|
THE WTO PANEL DECISION ON U.S. CLEAN AIR ACT REGULATIONS
|
By Steve Charnovitz *
On January 17, 1996, a panel set up by the new World Trade Organization (WTO) issued a decision in the first trade dispute brought to the WTO. This case involves a challenge to American implementation of the Clean Air Act. The panel held that the U.S. regulations, which control harmful emissions from gasoline, violate international trade rules. This decision is now under appeal.
It is not surprising that the first decision handed down by the WTO involves the environment. Many trade analysts are worried that protectionist-inspired trade barriers will increasingly be clothed as environmental measures. In reaction, many environmentalists are worried that the WTO itself will become a barrier to effective ecological protection. According to Elizabeth Dowdeswell, the executive director of the U.N. Environment Program, ``national governments are finding that when playing by liberal trade rules, their autonomy is increasingly constrained by what amounts to a principle of non-interference in the sovereignty of the marketplace.'' 1
To facilitate understanding of the panel's report, it may be helpful to review some key terms. Before the WTO came into being in January 1995, international trade rules were administered by the General Agreement on Tariffs and Trade (GATT). The term ``GATT'' applied both to the international organization and to its trade rules. GATT as an organization has now been absorbed by the WTO. But GATT rules continue as central elements of the comprehensive WTO treaty.
The dispute at the WTO concerns the determination of the baseline. EPA provides three ``historic'' methods for domestic refiners to use. Domestic refiners that began operations after mid-1990 will be unable to use any of these objective methods. Therefore, they are assigned a statutory baseline of the average U.S. gasoline quality in 1990. Because importers (and blenders using imports) might not have adequate and reliable data to use ``historic'' methods, EPA requires that they use the statutory baseline. 2
On behalf of these importers, Venezuela and Brazil complained to the WTO that EPA was applying a different standard to imports than to domestic production. Because foreign refiners do not enjoy individual baselines, most of them_especially the ``dirtier'' producers_are disadvantaged from being assigned the statutory baseline.
EPA's distinction between domestic and foreign producers led to a sharp protest from Venezuela, which subsequently lodged a GATT complaint. In response, the U.S. Department of State negotiated an agreement with Venezuela in early 1994 that EPA would change the regulation if Venezuela withdrew its complaint. EPA followed through with new proposed rules in May 1994 (59 FR 22800) that would have permitted foreign refiners to seek individual baselines for reformulated gasoline. EPA sought public comments and alternative suggestions for treating foreign refiners ``in a manner consistent with provisions of the GATT.''
U.S. environmentalists were unenthusiastic about this deal and the way that the State Department was dictating EPA's rule-making process. Seizing an opportunity to increase market advantage, some large oil corporations (aided by some environmental groups) lobbied the U.S. Senate to stop EPA from implementing the anticipated rules change. Congress responded in September 1994 by forbidding EPA to spend money to implement the proposed regulation. This appropriations rider (108 Stat. 2322) expired in September 1995, but there is current congressional report language indicating that EPA should continue its inaction.
While many public statements were made by U.S. officials, two are especially noteworthy. In April 1994, EPA Assistant Administrator for Air Mary Nichols told a Senate committee that she was motivated by a desire to lean in the direction of doing something that would favor the competitiveness of U.S. petroleum companies vis-a-vis Venezuelan companies. In September 1994, Representative Philip Crane (R-Illinois, now chairman of the Ways and Means trade subcommittee,) warned during House debate that the EPA regulation violated the GATT.
Following congressional restraining action, Venezuela renewed its complaint at the GATT. When it became clear that the WTO would go into force in January 1995, Venezuela decided to refile its complaint at the WTO, rather than the GATT, because of the significant advantages to complaining countries under the new rules. Unlike the GATT, where panel reports required unanimous agreement for adoption, the WTO provides for automatic adoption of panel reports unless every country opposes adoption.
