Reproduced, with permission, from:
Charnovitz, S. 1992. GATT and the environment: Examining the issues. International Environmental Affairs 4 (3): 203-33.
STEVE CHARNOVITZ
Editor's Note. This article is based on a paper presented at a conference, International Trade and Sustainable Development, jointly sponsored by the Centre for Trade Policy and the University of Ottawa Faculty of Law, May 14, 1992.
One year ago, a dispute panel under the General Agreement of Tariffs and Trade (GATT) Found that a U.S. marine mammal conservation law violated international trade rules.[1] This decision[2]--probably the most controversial in GATT's 44-year history--confirmed the fears in some camps that trade rules could hinder environmental efforts. So far, this "Tuna-Dolphin" decision has not been adopted by the GATT Council. But its reverberations continue to be felt in both international trade and environmental policy-making.
The Tuna-Dolphin case has assumed an importance beyond the American dolphin conservation program. In raising the issue of what ecological measures are permitted under the GATT, the panel decision concretizes many of the issues and concerns that underlie the "trade and environment" debate. For example, how does one distinguish between economic protectionism and legitimate environmentalism? May governments be paternalistic about the environments of other countries? May governments be paternalistic about the environments of other countries? Who has competence to impose regulations related to the global commons? Should an international organization be able to override national sovereignty in health or environmental matter? Where does the GATT fit into the hierarchy of international law? What forms of adjuication are appropriate for disputes with important non-commercial dimensions? Is there any way to accommodate the differing points of view about the environment between wealthy and poor countries?
Much has been said about trade and the environment over the past two years. But little progress is being made in reconciling the competing positions. While the debate has generally been constructive, sometimes the trade and environmental policy communities talk past one another.[3] The burgeoning theoretical literature--in economics, law, and ecology--complicates a bridging of the various perspectives.
The ongoing multilateral trade talks add a normative layer to this debate: If GATT rules interfere with environmental protection, how should those rules be changed? If environmental vehicles are being hijacked by protectionists, how can this be prevented?
The purpose of this article is to seek a modicum of synthesis by focusing on a few of the central propositions in the debate. It is my contention that some of the most strongly held views (particularly the Geneva orthodoxy) are incorrect and, in fact, are barriers to resolving the conflicts between GATT and the environment. To support this point, I will analyze and critique several of these key assumptions and arguments--especially those in the Tuna-Dolphin decision and in the GATT Secretariat's recent Report on Trade and the Environment. I will also offer my own recommendations for improving GATTs interaction with environmental issues.
What the [GATT] rules do constrain is attempts by one or a small number of countries to influence environmental policies in other countries not by persuasion and negotiation, but by unilateral reductions in access to their markets.[4]
In February 1992, the GATT Secretariat issued its second major report on trade and the environment. Lauded in trade policy circles, the GATT report, according to The Economist, shows that the GATT is "fighting back" against "often ill-informed criticism from environmentalists, especially in America."[5] The GATTs attack begins with a salvo against the use of trade measures to influence environmental policies in other countries. Whether in the form of laws that seek "to change another's environmental behaviour" or that "attempt to force other countries to adopt domestically-favoured practices and policies," such measures (according to the report) violate the GATT.[6]
The most important point to note about this proposition is its iconoclasm. Twenty-one years ago, in the GATTs first major report on trade and the environment, the Secretariat propounded a different view.[7] That report states that a shared resource, such as a lake or the atmosphere, which is being polluted by foreign producers may give rise to restrictions on trade in the product of that process justifiable on grounds of the public interest in the importing country of control over a process carried out in an adjacent or nearby country.[8] [Emphasis in this and all other extracts and citations is added.]
The iconoclasm of the new GATT thesis becomes more striking when it is recalled that national trade measures have long been used to influence other countries. In 1906, for example, the United States banned the landing and sale of sponges from the Gulf of Mexico gathered by certain armful methods--namely, diving or using a diving apparatus.[9] The purpose of this law was to conserve sponge bets in international waters that were vital to American industry. In 1921, Great Britain prohibited the importation of plumage of any bird.[10] The purpose of this law was to stem the widespread destruction of birds due to the feather trade. In both cases, a nation used trade restrictions to influence environmentally sensitive actions beyond its territorial borders.
Until recently, few would have thought that laws of this type were GATT-illegal. There is, after all, very little in the GATT concerning the intent of a law.[11] It would not seem to matter who might be influenced by a border measure so long as the method of regulation meets the relevant GATT rules (that is, Articles I, II, III, and XI).
What makes the GATT report so unsettling is the suggestion that any environmental import standard that influences foreign behavior may be GATT-inconsistent. It would be one thing for the GATT secretariat to criticize trade sanctions (for example, penalties on unrelated products) used to change environmental behavior.[12] It is quite another to denounce standards establishing non-discriminatory conditions for importation. Because virtually every environmental regulation or standard can influence foreign exporters, the new GATT thesis has radical implications.
