set its MRL at the level recommended by Codex. Also if a Member had chosen not to allow any residue of a dangerous substance in food, this did not mean that it was obliged to base its protection on the concept of an MRL recommended by Codex, if one existed for that substance. The maintenance of a higher level of protection than that afforded by a Codex standard was explicitly allowed by Article 3.3.

IV.88 The European Communities claimed that the two situations referred to in Article 3.3 as "scientific justification" and "consequence of the level of sanitary or phytosanitary protection a Member determines to be appropriate ..." clarified the circumstances under which a Member may derogate from the provisions of Article 3.1. In order to understand properly the two possibilities envisaged, it was crucial to note the distinction between appropriate level of sanitary protection, on the one hand, and sanitary or phytosanitary measures introduced or maintained to achieve the level of protection, on the other. This distinction was important, because paragraph 6 of the preamble to the SPS Agreement made it plainly clear that harmonization of sanitary measures with international standards, guidelines and recommendations did not require Members to change the appropriate level of protection of human, animal or plant life or health which they had been applying before the entry into force of the SPS Agreement.

IV.89 The "scientific justification" was applicable to the first type of measures. This followed clearly from the text of Article 3.3, which had a comma (,) after justification and then started with the word "or". A "scientific justification" existed when the international standard, guideline or recommendation was inadequate, faulty or obsolete from the scientific point of view. The footnote to paragraph 3 clarified this very well. The first situation appeared to concentrate on the scientific dimension of the international standard, guideline or recommendation. A Member might determine that measures based on an international standard could not achieve its level of protection, because such standard was based, for example, on only part of the available scientific evidence, or was based on faulty scientific evidence, or was out of date because it had been overtaken by more recent evidence, or was based on different climatic or environmental conditions than those prevailing in the Member concerned. The scientific justification referred only to measures, not to the level of protection. This was also in conformity with the other provisions of the SPS Agreement, in particular Article 2.2.

IV.90 The European Communities further argued that an international standard, guideline or recommendation implied a certain level of sanitary protection. There might be several types of measures which could achieve that level of protection. The objective of the SPS Agreement was to harmonize such possible measures. But the objective of the Agreement was not to harmonize the level of protection a Member considered appropriate in its territory. Therefore, if the level of protection implied by the international standard and the level deemed appropriate by a Member were different, the measures to achieve that Member's level were also bound to be different, especially when its level was higher than that implied by the international standard.

IV.91 The second situation clarified that a Member was in any case entitled to introduce or maintain measures which aimed at achieving the level of sanitary protection it deemed appropriate in its territory, and that such measures resulted in a higher level of protection than that which would result by measures based on the relevant international standard. In this case, when determining its appropriate level of protection, the Member had to comply with paragraphs 1 to 8 of Article 5. Paragraphs 1, 2, 3, 6, 7 and 8 dealt with the assessment of the risk and the type of measures a Member may take. Paragraphs 4 and 5 dealt with the issue of determining the level of protection. But paragraphs 4 and 5 did not indicate how the level of sanitary protection in itself was to be determined. The European Communities recalled its argument that the level of protection was not a scientific judgment and a Member was free to apply the level of sanitary protection it deemed appropriate in its territory. Paragraph 4 (which was not mandatory) and paragraph 5 (which was only an objective to be implemented through guidelines yet to be developed) attempted to place only certain constraints on the liberty of Members to implement the level of protection they had determined. In other words, they should (not shall) take into account the objective of minimizing trade effects and shall avoid arbitrary or unjustifiable distinctions in the levels, if such distinctions result in discrimination or disguised restriction on international trade. Therefore, none of the paragraphs of Article 5 dealt with setting the level of protection per se.