In March 1995, Venezuela requested a dispute resolution panel. The panel was established in April, and Brazil later joined as a co-plaintiff. In addition to hearing the parties, the panel received testimony from the European Union and Norway. The EPA rule was defended by a U.S. team headed by a civil servant from the Office of the U.S. Trade Representative (USTR).
The panel contains three men from the trade community with excellent qualifications to consider a normal commercial dispute. The chairman is Joseph Wong from Hong Kong. He has held several positions in trade relations in the Hong Kong government, including representative to the GATT in Geneva. He is currently a director of home affairs. The second panelist is Crawford Falconer from New Zealand. He has held several positions in New Zealand's Ministry of Foreign Affairs and Trade, including director of the trade negotiations division. He is currently a division head in the Trade Directorate of the Organization for Economic Cooperation and Development. The third panelist is Kim Luotonen from Finland. He has held several positions in Finland's Ministry of Foreign Affairs, including head of the GATT Section and deputy representative to the GATT in Geneva. He is currently director for international economic organizations at the ministry. All three panelists had previously served on GATT panels.
The panel did not hold a public hearing on the case and did not solicit briefs or memorials from non-governmental groups. The panel chose not to establish the ``expert review group'' provided for in the new WTO dispute rules (Article 13). Such a group can be used by a panel to secure technical and scientific advice. USTR did not ask for such a group, nor did it submit to the panel any briefs prepared by environmental groups.
The panel issued its report to all member governments on January 29. 3 In line with a previous court order, pursuant to the Freedom of Information Act, the U.S. Trade Representative released the report to the public. This appears to violate WTO rules, which require that panel reports be kept confidential until final adoption. The WTO itself has not released the panel's report. Indeed, it has not even made a public announcement about the panel's decision.
GATT Article III:4 provides that imported products ``shall be accorded treatment no less favorable than that accorded to like products of national origin in respect of all laws, regulations, and requirements affecting their internal sale.'' The panel sought to determine whether the EPA regulation treated imported gasoline less favorably than domestic gasoline. The panel reached its decision in two steps. First, the panel determined that chemically identical imported and domestic gasoline are ``like'' products because they have the same end uses and are substitutable. Second, the panel determined EPA is treating the imported product less favorably by not giving its refinery the possibility of establishing an historic baseline.
In reaching this decision, the panel considered_and rejected_several defenses presented by the United States. One defense was that not all imported gasoline was disadvantaged by the statutory baseline. The panel conceded this was true, but held that GATT precedent dictated that the importing government could not balance a more favorable treatment of some foreign gasoline with less favorable treatment of other foreign gasoline. Another U.S. defense was that an importer could bring in any particular batch of foreign gasoline so long as it maintained the required annual average. The panel conceded this was true, but held that the EPA rule provided a more favorable sales condition for using domestic gasoline. In other words, the EPA rule denies foreign producers an equal opportunity to compete in the American market.
If that were all the panel had stated, it would have been unassailable. But the panel said more about the meaning of national treatment. Its decision could be viewed as establishing new contours for Article III. It seems to be a significant deviation from recent panel decisions affirming the GATT-legality of domestic regulations.
The current EPA rule treats foreign gasoline less favorably merely because of its foreignness. But what if the EPA rule had been written differently, using neutral criteria? For example, what if EPA had permitted all refiners (foreign and domestic) to qualify for a baseline, assigning the statutory baseline only to those that lacked sufficient reliable evidence. Although the panel does not rule on this hypothetical, its decision can be read as implying that this approach would violate Article III. 4
Since a significant quantity of domestic gasoline_in the same volume range of imported gasoline_was assigned the statutory baseline (e.g., new suppliers), USTR claimed that foreign producers are treated no less favorably than ``similarly situated'' domestic producers. The simplest and orthodox response to that claim by the panel would have been that ``national treatment'' does not mean average treatment (which the EPA rule approximates). Nor does it mean treatment as favorable as some domestic subclass receives. National treatment means that the best treatment offered to domestic products must also be offered to foreign products.
