RESTRICTED

WORLD TRADE WT/DS2/R

29 January 1996

ORGANIZATION


(96-0326)

image


United States - Standards for Reformulated and Conventional Gasoline


Report of the Panel

Table of Contents

  1. INTRODUCTION 1

  2. FACTUAL ASPECTS 2

    1. The Clean Air Act 3

    2. EPA's Gasoline Rule 3

      1. Establishment of Baselines 3

      2. Reformulated Gasoline 4

      3. Conventional Gasoline (or "Anti-Dumping Rules") 4

    3. The May 1994 Proposal 5

  3. MAIN ARGUMENTS 5

    1. General 5

    2. The General Agreement on Tariffs and Trade 6

      1. Article I - General Most-Favoured-Nation Treatment 6

      2. Article III - National Treatment on Internal Taxation

        and Regulation 7

        1. Article III:4 7

        2. Article III:1 14

      3. Article XX - General Exceptions 15

      4. Article XX(b) 15

        1. "Protection of Human, Animal and Plant Life

          or Health" 15

        2. "Necessary" 15

      5. Article XX(d) 21

      6. Article XX(g) 22

        1. "Related to the conservation of exhaustible natural resources..." 22

        2. "... made effective in conjunction with restrictions

          on domestic production or consumption" 23

      7. Preamble to Article XX 23

      8. Article XXIII - Nullification and Impairment 25

        ii

    3. Agreement on Technical Barriers to Trade 25

      1. Article 2 - Preparation, Adoption and Application of Technical Regulations by Central Government Bodies 25

        1. Whether or Not the Gasoline Rule is a Technical

          Regulation 25

        2. Article 2.1 27

        3. Article 2.2 27

      2. Article 12 - Special and Differential Treatment of Developing Country Members 28

  4. SUBMISSIONS BY INTERESTED THIRD PARTIES 28

    1. The European Communities 28

    2. Norway 31

  5. INTERIM REVIEW 31

  6. FINDINGS 32

    1. Introduction 32

    2. Article III 33

      1. Article III:4 33

      2. Article III:1 37

    3. Article I:1 37

    4. Article XX(b) 38

      1. Policy goal of protecting human, animal or plant life or health 38

      2. Necessity of the inconsistent measure 39

    5. Article XX(d) 42

      1. Securing compliance with consistent laws or regulations 43

      2. Other conditions 43

    6. Article XX(g) 43

      1. Policy goal of conserving an exhaustible natural resource 44

        iii

      2. Measure "related to" the conservation of an exhaustible natural resource; and made effective "in conjunction" with restrictions

        on domestic production or consumption 45

    7. Article XXIII:1(b) 46

    8. Applicability of the Agreement on Technical Barriers to Trade 46

  7. CONCLUDING REMARKS 46

  8. CONCLUSIONS 47

  1. INTRODUCTION


      1. On 23 January 1995, the United States received a request from Venezuela to hold consultations under Article XXII:1 of the General Agreement on Tariffs and Trade 1994 ("General Agreement"), Article 14.1 of the Agreement on Technical Barriers to Trade ("TBT Agreement") and Article 4 of the Understanding on Rules and Procedures Governing the Settlement of Disputes ("DSU"), on the rule issued by the Environmental Protection Agency on 15 December 1993, entitled "Regulation of Fuels and Fuel Additives - Standards for Reformulated and Conventional Gasoline" (WT/DS2/1). The consultations between Venezuela and the United States took place on 24 February 1995. As they did not result in a satisfactory solution of the matter, Venezuela, in a communication dated 25 March 1995, requested the Dispute Settlement Body ("DSB") to establish a panel to examine the matter under Article XXIII:2 of the General Agreement and Article 6 of the DSU (WT/DS2/2). On 10 April 1995, the DSB established a panel in accordance with the request made by Venezuela. On 28 April 1995, the parties to the dispute agreed that the Panel should have standard terms of reference (DSU, Art. 7) and agreed on the composition of the Panel as follows:


        Chairman: Mr. Joseph Wong Members: Mr. Crawford Falconer

        Mr. Kim Luotonen


      2. On 10 April 1995, Brazil requested the United States to hold consultations under

        Article XXII:1 of the General Agreement, Article 14.1 of the TBT Agreement and Article 4 of the DSU on the rule issued by the Environmental Protection Agency on 15 December 1993 entitled "Regulation on Fuels and Fuel Additives - Standards for Reformulated and Conventional Gasoline" (WT/DS4/1). Consultations between Brazil and the United States were held on

        1 May 1995 without resulting in a satisfactory solution of the matter. In a communication dated 19 May 1995, Brazil requested the DSB to establish a panel to examine the matter pursuant to Article XXIII of the General Agreement, Article 14 of the Agreement on Technical Barriers to Trade and Article 6 of the DSU. On 31 May 1995, the DSB established a Panel in accordance with the request made by Brazil.


      3. On 31 May 1995, pursuant to Article 9 of the DSU in respect of multiple complainants , the DSB decided, with the agreement of all the parties, that for practical reasons this matter be examined by the Panel already established at the request of Venezuela on 10 April 1995. The date of the constitution of the Panel, namely 28 April 1995, remained unchanged.


      4. Due to the additional task given to the Panel, the DSB agreed upon, at the same meeting, the following terms of reference:


        "To examine, in the light of the relevant provisions of the covered agreements cited by Venezuela in document WT/DS2/2 and by Brazil in document WT/DS4/2, the matters referred to the DSB by Venezuela and Brazil in those documents and to make such findings as will assist the DSB in making the recommendations or in giving the rulings provided for in those agreements".


      5. The Chairman of the DSB recalled Article 9.2 of the DSU which provides that "the rights which the parties to the dispute would have enjoyed had separate panels examined the complaints are in no way impaired".


      6. Australia, Canada, the European Communities and Norway reserved their rights to participate in the Panel proceedings as third parties. Only the European Communities and Norway presented arguments to the Panel.

        WT/DS2/R

        Page 2


      7. The Panel met with the parties to the dispute from 10 to 12 July 1995 and from 13 to 15 September 1995. It met with the interested third parties on 11 July 1995.


      8. On 21 September 1995, the Chairman of the Panel informed the DSB that the Panel would not be able to issue its report within six months. The reasons for that delay are stated in document WT/DS2/5.


    1..9 The Panel issued its interim report to the parties on 11 December 1995. Following a request made by the United States pursuant to Article 15.2 of the DSU, the Panel held a further meeting with the parties on 3 January 1996.


    1.10 The Panel issued its final report to the parties to the dispute on 17 January 1996.


  2. FACTUAL ASPECTS


    1. The Clean Air Act


        1. The Clean Air Act ("CAA"), originally enacted in 1963, aims at preventing and controlling air pollution in the United States. In a 1990 amendment to the CAA1, Congress directed the Environmental Protection Agency ("EPA") to promulgate new regulations on the composition and emissions effects of gasoline in order to improve air quality in the most polluted areas of the country by reducing vehicle emissions of toxic air pollutants and ozone-forming volatile organic compounds. These new regulations apply to US refiners, blenders and importers.


        2. Section 211(k) of the CAA divides the market for sale of gasoline in the United States into two parts. The first part, which covers approximately 30 percent of gasoline marketed in the United States, consists of the nine large metropolitan areas that experienced the worst summertime ozone pollution during the period 1987-1989, plus any areas that do not meet national ozone requirements and are added at the request of the governor of the state. These areas are referred to as ozone "nonattainment areas", and in this part of the United States only "reformulated gasoline" may be sold to consumers. In the rest of the United States, "conventional gasoline" may be sold to consumers.


        3. Section 211(k)(2)-(3) of the CAA established certain compositional and performance specifications for reformulated gasoline. The oxygen content must not be less than 2.0 percent by weight, the benzene content must not exceed 1.0 percent by volume and the gasoline must be free of heavy metals, including lead or manganese. The performance specifications of the CAA require a 15 percent reduction in the emissions of both volatile organic compounds ("VOCs") and toxic air pollutants ("toxics") and no increase in emissions of nitrogen oxides ("NOx"). These requirements are measured by comparing the performance of reformulated gasoline in baseline vehicles (representative model year 1990 vehicles) against the performance of "baseline gasoline" in such vehicles. Section 211(k)(10) of the CAA defines the specifications of baseline gasoline sold in the summer, which is the high ozone season, and leaves the specifications of winter baseline gasoline to be determined by EPA. It provides, however, that the specifications for winter gasoline shall be those of the industry average gasoline sold in 1990. For the year 2000 and beyond, the CAA requires that new reformulated gasoline requirements be developed that require a 20-25 percent reduction in emissions of VOCs and toxics, depending on EPA' s considerations of feasibility and cost.


          image


          142 U.S.C. §7545(k).

          WT/DS2/R

          Page 3


        4. The CAA also sets requirements for conventional gasoline, which ensure that each refiner' s, blender' s or importer' s conventional gasoline sold in the rest of the country remains as

          clean as it was in 1990. This programme is known as "anti-dumping rules" because it is designed to prevent refiners, blenders or importers from dumping into conventional gasoline fuel components that are restricted in reformulated gasoline and that cause environmentally harmful emissions . To accomplish this, section 211(k)(8) of the CAA provides that no refiner, blender or importer of gasoline may sell conventional gasoline that emits VOCs, toxics, NOx or carbon monoxide ("pollutants") in greater amounts than the gasoline sold in the United States by that refiner, blender or importer in 1990. In order to implement this provision, separate individual baselines must be established for refiners, blenders or importers based on the gasoline they sold in 1990. That permits determination of whether the emissions from a refiner' s, blender' s and importer' s conventional gasoline (post-1994 gasoline) are greater than the emissions from its 1990 gasoline. If, however, EPA determines that no adequate and reliable data exist regarding the composition of such 1990 gasoline sold by a refiner, blender or importer, the statutory baseline gasoline is applied. The statutory annual baseline values are calculated using a seasonal weighting of the statutory summer baseline, as defined in the CAA, and the statutory winter baseline, as determined by EPA.


    2. EPA's Gasoline Rule


      1. Establishment of Baselines


          1. The CAA directed EPA to determine the quality of 1990 gasoline, to which reformulated and conventional gasoline would be compared in the future: these determinations are known as "baselines". EPA set historic baselines for individual entities, and established a statutory baseline, intended to reflect average US 1990 gasoline quality, which would be used instead of the historic individual baselines for those entities who were determined to be lacking adequate and reliable data regarding the quality of the gasoline they produced in 1990.


          2. EPA' s final rule2 ("Gasoline Rule") requires any domestic refiner, which was in operation for at least 6 months in 1990, to establish an individual refinery baseline, which represents the quality of gasoline produced by that refiner in 1990. The rule establishes three methods for the purpose of determining a domestic refiner' s individual historic baseline. Under Method 1, the refiner must use the quality data and volume records of its 1990 gasoline. However, as acknowledged by EPA at the time, it was not anticipated that many domestic refiners would have all the data necessary to establish an individual baseline based entirely on actual 1990 data. If Method 1 type data are not available, a domestic refiner must use its 1990 gasoline blendstock quality data and 1990 blendstock production records (Method 2). In the event that neither one of these two methods is available, a domestic refiner must turn to Method 3 type data which consist of its post-1990 gasoline blendstock and/or gasoline quality data modeled in light of refinery changes to show 1990 gasoline composition . Domestic refiners are not permitted to choose the statutory baseline.


          3. An importer which is also a foreign refiner must determine its individual baseline using Methods 1, 2 and 3 if it imported at least 75 percent, by volume, of the gasoline produced at its foreign refinery in 1990 into the United States in 1990 (the so-called "75 % rule")3.



            image


            240 CFR 80, 59-Fed. Reg. 7716 (16 February 1994).

            340 CFR 80.91(b)(ii).

            WT/DS2/R

            Page 4


          4. Certain entities are, however, automatically assigned to the statutory baseline. Firstly, refineries which began operation after 1990 or were in operation for less than 6 months in 1990 are required to use the statutory baseline. Secondly, importers and blenders are assigned the statutory baseline unless they can establish their individual baseline following Method 1. If actual 1990 data are not available, which is, as for domestic refiners, anticipated by EPA, importers and blenders are assigned to the statutory baseline. EPA considers that blenders which produce gasoline by combining gasoline blendstocks purchased from many sources cannot determine with accuracy the quality of their 1990 gasoline using Methods 2 and 3. Similarly, EPA considers that importers cannot use Methods 2 and 3, because these methods inherently apply only to refineries and because of the extreme difficulty in establishing the consistency of their gasoline quality over time.


      2. Reformulated Gasoline


          1. Regarding the implementation of the regulations for reformulated gasoline, EPA proposes a two-step approach. From 1 January 1995 to 1 January 1998, EPA enforces an interim programme called the "Simple Model". Under this programme, reformulated gasoline sold in the United States by domestic refiners will be subject to requirements established with reference to the individual baseline for certain gasoline qualities and requirements specified in the Gasoline Rule for other gasoline qualities. More specifically, the parameters sulphur, olefins and T-90 are measured against each US refiner' s individual 1990 baseline and must be maintained at or below these 1990 levels (these are called "non-degradation requirements"). The requirements regarding four other gasoline qualities (Reid Vapour Pressure, oxygen, benzene and toxics performance) are specified by EPA in the Gasoline Rule4. Importers of foreign gasoline also have to comply with the requirements set out in the final rule regarding Reid Vapour Pressure, oxygen, benzene and toxics performance. However, importers cannot use individual 1990 baseline for sulphur, olefins and T-90, but have to comply with levels specified in the statutory baseline for these parameters. Under the Simple Model, requirements for sulphur, olefins and T-90 must be met on an annual average basis. EPA adopted the individual baseline approach for these parameters in the Simple Model because at the time it was formulating its regulation, it considered that the available data regarding sulphur, olefins and T-90 did not permit an assessment of the precise effects of these components on the emissions level of gasoline. Given this uncertainty, EPA did not want to require refiners immediately to make refinery changes which might later prove to be unnecessary, given the greater flexibility provided by the Complex Model.


          2. As of 1 January 1998, EPA will enforce the "Complex Model", which will apply the same emissions reduction requirements to all producers of reformulated gasoline. The individual baselines for sulphur, olefins and T-90 will no longer apply.


      3. Conventional Gasoline (or "Anti-Dumping Rules")


          1. The 1990 Amendment to the CAA requires that, as of 1 January 1995, each refiner' s, blender' s or importer' s conventional gasoline sold in the United States be no more polluting than the gasoline sold by that refiner, blender or importer in 19905. EPA requires domestic refiners to measure non-degradation requirements for conventional gasoline against their individual baselines while importers of foreign gasoline are assigned to the statutory baseline. However, in this programme, the non-degradation requirements apply to all conventional gasoline requirements, and not only to sulphur, olefins and T-90. Requirements must be met on an average annual basis.


            image


            440 CFR 80.41.

            542 U.S.C. § 7545 (CAA 211(k)(8)).

            WT/DS2/R

            Page 5


            The Gasoline Rule limits ("caps") the volume of conventional gasoline that is subject to an individual baseline to the volume of gasoline produced in 1990 by that entity; all conventional gasoline produced in excess of the specific volume cap is measured against the statutory baseline.


          2. Domestic refiners and importers of conventional gasoline, unlike those of reformulated gasoline, will still be subject to different baselines after the entry into force of the Complex Model in 1998.


    3. The May 1994 Proposal


        1. In view of the comments made by interested parties during the rulemaking process of the final Gasoline Rule, EPA proposed, in May 1994, to amend the reformulated gasoline regulation in order to define criteria and procedures by which foreign refiners could establish individual refinery baselines in a manner similar to that required for domestic refiners6. Pursuant to this proposal, foreign refiners would be allowed to establish an individual baseline using Methods 1, 2 or 3. If the individual baseline was approved by EPA, importers could use it for the purpose of certifying the portion of reformulated gasoline imported from that particular refinery into the United States. However, the use of individual foreign refinery baselines would be subject to various additional strict requirements, aiming at ensuring the accuracy and respect of the foreign refinery' s individual baseline with respect to gasoline shipped to the United States and verifying the refinery of origin. Furthermore, it would not apply to conventional gasoline. After a public comment period, the US Congress enacted legislation in September 1994 denying funding to EPA for implementation of the May 1994 Proposal.


  3. MAIN ARGUMENTS


    1. General


        1. Venezuela and Brazil requested the Panel to find that the final rule promulgated by the United States' Environmental Protection Agency ("EPA") on 15 December 1993 and entitled "Fuels and Fuel Additives - Standards for Reformulated and Conventional Gasoline" ("Gasoline Rule") was:


          1. contrary to Articles I and III of GATT 1994;

          2. not covered by any of the exceptions under Article XX of GATT 1994;

          3. contrary to Article 2 of the Agreement on Technical Barriers to Trade.


        2. Venezuela additionally requested the Panel to find that the Gasoline Rule nullified and impaired benefits accruing to Venezuela under the General Agreement within the meaning of Article XXIII:1(b).


        3. Accordingly, Venezuela and Brazil asked the Panel to recommend that the United States take all necessary steps to bring the Gasoline Rule into conformity with its obligations under the General Agreement and the TBT Agreement. Venezuela requested the Panel to recommend that the United States amend the Gasoline Rule to provide treatment for gasoline imports no less favourable than that accorded to US produced gasoline.


        4. The United States requested the Panel to find that the Gasoline Rule was:


          image


          640 CFR 80 (59 Fed. Reg. 22800, 3 May 1994).

          WT/DS2/R

          Page 6


          1. consistent with Articles I and III of the General Agreement 1994;

          2. falling within the scope of Article XX (b), (d), and (g) of GATT 1994;

          3. consistent with the Agreement on Technical Barriers to Trade.