If unilateral trade measures used to achieve environmental aims are GATT-illegal, then unilateral trade measures used for other aims are GATT-questionable. For instance, antidumping duties are employed to dissuade the practice of price discrimination.[13] Countervailing duties are employed to influence foreign subsidy policies. Long before the GATT existed, many countries had laws promoting respect for intellectual property rights by threatening an embargo against infringing imports.[14] Why are these kinds of influence appropriate while environmental influence is not?
The U.S. ban on goat cheese from unpasteurized milk is a simple product standard. But it also influences the production patterns of European cheese producers. Is that improper? Actually, any of the minute distinctions in a country's tariff schedule could be construed as an attempt to influence the investment and production decisions of potential foreign exporters.
It can be argued that all of the trade measure illustrated above are intended to influence foreign behavior. It can also be argued that none of them is. The problem with the GATT Secretariat's thesis is that there is no consistent way to draw a line between standards that seek influence and standards that don't.[15] The reason why such a line cannot be drawn is that any tax, standard, or regulation can change the incentive structure for foreign exporters (as well as for domestic producers).[16]
Classifying trade measures
Separating trade measures into two groups--influencing and non-influencing-is not feasible.[17]But distinguishing trade measures by the degree of influence can be a useful avenue for classifying process standards. Three categories might be used:
In the latter two categories, no tuna is allowed, even when a particular catch is fished benignly.[20]
Some analysts have suggested that all trade restrictions based on process standards (sometimes called PPMs[21]) should be disallowed by the GATT. But it won't do merely to invalidate process standards in favor of simple product standards. While there is a basic difference between standards relating to the processing of a product (for example, how it is grown, manufactured, or extracted) and standards relating to the characteristics of a product (that is, purity, size, design, etc.), process standards are sometimes needed to screen the quality of products for health or religious reasons. Nor would it help to try to gauge the intrusiveness of a standard since that is totally subjective. Any product specification--for example, using the metric system--may be too intrusive for someone.
Although it is commonly presented as a pivotal distinction in the trade-and-environment debate, the issue of "influence" shrinks upon close examination. The next two sections will discuss issues that really are pivotal: unliateralism and extrajurisdictionality.
[The GATT] protects trade relations from degenerating into anarchy through unilateral actions in pursuit of unilaterally defined objectives, however valid they may appear.[22]
The GATT Secretariat dislikes unilateralism. In its 35-page report, the term "unilateral" appears 25 times, and never in a favorable light. The GATT's campaign against unilateralism is having some impact. Earlier this year, the UN Conference on Trade and Development adopted a resolution stating that "Unilateral actions to deal with environmental challenges outside the jurisdiction of the importing country should be avoided."[23] The Rio Declaration repeats this statement.[24]
There is an important difference between unilaterally defined and multilaterally defined standards.[25] Nearly everyone would agree that, ceteris paribus, multilateral standards are much better. The continuing progress in attaining harmonized policies for the environment (for example, the Montreal Protocol) and for international commerce (for example, the Brussels Tariff Nomenclature) are certainly very positive developments.
But glorifying multilateral agreements is easier than obtaining them.[26] Since the enactment of the Marine Mammal Protection Act in 1972, the United States has sought an international agreement to protect dolphins from dangerous fishing practices.[27] Until recently, little progress was made.[28] The difficulty in achieving these kinds of treaties has long been recognized.[29]
Faced with a choice between doing nothing (while waiting for an international consensus) and taking action, many nations have opted to impose unilaterally defined standards for internal and external commerce.[30] One can characterize such action as "eco-imperialism," "gunboat environmentalism," economic "righteousness," or "green vigilantism."[31] But name-calling is not likely to stem the incidence of standard-setting. It is probably true that larger countries (with larger markets) are more likely to see their standards attained than are smaller countries.[32] This asymmetry may seem unfair to the smaller countries. But it can also lead the larger countries to feel more responsible for the impact of their consumption patterns.
It is fortunate---since national standards are inevitable---that unilateralism can be good for the environment. For nearly 100 years, there has been a fruitful interplay between unilateral measures and international environmental treaties. For example, the U.S. ban of 1897 on fur seal imports led to the international treaty on seals and sea otters (enforced by trade controls) of 1911.[33] The U.S. ban of 1969 on the importation of endangered species---along with similar action by other nations---spurred the Washington Convention (on International Trade in Endangered Species of Wild Fauna and Flora, or CITES) of 1973.[34] The U.S. government threat (beginning in 1988) to impose trade sanctions against Korea and Taiwan for failing to cooperate in driftnet fishing negotiations and the U.S. ban on the importation of driftnet-caught fish (beginning in 1991) were instrumental in gaining support for and adherence to three U.N. resolutions calling for a moratorium on the use of large-scale driftnets.[35]
The U.S. embargo on Mexican tuna has contributed to the reversal of Mexico's longtime intransigence regarding an intergovernmental agreement on dolphin protection.[36]
Although one might anticipate that unilateralism will continue to be good for the environment, the future is always an open question. The GATT report states that the unilateral use of "negative incentives...reduces the prospects for inter-governmental cooperation on future problems."[37] But the report offers little evidence to support that conclusion. Still, there is a danger that such predictions can be self-fulfilling.