IV.92 The European Communities observed that in 1989 the detailed provisions of the SPS Agreement were not in place and the GATT 1947 case law, in particular the Thai Cigarettes Panel report of 1990, had clearly acknowledged the right of Members to take sanitary or phytosanitary measures provided the provisions of the "chapeau" of Article XX were respected. On 1 January 1995, the WTO Agreement and its annexes had entered into force. Still at that time there were no Codex standards, which were adopted only in July 1995. The EC level of protection for these hormones had been the same both before and after the entry into force of the SPS Agreement. The 6th paragraph of the preamble of the SPS Agreement also provided that in harmonizing their measures to international standards Members were not required to change their appropriate level of protection of human, animal or plant life or health. Therefore, the SPS Agreement could not force the European Communities to change the level of protection it had been applying legally since 1989. If the measures of a Member did not follow international standards, there should be no negative inference that they violated the provisions of the SPS Agreement or GATT. The burden of proof was on the complaining party to establish the alleged violations. Since Canada argued that the European Communities had no scientific justification, it needed to prove on the basis of the scientific evidence of today how the European Communities could achieve its higher level of protection, if it had based its measures on the standards of Codex. The European Communities concluded that when it decided to maintain its level and its measures for these hormones, there was no international standard to compare them with. In addition, on 1 January 1995, the European Communities clearly had a scientific justification for its measures. In 1988, the European Communites had examined the available scientific information and found it deficient and inadequate to guarantee its level of sanitary protection. In addition, at the time of entry into force of the SPS Agreement, the European Communities had decided to maintain its measures because it was clear that the evidence on which the 1988 JECFA report was based out of date. At that time, the European Communities had decided that the draft JECFA MRLs were not sufficient to achieve its appropriate level of sanitary protection.

IV.93 The European Communities claimed that the legal question of justification arose only after July 1995, when the Codex Commission adopted the recommended MRLs for the two synthetic hormones and decided not to adopt MRLs for the three natural hormones. The decision of the Codex Commission of July 1995 made no difference from the scientific point of view, because it was based on evaluation of scientific information dating from 1987 or earlier. This scientific information had been reviewed and evaluated several times by the relevant bodies of the European Communities as regarded these hormones for growth promotion; and the European Communities had constantly rejected it on scientific as well as other grounds. Such other factors included conditions of use of these hormones for growth promotion, potential for misuse, effectiveness of detection and control, synergistic effects, long-term exposure, etc. It should also be borne in mind that JECFA and Codex had never evaluated these hormones when used for therapeutic or zootechnical purposes.⁶³ The European Communities had decided again to maintain its measures in force because it had both a scientific justification and because its appropriate level of sanitary protection for these hormones was the same before and after that date, i.e. higher than that which could be achieved by measures based on Codex standards. The scientific information which had been emerging since the mid 1980's had reinforced the scientific basis upon which the European Communities had already decided to prohibit the use of these hormones for growth promotion. The newly emerging evidence supported clearly the view that these hormones (or at least some of them) were genotoxic and carcinogenic and, therefore, no tolerance levels could be fixed. As Dr. Lucier and the scientists advising the European Communities had said, the JECFA Report of 1988 was clearly out of date at that time. This report was certainly in need of review in July 1995, in April 1996 and even more today.

IV.94 The European Communities explained that it had started the process of reviewing the scope of application of Directive 88/146 since 1993, in order to expand its application to new substances (like beta-agonists) and to reinforce its provisions on control, testing and, in particular, to increase the fines and penalties for violators. This reinforcement in the legislation was requested by the European Parliament, in the light of the reports it had prepared on these substances. Whilst this proposal was being discussed, the SPS Agreement had entered into force and, consequently, the European Communities had reviewed its measures and had decided to maintain the prohibition on the use of these hormones for growth promotion. While the proposal which had led to the adoption of Directive 96/22 was pending, the European Communities had also organized the 1995 Scientific Conference in December 1995, mindful of its international obligations under the SPS Agreement. This conference was in itself not a risk assessment, in the sense of Articles 5.1 and 5.2, because the concept of risk assessment, as explained, was much wider and encompassed the other elements mentioned therein. This was made clear from the discussion with the scientists. But on the basis of some of its scientific conclusions, together with the other pieces of scientific information which were already available on these hormones, and after consulting with the appropriate legislative bodies, the European Communities had decided to maintain the prohibition on the use of these hormones for growth promotion (i.e. it had decided to maintain the measures) and also to expand their coverage to other substances and to increase controls and the levels of fines. This was contained in Directive 96/22 of April 1996, which would enter into force on 1 July 1997.