    2. The General Agreement on Tariffs and Trade


      1. Article I - General Most-Favoured-Nation Treatment


          1. Venezuela and Brazil argued that the rule allowing an importer which was also a foreign refiner to establish its individual baseline, provided that it imported into the United States at least 75 percent of the gasoline produced at that refinery in 1990 ("75 % rule"), granted an advantage to gasoline exported from certain third countries in violation of Article I of the General Agreement.


          2. Venezuela argued that the 75 % rule applied only to a fixed, finite and easily determinable group of countries, determined only by historical facts. Hence, no importer or foreign refiner could take any action that would alter its ability to benefit from this rule. According to information submitted to Venezuela by the United States, refineries based in Canada only were likely to meet the criteria. A previous panel report had found that the European Community' s meat quality regulations requiring certification of imported beef by the US Department of Agriculture were inconsistent with Article I of the General Agreement because "[E]xports of like products of other origin than that of the United States were in effect denied access to the EEC market considering that the only certifying agency authorized to certify the meat ... was a United States agency mandated to certify only meat from the United States"7. As a matter of fact, only certain countries could benefit from this provision. This was, accordingly, a breach of Article I.


          3. Brazil submitted that the two criteria contained in the 75 % rule, i.e. the owner relationship between the importer and the foreign refiner and the percentage of gasoline it imported into the United States, were not neutral but had been chosen so as to suit a particular category of countries and therefore constituted an "advantage" within the meaning of Article I. These criteria had no link with the characteristics of gasoline as a product. Thus, the 75 % rule applied a different and more favourable standard to imports from some foreign refineries than it applied to imports from refineries in other countries.


          4. The United States replied that the 75 % rule did not provide an "advantage" to the products of any particular country. The 75 % rule would have applied to any importer that could meet its two objective criteria, regardless of the country of origin of the gasoline. The

            United States specified that the 75 % requirement represented the minimally acceptable value to ensure that the importer' s individual baseline determined under Methods 1, 2 and 3 would be accurate. (In fact, most foreign refiners were unlikely to export more than 30 % of their gasoline). Secondly, the criterion requiring that the importer and foreign refiner be the same entity eliminated the enforcement concerns arising in relation with the establishment of an individual baseline for a foreign refiner. The United States considered that Venezuela' s invocation of the panel report "European Economic Community - Imports of Beef From Canada" was inapposite because, in that dispute, the EC regulation at issue expressly listed the US Department of Agriculture (USDA) as the only certifying authority for the meat in question, and USDA was only authorized to certify US meat. In that case, the certifying process itself guaranteed that only US beef would be certified, thereby expressly favouring US beef over that of all other countries. In contrast, the 75 % rule expressly requires any importer that meets its objective criteria to establish


            image

            7"European Economic Community - Imports of Beef From Canada", BISD 28S/92 (adopted on 10 March 1981), paragraph 4.2(a).

            WT/DS2/R

            Page 7


            an individual baseline for its gasoline. The United States also noted that the regulatory deadline for individual baseline applications under the 75 % rule had elapsed without any company meeting the criteria. The 75 % rule had no application and could therefore not be inconsistent with any provisions of the General Agreement.


          5. Venezuela considered that the United States interpreted too narrowly the panel report "EEC - Imports of Beef From Canada" when saying that the favoured country must be expressly identified in order for the regulation to violate Article I. A rule violated Article I when it stipulated, like the 75 % rule, that the products of only some countries could qualify.


          6. Venezuela and Brazil considered that the fact that the 75 % rule had no application should not prevent the Panel from ruling on it. Venezuela considered that the mere existence of such a regulation might have inhibiting effects on commercial and investment decisions. Thus, the possibility of its future application was sufficient to establish an Article I violation. Brazil added that a clear ruling on the 75 % rule was necessary because it would dissuade countries from designing future standards that, being neutral at first sight, were in fact designed to fit only the precise situation of their own multinationals, thus threatening the integrity of Article I.


      2. Article III - National Treatment on Internal Taxation and Regulation


        1. Article III:4


            1. Venezuela and Brazil stressed that they were not questioning the right of the United States to enact stringent environmental standards and regulations in order to improve air quality within the US territory provided these standards and regulations treated imported products no less favourably than domestic like products.


            2. Venezuela and Brazil argued that the Gasoline Rule, by denying foreign refiners the possibility to establish an individual baseline, violated Article III:4 because it accorded less favourable treatment to imported gasoline, both reformulated and conventional , than to US gasoline. The Gasoline Rule required imported gasoline to conform with the more stringent statutory baseline when US gasoline had to comply only with a US refiner' s individual baseline. Practically, this meant that imported gasoline with certain parameter levels above the statutory baseline could not be directly sold in the US market whereas gasoline with these same qualities produced in a US refinery could be freely sold on the US market provided that it conformed with that refiner' s individual baseline. In order to accommodate this situation, foreign refiners had two options: (i) make expensive investments and changes to their refineries in order to produce gasoline conforming to the more stringent statutory baseline, or (ii) supply at a lower price gasoline to an importer that could average that gasoline with other gasolines (if such other gasolines exist in sufficient amount) in order to meet, over an annual period, the requirements of the statutory baseline. Both options adversely affected the conditions of competition for imported gasoline and afforded protection to domestic production in a manner contrary to Article III. Furthermore, these adverse competitive effects were precisely what EPA intended to avoid for US refiners by granting them individual baselines. Brazil added that it was up to the United States to demonstrate that its discriminatory system did not treat imports less favourably.


            3. Venezuela and Brazil held that officials from the US government had acknowledged on various occasions that the Gasoline Rule discriminated against imported gasoline and accorded more favourable treatment to domestically produced gasoline. Venezuela added that another US government official had publicly stated that such discrimination was intentionally endorsed as a means of affording protection to US gasoline. These statements showed that the Gasoline Rule discriminated both in effect and in intent against foreign refiners. Venezuela and Brazil further

              WT/DS2/R

              Page 8


              argued that EPA' s 1994 proposed amendments to the Gasoline Rule ("1994 Proposal") acknowledged that the discriminatory treatment of imported gasoline was inconsistent with the United States' obligations under the General Agreement. Venezuela and Brazil argued that the 1994 Proposal would have partly eliminated the discrimination by providing for the establishment of individual baselines by foreign refiners of reformulated gasoline; however, the discriminatory treatment of conventional gasoline would have continued.


            4. Venezuela noted that "Petroleos de Venezuela, S.A." ("PDVSA") had already made costly adjustments to its production in order to meet the statutory baseline requirements and had accelerated its programme of investments with a view to complying with the Complex Model requirements. These adjustments had reduced the volume and value of Venezuela' s current and anticipated gasoline exports to the United States below the levels that would have prevailed if PDVSA were allowed to establish its individual baseline. These adjustments interfered with PDVSA' s investment programme, obliging it to focus on production for the US gasoline market and adversely affecting other important investment projects.


            5. Brazil stated in addition that application of the statutory baseline to foreign refiners and domestic importers was discriminatory in several respects. First, the flexibility given to domestic refiners in establishing individual baselines had the effect that many of them were allowed emissions levels higher than those permitted by the statutory baseline. Secondly, the statutory baseline was more stringent than the average of the individual baselines for refineries located in the Eastern and Gulf Coast states (where virtually all Brazilian gasoline was sold) because of the inclusion in the national average of the strict 1990 Californian standards. The Gasoline Rule also favoured imports by domestic refiners over imports by importers who were not domestic refiners. Domestic refiners whose current production was "cleaner" than their individual baseline could import gasoline with parameter levels above the statutory baseline, could blend it with their own cleaner production and sell it on the US market as long as the mixture conformed with their individual baseline. Importers who were not domestic refiners had to conform to the statutory baseline in all instances. Thus, the Gasoline Rule affected the distribution of gasoline in the United States by channelling imports to domestic refiners who had an incentive to take advantage of their privileged position by demanding lower prices from foreign refiners.


            6. Brazil stated that the same gasoline that it used to export to the United States market as "finished" gasoline was, since the entry into force of the Gasoline Rule, considered only as "blendstock8", which was sold at a lower price. Thus, Brazil had not been able to export "finished" conventional gasoline to the US market since 1 January 1995. Brazilian refiners were not currently producing reformulated gasoline .


            7. The United States replied that the Gasoline Rule did not treat imported gasoline less favourably than domestic gasoline overall. The environmental goal of the Gasoline Rule was to regulate the overall quality of the gasoline sold in the United States. Each importer had to satisfy on average the statutory baseline, which approximated average gasoline quality consumed in the US in 1990, and domestic refiners had to satisfy on average their 1990 individual baselines, which overall roughly represented 1990 US gasoline quality. Hence, overall domestically produced gasoline had to be at least as clean as foreign gasoline since roughly half of domestic gasoline would be "cleaner" and roughly half would be dirtier than gasoline using the statutory baseline. The United States supplied the Panel with data documenting the number of domestic refiners with baseline values both above and below the statutory baseline for specific parameters and emissions levels upon which compliance with the non-degradation requirements was based. This analysis


              image


              8Blendstock is unfinished gasoline which has to be blended in order to be sold as finished gasoline.

              WT/DS2/R

              Page 9


              showed that five domestic refineries had individual baselines that were below the annual statutory baseline for all fuel parameters and emissions levels, and three domestic refiners had individual baselines that were above the annual statutory baseline for all fuel parameters and emissions levels. Thus, most refiners had individual baselines with several parameters above the corresponding statutory values and several below. The United States considered that a previous panel report had recognized that "there may be cases where application of formally identical legal provisions would in practice accord less favourable treatment to imported products and a contracting party might thus have to apply different legal provisions to imported products to ensure that the treatment accorded them is in fact no less favourable"9. Since the majority of importers did not have the necessary data to use Methods 1, 2, or 3, they would be precluded from supplying the US market as they would be unable to establish an individual baseline. In fact, the Gasoline Rule granted more favourable treatment to imports since identical treatment would have in practice excluded imported gasoline from the US market.


            8. The United States argued that the Gasoline Rule applied to imported gasoline and not to foreign refiners producing gasoline. Moreover, foreign refiners were not required to produce gasoline that met any baseline at all, but could produce gasoline which was cleaner or dirtier than the statutory baseline. The baseline establishment rule focused on the importer of foreign gasoline because the United States was not attempting to regulate the conduct of foreign companies or those of other overseas entities; the importer was the first entity, within US territory, that had control over the quality of gasoline imported into the United States. Thus, foreign refiners were subject only to the independent purchasing decisions of US importers who had to balance the products of one or more foreign refiners with that of another in order to comply with the statutory baseline. The fact that no single batch of gasoline would be deemed as non-complying provided

              additional flexibility to both importers and foreign refiners. Moreover, the complainants ' focus on equal treatment of individual foreign refiners was misplaced since the General Agreement applied to the imported product and not to the producer.


            9. The United States argued that gasoline from importers was treated similarly to gasoline from similarly situated domestic parties. For instance, imported gasoline was treated identically to gasoline produced by domestic refiners with limited 1990 operations or to gasoline produced by US blenders whose business entailed a lack of consistency of sources and quality of the gasoline produced. These producers had in common with the importers the inability to establish an accurate individual baseline because their business characteristics or history were such that they could not determine the quality of their gasoline as required by Methods 2 and 3. Although theoretically, the 1990 gasoline quality of an importer might be established by first determining the refineries of origin for all of the gasoline imported by that importer in 1990, and then obtaining accurate and verifiable information on the quality of that subset of 1990 gasoline produced at the refinery and exported to the particular US importer, the United States expected that only a very limited number of importers would be able to establish an individual baseline using such a procedure. In addition, there were significant problems associated with establishing the 1990 gasoline quality of foreign refiners: tracking the refinery of origin of the imported gasoline, establishing the quality of the small subset of gasoline shipped to the United States and lack of adequate enforcement capacity. According to the United States, these factors made it very difficult to verify the accuracy or reliability of claims regarding a foreign refiner' s 1990 gasoline quality for that purpose. On the other hand, gasoline produced by domestic refiners was made from crude oil whose quality could easily be documented, as were the characteristics of the physical plant and operational procedures. Thus, the quality of the gasoline produced at such a domestic refinery could be accurately assessed. The United States further argued that imported


              image


              9"United States - Section 337 of the Tariff Act of 1930", BISD 36S/345 (adopted on 7 November 1989).

              WT/DS2/R

              Page 10


              gasoline and domestically produced gasoline were in the same position with regard to the flexibility for complying with their respective baselines. As various qualities of gasoline were available in the market, some above and some below the statutory baseline, an importer had complete flexibility to select gasoline from different sources and mix them in order to reach the annual average quality required by the statutory baseline. By contrast, a domestic refiner was constrained by its refinery equipment and crude oil supplies.


            10. The United States considered that Venezuela was incorrect in its claim that the US government official' s statement demonstrated that the Gasoline Rule had a protectionist purpose. The statement in question actually reflected the US government' s commitment when it issued the final Gasoline Rule to continue addressing the issue of how imports were treated, and the US government' s ongoing concern that environmental protection not be compromised. This statement also showed that the government official' s objective was adoption of the best environmental provision that was fair to all parties concerned. The United States equally rejected Venezuela and Brazil' s argument that the 1994 Proposal implied recognition that the Gasoline Rule discriminated against imports. The 1994 Proposal was a continuation of EPA' s prior attempts to develop criteria that would protect the environment, minimize disruption to producers, and treat similarly situated parties alike. The fact that EPA was willing to make this attempt was neither a determination that the Gasoline Rule was flawed nor a determination that the 1994 Proposal was feasible. This proposal received largely negative public comment and was rejected in the end, including by Venezuela. PDVSA itself had objected that the proposed conditions related to gasoline tracking were basically unworkable.


            11. The United States disagreed with Venezuela' s claim about decreased imports. Data from the US Energy Information Administration showed that current import volumes had not significantly decreased from historical levels. Furthermore, import volume and domestic production fluctuated greatly depending on market conditions , irrespective of US regulatory action. More specifically, Venezuela' s share of the US import market rose from 11.5 to

          18.5 percent in the first five months of 1995 compared to the same period in 1994. Regarding the claim by Venezuela that the Gasoline Rule had obliged its refineries to make burdensome investments, the United States noted that it was impossible to judge to what extent PDVSA' s investments were made in reaction to regulations, such as the Gasoline Rule, in force in any particular export market. If, however, PDVSA' s investments were related to the Gasoline Rule, they were more likely due to the need to comply with the Complex Model starting in 1998 than to the 1995-1998 non-degradation requirements' programme. With respect to compliance with the Simple Model requirements, the United States considered that PDVSA could upgrade the portion of its gasoline output sent to the United States by simply adding to it additives such as oxygenates. In addition, the amount of complying gasoline that could be produced, or the cost of producing it, was highly dependent on the fraction of gasoline made to meet a particular specification. While a domestic refiner had to produce all its gasoline output within certain limits, foreign refiners typically produced only a small fraction of their gasoline output for the US market. Thus foreign refiners also had the flexibility to select their cleanest blendstocks for the US market. Such approaches did not require refinery modifications. There was no reason that Brazil could not adjust in the same manner to the Simple Model requirements. The United States also noted that reported internal turmoil in Brazil' s refining sector, a month-long strike beginning in November 1994, and apparently recurring labour problems this year suggested that Brazil' s decreased exports to the United States could hardly be attributed to US regulations.


            1. Venezuela agreed with the United States that Article III applied to imported gasoline and not to the foreign refiners. The Gasoline Rule discriminated against imported gasoline by applying the statutory baseline to such gasoline while applying individual baselines to US gasoline.

              WT/DS2/R

              Page 11


              The United States wrongly introduced the concept of "similarly situated parties" as a basis for arguing that imported gasoline and US produced gasoline were not "like product". Imported and domestic gasolines had the same tariff classification , served the same end use and the same end users and were indistinguishable from the commercial standpoint; thus, all gasoline was a like product. The concept "similarly situated parties" was new to GATT and lacked a legal basis.

              Moreover, Venezuela considered that these parties were not "similarly situated". Importers, who obtained finished gasoline for distribution to other wholesalers or retailers were not "similarly situated" to blenders, who produced gasoline by mixing gasoline components produced by others. It was more appropriate to compare them to "jobbers" who obtained finished gasoline for distribution to other wholesalers or retailers and who used the individual baselines associated with the gasoline they acquired. Foreign refiners were "similarly situated parties" with respect to US refiners in that the reasons given by the United States as to why US refiners can establish their own baselines apply equally to foreign refiners.


            2. Venezuela argued that the United States did not deny the existence of differential treatment for imported gasoline. Thus, it had to assume the burden of proving that such treatment was no less favourable to the imported product. According to past panel reports, the test was not whether the rules were different but whether such differences accorded no less favourable treatment to imported products. Venezuela considered that such a demonstration had not been made.

              Venezuela disagreed with the interpretation given by the United States to the panel report "United States - Section 337 of the Tariff Act of 1930" and considered that panels interpreted the words "treatment no less favourable" contained in Article III:4 as calling for effective equality of opportunities for imported products. The United States' assertion that it was "incumbent upon the importer to balance the products of one or more foreign refiners with that of another" was contrary to that established understanding of Article III:4. No such equality could exist if the very ability of a producer/exporter to introduce his product into the importer' s market depended on the subsequent decisions of the importer to buy additional product produced and exported by another person. The opportunity to import a like product could not be conditional upon the importer' s willingness to run a risk of not finding below-statutory baseline gasoline in order to

              average it with above-statuto ry baseline gasoline. Venezuela considered that the reasoning of two previous panel reports10 regarding the loss of competitive opportunities for imported products was applicable to this case and led to the conclusion that imported gasoline should have the same distribution opportunities available to US produced gasoline, including the ability to be sold directly into commerce with the application of an individual baseline.


            3. Venezuela considered that the issue at stake was not averaging, a technique which was also available to domestic refiners, but the difference between the requirements imposed on imported gasoline and the requirements imposed on US gasoline. In order to regulate the average quality of gasoline in the United States, the Gasoline Rule regulated every batch of gasoline produced in or imported into the US. The fact that the very same gasoline with identical characteristics would be treated differently under the Rule if produced by a US refiner as opposed to a foreign refiner was precisely the "less favourable treatment" prohibited under Article III:4.