Unilateralism is also good for the environment because it assists sovereign nations in achieving their own ecological goals. Since nations face different environmental challenges and have different values and temporal preferences, it is natural that countries will want to formulate their own standards for production, consumption, and disposal---which could apply to imported as well as domestically produced goods. A world where countries marched in environmental lockstep would depress standards to the lowest common denominator.[38]
There is also another reason to allow each country to fashion its own standards for what its citizens produce and consume---namely, the value of competition (that is, competing on the quality of governmental regulation). Since the "proper" level of environmental protection is rarely apparent, one way to determine it is by "letting a hundred flowers blossom." Any country that strategically manipulates imports by imposing unreasonably high environmental regulations should see its standard of living fall.
The considerations that led the panel to reject an extrajurisdictional application of Article XX(b) therefore apply also to Article XX(g).[39]
Last year, in the Tuna-Dolphin dispute, the GATT panel determined that "extrajurisdictional" trade restrictions were not included within the scope of GATT Article XX (General Exceptions).[40] In addition to its impact on the vitality of dolphins, this decision has implications for a broad range of environmental treaties and laws that are equally extrajurisdictional.
What is extrajurisdictionality?
Because the core of its decision rests on the concept of extrajurisdictionality, one might think that the GATT panel---in inventing the term---would have paused to define it. Since the panel did not, one can only induce from context that "extrajurisdictionality" means a law concerning activities that occur outside one's country.[41] Whether the term covers a law applying simultaneously to domestic and non-domestic activities remains unclear.[42] Also unclear is the exact boundary of a "domestic" or "jurisdictional" objective.[43]
One thing that extrajurisdictionality does not mean is extraterritoriality.[44] Extraterritorial laws impose domestic standards on transactions occurring in foreign countries. For example, the recent decision by the administration of President George Bush to apply U.S. antitrust law to Japanese companies in Japan is an application of extraterritoriality. In addition, laws that regulate foreign use of domestic-origin goods or the behavior of domestic corporations abroad are extraterritorial.[45] Although the Tuna-Dolphin panel did not confuse the two issues, "extraterritoriality" is commonly misused to describe standards or conditions for voluntary commerce.[46]
Article XX(b)
As the Tuna-Dolphin panel stated, Article XX(b) "refers to life and health protection generally without expressly limiting that protection to the jurisdiction of the contracting party concerned."[47] The panel could have stopped with that textual explication. Instead, the panel decided to examine the history of Article XX(b). Unfortunately, the panel presented an incomplete and misleading reading of that history.
The panel's conclusion that Article XX(b) cannot be extrajurisdictional is based on the fact that during one of the preparatory sessions of the UN Conference on Trade and Employment, a proposed amendment to Article XX(b) that might have precluded extrajurisdictionality was dropped.[48] (The UN Conference of 1946-1948 wrote the Charter for the International Trade Organization (ITO), as well as the GATT.[48]) As several commentators have noted, this line of reasoning is weak.[50] More importantly, the panel fails to take into account either the historical context of the "life and health" exception in trade treaties or the laws in existence in 1947 that might have motivated such an exception.
Trade treaties have provided exceptions for the protection of humans, animals, and plants since the late 19th century.[51] There is ample indication that these exceptions were understood as applying to extrajurisdictional laws. For example, in the International Convention for the Abolition of Import and Export Prohibitions and Restrictions (of 1927), the article listing exceptions (on which GATT Article XX is based) includes measures to preserve animals and plants from "degeneration or extinction."[52] This language was needed to assure that the "abolition" would not apply to the contemporaneous controls on the importation of birds, seals, salmon, halibut, wildlife trophies, etc.
There was very little ITO preparatory debate on the scope of Article XX. It is sometimes suggested that the GATT's authors never contemplated extrajurisdictional use. A better interpretation, I believe, is that they understood that Article XX(b) would apply to extrajurisdictional measures, but considered that point so obvious that it did not engender debate. Certainly, the record fails to show anyone at the UN Conference suggesting that Article XX(b) should not apply extrajurisdictionally. Moreover, it seems evident that the United States---whose 1946 draft of Article XX(b) emerged unscathed in the GATT---perceived its text as covering U.S. import prohibitions in effect at that time, which included extrajurisdictional measures. For instance, the United States had enacted a law in 1936 to ban the importation of certain whale species.[53]
In light of the criticism of the Tuna-Dolphin report, some trade officials have proposed a broader version of jurisdictionality. That is, Country A can invoke Article XX(b) to cover any production (no matter where it is located) that directly affects the life or health of people in Country A, to cover any production in Country A (even when exported), or to cover living organisms in the global commons. Conversely, Article XX(b) cannot be invoked to cover production occurring in foreign jurisdictions that does not directly affect the people of Country A. Although this alternative would be far better than the Tuna-Dolphin decision, there would continue to be disagreements as to what directly affects the people of Country A.[54]
Article XX(g)
The Tuna-Dolphin panel's conclusion that there is no extrajurisdictionality in Article XX(g) was not premised upon an analysis of the GATT's preparatory history. Instead, the panel relied upon a scholastic argument based on an interpretation of Article XX(g) suggested in a previous GATT case.[55] One can object to the panel's argument,[56] but more revealing is what the panel did not say.