But the panel does not give that orthodox answer to a clear instance of facial discrimination. Instead, the panel states that the EPA rule was linked to certain differences in the characteristics of refiners, blenders, and importers. This was problematic, in the panel's view, because Article III:4 ``deals with the treatment to be accorded to like products; its wording does not allow less favorable treatment dependent on the characteristics of the producer and the nature of the data held by it.''
In expounding Article III, the panel declares that goods should be treated ``on the objective basis of their likeness as products.'' Exposing imported goods to ``variable treatment according to extraneous factors,'' the panel said, would create great instability and uncertainty that would be ``fundamentally inconsistent with the object and purpose of Article III.'' It is unclear what the panel means by ``extraneous'' factors. Is the cleanliness of a refinery's gasoline output in 1990 an extraneous factor? Assume that Refiners A and B have reliable 1990 data and Refiners Y and Z do not. Is it enough that A and B are treated equally and Y and Z are treated equally? Or would a WTO panel view assignment of the statutory baseline to Y or Z as less favorable treatment than A and B receive?
If the gasoline panel is declaring that regulatory distinctions linked to a producer are inconsistent with Article III, that could have profound consequences for national regulatory freedom. Previous panels have not interpreted Article III so broadly. For example, the U.S.-Canada Alcoholic Beverages panel of 1992 stated that ``the purpose of Article III is not to prevent contracting parties from differentiating between different product categories for policy purposes unrelated to the protection of domestic production.'' 5
In the U.S.-European Community ``Section 337'' dispute of 1989, the panel found an Article III violation in the different procedures used by U.S. tribunals to address patent infringements by imported and domestic goods. But the panel evinced no doubt that Article III would allow a distinction between a product made by a manufacturer who holds a patent or license to use a particular processing method and a similar product made by a manufacturer who does not hold such rights. In other words, the Section 337 panel recognized that Article III does not invalidate all regulatory distinctions between substitutable products.
A recent panel report not adopted by the GATT Council seems to concur that Article III should be a limited discipline. In 1994, the U.S.-European Community Automobile Taxes panel stated that the purpose of Article III ``is not to prohibit fiscal and regulatory distinctions'' applied so as to achieve goals other than protecting domestic production. 6 Applying this distinction to the contested U.S. laws, the panel found that imported and domestic automobiles with identical fuel economies could be taxed differently based on the fuel economy of their model type. By contrast, the panel found that the Corporate Average Fuel Economy law violated Article III because it required separate fleet averaging for U.S.- and non-U.S.-made automobiles.
The European Commission blocked adoption of the report by the automobile taxes panel. Last year, the Commission complained that this report ``constitutes a backward step in the interpretation of GATT Article III that risks opening the door for inventive tax and regulatory authorities to discriminate against imported products.'' 7
The gasoline panel seems to agree with the European Commission.
In summary, the gasoline panel is on solid ground in finding the EPA rule a violation of Article III. It is an importer performance requirement that does not apply to all domestic producers. Yet in interpreting Article III, the panel takes a less deferential stance toward national regulation than previous panels have taken. The panel seems to suggest that withholding an individual baseline from a refiner because it lacks the requisite qualifying data would be a violation of GATT Article III. 8 The panel seems to place limits on the regulatory classification a government may use to safeguard its domestic environment.
At the time the WTO came into force, Article XX(b) had been invoked in three environment- or health-related GATT disputes. This defense was rejected by the panel each time. The gasoline panel continues the pattern of rejecting the Article XX(b) defense.
After agreeing that the Clean Air Act fit within Article XX(b), the panel considered whether EPA's distinction between foreign and domestic refinery baselines is ``necessary'' to protect life or health. Relying upon recent GATT decisions, the panel established the following test: A measure is necessary only if there is no ``reasonably available'' alternative measure that is GATT-consistent or less inconsistent with the GATT. Since another GATT precedent assigns the burden of proof to the party invoking an Article XX exception, the burden for the United States involves proving a negative. That is, USTR must prove to the panel the non-existence of alternative measures that could meet the U.S. life or health objective.