            4. Venezuela rejected the United States' assertion that a foreign refiner that produced gasoline in 1990 with properties of sulphur, olefins and T-90 above the statutory baseline only had to mix "additives such as oxygenates" to upgrade its gasoline; in fact, the very composition of the foreign gasoline had to change. Venezuela denied the United States' claim that Venezuela had rejected the 1994 Proposal. Venezuela had simply explained why some means by which the EPA



              image


              10"United States - Measures Affecting Alcoholic and Malt Beverages", BISD 39S/206 (adopted on 19 June 1992), and "Canada - Import, Distribution and Sale of Certain Alcoholic Beverages", BISD 39S/27 (adopted on 18 February 1992).

              WT/DS2/R

              Page 12


              proposed to achieve certain ends were not workable from a practical standpoint, and presented alternatives to achieve the same ends in a more practical and workable manner.


            5. Brazil argued that the alleged benefit to imports deriving from the fact that "roughly half" of the domestic gasoline must be "cleaner" than imported gasoline (which, in Brazil' s view had not been demonstrated) did not overcome the less favourable treatment accorded to imports deriving from the fact that "roughly half" of the domestic gasoline was permitted to be "dirtier" than imported gasoline. This statement by the United States implicitly admitted the discrimination contained in the Gasoline Rule which required imported gasoline to be cleaner than half of the domestically produced gasoline. Brazil noted that a previous panel report had rejected any notion of balancing more favourable treatment of some imported products against less favourable treatment of other imported products11. The same rationale applied to any notion of balancing more favourable treatment vis-à-vis some domestic products against less favourable treatment vis- à-vis other domestic products. Similarly, another panel report12 had concluded that the exposure of a particular imported product to a risk of discrimination constituted , by itself, a form of discrimination. The Gasoline Rule exposed all imports to less favourable treatment than the treatment accorded to what the United States had described as "roughly half" of the domestic production. Thus, in this dispute as in previous disputes, the panel should reject the notion of "balancing". Brazil further argued that the United States had not in any event demonstrated that the average of the individual baselines was equivalent to the statutory baseline. Domestic refiners were allowed to use post 1990 (Method 3) data to establish their individual baselines whereas the statutory baseline was presumably established based on actual 1990 data. This contradicted the US argument that importers were subject to an overall average standard which was equivalent to that imposed on domestic refiners because baselines based on data from different time periods could not be considered to be equivalent.


            6. Brazil argued that United States' reference to the "Section 337" panel report was irrelevant since it had not demonstrated in this particular case that the application of formally identical legal provisions would in practice accord less favourable treatment to imported products. The United States only "believed" that granting individual baselines to foreign refiners would disadvantage them. Similarly, the United States had not demonstrated why US importers would not be able to establish an individual baseline following Method 3, based on post-1990 gasoline blendstock, since according to the United States, importers were blenders and used blendstocks. Finally, the comparison made by the United States between a foreign refiner and a US importer was inaccurate, for a foreign refiner had to be compared with a US domestic refiner.


            7. Brazil rejected the argument that imported gasoline was treated similarly to gasoline from similarly situated parties. Imported gasoline produced by foreign refiners who had unlimited operations in 1990 was treated like gasoline produced by blenders and domestic refiners who had limited operations in 1990, whereas it should have been treated like gasoline produced by comparable US domestic refiners. Brazil further noted that most US blenders who were not domestic refiners were in fact importers, and hence should not, as argued by the United States, be considered as two separate categories. Thus, discriminating between domestic refiners and domestic importers in practice resulted in discriminating against imports. Moreover, the fact that domestic refiners were allowed to import blendstock, mix it with their own production and measure the compliance of the final product against their individual baseline, whereas neither blenders nor importers were allowed to do this with respect to their own blendstock, showed that the system did not treat like products in a similar way.


              image

              11"United States - Section 337 of the Tariff Act of 1930", BISD 36S/345, (adopted on 7 November 1989).

              12"EEC - Payments and Subsidies Paid to Processors and Producers of Oilseeds and Related Animal-Feed Proteins", BISD 37S/86 (adopted on 25 January 1990).

              WT/DS2/R

              Page 13


            8. Brazil agreed with the United States that Article III applied to gasoline and not to the producer of gasoline. In this particular case, the standard applied to gasoline had been determined with reference to the producer of gasoline. Brazil was not questioning this policy choice as such, but the fact that a different and less favourable standard applied to imported products. In the case of Brazil, this discrimination was illustrated by the fact that the gasoline exported by Brazil as "finished" conventional gasoline until 1 January 1995 was now considered as a mere "blendstock" because it did not meet the statutory baseline requirements. Blendstock gasoline commanded a lower price because the buyer had to mix it with "cleaner" gasoline in order to comply with the statutory baseline. In conclusion, Brazil stated that the fact that the product produced by a Brazilian refiner could not be sold in the United States as finished gasoline while the identical product produced by a US refiner complying with its individual baseline could be sold as finished gasoline constituted precisely the less favourable, discriminatory treatment Article III was intended to prohibit.


            9. The United States argued that the complainants improperly focused on developing foreign refinery baselines, and had not demonstrated how foreign refiners could accurately ascertain the quality of the subset of their total gasoline production exported to the United States, even if they were able to establish the quality of their total 1990 output by using Methods 1, 2 or 3. Even assuming that this problem could be overcome by Venezuela and Brazil, it remained for other foreign refiners. By focusing on foreign refiners' baselines rather than on the more logical establishment of importers' baselines, the complainants appeared to seek a commercial advantage for their national oil companies over other foreign gasoline suppliers. This could favour those importers who had commercial ties to such refiners over others, and thereby distort competitive conditions among importers. The General Agreement did not require the United States to accord rights on foreign soil to foreign refiners, as opposed to WTO Members, especially if issues relating to imports could better be addressed through regulating importers who were located on US territory.


            10. The United States further argued that the references made by Venezuela to previous panel reports were irrelevant because the situation currently under examination was different. The question posed to this Panel was not whether imports and domestic products were treated identically (the United States recognized that it was not the case), but whether that treatment was less favourable under Article III. Given the fact that, for the reasons stated above, an identical treatment would have precluded most importers from marketing gasoline in the United States, the specific provisions applying to imports did not violate Article III. The United States also rejected Brazil' s claim that blenders were in fact importers and maintained that they did fall into two different categories, even if similarly situated in other respects. (Of the registered entities that were blenders or importers, over a third were solely blenders and over a third were solely importers.) The United States considered that the difficulties supposedly experienced by Venezuela and Brazil to export gasoline to the US market were groundless. Firstly, importers could easily find large amounts of gasoline with low levels of sulphur and olefin, thus "offsetting" Venezuela' s high sulphur and olefin content, because of the typical configurations of refineries outside of the United States. This was evidenced by the clean imports of reformulated gasoline this year and by the fact that 1995 exports from Venezuela did not reflect an identifiable impact of the Gasoline Rule. Secondly, Brazil' s choice not to export finished gasoline, but to export blendstock instead, was unrelated to US environmental regulations since the Gasoline Rule did not contain any requirements as to whether importers imported gasoline on the one hand, or blendstock on the other. The conventional and reformulated gasoline requirements did not aim at establishing whether a product was or was not gasoline. Rather, to be sold as "gasoline", commercial contracts generally required a product to meet certain standards established by the American Society of Testing and Materials. Regarding the alleged reduced export volumes from the complainants , the United States noted that import levels in a given market were generally

              WT/DS2/R

              Page 14


              sensitive to various factors (US demand, exporting country' s supply and demand, refinery cost structure, gasoline market conditions in other competing exporting countries, for instance), and that worldwide exports of gasoline to the United States had been following a downward trend over the last five years. It was equally difficult to know the exact role played by the Gasoline Rule with regard to the investment programme of Venezuelan refineries since any refiner operating at the world level needed a significant reformer capacity. Venezuela' s investment in such a reformer unit likely reflected overall Venezuelan market strategy. Moreover, whatever production limitations PDVSA might have with respect to one particular refiner, blending of feedstock from several refineries made higher levels of statutory-quality gasoline on a per-shipment basis possible. In addition, the US government had studied refinery cost structure shortly after passage of the Clean Air Act and found that refiners whose production of reformulated gasoline was about 30% or less of their total gasoline production could produce the reformulated gasoline at little or no incremental cost (i.e. investment costs) because of their ability to select among blendstocks.


            11. The United States argued that, with respect to the particular baseline to be employed by domestic refiners that import, the Rule' s requirement that refiners use their own baselines applied only to conventional gasoline, and up to the volume that the refiner imported in 1990. The purpose of the provision was to prevent a domestic refiner with a baseline more stringent that the statutory baseline from avoiding that stringent baseline by exporting gasoline produced at the domestic refinery and then reimporting that same gasoline under the statutory baseline, a concern raised to EPA in public comments.


            12. Venezuela argued it was not focusing on foreign refiners but on the situation of imported gasoline. The foreign refiner had become an issue only because the characteristics of

              US produced gasoline set by the Gasoline Rule was determined, in large part, by the historical quality levels of the individual US producers. Venezuela considered that it was misleading to engage, as the United States did, in a discussion of the relative situation of importers and domestic refiners when comparing the respective treatment of domestic and imported products. The focus of the analysis had to remain on gasoline as a product. The previous panel cases cited by Venezuela were relevant because the situations were similar. Venezuela disagreed with the United States' argument that importers could easily obtain suitable offsetting gasoline for Venezuela' s gasoline quality. The data presented by the United States to the Panel regarding the properties of gasoline imported in 1995 suggested exactly the contrary: by showing that the maximum values of sulphur, olefins and T-90 of imported gasoline were essentially at the statutory baseline, these data confirmed that above-statuto ry foreign gasoline was not purchased by US importers. Thus, foreign refiners were effectively obliged to comply with the statutory baseline requirements if they wanted to sell gasoline in the United States.


        2. Article III:1


            1. Venezuela and Brazil claimed that discriminatory baseline requirements contained in the Gasoline Rule violated Article III:1 for the same reasons they violated Article III:4. By distorting the conditions of competition for foreign gasoline, both reformulated and conventional , they were applied "so as to afford protection to domestic production". Venezuela also noted that

              Article III:1 was a more general provision than Article III:4. Thus, it would not insist on the panel ruling on Article III:1 if the Panel found the Gasoline Rule to be inconsistent with Article III:4.


            2. The United States replied that, for the reasons expressed under Article III:4, the Gasoline Rule did not afford protection to domestic production. Furthermore, since Article III:1 had by itself only an exhortatory character, it was not amenable to a finding of "violation" in a dispute settlement proceeding. A Panel had never found an independent violation of Article III:1.

              WT/DS2/R

              Page 15


            3. Brazil argued that previous disputes13, to which the United States had been a party, involved a violation of Article III:1.


      3. Article XX - General Exceptions


          1. The United States argued that the Gasoline Rule fell within the scope of Article XX whether or not it was consistent with other provisions of the General Agreement. Not all measures described by Article XX were inconsistent with the General Agreement. However, if the Panel accepted that the Gasoline Rule was consistent with other provisions of the General Agreement, in particular Article III, it did not need to decide whether the measures at issue also fell under Article XX. Article XX guaranteed in any event that these measures were not inconsistent with the General Agreement.


          2. Venezuela and Brazil considered that the issue at stake under Article XX was not whether the CAA or the regulations implementing it were necessary, but whether it was necessary to accord foreign gasoline less favourable treatment , which, they argued, was the situation in this case. Venezuela argued further that Article XX provided limited and conditional exceptions from obligations under other provisions of the General Agreement, and the burden was on the party invoking that provision to justify the application of any of the enumerated exceptions. The United States lacked the factual and legal support necessary to carry that burden with respect to any of its claims under Article XX.


      4. Article XX(b)


        1. "Protection of Human, Animal and Plant Life or Health"


            1. The United States argued that it was well established that air pollution, and in particular ground-level ozone, presented health risks to humans, animal and plants. Toxic air pollution was a cause of cancer, birth defects, damage to the brain or other parts of the nervous system, reproductive disorders and genetic mutation. It could affect not only people with impaired respiratory systems, but healthy adults and children as well. Ozone was also responsible for agricultural crop yield loss in the United States. Vehicular air toxic emissions accounted for approximately 40 to 50 percent of total air toxic emissions . The Gasoline Rule provisions sought to control toxic air pollution from mobile sources by addressing the fuel that creates these emissions . Thus, its aim was to protect public health and welfare by reducing emissions of toxic pollutants, VOCs and NOx for reformulated gasoline, and to avoid degradation of air quality for emissions of NOx and toxic air pollutants for conventional gasoline. Therefore, the Gasoline Rule fell within the range of policies specified in Article XX(b).


        2. "Necessary"


            1. The United States argued that the non-degradation requirements for both reformulated and conventional gasoline were necessary to protect human, animal and plant life or health. Using individual baselines for conventional gasoline was the quickest and fairest way to achieve the programme' s environmental goal, which was to ensure the maintenance of US 1990 gasoline quality in the cleaner areas without affecting the speedy and cost-effective implementation of the reformulated gasoline programme in the most polluted areas, and without causing major disruptions in the domestic production of conventional gasoline. If a single baseline were used for all conventional gasoline, then all producers whose gasoline was dirtier than this baseline for


              image

              13"Canada - Administration of the Foreign Investment Review Act", BISD 30S/140 (adopted on 7 February 1984) and United States - Measures Affecting Alcoholic and Malt Beverages" BISD 39S/206 (adopted on 19 June 1992).

              WT/DS2/R

              Page 16


              certain gasoline qualities would need to change the characteristics of their production to meet the standard for those qualities, and those producers whose gasoline was cleaner than the baseline could degrade down to the baseline. The result would be the same overall average for gasoline, but large segments of gasoline producers would have been required to make changes to their conventional gasoline production. Where future production exceeded their 1990 output, refiners must meet the statutory baseline. With respect to reformulated gasoline, individual baselines were used for a three year transition period and applied to three gasoline qualities -sulphur, olefin and T-90- which were required to preserve their average US 1990 levels because EPA lacked data about their precise emission effects. This approach avoided requiring large segments of producers to make changes in their gasoline in order to meet a single requirement, whereas it was not clear whether and how any such change was needed to avoid emissions increases. However, all reformulated gasoline refiners must have begun to adjust their operations in order to meet the new reformulated gasoline requirements that would be in effect in 1998 under the Complex Model.

              All regulated gasoline qualities would then be measured against the statutory baseline. Thus, the baseline system protected air quality in the most practical and cost-effective manner, while taking the best account of the various producers' characteristics.


            2. The United States argued that the individual baseline approach was however not possible with all producers, in particular, refiners that were only producing during part of 1990, blenders and importers. These categories of producers were in a different situation since they lacked the data necessary to use Methods 1, 2 and 3, and requiring them to establish an individual baseline, like domestic refiners, would have had the effect of precluding them from the US market. Thus, assigning importers to the statutory baseline ensured that they would not be forced out of the market while treating similarly situated parties alike. Moreover, even if in some cases importers might be able to establish individual baselines derived from foreign refiner information, giving importers a choice as to which baseline to use would inevitably have undermined the air quality objective of the regulation since business incentives would have induced them to use the cheapest and least stringent option, which would also have been the most polluting one. Taking into account these concerns over gaming, EPA had determined that no other option was feasible without having adverse effects on trade. The United States stressed that the Gasoline Rule applied to the importer and not to the foreign refiner. Given that there had traditionally been a variety of gasoline and blendstock qualities available on the market, importers were likely to have the flexibility to import gasoline from various sources, some with levels above and others below the statutory baseline, as long as the annual average for the importer met the statutory baseline.


            3. The United States argued that it was not feasible to give individual baselines to foreign refiners for various reasons. First, gasoline was a fungible international commodity and a shipment of gasoline arriving in a US port generally contained a mixture of gasoline that had been produced at several foreign refineries. Therefore it would be very difficult, if not impossible, to determine the refinery of origin of a shipment of gasoline for the purpose of establishing an individual baseline. Second, the difficulty of identifying the refinery of origin would also favour potential gaming of the system since the foreign refiner could be tempted to claim the refinery of origin for each shipment of imported gasoline that would present the most benefits in terms of the baseline restrictions. The third reason related to the difficulties of the United States to exercise enforcement jurisdiction over foreign refiners. The Gasoline Rule could not be enforced simply by examining the product at the border but required EPA to audit the facilities of refineries in order to verify, inter alia, that the data provided to establish the individual baselines were accurate as well as to ensure future compliance. EPA also needed other enforcement tools such as criminal penalties, civil enforcement proceedings or court warrants, that would not be readily available to use outside US territory against a refinery located on foreign soil. Holding the importer accountable for the conduct of the foreign refiner with whom he has not colluded could have been an unfair solution. The United States recalled that the panel report "United States -

              WT/DS2/R

              Page 17


              Section 337 of the Tariff Act of 1930" had acknowledged that a measure might need to provide apparently less favourable treatment to imports in situations where it "might be considerably more difficult to identify the source of infringing products or to prevent circumvention of orders limited to the products of named persons"14, than for US products. The present case offered similar differences in enforcement needs and capabilities with respect to the identification of the source of gasoline.


            4. The United States disagreed with Venezuela' s assertion that the May 1994 Proposal demonstrated the feasibility of individual baselines for foreign refiners. This Proposal was a continuation of EPA' s efforts to develop criteria that would protect the environment, minimize disruption to producers and treat similarly situated gasoline alike, taking into account the comments and concerns expressed by interested parties. The fact that EPA made this attempt was neither a determination that the Gasoline Rule was flawed nor a determination that the 1994 Proposal was feasible. The 1994 Proposal contained several strict conditions governing the establishment of individual baselines for foreign refiners, which showed that concerns still existed. Moreover, it applied only to reformulated gasoline because the indefinite application of individual baselines for conventional gasoline and the expectation that many more foreign refiners would supply the conventional market indicated that the environmental risk associated with allowing this option were too great to justify even its proposal. In public comments, the 1994 Proposal had been criticised as favouring a small group of importers over all the others. PDVSA and other foreign refiners had objected that the proposed conditions , inter alia those related to gasoline tracking, were unworkable. For these reasons, the United States rejected Venezuela' s assertion that EPA would have finalized this Proposal except for the action by Congress.