IV.95 In the view of the European Communities, Canada had to show on the basis of today's scientific evidence that the measures the European Communities maintained in place after July 1995 violated the provisions of the SPS Agreement. However, instead of using the scientific evidence of today, Canada had been using the scientific evidence of JECFA 1988 (and the much earlier evidence of the Lamming Committee reports). This scientific evidence was clearly out of date, as all the scientists advising the European Communities (and some of the scientists advising the Panel) had pointed out in their reports. As Dr. Ritter had said in one of his replies, "the current weight of scientific evidence would, however, appear to continue to support the view ... that at present there is no evidence for possible health risks to the consumer due to the use of the natural sex hormones for growth promotion ... ". But this was the personal view of Dr. Ritter about the current weight of scientific evidence. Since 1989, the scientific and legislative bodies of the European Communities viewed the weight of scientific evidence differently. The SPS Agreement did not require unanimity in deciding what was the relevant scientific evidence before a Member was allowed to take sanitary measures; people did not have to start dying before a preventive measure could be taken.

IV.96 The European Communities considered therefore that it had done all it could to conform with its international obligations. It had reviewed the measures and had decided to maintain them, in accordance with the provisions of the SPS Agreement, because it had a scientific justification. The other grounds for maintaining them included conditions of use of these hormones for growth promotion, potential for misuse, effectiveness of detection and control, synergistic effects, long-term exposure, etc. It had also reviewed, even if it did not have to, its level of protection in this case, and had decided to maintain it as it was before the entry into force of the SPS Agreement with regard to the use of these hormones for growth promotion. It could not achieve its level of protection by other measures based on the standards adopted by Codex in July 1995, which would have been significantly less restrictive on trade. The extremely small number of controls performed by Canada and the number of violations still discovered within this extremely low number of controls demonstrated that there was no other measure which could achieve its level of protection and which was technically and economically feasible and, in addition, significantly less restrictive on trade. At least Canada had not been able to demonstrate the existence of such an alternative measure. The European Communities had also shown that any type of labelling in this case did not work.

(h) Article 5 of the SPS Agreement

IV.97 Canada noted that an understanding of how scientists evaluate risk is essential to an overall understanding of the safety of veterinary drugs used in animal husbandry. Risk analysis had been defined by the Joint FAO/WHO Expert Consultation as meaning "... a process consisting of three components: risk assessment, risk management, and risk communication".⁶⁴ At the end of the process, a sanitary measure might be put in place to control an identified health risk(s). The type of measure chosen related to the severity of the risk and the least trade restrictive risk management option(s), as identified during the risk analysis process. In its most recent draft, Codex had defined risk as "... a function of the probability of an adverse health effect and the severity of that effect, consequential to a hazard(s) in food".⁶⁵ Codex acknowledged that hazards were present in food as a result of chemical contaminants (e.g., pesticides and veterinary drug residues), microbiological contaminants, or naturally occurring substances (e.g., toxins) found in food. These hazards presented a risk to human health. Risk assessment, which was one component in the risk analysis process, permitted an objective evaluation of these hazards.

IV.98 Canada noted that the SPS Agreement required that a chosen sanitary measure be based on an appropriate risk assessment. Risk assessment was a process that recognized the inherent uncertainties in conducting a scientific evaluation of the effects certain hazards posed to human health. Thus, the risk assessment process was designed to be conservative and required extensive testing and analysis when evaluating potential human health hazards. Risk assessment was a specific component of the risk analysis process which, conducted by scientists, included well defined procedures that had been described by Codex and JECFA. In the most recent report of the Joint FAO/WHO Expert Consultation on the application of risk analysis to food standard issues, the risk assessment process was defined as having four components:

(i) hazard identification

(ii) hazard characterization

(iii) exposure assessment

(iv) risk characterization.⁶⁶

IV.100 Hazard Identification was the identification of biological, chemical and physical agents capable of causing adverse health effects, and which might be present in a particular food or group of foods.⁶⁷ Canada indicated that in dealing with veterinary drug residues, the goal was to identify potential adverse health effects in humans associated with exposure to a veterinary drug. The qualitative likelihood of such effects occurring in exposed human populations, and the certainty or uncertainty associated with such effects, were evaluated using available data. These data might be derived from a number of sources, such as epidemiological studies or animal toxicological studies.⁶⁸ If there was any evidence of a hazard, then the hazard characterization process of the risk assessment was undertaken.