The panel ignored the ITO preparatory history of Article XX(g), which demonstrates rather clearly that the GATTs authors did not want to hinder international fish and wildlife conservation efforts.[57] It is true that almost all of this history was in the context of the provision similar to Article XX(g) in the commodities chapter (of the ITO Charter). But in considering its scope, there is no reason to presume that the drafters were environmentally cosmopolitan in one part of the ITO Charter and environmentally nativistic in another.
Ecological objections
Even if the Tuna-Dolphin panel were correct about the original meaning of Article XX, there would still be good ecological reasons to reject jurisdictionality as a GATT principle. Although both the Tuna-Dolphin decision and the GATT Report attempt to distinguish between a nation's own environment and the rest of the world's environment, this segregation is unhelpful in dealing with natural resources not located in any country's jurisdiction (for example, the ozone layer) or with resources that migrate (for example, birds).[58] If no country is permitted to take extrajurisdictional action, then much of our biosphere would be unreachable by environmental trade measures.[59]
Environmentalists also argue that even when living organisms lie within the territory of a particular country, other countries ought to be able to ensure that their own actions (for example, importing) do not indirectly harm endangered animals and plants. There are important medical reasons to preserve biodiversity. But there are also important moral reasons. Geopolitical boundaries should not override the word of God who directed Noah to take two of every living creature into the Ark "to keep them alive with you."[60]
The Panel considered that if the broad interpretation of Article XX(b) suggested by the United States were accepted, each contracting party could unilaterally determine the life or health protection policies from which other contracting parties could not deviate without jeopardizing their rights under the General Agreement.[61]
In determining the scope of Article XX, the Tuna-Dolphin panel took into consideration the "consequences" of accepting the U.S. government's interpretation.[62] One serious consequence, according to the panel, would be that the United States could unilaterally set standards for other countries from which they "could not deviate without jeopardizing their rights under the General Agreement." But the panel's analysis rests on a petitio principii fallacy. That is, the panel assumes what it tries to prove.
If contracting parties had GATT rights to export without impediment, then it would be clear that unilaterally minded nations could not impose their own import standards. But the GATT does not guarantee the acceptability of one's exports. The GATT has rules against import bans (Article XI), but also has exceptions to those rules (Article XX). Both the rules and the exceptions have to be considered together to ascertain, in any particular dispute, whose "rights" should be upheld. Thus, the Tuna-Dolphin panel errs in assuming that a country facing foreign import standards automatically has rights being violated.
The case in which the Tuna-Dolphin panel slips into this logical fallacy betrays a serious problem of Article XX adjudication over the past several years--that is, the practice of assigning the burden of proof to the party relying upon an Article XX exception.[63] This is not the only way to conduct adjuication.[64] The burden of proof could be shifted to the party alleging an improper trade barrier.
Regardless of which side should have the burden of proof, the current procedures have been unfair. In case after case, GATT panels have narrowed the Article XX defense while ruling against each defendant on the grounds that it "had not demonstrated to the panel" one or another of the ever-expanding list of qualifications for using Article XX.[65] Since these increasingly stringent tests are being created by panels on an ad hoc basis, national authorities may not know whether, at any given time, their environmental policies conform to the Article XX standards. This unpredictability also makes it difficult for governments do defend their policies to GATT panels.
In considering how the GATT's General Exceptions ought to be properly applied, it should be recognized that Article XX does not create or confer rights to restrict trade. It acknowledges such rights. The categories in Article XX are not potential exemptions to GATT discipline. They are exceptions to GATT dominion.[66]
Although the parties signing the GATT agreed to curb their trade restrictions, they drew a line at health-related controls so long as these were neither discriminatory nor protectionism in disguise. Had Article XX not been part of the GATT, the GATT would not have existed. The fact that nearly every treaty on trade in this century has included an exception for health measures demonstrates the unwillingness of nations to yield sovereignty in this area.[67] When U.S. Secretary of State Cordell Hull, in 1933, first suggested an international agreement to reduce tariffs and other trade barriers, his plan provided for the "exceptions generally admitted in existing treaties, for purposes of safety, sanitation, plant and animal protection, morals, etc..."[68] In their efforts to delimit Article XX, several recent GATT panels have blasted at the foundations on which the GATT was built.