            5. The United States considered that Venezuela' s citation of the testimony of a US government official was inapposite. This testimony reflected the US government' s commitment when it issued the final Gasoline Rule to continue addressing the issue of how imports were treated together with the concerns that environmental protection not be compromised and that the provisions be fair to all the parties affected. Contrary to Venezuela' s argument, this testimony did not show that protectionism underlay the Gasoline Rule' s treatment of imports. The statement of a US government official defending the proposed use of foreign refiner baselines only illustrated that US regulators wished to obtain a mutually satisfactory solution with Venezuela. Moreover, the Rule explicitly stated that its motivation was environmental protection, not protectionism.


            6. Venezuela argued that Article XX(b) was not applicable because the United States had not demonstrated that there were no less trade-restrictive means to achieve its health policy objectives. The Gasoline Rule' s discriminatory baseline requirements were not, therefore, "necessary" within the terms of Article XX(b). Venezuela considered that there were less trade-restrictive alternatives to the Gasoline Rule discriminatory baseline requirements that would achieve the same objective. One such alternative was to authorize the use of individual baselines by foreign refiners for both reformulated and conventional gasoline. Another alternative was to require all US

              gasoline producers to meet the statutory baseline requirements. A third alternative, in Venezuela' s view, would have been to enforce the Complex Model as of 1995, rather than 1998, so as to treat both US and imported reformulated gasoline equally from the beginning. A fourth alternative would have been to authorize the use of foreign refiner individual baselines and, if compensating emissions reductions were necessary, to spread the burden of such compensation equally across all gasoline, US and imported alike.



              image


              14BISD 36S/345, § 5.32, adopted 7 November 1989.

              WT/DS2/R

              Page 18


            7. Venezuela considered that, contrary to the US argument, the use of a foreign refiner baseline was feasible. It was feasible for a foreign refiner to develop an individual baseline, relying on the same types of records and data as US domestic refiners. In the case of Venezuela, PDVSA had all the records necessary to accurately determine an individual baseline in conformity with the requirements applicable to the US refiners, as had been confirmed by the firm

              Turner, Mason and Co. which served as an independent, EPA-approved auditor. The foreign refiner' s individual baseline would be submitted to EPA for approval and to correct any possible mistakes before the product could be imported into the United States. EPA could then require a foreign refiner, as a requirement for establishing and maintaining its individual baseline, to appear before the agency and/or to make available to it production records and any other reasonable information aimed at ensuring the accuracy of the baseline. Then, enforcement would only be relevant in verifying the characteristics of gasoline as it entered the United States. This kind of verification was routinely performed on many types of imported products and, in the case of gasoline, compliance for each shipment could be determined at the port of entry by testing the shipment and comparing its fuel properties with the individual baseline of the foreign refiner.

              Furthermore, since the importer was currently liable towards EPA for gasoline that did not conform to the statutory baseline, it was similarly possible to make him liable for gasoline that did not conform to the foreign refiner' s individual baseline. Under US customs law, precedents existed where an importer was liable for an imported product that did not meet certain standards, and subject to civil and criminal sanctions. Venezuela cited, inter alia, the case of importers being liable for products not meeting the safety standards promulgated by the Consumer Product Safety Commission. This demonstrated that United States' concerns about enforcement mechanisms outside its territory and importer' s liability for the conduct of a foreign company

              were without merit. Venezuela rejected the argument that it was not possible to determine the refinery of origin and noted that during the consultations leading to the 1994 Proposal, PDVSA had suggested several alternatives to deal with this problem. Moreover, the concerns expressed by the United States about foreign gasoline being mixed with "dirty" gasoline before being

              imported into the United States should apply equally to US produced gasoline being degraded after it left a US refinery. Venezuela also strongly denied the assertion that it had rejected the 1994 Proposal since it had clearly supported EPA' s proposal to allow foreign refiners to establish their individual baselines and had continued to work with EPA to that end, presenting alternatives and explaining why particular provisions of the Proposal were unnecessarily burdensome and unworkable.


            8. Lastly, Venezuela argued that United States' concern about "gaming", which was based on the assumption that foreign refiners with "cleaner" gasoline would select the statutory baseline rather than establishing their own baseline, thus affecting the air quality, was purely speculative for several reasons. EPA had itself recognised that it did not have data about the average quality of gasoline imported in 1990, and thus could not know whether a significant amount of that gasoline imported was "cleaner" than the statutory baseline. Available data indicated that foreign refiners were not likely to "degrade" from their 1990 gasoline quality and the impact of any such activity would at most be de minimis. Moreover, US statistics showed that, from January to March 1995, imported gasoline represented less than two percent of total US gasoline

              consumption . Thus, even accepting (which Venezuela did not) the United States' argument that half of the foreign refiners produced gasoline in 1990 that was below the statutory baseline and half of the foreign refiners produced gasoline that was above the statutory baseline, possibilities for gaming would arise for less than one percent of total US gasoline consumption . The practical impact of gaming was too small to justify discrimination against imported gasoline under

              Article XX(b). Finally, several aspects of the Gasoline Rule are incompatible with the professed US concern regarding the possible environmental impact of potential "gaming" : for example, the fact that there was no limitation on the volume of reformulated gasoline a US refiner could produce under its individual baseline, or the fact that in a particular geographical area, the

              WT/DS2/R

              Page 19


              emissions from gasoline would vary, depending on which refiners were supplying it, and as a consequence, the emissions levels could exceed the statutory baseline. Moreover, EPA had recently proposed several amendments to the Gasoline Rule, such as provisions permitting upward adjustment in baseline levels because of a US refiner' s inability to acquire low sulphur content crude oil that was available in 1990, which equally undermined its environmental objectives.

              Venezuela concluded that the United States had not met the burden of the proof required by Article XX.


            9. Brazil did not disagree with the purpose of the United States which was to address the problem of air pollution in order to protect human, animal and plant life and health. However, Brazil considered that the Gasoline Rule programme did not satisfy the requirements of

              Article XX(b), because the burden of achieving this purpose was placed disproportionately on imported gasoline. All imported gasoline had to meet the 1990 average expressed in the statutory baseline whereas half of the domestic refineries could sell gasoline which did not meet the statutory baseline. The concerns expressed by the United States about the negative impact of imposing a single statutory baseline on domestic refiners could not justify a violation to the national treatment obligation for the following two reasons. First, EPA did not want to impose on domestic refiners whose gasoline was dirtier than the statutory baseline the burden of changing their production characteristics, but it imposed precisely this requirement on foreign refiners.

              Secondly, the United States had not demonstrated to the Panel why domestic refiners whose production was cleaner than the statutory baseline would downgrade to the baseline. And even assuming that such a downgrade would occur, the overall air quality would not change as long as the refiners with "dirtier" gasoline were required to upgrade to the statutory baseline. Brazil considered that a rule establishing the statutory baseline as a minimum with the additional requirement that those refiners who produced gasoline above the statutory baseline continue to do so was another option which would take care of the downgrading concern while at the same time improving air quality in the United States and eliminating discrimination against imports.


            10. Brazil further argued that the United States had not explained why importers could not establish an individual baseline, especially using Method 3 since importers presumably maintained records of their imports and thus could have data on their 1990 imports. Even assuming that it was necessary to assign importers to the statutory baseline, this did not explain the failure to provide for individual baselines for foreign refiners. Brazil considered that the United States had not demonstrated that foreign refiners did not have sufficient data to establish their own baselines. In that context, the United States referred only to "difficulties" but, according to Brazil, mere "difficulties" did not create necessity within the meaning of Article XX(b). Moreover, assuming that these difficulties were insurmountable, they would nevertheless not allow the United States to discriminate against foreign gasoline since there was an alternative measure, reasonably available, which was the requirement that all gasoline, domestically produced and imported, meet the same statutory baseline, as Brazil had noted above.


            11. Brazil considered that the United States had presented no factual basis to support its concern that a foreign refiner would "game" the system if given the choice between the statutory and the individual baseline. Besides, this opportunity for "gaming" could be eliminated by simply assigning all refiners, domestic or foreign, to the same baseline, statutory or individual.

              Regarding the use of individual baselines by foreign refiners, the United States had never made any attempts to investigate or determine empirically whether the calculation and enforcement of such baselines were possible. However it merely insisted that these problems were insurmountable and, therefore, the statutory baseline had to be applied to imported gasoline.

              Finally, the fact that numerous parties had objected to particular aspects of the 1994 Proposal did not mean that non-discriminatory baselines for foreign refiners were not possible. Brazil

              WT/DS2/R

              Page 20


              concluded that the United States had not demonstrated why it was not possible to permit foreign refiners of both conventional and reformulated gasoline to use their own baselines.


            12. Brazil argued that the discrimination under the General Agreement or the TBT Agreement was not justifiable even assuming that the use of foreign refiners' individual baselines was impossible. If it were impossible to assign individual baselines to foreign refiners, the United States would then be justified in using individual baselines for domestic refiners only if no other, non-discriminatory measure were available. A WTO Member was not permitted to review several options, select one in which discrimination was unavoidable, and then plead that the selected option required discrimination. Under Article III of the General Agreement -but also under Article I of the General Agreement and Article 2 of the TBT Agreement- a WTO Member was obliged, when the policy option involved discrimination, to choose another option when one was available. In this particular case, there was such an available alternative, which was to apply the statutory baseline to all producers of gasoline.


            13. The United States maintained its arguments regarding the impraticability of foreign refiner' s baseline. It argued that compliance with requirements based on foreign refinery baselines could not be established only by sampling gasoline on its arrival at a US port of entry, because it would be necessary to determine the refinery of origin for such imported gasoline.

              This type of determination would be difficult, if not impossible, due to the fungible mixing of gasoline that occurs before arrival in a US port of entry. The enforcement provisions of other US statutes cited by Venezuela to negate this concern were inapposite, because those statutes all involved matters that could be resolved by inspection of the product by Customs officials at the border. The United States also noted that there was no analogous concern with identifying the source of gasoline produced at domestic refineries, because domestic gasoline was regulated at the refinery gate, which left no questions of which refinery produced any particular batch of gasoline. The United States further argued that the potential environmental effect of "gaming" could be, under a reasonable scenario, an annual increase in NOx emissions from imported gasoline by 5.6 to 7 percent (about 115,000 short tons). US analyses of foreign refinery configurations showed that because of low fluid catalyst unit capacity among foreign refiners, sulphur and olefin levels in imports were likely to be low compared to the US statutory baseline, thus leaving ample room for gaming and degradation. Moreover, the "gaming" incentives for foreign "cleaner" refiners were not hypothetical: various changes since 1990 in physical plants and operating procedures could change the economic calculus for producing gasoline of a specified quantity, and the quality of the crude used in refining could change. In these conditions , a refiner might choose to degrade the sulphur, T-90 or other characteristics if it proved to be economical. The United States emphasized that there were no regulatory requirements on foreign refiners, who had ample flexibility, not available to domestic refiners, to select among blendstocks. The United States concluded that, contrary to Venezuela' s and Brazil' s claim, Article XX did not require adoption of the statutory baseline as a national standard even if the difficulties associated with the establishment of individual baselines for importers were insurmountable. Application of the statutory baseline to domestic producers of reformulated and conventional gasoline in 1995 would have been physically and financially impossible because of the magnitude of the changes required in almost all US refineries; it thus would have caused a substantial delay in the programme.

              Weighing the feasibility of policy options in economic or technical terms in order to meet an environmental objective was a legitimate consideration, and did not, in itself, constitute protectionism, as alleged by Venezuela and Brazil. Article XX did not require a government to choose the most expensive possible way to regulate its environment. In the case at hand, it was not necessary to assign domestic refiners to the statutory baseline for the non-degradation requirements for the reasons stated above.

              WT/DS2/R

              Page 21


            14. The United States argued that the lack of a volume limitation for the use of individual baselines under the reformulated gasoline programme was not expected to affect the success of that program. US data showed that refineries with the highest olefin and sulphur levels in their baselines (i.e. the dirtiest baselines) as a group had not extended their market share after the start up of the reformulated gasoline programme. This was consistent with EPA' s original expectations that the short time period during which individual baselines were used in the reformulated programme would not provide an incentive for refiners to revise their investment and production decisions based on whether their baselines were above or below the statutory baseline. The United States also argued that the various baseline adjustments allowed under the Gasoline Rule either redressed disadvantages occurring as a result of government requirements, or dealt with situations where the US government did not have data for a full year' s representative operations.


            15. Venezuela argued that the examples posed by the United States in the attempt to show an increase of average NOx emissions from imported gasoline because of potential gaming were flawed, leading to exaggerated results because they had relied on the Complex Model, which was not in use for 1995-1997, and on the assumption that half of the imported gasoline in 1990 had properties below and half had properties above the statutory baseline. Regarding this last assumption , the United States had conceded that it simply did not know the properties of the pool of imported gasoline in 1990, and had failed to present evidence that foreign refiners which might have exported to the United States gasoline cleaner than the statutory baseline would have an incentive to degrade down to the statutory baseline. Venezuela rejected this assumption and stated that there was no economic incentive for a refiner to operate its refinery in a less than optimal manner to increase the level of a fuel property such as sulphur or olefins for the sole purpose of making "dirtier" gasoline.


      5. Article XX(d)


          1. The United States considered the Gasoline Rule' s baseline establishment system was necessary to enforce the non-degradation requirements aiming at preventing deterioration of air quality. The non-degradation requirements ensuring that gasoline sold in the United States did not become more polluting than in 1990 were "laws or regulations which are not inconsistent with the provisions of the General Agreement". They were measures for which, pursuant to Article XX(g) and XX(b), "nothing in [the General Agreement] shall be construed to prevent the adoption or enforcement by any contracting party". For the reasons stated under Article XX(b), the baseline establishment rules were necessary to ensure that there was no degradation in gasoline or air quality. If importers were allowed to use several baselines, depending on which foreign refiners chose to use them, "gaming" could occur, and result in a deterioration of overall air quality. Therefore, the Gasoline Rule fell within the scope of Article XX(d).


          2. Venezuela considered that the United States had not clearly established which were the "laws or regulations" which were not inconsistent with the General Agreement and with which compliance was secured, and hence had failed to demonstrate such consistency. Venezuela noted that a previous panel had found that a measure was deemed to "secure compliance with" only if it was effective to "enforce obligations" under laws or regulations consistent with the General Agreement, as opposed to ensuring the broader attainment of an objective15. When stating that "the baseline establishment rules are necessary to ensure that there is no degradation in gasoline and air quality", the United States precisely referred to an objective, instead of identifying any obligation of the non-degradation requirements that the discriminatory baseline requirements were


            image


            15"EEC - Regulation on Imports of Parts and Components", BISD 37S/132, para. 5.17-5.18 (adopted on 16 May 1990).

            WT/DS2/R

            Page 22


            necessary to enforce. Moreover, for the reasons expressed under Article XX(b), the Gasoline Rule was not necessary. Thus, the United States did not meet the requirements of Article XX(d).


          3. Brazil considered that, for the reasons it had already developed under Article XX(b), the United States failed to demonstrate that the Gasoline Rule was "necessary" to secure compliance with the Clean Air Act, within the meaning of Article XX(d). As Brazil had previously indicated, there were non-discriminatory alternatives available to the United States.


      6. Article XX(g)


          1. The United States argued that, as a programme intended to preserve clean air, the Gasoline Rule fell within the scope of Article XX(g).


            1. "Related to the conservation of exhaustible natural resources..."


          2. The United States argued that clean air was an exhaustible resource within the meaning of Article XX(g) since it could be exhausted by the emissions of pollutants such as VOCs, NOx and toxics. In the most polluted areas, it could become chronically contaminated and remain so over long periods of time. Air containing pollutants could move long distances to contaminate other airsheds. Moreover, by stopping air degradation, the CAA also protected other exhaustible natural resources such as lakes, streams, parks, crops and forests, which were affected by air pollution. Thus, the objectives underlying the reformulated and conventional gasoline programmes fell within the range of policies to preserve both clean air and, consequently , other natural resources.


          3. Venezuela noted that it shared with the United States a concern for the impact of dirty air on health, but claimed that the United States' arguments regarding the applicability of

            Article XX(g) to this case were both factually and legally erroneous. Recalling past panel reports, Venezuela considered that the exceptions provided for by Article XX had to be interpreted narrowly, in a manner that preserved the basic objectives and principles of the General Agreement16. Noting that the original purpose of Article XX(g) was to permit exceptions to otherwise applicable prohibitions or restrictions on the export of tradeable goods that could be exhausted as a result of their exploitation , Venezuela doubted that clean air was an exhaustible natural resource within the meaning of article XX(g). Venezuela considered that clean air was a "condition" of air that was renewable rather than a resource that was exhaustible, such as petroleum and coal. There was no textual basis for expanding the scope of Article XX(g) to cover renewable "conditions " of resources as opposed to exhaustible natural resources.


          4. Venezuela noted that under established GATT jurisprudence, a measure "related to" the conservation of an exhaustible natural resource only if it was "primarily aimed at" conserving that resource17. The United States had not even attempted to argue that the Gasoline Rule' s discriminatory requirements, which were the measure at issue in the dispute, were "primarily aimed at" conservation, but had merely attempted to justify that the reformulated and conventional gasoline requirements fell under Article XX(g). Furthermore, the United States had identified only the protection of health as the primary objective for the reformulated and conventional gasoline requirements, which was irrelevant to an Article XX(g) analysis. Venezuela noted that it


            image


            16"United States . Restrictions on Imports of Tuna", DS29/R, 16 June 1994 (unadopted), "Canada - Administration of the

            Foreign Investment Review Act", BIDS 30S//140, para. 5.20 (adopted on 7 February 1984), "United States - Section 337 of the Tariff Act of 1930", BISD 36S/345 (adopted on 7 November 1989).