USTR argued that there were no such alternative measures. EPA had considered giving foreign refiners individualized baselines, but rejected this option because it did not have reliable data to establish such baselines or to verify the refinery of origin for imported gasoline. Moreover, EPA did not want to allow foreign refiners to choose between a statutory and an individual baseline because that could permit degradation to a dirtier statutory baseline.
The panel was not convinced by USTR's arguments. It declared that alternative measures would be available. For example, the panel suggested the possibility of assigning the statutory baseline to all producers. (This would require an amendment to the Clean Air Act.) It also suggested that foreign refiners could be given individual baselines.
For any hypothetical regulatory measure, there are three important factors: Is the achievement of the government's environmental objective feasible using this measure? Is the measure consistent with the GATT, or less inconsistent than the measure being complained of? Is the measure reasonably available to policy-makers? Unfortunately, the panel did not address these factors in a satisfactory way.
Regarding feasibility, the panel made two key factual findings. First, that a determination of gasoline origin ``would often be feasible.'' Second, that EPA had access to sufficient data for verification of foreign refinery baselines. While these findings may be correct, they are worrisome because the panel members do not seem to have the technical competence for rendering such judgments. Notably, the panel did not establish an experts advisory group. Moreover, the panel's suggestion that the EPA could follow the evidentiary practices used in U.S. anti-dumping investigations is strange advice given the arbitrariness of those investigations.
Although it pointed out that most domestic gasoline does not meet the statutory baseline, the panel did not assess the likely environmental impact of granting individual baselines to foreign refiners. This is a significant omission in the panel's suggestion that a feasible alternative existed. According to EPA's staff, importing more foreign gasoline might lead to more NOx but less air toxics. The panel apparently saw no need to understand the pollution implications of its decision.
The panel does not point to any alternative as being GATT-consistent. The main alternative that it gives attention to would be for EPA to give foreign refiners the opportunity to qualify for individual baselines 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 inadequate data. The panel declares that this alternative ``would entail a lesser degree of inconsistency'' with the GATT as compared to the current EPA rule. But the panel does not explain how it weighs relative GATT inconsistency.
The panel does not separately address the ``reasonably available'' factor. So far, GATT jurisprudence has been silent as to how political constraints should be evaluated. For example, what if USTR had argued that individual baselines were not reasonably available because Congress had forbade spending for this? Most trade law analysts would reject that defense on the grounds that GATT obligations are owed by the U.S. government as a whole, not EPA.
Another issue that is unclear is whether a complaining country can point to alternatives that rely on totally different policy approaches. For example, the U.S. government might be able to achieve its clean air goals through a higher tax on gasoline. Could a panel consider such a tax reasonably available to the United States?
The first issue considered by the panel was whether this exception applied. Venezuela argued, with some validity, that clean air is a renewable rather than an exhaustible resource. But the panel, relying on recent GATT precedents, held that renewable resources could also be exhaustible. In another act of definition, the panel held that clean air was a ``resource'' because it had value. This could be cited by a future WTO panel in holding that certain life forms are not resources because they lack market value.
The next issue considered was whether the EPA rule was ``related'' to the conservation of exhaustible natural resources. Following recent GATT precedent, the panel explained that this meant that the U.S. measure had to be ``primarily aimed'' at such conservation. The panel did not consider the intention behind the EPA gasoline rule. Rather, it considered whether the ``precise aspects'' of the rule that violated GATT Article III were primarily aimed at conservation.
In an elliptical judgment, the panel concludes that these precise aspects are not primarily aimed at conservation. The panel seems to base its decision upon a factual determination that consistency with Article III would not prevent EPA from attaining ``the desired level of conservation of natural resources under the gasoline rule.'' Yet even assuming this factual determination is correct, it is not clear why that matters. Article XX(g) does not require that any deviation from Article III be ``necessary.''