...the provisions of the General Agreement impose few constraints on a contracting party's implementation of domestic environmental policies.[69]GATT rules, therefore, place essentially no constraints on a country's right to protect its own environment....[70]
It is said that the GATT does not impose very many constraints on national environmental laws. Yet the few that the GATT does impose could interfere with scores of existing laws that rely on trade instruments. Furthermore, the GATTs few constraints are rapidly tightening.
The mutating "necessary" test
GATT Article XX(b) provides an exception for measures "necessary to protect human, animal or plant life or health."[71] The term "necessary" received little attention at the ITO preparatory meetings.[72] There is no indication that the drafters contemplated disputes regarding sanitary measures turning on the meaning of "necessary." Although the importance of guarding against an abuse of health standards was certainly a topic of discussion, it was Article XX's headnote that was viewed as providing most of the needed discipline.[73]
The ITO documentation suggests that disputes under Article XX(b) were to be resolved on the basis of a scientific test.[74] A recent legal challenge against a U.S. Environmental Protection Agency (EPA) regulation on asbestos presents a good example of the way GATT's authors seemed to anticipate that disputes over health restrictions would be framed. Canada claimed that the EPA ban on asbestos (which also applies to imports) "is not supported by the international scientific evidence, and is therefore not 'necessary' within the meaning of Article XX of the GATT."[75] (The point here is not the scientific merit, or lack thereof, of the Canadian position, but rather that Canada presented a science-based argument.)
How should the GATT Council deal with situations of this type where there is significant scientific uncertainty? So far, Canada has not taken this complaint to the GATT.[76] If Canada does, then a GATT panel could be asked to decide whether there is enough evidence that asbestos is harmful. Yet the GATT lacks criteria for making such a determination. A few months ago, the Business Council for Sustainable Development recommended that "where environmental threats are particularly serious or irreversible, GATT should adopt the precautionary principle, erring on the side of prudence."[77]
For the first four decades of GATT history, the discipline for the health exception, insofar as it existed, was assumed to be science-based. But a few years ago, a GATT panel invented a new scheme for interpreting "necessary."[78] Under this test, a health-related trade measure would be considered "necessary" under Article XX(b) only if there were no alternative measures less inconsistent with the GATT which a country could reasonably be expected to employ to achieve its policy objectives.[79] This parsing has come to be known as the least GATT-inconsistent test.[80]
Applying this test requires policy analysis.[81] First, the panel needs to determine whether there are alternative measures that would be at least as effective in achieving the country's environmental goals. Then the panel must examine such alternatives to see if any of them are less GATT-inconsistent than the trade measure in dispute. If so, then the disputed measure will not qualify under Article XX(b).
Because this test is so open-ended, there is a danger of "runaway" GATT panels second-guessing national laws. Virtually any trade measure (for example, a ban on hormone-fed beef) could be replaced by a labeling requirement on the grounds that "consumer choice" is less GATT-inconsistent.[82] There are two main problems with relying on labels to achieve environmental or health goals. One is that consumers may act rationally in calculating that their individual purchase of environmentally unfriendly products (for example, chlorofluorcarbons, or CFCs) would have only a negligible effect on the ecosystem.[83] The other is that consumers may act irrationally by not properly weighing the implications of low probability events. Although there are some instances where governments mandate labels for unsafe food or drugs (for example, cigarettes), the more common approach is proscription.
The Uruguay Round is considering a third hurdle for environmental trade measures--a least trade-restrictive test.[84] Two of the proposed agreements in the round, the "Standards Code" and the "Sanitary and Phytosanitary Decision," would impose this type of test for product standards and regulations.[85] Under the draft Standards Code, regulations "shall not be more trade restrictive than necessary to fulfil a legitimate objective..." The difference between the least GATT-inconsistent and the least trade-restrictive tests is that the former uses a legal scale while the latter uses an economic scale. Imposing these two tests interdependently would significantly tighten GATT's discipline.
Perhaps out of impatience with sluggish multilateral negotiations, a recent GATT panel decided to adopt the least trade-restrictive test. In the U.S. Alcoholic Beverages case (Beer II), the panel found that certain state laws could not meet the "necessary" test under Article XX(d)[86] because they were not the "least-trade restrictive" enforcement measures available.[87]
Although the panel's unilateral attempt to create a new GATT standard is disturbing, even more disturbing is the manner in which the panel administers its newly minted standard. Consider one example: the issue is whether the laws of five states violate the GATT by requiring that alcoholic beverages be imported into the state by common carrier. The Bush administration did not contest that these state laws violate GATT Article III (since in-state producers can use their own transportation), but argued that these laws could be justified under Article XX(d). The panel rejected this Article XX(d) defense by declaring that
the United States has not demonstrated that the common carrier requirement is the least trade restrictive enforcement measure available to the various states and that less restrictive measures, e.g. record-keeping requirements of retailers and importers, are not sufficient for tax administration purposes.[88]
But the panel rationalized that since "not all fifty states of the United States maintain common carrier requirements...[i]t thus appeared to the panel that some states have found alternative, and possibly less trade-restrictive, and GATT-inconsistent, ways of enforcing their tax laws."[89]
In other words, the panel concludes that the mere existence of unharmonized state laws shows that alternative methods are available for enforcement. The panel did not consider whether the five states had the same tax goals, or the same alcohol policy goals, as the 45 other states. The panel did not consider whether the five states might have special needs for their laws that do not exist in the 45 other states. The panel did not consider whether the alternative methods used in the 45 other states would be effective in achieving the policy goals of the five states. The panel did not consider any differences in health objectives among the five states. The panel did not even consider whether any of the 45 alternative state measures were, in actuality, less trade-restrictive. One can only hope that this panel's cursory technique will not become the model for implementing the "least trade-restrictive" rule of the Uruguay Round.