            17"Canada - Measures Affecting Exports of Unprocessed Herring and Salmon", BISD 35S/98, para. 4.6 (adopted on 22 March 1988).

            WT/DS2/R

            Page 23


            had previously demonstrated to the Panel that the Gasoline Rule methodology contained loopholes which undermined its own conservation objectives, thus confirming that the discriminatory baseline system could not be "primarily aimed at" the conservation of an exhaustible natural resource.


          5. The United States disagreed with the claim that clean air was not an exhaustible natural resource within the meaning of Article XX(g). The United States maintained that air was undoubtedly a natural resource which could be exhausted if it was rendered unfit for human, animal or plant consumption . This was similar to the recognition in previous panel proceedings that fish were an "exhaustible natural resource" since their populations could be depleted or rendered extinct18.


            b) "... made effective in conjunction with restrictions on domestic production or consumption"


          6. The United States considered that the Gasoline Rule restricted domestic production of gasoline by requiring manufacturers to limit their production of gasoline so that over the course of the year the average of particular components of the gasoline did not exceed certain maximum levels. It also restricted domestic consumption by ensuring that the average of those components of gasoline sold did not exceed certain maximum levels.


          7. Venezuela rejected this argument because it considered that the United States had not shown that the discriminatory baseline requirements were "primarily aimed at rendering effective" restrictions on domestic production or consumption of clean air, the "natural resource" to be conserved by the Gasoline Rule. The United States had only referred to restriction on domestic production and consumption of gasoline.


          8. Brazil argued that, even assuming that clean air was an exhaustible natural resource, the Gasoline Rule did not restrict domestic production or consumption of clean air. At best, the Gasoline Rule sought to increase production if not consumption of clean air, not to restrict it. Moreover, neither the CAA nor the Gasoline Rule restricted in any way the quantity of gasoline that could be produced or consumed in the United States, but merely regulated its quality. Since neither the production nor the consumption of air or gasoline was restricted by the CAA or the Gasoline Rule, the Gasoline Rule did not fall under Article XX(g).


          9. The United States argued that the Gasoline rule did restrict domestic consumption of clean air through its restriction of emissions that polluted the air. This was similar to restrictions applied on cars in order to conserve fuel. In this case, the Gasoline Rule' s application to imports

            - including the baseline rules- was primarily aimed at rendering effective restrictions on domestic production of dirty air, or conversely the consumption of clean air, through regulation of the gasoline that caused air pollution.


      7. Preamble to Article XX


          1. The United States argued that, as it had demonstrated in the discussion concerning Article III, the Gasoline Rule applied equally to similarly situated parties. Importers and blenders were required to meet the parameters of 1990 average US gasoline because they could not ascertain the refinery of origin and the quality of the gasoline they marketed in 1990. This avoided the alternatives of either "gaming" problems or excluding most imported gasoline from


            image

            18"Canada - Measures Affecting Exports of Unprocessed Herring and Salmon", BISD 35S/98 (adopted on 22 March 1988) and "United States - Prohibition of Imports of Tuna and Tuna Products from Canada, BISD 29S/91 (adopted on 22 February 1982).

            WT/DS2/R

            Page 24


            the market. Unlike domestic refiners, importers had the flexibility to rely on a variety of sources so as to meet an annual average quality of gasoline. Moreover, for each of the requirements, about half of US gasoline produced by domestic refiners had to be cleaner in certain respects than the annual average gasoline quality supplied by importers. In addition, a portion of the US gasoline market was being supplied with gasoline by domestic refiners which had to meet the statutory baseline because their gasoline could not be presumed to have been part of the gasoline pool in 1990. But to the extent that the enforcement conditions differed between the United States and other countries, the "same conditions " did not prevail in the United States and in other supplying countries. Accordingly, any differences in treatment were neither arbitrarily nor unjustifiably discriminatory, but were based on valid, legitimate policy reasons.


          2. The United States further argued that the Gasoline Rule did not constitute a disguised restriction on trade since its objective was to ensure no degradation from 1990 levels for emissions and air pollutants, a health objective that had nothing to do with a restriction on trade. The provisions were transparent and imposed the same overall requirements, stemming from the same objective, on imported as on domestic gasoline. The evolution of the provisions at issue demonstrated that treatment of imports had nothing to do with the fundamental structure of the part of the rule that was being contested. For conventional gasoline, the CAA prescribed individual baselines on a producer-specific basis for refiners, blenders and importer. At the time the CAA Amendments were introduced, imports did not figure significantly in the debate, not surprisingly in the light of their small (2 to 6 percent) share of the US gasoline market. In view of the uncertainty of the emissions effects of these parameters irrespective of the source of gasoline, in 1991 EPA agreed to regulate, with respect to reformulated gasoline, on the basis of individual baselines for the three non-degradation requirements.


          3. Venezuela argued that the 75 % Rule which applied to only a few refineries that were historically determined did grant an advantage to gasoline imported by the United States from certain third countries, as opposed to gasoline imported from Venezuela. Thus, the Gasoline Rule constituted a means of arbitrary and unjustifiable discrimination between countries where the same conditions prevailed. Referring to past panel reports19, Venezuela considered that the reference "where the same conditions prevail" did not relate to the national treatment obligations, as had been argued by the United States, but only to the most-favoured-nation obligation of the General Agreement. Moreover, as had been previously argued by Venezuela, the discriminatory baseline requirements of the Gasoline Rule were not justified by environmental concerns, but intended to distort the conditions of competition in favour of US gasoline against imported gasoline. Hence, the Gasoline Rule was a disguised restriction on international trade, within the meaning of the Preamble of Article XX.


          4. Brazil rejected the arguments given by the United States and argued that by discriminating between the United States and all other countries, and by discriminating among third countries based upon the criteria of ownership and quantity of exports, the Gasoline Rule constituted a means of arbitrary or unjustifiable discrimination between countries where the same conditions prevailed. Since the discrimination of imported products was so blatant, Brazil considered that the restrictions on trade were not disguised.


            image


            19"United States - Imports of Certain Automotive Spring Assemblies", BISD 30S/107, para. 55 (adopted on 26 May 1983) and "United States - Prohibition on Imports of Tuna and Tuna Products from Canada", BISD 29S/91, para. 4.8 (adopted on

            22 February 1982).

            WT/DS2/R

            Page 25


      8. Article XXIII - Nullification and Impairment


          1. Venezuela argued that, in addition to its violation claim under XXIII:1(a), it was making an alternative claim of nullification and impairment under XXIII:1(b). The discriminatory baseline requirements had resulted in shipments of approximately thirty-three thousand fewer barrels of Venezuelan gasoline to the United States per day than would be possible absent the discrimination. The price of Venezuelan gasoline and its share in the US market, as well as the investment programme for Venezuelan refineries, had also been adversely affected. Venezuela was aware that statistical evidence of adverse trade effects was not the basis for a finding of nullification or impairment under Article XXIII:1(b). Nevertheless, it wished to emphasize that by so affecting trade volumes, prices received for Venezuelan gasoline, Venezuela' s share in the US market and PDVSA' s investment programme, the Gasoline Rule had distorted the conditions of competition for trade in the United States compared to the conditions reasonably expected by Venezuela under the General Agreement. Venezuela said that if the Panel found nullification or impairment under Article XXIII:1(a), it needed not make a ruling on non-violation nullification or impairment under Article XXIII:1(b).


          2. In responding to other claims by Venezuela and Brazil, the United States contested generally the allegation that there was any identifiable impact on 1995 Venezuelan exports attributable to the Gasoline Rule. Venezuela' s exports to the United Sates had steadily declined over the last five years, and its decrease in exports in 1995 was entirely consistent with the earlier decline.


    3. Agreement on Technical Barriers to Trade


      1. Article 2 - Preparation, Adoption and Application of Technical Regulations by Central Government Bodies


        1. Whether or Not the Gasoline Rule is a Technical Regulation


            1. Venezuela and Brazil submitted that the Gasoline Rule was a "document" which laid down "product characteristics" and "with which compliance was mandatory" for both conventional and reformulated gasoline. Therefore, it was a "technical regulation" within the meaning of Annex I of the TBT Agreement.


            2. The United States replied that the non-degradation requirements contained in the Gasoline Rule did not specify particular product characteristics, and therefore did not meet the TBT Agreement' s definition of a "technical regulation". Shipments of gasoline of widely varying characteristics could be sold by a particular entity, the only requirement being that at the end of the year, the average of certain of their chemical ingredients fell below certain levels. Thus, these provisions were requirements on companies, not on products, and compliance was measured on a company level for importers and blenders, and on a refinery level for domestic refiners, but not on a product basis. These provisions constituted requirements on total annual sales, but were not technical regulations within the meaning of the TBT Agreement. Therefore, the TBT Agreement did not apply to this dispute.


            3. Venezuela maintained that EPA' s regulation implementing the CAA through the baseline setting mechanisms precisely established product characteristics for gasoline consumed in the United States and was therefore a "document which lays down product characteristics" within the meaning of the definition contained in Annex I of the TBT Agreement. Moreover, the

              United States itself admitted this fact when saying that the Rule dealt with "chemical ingredients". Venezuela was also of the view that averaging did not make any difference for the purpose of the

              WT/DS2/R

              Page 26


              TBT Agreement since any averaging techniques required examination of the properties of each individual gasoline shipment. Excluding from the coverage of the TBT Agreement regulations relying on averaging would open a gaping loophole. Under this interpretation, the obligation of the Agreement could be avoided by averaging. Venezuela considered that the United States wanted to avoid examination under the TBT Agreement in order to escape the requirements contained in Article 2.2.


            4. Brazil objected to the United States' argument that the Gasoline Rule was not a technical regulation within the meaning of the TBT Agreement. Brazil considered that the language of the CAA and that of the Gasoline Rule, referred to the establishment of product standards for gasoline when determining the fuel properties for the statutory baseline and the individual baselines. These product standards applied to gasoline were mandatory. The fact, argued by the United States, that no particular shipment of gasoline needed to meet any precise standards since the requirements were measured on an annual average basis, and thus that the "product" was the annual quantity of gasoline produced, blended, or imported, rather than each sub-unit, was irrelevant. Annual production in this case was simply the unit of production to which the standard was applied.

              Brazil noted that if the United States were correct in its assertion that the individual baselines applied to refiners and not to gasoline, the discrimination would then be even more apparent because foreign refiners had no baseline. In this case a mandatory requirement would apply only to imported gasoline while, under the logic of the United States, no requirement would apply to domestic gasoline, as distinct from domestic refiners. However, the enforcement and surveillance system provided for by EPA in the Gasoline Rule in order to regularly check the quality of gasoline and its property at the refinery level argued in favour of a technical regulation setting forth product characteristics. Moreover, the United States' own statements to the Panel acknowledged this fact when declaring that the "requirements" of the Gasoline Rule were "necessary to protect human, animal and plant life or health". In conclusion, Brazil considered that a rule which obliged imported gasoline that did not meet the statutory baseline to be blended with gasoline that exceeded these requirements in order to meet the mandatory statutory requirements was a "document" with mandatory product characteristics.


            5. The United States argued that the TBT Agreement had been designed to elaborate on the disciplines of Article III of the General Agreement for a very specific subset of measures (technical regulations, standards and conformity assessment procedures). The fact that a measure was in writing, mandatory and applied to products did not make it a technical regulation. Excise taxes, for instance, met all these criteria but were not "technical regulations". Similarly, the term "technical regulation" was not so broad as to cover all government regulatory actions affecting products. For example, government regulations requiring factory smokestacks to have devices to reduce emissions were not technical regulations, though they were in writing, mandatory and specified "characteristics". Contrary to what was argued by the complainants , there were no minimum or maximum content or emissions requirements applied with respect to the non- degradation requirements for individual shipments of either reformulated or conventional gasoline under the Simple Model. A shipment or even sale of gasoline was not required to meet specific product characteristics with respect to the non-degradation requirements at issue. The Gasoline Rule was not setting uniform criteria in terms of gasoline characteristics; standardization was neither the purpose nor the result of the regulation. The United States concluded that the complainants were interpreting the term "technical regulation" out of context and such an interpretation, if accepted, would introduce into the TBT Agreement many measures which were in fact not intended to be covered. The United States also argued that Brazil' s view that a "product" in this case be defined as an entire year' s production, rather than a shipment or a batch, would be a radical departure from the concept of "product" under the WTO and was without basis in the WTO.

              WT/DS2/R

              Page 27


        2. Article 2.1


            1. Venezuela argued that Article 2.1 of the TBT Agreement incorporated the obligations of national treatment and MFN set forth in Articles III and I of the General Agreement. Venezuela and Brazil argued that, as a technical regulation within the meaning of the TBT Agreement, the Gasoline Rule laid down product characteristics for imported Venezuelan and Brazilian gasoline that gave less favourable treatment than that provided to imports from certain third countries and to US gasoline. Thus, it violated the obligations of national treatment and MFN treatment contained in Article 2.1 of the TBT Agreement.


        3. Article 2.2


            1. Venezuela and Brazil claimed that the Gasoline Rule created unnecessary obstacles to international trade in violation of Article 2.2 of the TBT Agreement.


            2. Venezuela considered that the Gasoline Rule violated Article 2.2 for two reasons. First, there was evidence that this Rule had been "prepared, adopted or applied with a view to ... creating obstacles to international trade". The United States did not intend to discriminate against imported gasoline when it initiated the regulatory process. However, crucial decisions involving the specific discriminatory aspects of the Rule were knowingly made both during the regulatory process and thereafter. The testimony under oath made in April 1994 by a government official to the United States Senate was evidence that the discrimination was intentionally adopted as a means of affording protection to gasoline produced in the United States.


            3. Second, the Gasoline Rule had the effect of creating an unnecessary obstacle to international trade because the more stringent requirements imposed on imported gasoline were

              not necessary to fulfil the stated objective of the Rule which is to improve air quality in the United States. In this respect, Venezuela considered that Article 2.2 of the TBT Agreement provided greater guidance with respect to the concept of "necessity" than Article XX of the General Agreement, especially its second sentence which expressly calls for a certain balance. Article XX spoke only of measures that are "necessary", which had been in previous cases strictly interpreted to mean that a measure is not necessary if it is not the least trade-restrictive measure reasonably available20. As explained in relation of Article XX(b), the Gasoline Rule clearly pursued a trade- restrictive approach despite the fact that alternatives consistent with the General Agreement were available, while the risks of non-fulfilment of any legitimate objective had been deliberately exaggerated. EPA itself had acknowledged that less trade-restrictive alternatives of achieving the air quality objective were possible and the 1994 Proposal, though not entirely consistent with the General Agreement, was one such alternative.


            4. Venezuela further considered that the risks of non-fulfilment of a legitimate objective had to be assessed against "scientific and technical information" which, in the case at hand, had never been provided despite various requests made by Venezuela, in particular under Article 2.5 of the TBT Agreement. Article 2.2 of the TBT Agreement required that the trade-restrictive elements of a technical regulation be eliminated unless "scientific and technical information" or other reliable factual data demonstrate that those elements were necessary to fulfil a legitimate objective. The United States had never submitted scientific evidence or technical data demonstrating that the different baseline requirements were necessary to fulfil the air quality objectives but had always relied on "gaming" as the justification. EPA had never attempted to analyze how much imported gasoline would be susceptible to gaming or whether, in case of gaming, the impact on health


              image

              20"United States - Section 337 of the Tariff Act of 1930", BISD 36S/345, adopted on 7 November 1989 and "Thailand - Restrictions on Importation of and Internal Taxes on Cigarettes", BISD 37/200, adopted 7 November 1990.

              WT/DS2/R

              Page 28


              objectives would be unacceptable. In that regard, Venezuela recalled that EPA itself had acknowledged that the environmental impact of gaming was speculative because it lacked "clear evidence" regarding the actual average quality of 1990 imported gasoline and did not know whether a significant amount of imported gasoline was "cleaner" than the statutory baseline.

              Moreover, so little gasoline was imported that the potential differential emissions -between individual and statutory baselines- would not have any significant impact on the average emission quality of the gasoline consumed in the United States.


            5. Brazil stated that, for the reasons already expressed under arguments relating to Articles I and III of the General Agreement and Article 2.1 of the TBT Agreement, the Gasoline Rule created "unnecessary obstacles to international trade" in a manner contrary to Article 2.2 of the TBT Agreement.


      2. Article 12 - Special and Differential Treatment of Developing Country Members


          1. Venezuela observed that Article 12 of the TBT Agreement imposed certain obligations on the United States with respect to developing countries. Venezuela did not seek any special treatment but merely wanted its gasoline to be held to the same baseline requirements as US gasoline. Venezuela stated that it was not asking for the Panel to rule under Article 12 but intended to point out that the discriminatory treatment affecting Venezuelan gasoline was particularly objectionable in the light of that provision.


  4. SUBMISSIONS BY INTERESTED THIRD PARTIES


    1. The European Communities


        1. The European Communities (the "EC") stated that, as an exporter to the United States of gasoline for automobiles and other fuel oils, it had a substantial interest in the matter before the Panel. In 1994, the total volume of EC-12 exports to the United States for gasoline represented 6' 423' 411 metric tonnes. This volume had increased since the enlargement of the EC, on

          1 January 1995. The EC declared that it did not contest the right of the United States to enforce legislation whose purpose was to protect human, animal or plant life or health. However, such measures had to be in conformity with the provisions of the WTO Agreement and not be applied so that imports from third countries were discriminated against, that the domestic industry was afforded protection, or that disguised restrictions were imposed on international trade.