The panel's decision-making approach to Article XX(g) is disturbing. It is hard to imagine any environmental law or regulation meeting its narrow test. Yet the authors of the GATT in 1947 included the XX(g) exception to excuse measures that would otherwise be inconsistent with trade rules. By applying a microscope to ``precise aspects,'' the panel is in danger of missing the broad picture. In other provisions of the GATT that provide for exceptions, the GATT Council looked at the broad picture. For example, GATT Article XXIV permits free trade agreements (e.g., the North American Free Trade Agreement) in which ``restrictive regulations of commerce'' are eliminated on substantially all of the trade involved. But the GATT did not require governments to justify every ``precise aspect'' of remaining commercial restrictions. No ``primarily aimed at'' test is applied to exclusions from free trade agreements.
On the other hand, USTR's invocation of Article XX(g) in this case seems inappropriate. Promoting cleaner gasoline is not a conservation program. Article XX(g) requires a link to domestic production or consumption goals, but there is no production or consumption of clean air. USTR also invoked GATT Article XX(d) which relates to the regulation of business practices. The panel quickly dismissed that argument and was right in doing so.
In summary, the panel considered whether the EPA rule qualified for GATT's exceptions for health and conservation and found that it did not. Since the terms of Article XX(b) and (g) were not met, the panel made no judgment as to whether the United States had complied with the conditions in Article XX's headnote. In reaching these judgments, the panel made some legal interpretations that may serve as precedents in future challenges to national environmental laws. The overall impact of this panel report will probably be to apply a tighter seal to Article XX's environmental window.
This action is significant in clarifying the relationship between the longtime GATT rules and the new TBT rules. During the public debate on the ratification of the WTO, some commentators suggested that the TBT agreement would help countries defend their environmental laws before a WTO tribunal. That viewpoint was wrong. The TBT agreement applies additional requirements for national laws on top of those already in the GATT. So if the panel had found that the EPA rule qualified for Article XX, it would have ruled on TBT since that can override an environmental defense.
It might also be noted that the panel made no mention of the language about the environment in the preamble of the new WTO agreement. During the U.S. debate on joining the WTO, several officials of the administration of U.S. President Bill Clinton pointed to this language as an important step toward greening world trade rules. It may take a few more panel decisions to know for sure what influence the preamble is having.
The most vociferous response has come from Republican presidential candidate Pat Buchanan, who called the decision ``a gross attack on the sovereignty of the United States.'' While several environmental groups expressed unhappiness with the panel's decision, the criticism has been muted because the facts of the case tilt so heavily against the United States. Most of the U.S. newspaper editorials about the decision have supported the panel.
The new WTO rules provide for the automatic adoption of panel reports. These rules are a bit murky as to whether the losing country has a formal obligation to change the offending law or regulation. If the losing country does not act to change its law within a reasonable period, then the rules are clear that the winning country may retaliate with trade sanctions. This differs from the previous procedure under GATT. For example, USTR lost the challenge to U.S. Corporate Average Fuel Economy standards in 1994. But the European Union was not able to retaliate because USTR blocked adoption of the report.
On February 20, Kantor announced that the USTR would appeal the gasoline decision. Under the new rules, appeals are supposed to be decided within 90 days. It is unclear whether Venezuela and Brazil will also appeal.
At least three interpretations developed by the panel are easily challengeable. First, the panel's broad reading of Article III would seem to suggest that even if EPA allowed foreign refiners with adequate data to establish their own baselines, there would still be an Article III violation when adequate data does not exist. Second, the tests utilized by the panel for applying Article XX(b) and (g) are unjustified by any language in the GATT and may be impossible to meet. 9 Third, the panel may have a duty to rule on the TBT complaint since it is separate from the GATT complaint. It would be useful to get these issues resolved because more environment-related disputes are expected over the next year.