Proportionality[90]
Another GATT constraint on national environmental measures is the principle of proportionality, especially as it has developed in the European Communities (EC).[91] Traditionally, the European Court of Justice has used a relative proportionality approach to require the means which least restricts the free movement of goods.[92] Yet in more recent adjudication (that is, the Danish bottle case), the court has moved toward an absolute proportionality test which considers a "balancing of interests between the free movement of goods and environmental protection..."[93]
EC jurisprudence has no automatic transferability to the GATT. Nevertheless, the concept of weighing commercial versus environmental objectives is gaining influence among trade policy specialists.[94] It has also appeared in adjudication under the Canada-U.S. Free Trade Agreement.[95] The Dunkel Text for the Standards Code would normally introduce the concept of "proportionality" into the GATT.[96]
Under the principle of absolute proportionality, the GATT would judge the acceptability of a national trade restriction by weighing its commercial costs against the environmental benefits.[97] There are two ways the GATT could do this.[98] First, by entering the minds of that country's policy-makers and using their national preference function.[99] Second, by using a transnational preference function in the manner of the EC.[100] Under either method, the GATT would be setting a maximum standard by deciding that a country could not value an environmental improvement any more than X cost in trade.[101]
A GATT omniscient enough to prescribe a maximum standard for environmental protection could also prescribe a minimum one. Certainly, a minimum standard would not be a necessary implication But if the GATT tells some countries what they cannot do to protect the environment, there will be countervailing pressure to dictate to other countries what they must do to protect it. If the GATT follows this approach, it would be deciding that a country could not value an increase in trade any more than Y cost in environmental degradation. Although this falls far short of harmonization, it would be a significant step toward policy convergence.
The issue of whether it is desirable to transform the GATT into an institution that would foster the coordination of environmental policies is beyond the scope of this article.[102] But the difference between international harmonization and regional harmonization should be noted. Even if one doubts the practicality of a GATT role in harmonization, one could still favor steps toward the convergence of environmental policies within any plurilateral trade agreement. Although environmental convergence is not a precondition of a regional trade agreement or customs union (or, for that matter, of a federal nation of states or provinces), the benefits of such convergence are becoming increasingly apparent.
GATT rules could never block the adoption of environmental policies which have broad support in the world community.[103]
Although a multilateral treaty is unlikely to violate the GATT, action by parties to implement such a treaty could be inconsistent with GATT obligations.[104] One reason why the Tuna-Dolphin panel invoked such an outcry among environmentalists is that the logic of the decision applies equally to numerous environmental treaties. Sometimes the GATT Secretariat has tried to play down this problem. For example, earlier this year, GATT's director general explained that: "If, in Rio, governments can negotiate environmental agreements with universal participation, then whatever trade provisions may be included in those agreements, no controversy need arise over them in GATT.[105] (Of course few, if any, multilateral agreements have universal participation.) At other times, GATT officials have acknowledged the latent conflicts. For example, the GATT report admits that the availability of Article XX for treaties like the Montreal Protocol is untested, and opines that the discriminatory provisions in such treaties may not be "necessary."[106]
The potential GATT inconsistencies of international environmental agreements like the Montreal Protocol, the Basel Convention, and the International Whaling Commission are too lengthy to be detailed here.[107] It should be noted, however, that despite the Tuna-Dolphin report, some discriminatory provisions continue to be adopted. At the 1992 CITES Conference, the parties recommended that endangered species trade with non-parties occur "only in special cases" and "only after consultation with the [CITES] Secretariat."[108] Yet in other arenas, the Tuna-Dolphin report seems to be having a chilling effect. For example, the International Convention for the Conservation of Atlantic Tunas recently backed away from a new trade-based enforcement mechanism because of potential GATT complications.[109]
Several approaches are being suggested for how environmental treaties might be reconciled with the GATT. None of them offers much promise for escaping the dilemma created by the Tuna-Dolphin report. Four will be discussed briefly:
1. ARTICLE XX(H).[110] This provision provides an exception from GATT rules for measures "undertaken in pursuance of obligations under any intergovernmental commodity agreement" if the agreement either conforms to a United Nations Economic and Social Council resolution of 1947 or is submitted to the GATT and not disapproved.[111] (A third option would exist if the GATT adopted criteria for commodity agreements, but this has not happened.[112]) No disputes have occurred regarding this exception.[113]
The term "commodity agreement" is not defined in the GATT. One of the various purposes for commodity agreements, according to the ITO Charter, is "to maintain and develop the natural resources of the world and protect them from unnecessary exhaustion."[114] Most international environmental treaties could be construed as commodity agreements. For instance, CITES could be viewed as regulating trade in the "commodity" of endangered species. The Montreal Protocol could be viewed as regulating trade in the "commodity" of CFCs and halons.