        2. The EC stated that, while agreeing with the United States that the application of formally identical provisions could, in certain cases, accord in practice less favourable treatment to imported products, it could not agree with the consequences which the United States seemed to draw from the findings of the panel report "United States - Section 337 of the Tariff Act of 1930" for the present case. It could not be concluded from that report that when it was not technically feasible to apply to imported products the rule applied to domestic products, it was then sufficient to find a workable rule which was sufficiently close to that applicable to domestic products, without changing that latter rule. The EC considered that the logic behind Article III of the General Agreement required Members to achieve effective non-discrimination or absence of protection. Such an objective should be achieved preferably by amending existing rules or re- formulating new rules which could be applied identically to domestic and imported goods.


        3. The EC argued that risks of violations of Article III:1 and 4 resulted from the fact that Methods 2 and 3 of establishing individual baselines were only available to domestic refiners.

          WT/DS2/R

          Page 29


          The EC did not want to discuss the accuracy of the arguments developed by the United States with respect to the feasibility of individual baselines for foreign refiners, but assumed, for the sake of argument, that Methods 2 and 3 could not be applied to imported gasoline in this case.

          Considering the explanations given by the United States as to what the statutory baseline represented, and assuming they were correct, the EC failed to see why US refiners could not be subject to the statutory baseline, like importers and blenders. Such a measure would have been in total conformity with Article III, paragraphs 1 and 4. In addition, it appeared from the information submitted by the main parties to the dispute that barely half of the US refiners had their individual baselines approved at the time of the entry into force of the Rule. While not affecting the existence of the violation, this fact proved that the application of the statutory baseline erga omnes would probably not affect significantly the competitive position of US refiners.


        4. The EC understood the concern expressed by the United States that certain importers and blenders, who had the flexibility to select gasoline from various sources, might have an advantage over US refiners if the statutory baseline were to be applied to all gasoline producers. However, this potential advantage was inherent to the averaging mechanism contained in the Gasoline Rule and was not a sufficient reason to introduce a system which would unavoidably favour certain US producers. Article III required that no less favourable treatment be given to imported products, not the contrary21. If one considered that a US refiner might have produced extremely "dirty" gasoline in 1990, the Gasoline Rule did not give an immediate incentive for US refiners to adapt their production, whereas increased access to US market for third country gasoline was dependent on a gradual approximation of their quality compared with the statutory baseline. Therefore, the Gasoline Rule entailed at the very least a serious risk of discrimination, which constituted , by itself, a form of discrimination22. According to past panel reports, the United States had to show that, despite the different treatment accorded to imported products, the no less favourable treatment standard of Article III was met.


        5. Regarding the 75 % Rule, the EC argued that the fact it was based on objective criteria, as argued by the United States, was not sufficient to avoid discrimination in the present case. A de facto discrimination in the application of the most-favoured-nation principle was possible, as the criteria used to grant that treatment were based on the situation in 1990. An importer meeting the required criteria after that date could not have invoked it. Hence, the Rule could only benefit certain countries where US companies had invested in local refiners before 1990. As acknowledged by the United States, this could have included Canada, where certain US companies owned refineries at that time. However, this could not have benefitted production of refiners located in countries where, for instance, the petroleum industry was mostly, if not exclusively, owned by the State in 1990. Therefore, the 75 % Rule was de facto not based on objective criteria, and hence contrary to Article I of the General Agreement. The EC further argued that the Panel should make a finding on this Rule despite the fact that it could no longer be invoked.

          A Panel should be guided in its examination by the content of the "matter" -the 75 % Rule in the present case- referred to it23. A determination by the Panel on the conformity of that rule with Article I would help avoid this kind of measure being found in the future in the legislation of a WTO Member.


          image


          21"United States - Section 337 of the Tariff Act of 1930", BISD 36S/345 (adopted on 7 November 1989).

          22"EEC - Payments and Subsidies Paid to Processors and Producers of Oilseeds and Related Animal Feed Proteins", BISD 37S/86 (adopted on 25 January 1990).

          23United States - Imposition of Countervailing Duties on Certain Hot-Rolled Lead and Bismuth Carbon Steel Products from France,

          Germany and the United Kingdom", SCM/185, 15 November 1994 (not adopted).

          WT/DS2/R

          Page 30


        6. As to whether the Gasoline Rule fell under the TBT Agreement, the EC stated that it agreed with the United States that the requirements on chemical ingredients did not need to be satisfied by each shipment and also that, the measures at issue being based on a yearly average, the importers remained free to import different varieties of gasoline provided that the annual average met the requirements. However, the EC doubted that a standard should be excluded from the scope of the TBT Agreement only for the reason that it required compliance on a yearly basis instead of on a shipment basis. It was clear that the importer had only to balance various qualities of gasoline in order to meet the statutory baseline. From the point of view of the exporting country, the Gasoline Rule created a clear incentive for adapting its production standards if it wanted to maintain or increase its share of the US market. Exporting refiners not adapting their production standards to US requirements (or at least not gradually narrowing the difference down to total compliance) would be unlikely to increase their sales in the US since importers had to blend or balance the "dirty" imported gasoline with "cleaner" gasoline. The "cleaner" gasoline being likely to be more expensive, importers would gradually cease to import "dirty" gasoline, thus obliging foreign refiners to meet the statutory baseline for each shipment. For this reason, the EC could not agree with the United States that the non-degradation requirements did not specify particular "product characteristics". Although no maximum content was set per shipment, the US methodology resulted in pushing the market to apply standards gradually closer to the averages referred to in the rule.


        7. The EC considered that, in certain circumstances, the US system would impose a clearly defined standard. For example, an exporter setting up its own importation network in the United States was likely to be obliged to immediately adapt its production to US standards in order not to have to import gasoline of different qualities at the same time as its own gasoline (with the related increased costs). Therefore, the EC believed that the Gasoline Rule imposed a technical regulation within the meaning of the TBT Agreement. The EC further considered that, if one were to agree with the US arguments, such an averaging system could represent a potential for circumvention of the TBT Agreement, mainly in the field of chemical products where it was used relatively frequently. If the mere use of an averaging requirement was sufficient to exclude the TBT Agreement from applying to certain environmental standards, then the increased legal protection resulting from Article 2.2 of that Agreement compared, for instance, with Article XX of the General Agreement, would no longer be available for the other Members.


        8. Referring to the findings of a previous panel, the EC considered that, from a procedural point of view, the United States was entitled to rely on Article III and on Article XX in the alternative24. However, Article XX, as an exception, had to be interpreted strictly, and the EC was of the view that the Gasoline Rule, in the way it was applied, did not meet the requirements of Article XX but imposed a disguised restriction on trade by allowing US refiners to continue producing "dirty" gasoline meeting their individual baselines, while imposing actual constraints on foreign producers to adapt their production to US standards. Such a protectionist effect would probably not be created if identical baselines were applied to both imported and domestically produced gasoline. Moreover, as demonstrated above, the application of the statutory baseline to both domestic and imported gasoline would have achieved the same aim without introducing discrimination between sources of supply. In any event, measures inconsistent with the General Agreement were not necessary to enforce the 1990 CAA amendment. Therefore, the EC considered that, even if the Gasoline Rule did not constitute a "means of arbitrary or unjustifiable discrimination between countries where the same conditions prevail", it was obviously a means to alleviate the restructuring efforts of the US refining industry while at the same time requiring foreign producers to adapt almost immediately their production. Hence, while officially pursuing


          image


          24"United States - Restrictions on Imports of Tuna", DS21/R, 3 September 1991, para. 5.22 (not adopted).

          WT/DS2/R

          Page 31


          environmental objectives, it introduced a disguised restriction on trade. The EC concluded that the baselines system was not proportionate and did not meet the necessity test of Article XX.


    2. Norway


        1. Norway stated that its reasons for reserving its rights as a third party in this case were in legal and practical terms very similar to those argued by Venezuela and Brazil in their respective requests for the establishment of a panel. The Gasoline Regulation denied national treatment to gasoline imported from Norway. Therefore, Norway supported Venezuela and Brazil' s request that the Panel find the Gasoline Rule to be inconsistent with Articles I and III of the General Agreement, and with Article 2 of the TBT Agreement.


        2. Norway said that the Norwegian State Oil Company ("Statoil") was experiencing a very difficult situation because of the way the Gasoline Rule operated. There was a considerable incentive for Statoil to be able to export to the US market as compared to other markets. In 1989, Statoil built the Mongstad refinery with the objective of selling to the United States some

      0.5 millions tons per year out of a total gasoline production of 2,5 millions tons per year. In 1990, Statoil sold a total of about 470' 000 tons of gasoline to the United States, out of which about 350' 000 tons came from Mongstad refinery. Since December 1994, Statoil had not exported gasoline from Mongstad to the United States.


      4.11 Norway argued that changing specifications was in the nature of the refining business. However, like Venezuela's and Brazil' s refineries, Statoil was affected by the discriminatory nature of the US regulation. Assigned to the statutory baseline for its exports of reformulated gasoline until 1998 and for conventional gasoline indefinitely, the Mongstad refinery would not competitively be able to produce any volumes of gasoline, based on current refinery configurations and investment plans. Norway considered that, if the Panel ruled in favour of Venezuela and Brazil, Statoil would, as an "Importer of Record" in 1990, be allowed to establish its individual 1990 baseline for the volume sold that year.


  5. INTERIM REVIEW


      1. On 18 December 1995, the United States requested the Panel to review in accordance with Article 15.2 of the DSU precise aspects of the interim report that had been issued to the parties on 11 December 1995, and to hold a meeting for that purpose. The Panel met with the parties on 3 January 1996 in order to hear their arguments concerning the interim report. The Panel carefully reviewed the arguments presented by the United States and the responses offered by Venezuela and Brazil.


      2. In respect of the interim report’s discussion of Article III, the United States argued that in several respects the interim report dealt with issues that were not disputed by the parties or were unnecessary to the Panel’s conclusion that aspects of the Gasoline Rule violated Article III:4. While the Panel did not agree with all the arguments made by the United States, it did revise the report to take into account those arguments with which it agreed and paragraphs 6.5 and 6.9 -

    6.11 of the findings reflect the Panel’s response.


    5.3 In respect of the interim report’s discussion of Article XX(b), the United States objected to the Panel’s use of specific terms which did not appear in the text of the provision, the description of the US argument, and the Panel’s analysis of alternative measures available to the United States. The Panel revised the report where it accepted the US arguments and paragraphs

    6.20 - 6.25 and 6.27 - 6.28 of the findings reflect the Panel’s response.

    WT/DS2/R

    Page 32


      1. In respect of the interim report’s discussion of Article XX(d), the United States objected to the Panel’s use of specific terms which did not appear in the text of the provision. The Panel accepted the US arguments and paragraph 6.31 of the revised findings reflects the Panel’s response.


      2. In respect of the interim report’s discussion of Article XX(g), the United States objected to the Panel’s use of specific terms which did not appear in the text of the provision, and the analysis of alternative measures available to the United States. Venezuela requested a change to the description of its argument under this provision. The Panel revised the report where it accepted the arguments of the US and Venezuela and paragraphs 6.35 - 6.36 and 6.40 - 6.41 of the findings reflect the Panel’s response.


      3. In respect of the interim report’s descriptive section, Venezuela and the United States suggested further changes which the Panel took into account in re-examining that part of the report. The Panel revised the descriptive section of the report where it accepted the need for these changes.


  6. FINDINGS


    1. Introduction


        1. The Panel noted that the dispute arose from the following facts. The Clean Air Act aims to control and reduce air pollution in the United States. The Act and certain of its regulations (the “Gasoline Rule”) set standards for gasoline quality intended to reduce air pollution, including ozone, caused by motor vehicle emissions . From 1 January 1995, the Gasoline Rule permits only gasoline of a specified cleanliness (“reformulated gasoline”) to be sold in areas of high air pollution. In other areas, only gasoline no dirtier than that sold in the base year of 1990

          (“conventional gasoline”) can be sold.


        2. The Gasoline Rule applies to refiners, blenders and importers of gasoline. It requires that certain chemical characteristics of the gasoline in which they deal respect, on an annual average basis, defined levels. In the Gasoline Rule some of these levels are fixed; others are expressed as “non-degradation” requirements. Under the non-degradation requirements, each domestic refiner must maintain, on an annual average basis, the relevant gasoline characteristics at levels no worse than its “individual baseline” — that is, the annual average levels achieved by that refiner in 1990. To establish an individual baseline, a refiner must show evidence of the quality of gasoline produced or shipped in 1990 (“Method 1"). If that evidence is not complete, then it must use data on the quality of blendstock produced in 1990 (“Method 2"). If these two methods do not result in sufficient evidence, the refiner must also use data on the quality of post-1990 gasoline blendstock or gasoline (“Method 3").


        3. Importers are also required to use an individual baseline, but only in the case (unlikely, according to the parties to the dispute) that they are able to establish it using Method 1 data. Unlike domestic refiners, they are not allowed to establish an individual baseline by using the secondary or tertiary data specified in Methods 2 and 3. If an importer cannot produce Method 1 data, then it must use a “statutory baseline” which the United States claims is derived from the average characteristics of all gasoline consumed in the United States in 1990. Some other domestic entities (such as refiners with only partial or no 1990 operations, and blenders with insufficient Method 1 data) are also assigned the statutory baseline. Exceptionally, importers that imported in 1990 at least 75 percent of the production of an affiliated foreign refinery are treated as domestic refiners for the purpose of establishing baselines. Since this dispute concerns only the

          WT/DS2/R

          Page 33


          Gasoline Rule’s non-degradation requirements, and not reformulated and conventional gasoline as such, the Panel will refer generally to “gasoline” in the course of its findings.


        4. Venezuela and Brazil claim that the Gasoline Rule violates the national treatment provisions of Article III:1 and 4 of the General Agreement and the most-favoured-nation provision of Article I. Venezuela claims in the alternative that the Gasoline Rule has nullified and impaired benefits under the non-violation provisions of Article XXIII:1(b). Venezuela and Brazil also claim that the Gasoline Rule violates Article 2 of the Agreement on Technical Barriers to Trade (the

          “TBT Agreement”). The United States rejects these claims and argues that the Gasoline Rule can be justified under the exceptions contained in Article XX, paragraphs (b), (d) and (g), which argument is rejected by Venezuela and Brazil. It also argues that the Gasoline Rule does not come within the scope of Article 2 of the TBT Agreement.


    2. Article III


      1. Article III:4


          1. The Panel proceeded to examine the claim that the Gasoline Rule violates Article III:4 of the General Agreement, which states:


            The products of the territory of any contracting party imported into the territory of any other contracting party shall be accorded treatment no less favourable than that accorded to like products of national origin in respect of all laws, regulations and requirements affecting their internal sale, offering for sale, purchase, transportation, distribution or use.


            The Panel noted that under this provision the complainants are required to show the existence of:

            1. a law, regulation or requirement affecting the internal sale, offering for sale, purchase, transportation, distribution or use of an imported product; and (b) treatment accorded in respect of the law, regulation or requirement that is less favourable to the imported product than to the like product of national origin. The Panel agreed with the parties that the Gasoline Rule was a law, regulation or requirement affecting the internal sale, offering for sale, purchase, transportation, distribution or use of an imported product. It proceeded therefore to consider whether the Gasoline Rule accorded less favourable treatment to imported products than to like products of national origin.


          2. The Panel noted the arguments of Venezuela and Brazil that imported gasoline was “like” domestic gasoline, but received treatment less favourable because imported gasoline was subjected to more demanding quality requirements than gasoline of US origin. The United States replied that gasoline from similarly-situated parties was treated in the same manner under the Gasoline Rule. Gasoline from importers was treated no less favourably than that from other domestic non- refiners such as blenders, or refiners who had only limited or no operations in 1990.


          3. The Panel observed that Article III:4 deals with treatment to be accorded to like products. However, the text does not specify exhaustively those aspects that determine whether the products are “like”. In resolving this interpretative issue the Panel referred, in conformity with Article 3.2 of the Understanding on Rules and Procedures Governing the Settlement of Disputes, to the Vienna Convention on the Law of Treaties, which states in Article 31 that “a treaty shall be

            WT/DS2/R

            Page 34


            interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose”.25


          4. The Panel proceeded to examine this issue in the light of the ordinary meaning of the term “like”. It noted that the word can mean “similar”, or “identical”. The Panel then examined the

            practice of the CONTRACTING PARTIES under the General Agreement. This practice was relevant since Article 31 of the Vienna Convention directs that “subsequent practice in the application of the

            treaty which establishes the agreement of the parties regarding its interpretation” is also to be considered in the interpretation of a treaty. The Panel noted that various criteria for the determination of like products under Article III had previously been applied by panels. These were summarized in the 1970 Working Party Report on Border Tax Adjustments, which had observed:


            With regard to the interpretation of the term ‘like or similar products’, which occurs some sixteen times throughout the General Agreement, it was recalled that considerable discussion had taken place . . . but that no further improvement of the term had been achieved. The Working Party concluded that problems arising from the interpretation of the terms should be examined on a case-by-case basis. This would allow a fair assessment in each case of the different elements that constitute a ‘similar’ product. Some criteria were suggested for determining, on a case-by-case basis, whether a product is ‘similar’: the product' s end-uses in a given market; consumers' tastes and habits, which change from country to country; the product' s properties, nature and quality.26


            These criteria had been applied by the panel in the 1987 Japan Alcohol case in the examination under Article III:2 of internal taxation measures. That panel had proceeded on a case-by-case basis, determining whether various alcoholic beverages were “like” on the basis of “their similar properties, end-uses and usually uniform classification in tariff nomenclatures.”27 The Panel considered that those criteria were also applicable to the examination of like products under Article III:4.