While the United States cannot raise procedural concerns about the panel process at the Appellate Body, it can bring them up in other WTO fora such as the Committee on Trade and Environment. There are four main problems with this panel report that exemplify defects in the WTO's dispute settlement process. First, the all-male panel contained three trade bureaucrats, hardly a neutral forum for judging an environmental law. Second, none of the panelists had any environmental regulatory experience so it is unclear whether they had the expertise to render the technical judgments that they did. Third, the panel did not hold a public hearing or entertain briefs from environmental or business groups. Fourth, the panel did not set up an expert review group to help it with making factual determinations. All in all, this is not a model process for how an international organization should adjudicate trade/environment disputes.
In pursuing both an appeal and these procedural initiatives, the Clinton administration could enhance its credibility if it admits that the current EPA rule is GATT-inconsistent and promises to change it quickly. So far, the administration has moved in the opposite direction. 10 By appearing to side both with protectionists who resist trade fairness and with isolationists who resist global order, the administration is undermining America's influence in international trade and environmental fora.
One of the most widely praised advances in the new WTO agreement is the tough dispute settlement system. A country whose law or regulation violates WTO rules will either have to amend the offending provision or suffer a sanction. Yet while greater discipline is useful to apply against protectionist trade measures, it may be counterproductive to apply against true environmental measures.
The EPA gasoline rule at issue here is discriminatory on its face. Its unfavorable treatment of foreign refiners is unquestionably a violation of GATT Article III. The Clinton administration should never have adopted this rule in the first place. Congress should not be blocking a correction.
Nevertheless, the WTO panel goes too far in its broad interpretation of GATT Article III disciplines. Many regulations are aimed at producer practices rather than at products. In seeking to disallow that, the panel could undermine environmental quality.
The panel's decision on GATT Article XX is troubling because the panel makes it too difficult for an environmental measure to qualify for a GATT exception. While the current EPA rule is probably not justified by Article XX, the methodology, logic, and interpretations used by the panel drain meaning from this important exception. The panel seems to have lost sight of the fact that Article XX was written into the GATT to assure that valid concerns about health and conservation could override the presumption in favor of open markets. When the WTO names unbalanced panels and countenances myopic adjudication, it strengthens the hand of politicians who seek to resist multilateralism.
* Steve Charnovitz is director of the Global Environment & Trade Study at Yale University. The opinions expressed in this article are those of the author and not of The Bureau of National Affairs, Inc.
1 Address to Conference on the Role of the United Nations in the 21st Century, Dublin, November 2, 1995.
2 Technically, importers may use their own baseline if they have qualifying ``Method 1'' data. Only one importer has met this exacting EPA standard.
3 WTO, ``United States-Standards for Reformulated and Conventional Gasoline,'' WT/DS2, January 29, 1996. The panel's findings are available at
http://www.gets.org/gets.4 Neither Venezuela nor Brazil argued that this would be an Article III violation.
5 This panel held that tax credits linked to the size of a beer brewery, even when offered to foreign breweries, would violate Article III. Many observers were surprised when the administration of former U.S. President George Bush yielded to the adoption of this decision.
6 For further discussion, see Steve Charnovitz, ``The GATT Panel Decision on Automobile Taxes,'' International Environment Reporter, November 2, 1994, p. 921.
7 European Commission, ``1995 Report on U.S. Barriers to Trade and Investment,'' May 1995, p. 40.
8 The panel states that assigning individual baselines to all refiners ``would not necessarily contravene Article I or other provisions of the General Agreement.'' This begs the question, however.
9 For further discussion of problems with Article XX tests, see Steve Charnovitz, ``Dolphins and Tuna: An Analysis of the Second GATT Panel Report,'' Environmental Law Reporter, October 1994, pp. 10579-80.
10 See ``Kantor Says U.S. Compliance with WTO Panel Not Under Consideration,'' Inside U.S. Trade, January 26, 1996, p. 7.