What makes this approach unsuitable is that it is so clearly inconsistent with the framework of the ITO, which sought to exclude wildlife treaties from being disciplined as commodity agreements.[115] Since the GATT's General Exception for commodity agreements was to be available only for agreements that met the disciplines of the ITO Commodities chapter, it would seem contradictory to grant exceptions for wildlife treaties without regard to that discipline.[116] In addition, a future panel might question why Article XX(h) can be any more extrajurisdictional than Article XX(b) or (g).
2. GATT WAIVER. Under Article XXV:5, the GATT may grant a waiver "in exceptional circumstances" by a supermajority consisting of more than half of all contracting parties and two-thirds of those voting. There are several problems with trying to accredit environmental treaties through waivers.
First, a GATT waiver is meant only for exceptional circumstances. Environmental treaties are increasingly unexceptional.
Second, a pending Uruguay Round "Understanding" would clarify the right of a GATT member to lodge a complaint even when a waiver exists.[117]
Third, the supermajority voting requirement is a high hurdle. Various forms of side payment might be needed to garner the requisite vote, and this may further polarize the GATT along North-South lines. The Uruguay Round provision that waivers be renewed annually could multiply the cost of such side payments.
Fourth, international environmental agreements often go into force with a small nucleus of countries that may fall far short of two-thirds of the GATT. For example, CITES went into force in 1975 with just ten countries. Now it has 115. Regional agreements might also have a difficult time gaining a GATT supermajority.
Fifth, although it is commonly suggested that widespread adherence to an environmental treaty (like CITES) would automatically translate into GATT approval, there may be situations when a government's environmental or fisheries ministry holds different views than its commercial or external affairs ministry.[118] Yet it will be trade officials who cast each country's vote in the GATT.
In addition to these procedural problems, there is a serious substantive concern---namely, how GATT determines whether a waiver is warranted. Is GATT going to weigh each treaty's objectives against other economic goals? Is GATT going to decide whether a treaty is "necessary"? Another troubling aspect of the waiver approach is the suggestion in the secretariat's report that the GATT should set "conditions designed to avoid abuse."[119] Since it is hard to apply conditions to an already-negotiated treaty, the GATT council may seek to insert itself into treaty negotiations. But the Council is ill-suited for such a role, for both political and institutional reasons.
3. GATT AMENDMENT. Many observers have inferred from the Tuna-Dolphin report that the panel invites an amendment to solve the GATT-environment conflict.[120] The panel's ambiguous comment has spurred numerous proposals for GATT amendments from environmental and trade experts.[121] But amending the GATT is difficult. It has not happened since 1965 and the requirement for unanimity is a formidable one.[122] By implying that amendments are a feasible course, GATT officials have raised environmentalists' expectations that are unlikely to be fulfilled. This may lead to further frustration and cynicism about the GATT.
4. OVERRIDING TREATIES. Another way out of the dilemma would be to determine that obligations of certain environmental treaties (like CITES) override obligations of the GATT. This could occur under the rules of international law regarding more recent treaties.[123] There are three main problems with this "trumping" approach. First, the legal issues are too complicated to settle the conflict in the public's mind.[124] Second, GATT members who are not parties to a particular environmental treaty cannot have their GATT rights revoked.[125] For example, there are 20 parties to GATT that are not parties to CITES.[126] Third, the Uruguay Round would reset GATT's effective date to 1993, thus making it the most recent "treaty."