          5. In light of the foregoing, the Panel proceeded to examine whether imported and domestic gasoline were like products under Article III:4. The Panel observed first that the United States did not argue that imported gasoline and domestic gasoline were not like per se. It had argued rather that with respect to the treatment of the imported and domestic products, the situation of the parties dealing in the gasoline must be taken into consideration. The Panel, recalling its previous discussion of the factors to be taken into account in the determination of like product, noted that chemically-identical imported and domestic gasoline by definition have exactly the same physical characteristics, end-uses, tariff classification , and are perfectly substitutable. The Panel found therefore that chemically-identical imported and domestic gasoline are like products under

            Article III:4.


          6. The Panel next examined whether the treatment accorded under the Gasoline Rule to imported gasoline was less favourable than that accorded to like gasoline of national origin. The Panel observed that domestic gasoline benefitted in general from the fact that the seller who is a refiner used an individual baseline, while imported gasoline did not. This resulted in less favourable treatment to the imported product, as illustrated by the case of a batch of imported


            image

            25Vienna Convention on the Law of Treaties, Art. 31.

            26L/3464, adopted on 2 December 1970, BISD 18S/97, 102, para. 18.

            27"Japan - Customs Duties, Taxes and Labelling Practices on Imported Wines and Alcoholic Beverages", BISD 34S/83, 115, para. 5.6 (adopted on 10 November 1987).

            WT/DS2/R

            Page 35


            gasoline which was chemically-identical to a batch of domestic gasoline that met its refiner' s individual baseline, but not the statutory baseline levels. In this case, sale of the imported batch of gasoline on the first day of an annual period would require the importer over the rest of the period to sell on the whole cleaner gasoline in order to remain in conformity with the Gasoline Rule. On the other hand, sale of the chemically-identical batch of domestic gasoline on the first day of an annual period would not require a domestic refiner to sell on the whole cleaner gasoline over the period in order to remain in conformity with the Gasoline Rule. The Panel also noted that this less favourable treatment of imported gasoline induced the gasoline importer, in the case of a batch of imported gasoline not meeting the statutory baseline, to import that batch at a lower price. This reflected the fact that the importer would have to make cost and price allowances because of its need to import other gasoline with which the batch could be averaged so as to meet the statutory baseline. Moreover, the Panel recalled an earlier panel report which stated that “the words ‘treatment no less favourable’ in paragraph 4 call for effective equality of opportunities for imported products in respect of laws, regulations and requirements affecting the internal sale, offering for sale, purchase, transportation, distribution or use of products.”28 The Panel found therefore that since, under the baseline establishment methods, imported gasoline was effectively prevented from benefitting from as favourable sales conditions as were afforded domestic gasoline by an individual baseline tied to the producer of a product, imported gasoline was treated less favourably than domestic gasoline.


          7. The Panel then examined the US argument that the requirements of Article III:4 are met because imported gasoline is treated similarly to gasoline from similarly situated domestic parties

            — domestic refiners with limited 1990 operations and blenders. According to the United States, the difference in treatment between imported and domestic gasoline was justified because importers, like domestic refiners with limited 1990 operations and blenders, could not reliably establish their 1990 gasoline quality, lacked consistent sources and quality of gasoline, or had the flexibility to meet a statutory baseline since they were not constrained by refinery equipment and crude supplies. The Panel observed that the distinction in the Gasoline Rule between refiners on the one hand, and importers and blenders on the other, which affected the treatment of imported gasoline with respect to domestic gasoline, was related to certain differences in the characteristics of refiners, blenders and importers, and the nature of the data held by them. However, Article III:4 of the General Agreement deals with the treatment to be accorded to like products; its wording does not allow less favourable treatment dependent on the characteristics of the producer and the nature of the data held by it. The Panel noted that in the Malt Beverages case, a tax regulation according less favourable treatment to beer on the basis of the size of the producer was rejected.29 Although this finding was made under Article III:2 concerning fiscal measures, the Panel considered that the same principle applied to regulations under Article III:4. Accordingly, the Panel rejected the US argument that the requirements of Article III:4 are met because imported gasoline is treated similarly to gasoline from similarly situated domestic parties.


          8. Apart from being contrary to the ordinary meaning of the terms of Article III:4, any interpretation of Article III:4 in this manner would mean that the treatment of imported and domestic goods concerned could no longer be assured on the objective basis of their likeness as products. Rather, imported goods would be exposed to a highly subjective and variable treatment according to extraneous factors. This would thereby create great instability and uncertainty in the conditions of competition as between domestic and imported goods in a manner fundamentally inconsistent with the object and purpose of Article III.


            image


            28"United States - Section 337 of the Tariff Act of 1930", BISD 36S/386, para 5.11 (adopted on 7 November 1989).

            29"United States - Measures Affecting Alcoholic and Malt Beverages", BISD 39S/206, para. 5.19 (adopted on 19 June 1992).

            WT/DS2/R

            Page 36


          9. The Panel considered that the foregoing was sufficient to dispose of the US argument. It noted, however, that even if the US approach were to be followed, under any approach based on “similarly situated parties” the comparison could just as readily focus on whether imported gasoline from an identifiable foreign refiner was treated more or less favourably than gasoline from an identifiable US refiner. There were, in the Panel’s view, many key respects in which these refineries could be deemed to be the relevant similarly situated parties, and the Panel could find no inherently objective criteria by means of which to distinguish which of the many factors were relevant in making a determination that any particular parties were “similarly situated.” Thus, although these refineries were similarly situated, the Gasoline Rule treated the products of these refineries differently by allowing only gasoline produced by the domestic entity to benefit from the advantages of an individual baseline. This consequential uncertainty and indeterminacy of the basis of treatment underlined, in the view of the Panel, the rationale of remaining within the terms of the clear language, object and purpose of Article III:4 as outlined above in paragraph 6.12.


          10. The Panel then noted the argument of the United States that the treatment accorded to gasoline imported under a statutory baseline was on the whole no less favourable than that accorded to domestic gasoline under individual refiner baselines. The United States claimed that the Gasoline Rule did not discriminate against imported gasoline, since the statutory baseline (by the nature of its calculation) and the average of the sum of the individual baselines both corresponded to average gasoline quality in 1990, and that domestic and imported gasoline was treated equally overall. The Panel noted that, in these circumstances, the argument that on average the treatment provided was equivalent amounted to arguing that less favourable treatment in one instance could be offset provided that there was correspondingly more favourable treatment in another. This amounted to claiming that less favourable treatment of particular imported products in some instances would be balanced by more favourable treatment of particular products in others. A previous panel had found that


            the "no less favourable" treatment requirement of Article III:4 has to be understood as applicable to each individual case of imported products. The Panel rejected any notion of balancing more favourable treatment of some imported products against less favourable treatment of other imported products. If this notion were accepted, it would entitle a contracting party to derogate from the no less favourable treatment obligation in one case, or indeed in respect of one contracting party, on the ground that it accords more favourable treatment in some other case, or to another contracting party. Such an interpretation would lead to great uncertainty about the conditions of competition between imported and domestic products and thus defeat the purposes of Article III.30


            The Panel concurred with this reasoning that under Article III:4 less favourable treatment of particular imported products in some instances could not be balanced by more favourable treatment of other imported products in other instances. The Panel therefore rejected the US argument.


          11. The Panel observed that, considered even from the point of view of imported gasoline as a whole, treatment was generally less favourable. Importers of gasoline had to adapt to an assigned average standard not linked to the particular gasoline imported, while refiners of domestic gasoline had only to meet a standard linked to their own product in 1990. Statistics on baselines bore out this difference in treatment. According to the United States, as of August 1995, approximately


            image

            30"United States - Section 337 of the Tariff Act of 1930", BISD 36S/345, para. 5.14 (adopted on 7 November 1989).

            WT/DS2/R

            Page 37


            100 US refiners, representing 98.5 percent of gasoline produced in 1990, had received EPA approval of their individual baselines. Only three of the refiners met the statutory baseline for all parameters. Thus, while 97 percent of US refiners did not and were not required to meet the statutory baseline, the statutory baseline was required of importers of gasoline, except in the rare case (according to the parties) that they could establish a baseline using Method 1.


          12. The Panel found that imported and domestic gasoline were like products, and that since, under the baseline establishment methods, imported gasoline was effectively prevented from benefitting from as favourable sales conditions as were afforded domestic gasoline by an individual baseline tied to the producer of a product, imported gasoline was treated less favourably than domestic gasoline.


      2. Article III:1


          1. The Panel then noted the arguments advanced by Venezuela and Brazil that the Gasoline Rule was applied “so as to afford protection to domestic production” contrary to Article III:1. The United States disagreed and argued in the alternative that Article III:1 was only hortatory and could not form the basis of a violation. The Panel examined first whether, after making a finding of inconsistency with Article III:4, it should make a finding under Article III:1. The Panel noted that the panel in the Malt Beverages case had examined a claim made under paragraphs 1, 2 and 4 of Article III. That panel had concluded that “because Article III:1 is a more general provision than either Article III:2 or III:4, it would not be appropriate for the Panel to consider [the complainant’s] Article III:1 allegations to the extent that the Panel were to find [the respondent’s] measures to be inconsistent with the more specific provisions of Articles III:2 and III:4.”31 The present Panel agreed with this reasoning, and therefore did not find it necessary to examine the consistency of the Gasoline Rule with Article III:1.


    3. Article I:1


        1. The Panel proceeded to examine the claim of Venezuela and Brazil that the Gasoline Rule violated the most-favoured-nation provision of Article I:1 by permitting an importer to use secondary evidence to establish an individual baseline, provided that in 1990 it imported at least 75 percent of the production from an affiliated foreign refinery. Venezuela and Brazil claimed that the rule targeted a small number of countries, and that the different treatment was based on criteria (ownership and proportion of product purchased) that had no link to the product, as required under Article I:1. The United States claimed the rule was based on objective criteria and, in any case, it was not applicable because no importer had qualified for the benefit before the deadline.


        2. The Panel observed that it had not been the usual practice of a panel established under the General Agreement to rule on measures that, at the time the panel’s terms of reference were fixed, were not and would not become effective. In the 1978 Animal Feed Protein case, the Panel ruled on a discontinued measure, but one that had terminated after agreement on the panel’s terms of reference.32 In the 1980 Chile Apples case, the panel ruled on a measure terminated before agreement on the panel’s terms of reference; however, the terms of reference in that case specifically included the terminated measure and, it being a seasonal measure, there remained the


          image


          31"United States - Measures Affecting Alcoholic and Malt Beverages", BISD 39S/206, 270, para. 5.2 (adopted on 19 June 1992).

          32"EEC - Measures on Animal Feed Proteins", L/4599, BISD 25S/49 (adopted on 14 March 1978). See also the Report of the Panel on "United States - Prohibitions of Imports of Tuna and Tuna Products from Canada", BISD 29S/91, 106, para. 4.3 (adopted on 22 February 1982).

          WT/DS2/R

          Page 38


          prospect of its reintroduction.33 In the present case, the Panel’s terms of reference were established after the 75 percent rule had ceased to have any effect, and the rule had not been specifically mentioned in the terms of reference. The Panel further noted that there was no indication by the parties that the 75 percent rule was a measure that, although currently not in force, was likely to be renewed. Finally, the Panel considered that its findings on treatment under the baseline establishment methods under Articles III:4 and XX (b), (d) and (g) would in any case have made unnecessary the examination of the 75 percent rule under Article I:1. The Panel did not therefore proceed to examine this aspect of the Gasoline Rule under Article I:1 of the General Agreement.


    4. Article XX(b)


        1. The Panel proceeded to examine whether the aspect of the baseline establishment methods found inconsistent with Article III:4 could, as argued by the United States, be justified under paragraph (b) of Article XX. The relevant parts of Article XX were as follows:


          Subject to the requirement that such measures are not applied in a manner which would constitute a means of arbitrary or unjustifiable discrimination between countries where the same conditions prevail, or a disguised restriction on international trade, nothing in this Agreement shall be construed to prevent the adoption or enforcement by any contracting party of measures:


          1. necessary to protect human, animal or plant life or health;


          The Panel noted that as the party invoking an exception the United States bore the burden of proof in demonstrating that the inconsistent measures came within its scope. The Panel observed that the United States therefore had to establish the following elements:


          1. that the policy in respect of the measures for which the provision was invoked fell within the range of policies designed to protect human, animal or plant life or health;


          2. that the inconsistent measures for which the exception was being invoked were

            necessary to fulfil the policy objective; and


          3. that the measures were applied in conformity with the requirements of the

            introductory clause of Article XX.


            In order to justify the application of Article XX(b), all the above elements had to be satisfied.


            1. Policy goal of protecting human, animal or plant life or health


        2. The Panel noted the United States argument that air pollution, in particular ground-level ozone and toxic substances, presented health risks to humans, animals and plants. The United States argued that, since about one-half of such pollution was caused by vehicle emissions , and the Gasoline Rule reduced these, the Gasoline Rule was within the range of policy goals described in Article XX(b). Venezuela and Brazil did not disagree with this view. The Panel agreed with the parties that a policy to reduce air pollution resulting from the consumption of gasoline was a


          image


          33"EEC - Restrictions on Imports of Apples from Chile", BISD 27S/98, (adopted on 10 November 1980).

          WT/DS2/R

          Page 39


          policy within the range of those concerning the protection of human, animal and plant life or health mentioned in Article XX(b).


            1. Necessity of the inconsistent measures


        3. The Panel recalled its finding in paragraph 6.16 that imported gasoline was treated less favourably than domestic gasoline, since, under the baseline establishment methods, imported gasoline was prevented from benefitting from as favourable sales conditions as were afforded domestic gasoline by an individual baseline tied to the producer of a product. The Panel then proceeded to examine whether the aspect of the Gasoline Rule found inconsistent with the General Agreement was necessary to achieve the stated policy objectives under Article XX(b). The Panel noted that it was not the necessity of the policy goal that was to be examined, but whether or not it was necessary that imported gasoline be effectively prevented from benefitting from as favourable sales conditions as were afforded by an individual baseline tied to the producer of a product. It was the task of the Panel to address whether these inconsistent measures were necessary to achieve the policy goal under Article XX(b). It was therefore not the task of the Panel to examine the necessity of the environmental objectives of the Gasoline Rule, or of parts of the Rule that the Panel did not specifically find to be inconsistent with the General Agreement.


        4. The Panel then turned to the arguments of the parties relating to that aspect of the Gasoline Rule found inconsistent with the General Agreement. The United States argued that not all entities dealing in gasoline could be assigned an individual baseline and, of those who could be assigned such a baseline, not all could use the same types of secondary or tertiary evidence (Methods 2 and 3) to establish it. Certain entities including importers, blenders and refiners which did not have continuous 1990 operations, were simply not in a position to furnish this secondary or tertiary evidence. Venezuela and Brazil argued on the other hand that foreign refiners should be accorded their own individual baselines under the Gasoline Rule using the same types of evidence, as easily available to them as to domestic refiners. Alternatively, they argued that importers should be able to use individual 1990 baselines established for the foreign refiners with whom they dealt. They noted that an EPA regulatory proposal had even been made along those lines in May 1994. The United States countered that such a proposal would not be feasible because of: (1) the impossibility of determining the refinery of origin for each imported shipment; (2) the incentive to “game” the system thereby handed to exporters and importers; and

          (3) the difficulty for the United States to exercise an enforcement jurisdiction with respect to a foreign refinery, since the Gasoline Rule required criminal and civil sanctions in order to be effective. The United States argued further against the use of foreign refiner baselines by citing

          “equity concerns” of importers that their use would favour those firms that dealt with Venezuelan product, and the existence of particular competitive conditions in the international market, including the flexibility maintained by foreign refiners.


        5. The Panel proceeded to examine whether the United States had in fact demonstrated that the inconsistent measures found to violate Article III:4 were necessary to achieve the stated policy objectives of the United States. The Panel noted that the term “necessary” had been interpreted in the context of Article XX(d) by the panel in the Section 337 case which had stated that:


          a contracting party cannot justify a measure inconsistent with another GATT provision as "necessary" in terms of Article XX(d) if an alternative measure which it could reasonably be expected to employ and which is not inconsistent with other GATT provisions is available to it. By the same token, in cases where a measure consistent with other GATT provisions is not reasonably available, a contracting

          WT/DS2/R

          Page 40


          party is bound to use, among the measures reasonably available to it, that which entails the least degree of inconsistency with other GATT provisions.34


          The same reasoning had been adopted by the 1990 Thai Cigarette panel in examining a measure under Article XX(b). That panel saw no reason not to adopt the same interpretation of

          “necessity” under Article XX(b) as under Article XX(d), stating that


          the import restrictions imposed by Thailand could be considered to be "necessary" in terms of Article XX(b) only if there were no alternative measures consistent with the General Agreement, or less inconsistent with it, which Thailand could reasonably be expected to employ to achieve its health policy objectives.35


          The Panel also noted that while several past panels examining issues under Article XX had identified alternative measures that were reasonably available and fully consistent with the General Agreement, they had also in other instances identified alternative measures that would be “less inconsistent” with the General Agreement. For example, the panel in the 337 case found that, while a general exclusion order applying to imported products was not “necessary”, a limited in rem order could be justified even though it too was inconsistent with Article III:4.36 Recalling its remarks in paragraph 6.22 above, the Panel considered that its task was thus to determine whether the United States had demonstrated whether it was necessary to maintain precisely those inconsistent measures whereby imported gasoline was effectively prevented from benefitting from as favourable sales conditions as were afforded to domestic gasoline by an individual baseline tied to the producer of a product. If there were consistent or less inconsistent measures reasonably available to the United States, the requirement to demonstrate necessity would not have been met.