IV.101 Hazard Characterization was the qualitative and/or quantitative evaluation of the nature of the adverse health effects associated with biological, chemical and physical agents which might be present in food. For chemical agents, such as veterinary drugs, a dose-response assessment was performed.⁶⁹ Canada added that in all cases, the chemicals being considered for hazard characterization were present at very low levels in foods, that is, parts per million ("ppm") or less. Therefore, to obtain adequate sensitivity in humans, animal toxicological studies were often conducted at very high levels, sometimes exceeding several thousand ppm's.⁷⁰ A safe level or Acceptable Daily Intake ("ADI") was derived from the experimental No Observable Effect Level ("NOEL") or the No Observed Adverse Effect Level ("NOAEL") by applying an appropriate safety factor. When data from long-term animal toxicity studies were available, a safety factor of 100 was generally applied. Larger safety factors, up to 1000, might be used in certain cases. This meant that there was no significant risk if the chemical was ingested at or below the ADI and the likelihood of adverse health effects was notionally zero.

IV.102 Canada submitted that traditionally, toxicologists have accepted the existence of thresholds for adverse effects with the exception of carcinogenicity. This was because genotoxic carcinogenic compounds had the ability to produce mutations in genetic material (DNA) leading to tumour formation. In recent years, however, it had been possible to discriminate between genotoxic carcinogens and non-genotoxic carcinogens. The latter were themselves not capable of producing mutations, although there might be an effect on cells that were already in the process of mutating. Hazard characterization now distinguished between genotoxic and non-genotoxic carcinogens.⁷¹ In principle, non-genotoxic carcinogens might be regulated using ADIs derived from an experimental NOEL or NOAEL, and by applying appropriate safety factors (the "threshold approach").

IV.103 For genotoxic carcinogens, the "NOEL-safety factor" approach was generally not considered a suitable method for setting ADIs. Two approaches were available: (i) to ban the chemical from commercial use, or (ii) to establish a level of risk that was sufficiently small to be deemed negligible or insignificant. For genotoxic carcinogens, in establishing a negligible level of risk, a quantitative risk assessment process was used.⁷² This approach had been used to establish a MRL for carbadox, which had a metabolite that was a known genotoxic carcinogen. Carbadox was permitted for use in the European Communities as a feed additive. The first approach, to ban the compound, had been adopted by several countries, including the European Communities, for nitrofurans, which were also known genotoxic carcinogens.

IV.104 Exposure Assessment was the qualitative or quantitative evaluation of the likely intake of biological, chemical or physical agents via food, as well as exposures from other sources if relevant.⁷³ Canada explained that this was usually done by examining the dietary intake of foods and determining if the theoretical dietary intake was below the recommended ADIs.

IV.105 Risk Characterization was the qualitative and/or quantitative estimation of the probability of occurrence, and severity, of known or potential adverse health effects in a given population based on hazard identification, hazard characterization and exposure assessment.⁷⁴

IV.106 Canada indicated that risk management had been defined by Codex as "... the process of weighing policy alternatives in the light of the results of risk assessment and, if required, selecting and implementing appropriate control options, including regulatory measures".⁷⁵ It was the view of the Joint FAO/WHO Expert Consultation that "... risk assessment of chemical hazards in foods usually results in the selection of risk management options to ensure that foodborne risks to consumers are not appreciable ('notionally zero')".⁷⁶ The setting of MRLs was a risk management option that was commonly used in controlling the risks arising from chemical contaminants (e.g., veterinary drugs) in foods. Coupled with monitoring and testing programs, it was an example of a comprehensive sanitary control measure that could be used to manage risk effectively.

IV.107 Canada noted that risk communication was defined by Codex as, "[t]he interactive exchange of information and opinions concerning risk among risk assessors, risk managers, consumers and other interested parties".⁷⁷ Canada submitted that risk communication must take place at all stages of the risk analysis process to ensure open, balanced and meaningful discourse between science experts, policy makers, farmers, industry, consumers and all other interested parties.

IV.108 The European Communities argued that the scientific evidence on which hormone growth promoters were approved, most of it gathered by drug companies and still secret, was obtained using primitive techniques which would not be acceptable today. For example, the Codex decision of 1995 on the growth promoting hormones had been taken on the basis of scientific papers dating back to the 1940s and, in one case, to 1939. The European Communities argued that the system described under "Application of Risk Analysis to Food Standards issues" had not been approved by Codex. The whole subject of risk analysis had been under discussion in various bodies of Codex for several years but, precisely because there was no international consensus on the subject, nothing had been agreed. Codex had not even been able to agree definitions of the various terms used, let alone procedures for their application. At the Codex Committee on General Principles (Paris, November 1996) it had been agreed to forward a draft list of definitions to the Codex Alimentarius Commission meeting in 1997 for it to agree to their use on an interim basis, pending further clarification. The European Communities submitted that a measure might be applied to protect against a suspected hazard, not only against "an identified health risk" as claimed by Canada. The type of measure need not relate to the severity of the risk, nor need it be the least trade restrictive option. The criteria for the choice of measures were set out in the SPS Agreement, which did not mention the "severity" of the risk. The same type of measure might be used for risks of different severity: for example vaccination was a measure widely used against risks of very different severity.

IV.109 The question of trade­restrictiveness was addressed in Article 5.6, which did not require the "least" trade­restrictive option; it required Members to ensure that their measures were "not more trade­restrictive than required to achieve their appropriate level of sanitary or phytosanitary protection, taking into account technical and economic feasibility". In the case of growth hormones, it might appear to be least trade­restrictive to allow trade in all meat from treated animals as long as no residues were present in it. However, once the technical and economic problems of ensuring the absence of residues (control over supply and use, testing of meat, etc.) were taken into account, this option would in fact be more trade­restrictive than the measures applied by the European Communities.

IV.110 Noting that there was a perfectly satisfactory definition of risk assessment already in the SPS Agreement, Annex A(4), which expressed risk as "the potential for adverse effects", the European Communities noted that disease and death were not the only adverse health effects which might result from the presence of a residue in food ­ there might also be unwanted physiological consequences. For example, the European Court of Justice, having referred explicitly to the constitution of the WHO, which defined health as "a state of complete physical, mental and social well-being that does not consist only in the absence of illness or infirmity", had concluded that a broad interpretation should be given accordingly to the concept of health contained in a provision of EC law.⁷⁸

IV.111 The European Communities recalled that there was no internationally agreed objective, scientific process which could deliver a mathematically precise risk assessment. The statement in Canada's submission that "there may be some uncertainties" was an understatement, particularly in the assessment of biological hazards to human health where direct testing on humans was not possible and individuals varied greatly in their responses to substances. Hazard identification consisted of exactly that ­ identification of a hazard. The processes of evaluating the likelihood (exposure assessment) and severity of effects (risk characterization) were later stages in risk assessment. The reason Canada tried to introduce these later steps into the hazard identification was that a Member needed only identify a hazard in order to set a level of protection. The SPS Agreement did not require a further assessment of the hazard in quantitative terms, as Canada pretended, in order for a Member to decide that it did not want any of the hazard present in its food.

TO CONTINUE WITH EC MEASURES CONCERNING MEAT AND MEAT PRODUCTS (HORMONES) COMPLAINT BY CANADA

⁵⁹ D.A. Wirth, "The Role of Science in the Uruguay Round and the NAFTA Trade Disciplines", 27 Cornell International Law Journal, 817-859, at 826 (1994). The European Communities noted also that the US had put it as follows in the SAA: "In the end, the choice of the appropriate level of protection is a societal value judgment .... The SPS Agreement imposes no requirement to establish a scientific basis for the chosen level of protection because the choice is not a scientific judgment".

⁶⁰ US Statement of Administrative Action, point 3(b).

⁶¹ The European Communities argued that the same applied also under the NAFTA agreement, whose Article 712, para. 2 provides:

"Right to Establish Level of Protection. Notwithstanding any other provision of this Section, each Party may, in protecting human, animal or plant life or health, establish its appropriate level of protection in accordance with Article 715".