        6. The Panel then examined whether there were measures consistent or less inconsistent with the General Agreement that were reasonably available to the United States to further its policy objectives of protecting human, animal and plant life or health. The Panel did not consider that the manner in which imported gasoline was effectively prevented from benefitting from as favourable sales conditions as were afforded to domestic gasoline by an individual baseline tied to the producer of a product was necessary to achieve the stated goals of the Gasoline Rule. In the view of the Panel, baseline establishment methods could be applied to entities dealing in imported gasoline in a way that granted treatment to imported gasoline that was consistent or less inconsistent with the General Agreement. If a single statutory baseline applying to all entities — refiners, blenders and importers — was not the chosen regulatory method, then importers could for example be permitted to use a gasoline baseline applicable to imports derived, when possible, from evidence of the individual 1990 baselines of foreign refiners with whom the importer currently dealt. Although such a scheme could result in formally different regulation for imported and domestic products, the Panel noted that previous panels had accepted that this could be consistent with Article III:4.37 The requirement under Article III:4 to treat an imported product no less favourably than the like domestic product is met by granting formally different treatment to the imported product, if that treatment results in maintaining conditions of competition for the imported product no less favourable than those of the like domestic product. Further, these conditions of competition referred to those conditions that were established by government measures and would not therefore include factors such as the “flexibility of individual producers” in this case. The Panel noted finally that a regulatory scheme using foreign refiner baselines, to


          image

          34"United States - Section 337 of the Tariff Act of 1930", BISD 36S/345, para. 5.26 (adopted on 7 November 1989).

          35"Thailand - Restrictions on Importation of and Internal Taxes on Cigarettes", BISD 37S/200, para. 75 (adopted on 7 November 1990).

          36"United States - Section 337 of the Tariff Act of 1930", BISD 36S/345, para. 5.32 (adopted on 7 November 1989).

          37"United States - Section 337 of the Tariff Act of 1930", BISD 36S/345, para. 5.11 (adopted on 7 November 1989).

          WT/DS2/R

          Page 41


          the extent that it did not distinguish between imported gasoline on the basis of its country of origin, would not necessarily contravene Article I or other provisions of the General Agreement, and that the United States, notwithstanding suggestions that certain importers might have equitable concerns, had not established the contrary.


        7. The Panel noted the claims of the United States that allowing importers or foreign refiners to use individual baselines in such a way was not feasible for the reasons listed in paragraph 6.23. The Panel was not convinced that the United States had satisfied its burden of proving that those reasons precluded the effective use of individual baselines in a manner which would allow imported products to obtain treatment that was consistent, or less inconsistent, with obligations under Article III:4. First, while the Panel agreed that it would be necessary under such a system to ascertain the origin of gasoline, the Panel could not conclude that the United States had shown that this could not be achieved by other measures reasonably available to it and consistent or less inconsistent with the General Agreement. Indeed, the Panel noted that a determination of origin would often be feasible. The Panel examined, for instance, the case of a direct shipment to the United States. It considered that there was no reason to believe that, given the usual measures available in international trade for determination of origin and tracking of goods (including documentary evidence and third party verification) there was any particular difficulty sufficient to warrant the demands of the baseline establishment methods applied by the United States.


        8. Second, the Panel did not agree that the United States had met its burden of showing that the “gaming” concern was an adequate justification for maintaining the inconsistency with Article III:4 resulting from the baseline establishment methods. It was uncertain if, or to what extent, gaming would actually occur, especially given the small market share of imported gasoline (approximately 3 percent). Moreover, the Panel noted that the Gasoline Rule did not guarantee in its regulation of US entities that gasoline characteristics subject to non-degradation requirements (i.e. those regulated by baselines), would remain at the 1990 average levels. For example, there was no volume cap on the production of reformulated gasoline by individual refineries, which meant that if producers of relatively dirtier gasoline expanded their relative share of production of reformulated gasoline, the national average level of pollutants subject to the non-degradation requirements would be greater than in 1990. Similarly, within the 1990 volume limitations, if the output of producers of relatively cleaner gasoline fell below 1990 levels, while output of others did not, national average levels of pollutants would be worse. Moreover, specific provisions of the Gasoline Rule permitted some refiners to produce dirtier gasoline than they produced in 1990 (e.g., certain producers of JP-4 jet fuel) and permitted others to request specific derogation from the Rule. The Panel stressed that it was not finding that such events would occur, only that they could under the Rule. Given that the Gasoline Rule did not therefore guarantee that gasoline characteristics subject to non-degradation requirements would remain at 1990 levels, the Panel considered that it was not consistent for the United States to insist that there could be no possible deviation from achieving those levels in respect of imports, when it had not deemed it necessary to be as exacting on its own domestic production. Moreover, slightly stricter overall requirements applied to both domestic and imported gasoline could offset any possibility of an adverse environmental effect from these causes, and allow the United States to achieve its desired level of clean air without discriminating against imported gasoline. Such requirements could be

          implemented by the United States at any time. The Panel concluded that the United States had not met its burden of showing that concern over gaming was an adequate justification for maintaining the inconsistency with Article III:4 resulting from the baseline establishment methods.


        9. Third, the Panel did not accept that the United States had demonstrated that there was no other measure consistent, or less inconsistent, with Article III:4 reasonably available to enforce compliance with foreign refiner baselines, or importer baselines based thereon. The imposition of penalties on importers was in the Panel’s view an effective enforcement mechanism used by the

          WT/DS2/R

          Page 42


          United States in other settings. In the view of the Panel, the United States had reasonably available to it data for, and measures of, verification and assessment which were consistent or less inconsistent with Article III:4. For instance, although foreign data may be formally less subject to complete control by US authorities, this did not amount to establishing that foreign data could not in any circumstances be sufficiently reliable to serve US purposes. This, however, was the practical effect of the application of the Gasoline Rule. In the Panel' s view, the United States had not demonstrated that data available from foreign refiners was inherently less susceptible to established techniques of checking, verification, assessment and enforcement than data for other trade in goods subject to US regulation. The nature of the data in this case was similar to data relied upon by the United States in other contexts, including, for example, under the application

          of antidumping laws. In an antidumping case, only when the information was not supplied or deemed unverifiable did the United States turn to other information. If a similar practice were to be applied in the case of the Gasoline Rule, then importers could, for instance, be permitted to use the individual baselines of foreign refiners for imported gasoline from those refiners, with the statutory baseline being applied only when the source of imported gasoline could not be determined or a baseline could not be established because of an absence of data. In the Panel' s view, because allowing for such a possibility was reasonably available to the United States and would entail a lesser degree of inconsistency with the General Agreement, the United States had failed to demonstrate the necessity of the Gasoline Rule' s inconsistency with Article III:4 on this matter.


        10. In view of the Panel’s finding that the aspect of the baseline establishment methods found inconsistent with Article III:4 was not “necessary” under Article XX(b), the Panel did not proceed to examine whether it met also the conditions in the introductory clause to Article XX.


    5. Article XX(d)


        1. The Panel proceeded to examine whether the aspect of the baseline establishment methods found inconsistent with Article III:4 could, as argued by the United States, be justified under paragraph (d) of Article XX. The relevant parts of Article XX were as follows:


          Subject to the requirement that such measures are not applied in a manner which would constitute a means of arbitrary or unjustifiable discrimination between countries where the same conditions prevail, or a disguised restriction on international trade, nothing in this Agreement shall be construed to prevent the adoption or enforcement by any contracting party of measures:


          (d) necessary to secure compliance with laws or regulations which are not inconsistent with the provisions of this Agreement, including those relating to customs enforcement, the enforcement of monopolies operated under paragraph 4 of Article II and Article XVII, the protection of patents, trade marks and copyrights, and the prevention of deceptive practices;


        2. The Panel recalled that the party invoking an exception under Article XX bore the burden of proving that the inconsistent measures came within its scope. The Panel observed that the United States therefore had to demonstrate the following elements:


          1. that the measures for which the exception were being invoked - that is, the particular trade measures inconsistent with the General Agreement - secure compliance with laws or regulations themselves not inconsistent with the General Agreement;

            WT/DS2/R

            Page 43


          2. that the inconsistent measures for which the exception was being invoked were

            necessary to secure compliance with those laws or regulations; and


          3. that the measures were applied in conformity with the requirements of the

            introductory clause of Article XX.


            In order to justify the application of Article XX(d), all the above elements had to be satisfied.


            1. Securing compliance with consistent laws or regulations


        3. The Panel proceeded to examine whether the aspect of the baseline establishment methods found inconsistent with the General Agreement secured compliance with a law or regulation not inconsistent with the General Agreement. The United States argued that the non-degradation requirements were laws and regulations not inconsistent with the General Agreement, and that the baseline establishment methods secured compliance with these. Venezuela argued that the United States had not clearly established which laws or regulations were not inconsistent with the General Agreement, and with which compliance was secured. Brazil considered that the US measures at most enforced a policy objective, not an actual obligation as required under Article XX(d).


        4. The Panel observed that, assuming that a system of baselines by itself were consistent with Article III:4, the US scheme might constitute , for the purposes of Article XX(d), a law or regulation “not inconsistent” with the General Agreement. However, the Panel found that maintenance of discrimination between imported and domestic gasoline contrary to Article III:4 under the baseline establishment methods did not “secure compliance” with the baseline system. These methods were not an enforcement mechanism. They were simply rules for determining the individual baselines. As such, they were not the type of measures with which Article XX(d) was concerned.38


            1. Other conditions


        5. The Panel observed that, in view of its finding that the less favourable treatment of imported gasoline under the baseline establishment methods accorded to importers did not “secure compliance” with the underlying baseline establishment rules, it did not need to consider also whether these methods were “necessary” to secure compliance and met the conditions in the introductory clause to Article XX.


    6. Article XX(g)


        1. The Panel proceeded to examine whether the part of the Gasoline Rule found inconsistent with Article III:4 could, as argued by the United States, be justified under paragraph (g) of Article XX. The relevant parts of Article XX were as follows:


          Subject to the requirement that such measures are not applied in a manner which would constitute a means of arbitrary or unjustifiable discrimination between countries where the same conditions prevail, or a disguised restriction on international trade, nothing in this Agreement shall be construed to prevent the adoption or enforcement by any contracting party of measures:



          image


          38"European Economic Community - Regulation on Imports of Parts and Components", BISD 37S/132, paras. 5.12 - 5.18 (adopted on 16 May 1990).

          WT/DS2/R

          Page 44


          (g) relating to the conservation of exhaustible natural resources if such measures are made effective in conjunction with restrictions on domestic production or consumption ;


          The Panel noted that as the party invoking an exception the United States bore the burden of proof in demonstrating that the inconsistent measures came within its scope. The Panel observed that the United States therefore had to demonstrate the following elements:


          1. that the policy in respect of the measures for which the provision was invoked fell within the range of polices related to the conservation of exhaustible natural resources;


          2. that the measures for which the exception was being invoked - that is the particular trade measures inconsistent with the General Agreement - were related to the conservation of exhaustible natural resources;


          3. that the measures for which the exception was being invoked were made effective

            in conjunction with restrictions on domestic production or consumption ; and


          4. that the measures were applied in conformity with the requirements of the

            introductory clause of Article XX.


            In order to justify the application of Article XX(g), all the above elements had to be satisfied.


            1. Policy goal of conserving an exhaustible natural resource


        2. The Panel noted the US argument that clean air was an exhaustible resource within the meaning of Article XX(g), since it could be exhausted by pollutants such as those emitted through the consumption of gasoline. Lakes, streams, parks, crops and forests were also natural resources that could be exhausted by air pollution. Measures to control air pollution were therefore measures to conserve exhaustible natural resources. Venezuela disagreed, considering that air was not an exhaustible natural resource within the meaning of Article XX(g); rather, its “condition” changed depending on its cleanliness. Article XX(g) was originally intended to cover exports of exhaustible goods such as petroleum and coal; to expand it to cover “conditions ” of renewable resources was not justified.


        3. The Panel then examined whether clean air could be considered an exhaustible natural resource. In the view of the Panel, clean air was a resource (it had value) and it was natural. It could be depleted. The fact that the depleted resource was defined with respect to its qualities was not, for the Panel, decisive. Likewise, the fact that a resource was renewable could not be an objection. A past panel had accepted that renewable stocks of salmon could constitute an exhaustible natural resource.39 Accordingly, the Panel found that a policy to reduce the depletion of clean air was a policy to conserve a natural resource within the meaning of Article XX(g).


          image


          39"Canada - Measures Affecting Exports of Unprocessed Herring and Salmon", BISD 35S/98, para 4.4 (adopted on

          22 March 1988). See also the same conclusion with respect to dolphins in the Report of the Panel on "United States - Restrictions on Imports of Tuna", circulated on 16 June 1994, DS29/R, para 5.13, not adopted.

          WT/DS2/R

          Page 45


            1. Measures “related to” the conservation of an exhaustible natural resource; and made effective “in conjunction” with restrictions on domestic production or consumption


        4. The Panel proceeded to examine whether the baseline establishment methods found inconsistent with Article III:4 were “related to” the conservation of clean air. Venezuela argued that past panels had interpreted “related to” to mean “primarily aimed at” the conservation of the resource. According to Venezuela, loopholes in the establishment of the baseline undermined its own conservation objectives, and the measure could not therefore be seen as “primarily aimed” at conservation.


        5. The Panel noted that the words “related to” did not in isolation provide precise guidance as to the required link between the measures and the conservation objective. However, the Panel agreed with the interpretation of this term in the report of the 1987 Herring and Salmon case, where the panel stated that


          as the preamble of Article XX indicates, the purpose of including Article XX:(g) in the General Agreement was not to widen the scope for measures serving trade policy purposes but merely to ensure that the commitments under the General Agreement do not hinder the pursuit of policies aimed at the conservation of exhaustive natural resources. The Panel concluded for these reasons that, while a trade measure did not have to be necessary or essential to the conservation of an exhaustible natural resource, it had to be primarily aimed at the conservation of an exhaustible natural resource to be considered as "relating to" conservation within the meaning of Article XX:(g).40 (emphasis added)


          For the same reasons, the Herring and Salmon panel decided that


          the terms "in conjunction with" in Article XX:(g) had to be interpreted in a way that ensures that the scope of possible actions under that provision corresponds to the purpose for which it was included in the General Agreement. A trade measure could therefore in the view of the Panel only be considered to be made effective "in conjunction with" production restrictions if it was primarily aimed at rendering effective these restrictions.41 (emphasis added)


        6. The Panel then proceeded to examine whether the baseline establishment methods could be said to be "primarily aimed at" achieving the conservation objectives of the Gasoline Rule. The Panel recalled the purpose of Article XX:(g), which had been expressed by the panel in the 1987 Herring and Salmon case as follows:


          [T]he purpose of including Article XX:(g) in the General Agreement was not to widen the scope of measures serving trade policy purposes but merely to ensure that the commitments under the General Agreement do not hinder the pursuit of policies aimed at the conservation of exhaustible natural resources.


          The Panel then considered whether the precise aspects of the Gasoline Rule that it had found to violate Article III -- the less favourable baseline establishments methods that adversely affected the conditions of competition for imported gasoline -- were primarily aimed at the conservation of


          image

          40"Canada - Measures Affecting Exports of Unprocessed Herring and Salmon", BISD 35S/98, para 4.6 (adopted on 22 March 1988).

          41Ibidem.

          WT/DS2/R

          Page 46


          natural resources. The Panel saw no direct connection between less favourable treatment of imported gasoline that was chemically identical to domestic gasoline, and the US objective of improving air quality in the United States. Indeed, in the view of the Panel, being consistent with the obligation to provide no less favourable treatment would not prevent the attainment of the desired level of conservation of natural resources under the Gasoline Rule. Accordingly, it could not be said that the baseline establishment methods that afforded less favourable treatment to imported gasoline were primarily aimed at the conservation of natural resources. In the Panel' s view, the above-noted lack of connection was underscored by the fact that affording treatment of imported gasoline consistent with its Article III:4 obligations would not in any way hinder the United States in its pursuit of its conservation policies under the Gasoline Rule. Indeed, the United States remained free to regulate in order to obtain whatever air quality it wished. The Panel therefore concluded that the less favourable baseline establishments methods at issue in this case were not primarily aimed at the conservation of natural resources.


        7. With respect to whether the baseline establishment methods could be said to be primarily aimed at "rendering effective restrictions on domestic production or consumption ", the Panel noted that it had not determined that the measures at issue were “restrictions”, and whether they were “on” domestic production or consumption . However, in light of its finding in paragraph

      6.40, the Panel did not proceed to determine this issue or whether the measure met the conditions in the introductory clause of Article XX.


    7. Article XXIII:1(b)


        1. The Panel then noted the claim by Venezuela under Article XXIII:1(b) that benefits accruing to it under the General Agreement had been nullified and impaired by the application of the Gasoline Rule, whether or not it conflicted with provisions of the General Agreement. In view of the finding by the Panel that the Gasoline Rule violated Article III:4 of the General Agreement, and could not be justified under Article XX (b), (d) and (g), the Panel concluded that it was not necessary to examine this additional claim.


    8. Applicability of the Agreement on Technical Barriers to Trade


        1. In view of its findings under the General Agreement, the Panel concluded that it was not necessary to decide on issues raised under the TBT Agreement.


  7. CONCLUDING REMARKS


    7.1 In concluding, the Panel wished to underline that it was not its task to examine generally the desirability or necessity of the environmental objectives of the Clean Air Act or the Gasoline Rule. Its examination was confined to those aspects of the Gasoline Rule that had been raised by the complainants under specific provisions of the General Agreement. Under the General Agreement, WTO Members were free to set their own environmental objectives, but they were bound to implement these objectives through measures consistent with its provisions, notably those on the relative treatment of domestic and imported products.

    WT/DS2/R

    Page 47


  8. CONCLUSIONS


    1. In the light of the findings above, the Panel concluded that the baseline establishment methods contained in Part 80 of Title 40 of the Code of Federal Regulations are not consistent with Article III:4 of the General Agreement, and cannot be justified under paragraphs (b), (d) and

      (g) of Article XX of the General Agreement.


    2. The Panel recommends that the Dispute Settlement Body request the United States to bring this part of the Gasoline Rule into conformity with its obligations under the General Agreement.