The European Communities indicated that Article 715 of NAFTA laid down a number of elements each Party should take into account in conducting risk assessment, including, inter alia, "relevant scientific evidence" (Art. 715(1)(b)). However, paragraph 3 of Article 715 made it clear that, in establishing its appropriate level of protection, each Party should only aim at minimizing negative trade effects and should avoid arbitrary or unjustifiable discrimination or disguised restriction on trade. The choice of the level and the measure to achieve the protection chosen, therefore, was not based on scientific grounds. For example, the former Canadian Minister for international trade, the honourable R. MacLaren had stated that :

⁶² "The agreements [i.e. the SPS and NAFTA] make clear that governments remain free to pursue legitimate regulatory objectives, such as consumer safety and health protection. Every government may establish the levels of protection that it considers appropriate. In other words, nothing in either the North American Free trade Agreement or the World Trade Organization Agreement constraints a government from determining the degree of tolerance or protection it wishes." (Statement made during as address at Cambridge University,United Kingdom, 17 July 1995).

⁶³ The European Communities argued that it followed that the evaluation the European Communities had made in 1993/94 of oestradiol-17b for therapeutic or zootechnical purposes was totally irrelevant to this dispute, because it was not made for, nor did it affect, the risk assessment the European Communities had made for the five of these hormones when used as growth promoters.

⁶⁴ "Application of Risk Analysis to Food Standards Issues: Report of the Joint FAO/WHO Expert Consultation, Geneva, Switzerland, 13-17 March 1995" (Geneva: WHO, 1995) p.6 [hereinafter: "Application of Risk Analysis to Food Standards Issues: Report of the FAO/WHO Expert Consultation"]. Canada submitted that this was the most recent comprehensive international report on risk analysis, but stressed that Codex definitions for risk analysis continued to evolve. In June 1996, Codex had issued revised definitions for interim use. These definitions were found in: Codex Alimentarius Commission, "Terms and Definitions Used in Risk Analysis," Joint FAO/WHO Food Standards Programme, June 1996, p.2 [hereinafter "Terms and Definitions used in Risk Analysis"].

⁶⁵ Ibid., "Terms and Definitions used in Risk Analysis".

⁶⁶ "Application of Risk Analysis to Food Standards Issues: Report of the FAO/WHO Expert Consultation", supra note 64.

⁶⁷ "Terms and Definitions used in Risk Analysis", supra note 57.

⁶⁸ "Application of Risk Analysis to Food Standards Issues: Report of the FAO/WHO Expert Consultation", supra note 64.

⁶⁹ Dose response assessment was defined as, "...the determination of the relationship between the magnitude of exposure (dose) to a chemical, physical or biological agent and the severity and/or frequency of associated adverse health effects": "Terms and Definitions used in Risk Analysis", supra, note 57.

⁷⁰ "Application of Risk Analysis to Food Standards Issues: Report of the FAO/WHO Expert Consultation", supra, note 64.

⁷¹ Canada explained that non-genotoxic carcinogens were referred to as "promoters"; that is, they did not cause cancer, but rather they could act as promoters in cells that had already been damaged. Non-genotoxic carcinogens were not capable of producing mutations. In contrast, genotoxic carcinogens were "initiators", and could cause mutations of DNA resulting in tumours in humans or animals.

⁷² "Application of Risk Analysis to Food Standards Issues: Report of the FAO/WHO Expert Consultation", supra, note 64.

⁷³ "Terms and Definitions used in Risk Analysis," supra, note 64.

⁷⁴ Ibid.

⁷⁵ Codex Alimentarius Commission, "Risk Analysis: Definitions, Procedures and Principles, Joint FAO/WHO Food Standards Programme, Codex Committee on General Principles - Twelfth Session", Paris, France, 25-28 November 1996 p.18 (hereinafter "Risk Analysis: Definitions, Procedures and Principles").

⁷⁶ "Application of Risk Analysis to Food Standards Issues: Report of the FAO/WHO Expert Consultation", supra, note 64.

⁷⁷ "Risk Analysis: Definitions, Procedures and Principles".

⁷⁸ Case C-84/94, United Kingdom v. Council, judgment of 12 November 1996 (1996 ECR I-5800, para. 15).