If the GATT contracting parties wished to permit environmental trade restrictions...they would need to agree on limits to prevent abuse. Since Article XX does not provide such limits, the Panel stated that it would be better to amend or supplement to provisions of the General Agreement or to provide a waiver...[127]
It is only a little surprising that the Tuna-Dolphin panel was unable to locate any limits in Article XX. After all, one of the most important limits---the "disguised restriction" proviso---has atrophied from inattention.[128] Various GATT panels have been so busy devising new interpretations that they have given short shrift to enforcing the limits already on the books. Yet the requirements in the Article XX headnote---namely, non-discrimination, national treatment, and no disguised restrictions---would, if properly policed, be perficient in weeding out illegitimate use of environmental trade measures. No additional limits, such as proportionality, are needed.[129]
Admittedly, this purist view of Article XX would lead to all kinds of unilateral, extrajurisdictional, and intrusive trade measures. But the GATT and the world economy would survive.[130] Indeed, the GATT might be strengthened if it did more to combat disguised protectionism and less to antagonize environmentalists. The unwillingness of the GATT Council to recognize the deep flaws in its parochial Article XX interpretation is undermining political support for the Uruguay Round.[131] It should be clear by now that preaching environmental abstinence is thinning the ranks of free traders.[132]
The United States has been the leading user of environmental trade measures. What would the world be like if all countries acted like the United States in this regard? For example, what if the EC blocked imports of fur from countries permitting the use of leg-hold traps?[133] Or if Country B banned cosmetics made using the Draize test (which squirts irritating liquids into the eyes of a helpless rabbit)? Or if Country C banned tuna from nations that kill any dolphins? Here is one answer: so long as such laws are non-discriminatory and national treatment is applied, the world would be a more salutary place for beavers, rabbits, and dolphins--at a cost consuming nations are willing to bear.[134]
Since the meaning of the limits in Article XX's headnote are not well defined, it would be useful for the GATT Council to develop guidelines on interpreting these rules.[135] The key issue is whether a questionable measure is crafted in such a way as to favor domestic over foreign suppliers. Several factors might be considered:
First, is the national measure applied to an unduly narrow range of products? For example, Ontario imposes a 10 cent tax on beer cans, but not on beer bottles, which Canadian producers predominantly use.[136] There could be a reasonable environmental distinction between cans and bottles. But if so, why does Ontario's tax not apply to cans of soft drinks, juice, soup, etc.?
Second, is the national measure restricted geographically? Mexico and Venezuela raised a valid concern about "discrimination" in pointing out that the U.S. tuna import ban applies only in the eastern tropical Pacific.[137] A similar concern could arise with a ban on the importation of unsustainably harvested timber.[138] Is the country that is interested in sustainable timber abroad doing all it can to sustain timber at home?
Third, do national environmental standards imply a greater risk aversion for goods that are imported? The new provision in the draft Sanitary and Phytosanitary Decision relating to the internal consistency of national risk-avoidance goals could be a useful approach to dealing with this problem.[139] But a foolish inconsistency is not necessarily protectionism.[140]
Fourth, is the environmental measure---in actuality---more burdensome to foreign producers and consumers than to domestic ones?[141] The more that a national measure shifts the costs abroad, the more questionable such a regulation becomes. This economic distinction can be illustrated by looking at three examples. Start with a U.S. law prohibiting the importation and domestic sale of dolphin-unsafe tuna. Such a law (if the standard is met) raises the cost of tuna to American consumers. Since the law applies only to foreign production bound for the American market, there would be no increase in the cost of tuna to foreign consumers.[142] Next, consider a U.S. law prohibiting the importation of tuna from any country that uses dolphin-unsafe fishing methods. Such a law (if the standard is met) raises costs for foreign as well as American tuna consumers. Last, consider a U.S. law threatening an embargo of widgets from any country that allows the internal sale of dolphin-unsafe tuna. Such a sanction (if the threat were effective) raises costs only for foreign tuna consumers.
Fifth, does a process standard include de facto trade performance requirements? For example, the Canadian forest products industry has complained that U.S. (state government) recycled content laws--which require that virgin pulp be mixed with waste paper--disproportionately affect countries like Canada with low quantities of waste paper output. In order to comply, Canada has to import waste paper for mixing.[143]
There are several steps the GATT should take to improve its interface with the environment.
Procedural
Tuna-Dolphin dispute
The Tuna-Dolphin decision should not be left in limbo. Delaying action on any report encourages other parties to do the same, and thereby weakens the dispute settlement process. The single most important action the GATT can take to improve the environment is to reject the Tuna-Dolphin report. Rejecting the report would not imply that the U.S. law is GATT-consistent, since surely it is not. Rather, it would be an admission that the panel's decision is fatally flawed. Unfortunately, this is unlikely to happen. For one thing, the GATT Council has never rejected a report. But more importantly, no nation has sided with the United States at the GATT Council.[151]
The recent suggestion by the EC that they will file a Tuna-Dolphin complaint against the United States offers an opportunity to secure a newly appointed panel to reconsider this dispute in light of what has been learned during the past year.[152] The United States could demonstrate its respect for the GATT by amending the Marine Mammal Protection Act to reform the dolphin kill-rate calculation procedures that clearly violate GATT rules.[153] This could be done without diminishing protection for dolphins. Alternatively, the import prohibition could be rewritten as a defiled-item standard.
New GATT rules
Although a resuscitated Article XX could cover most legitimate environmental trade measures, there are a few areas---beyond the scope of this article--where a revision in GATT rules might be desirable:
A consensus is building that after the Uruguay Round, the GATT should move quickly to address environmental issues.[159] But there is no consensus on what needs to be done. At one end of the spectrum is the idea of broadening GATTs mission to include the goal of "sustainable development."[160] At the other end is the idea of tightening the GATT so that (nth-best) trade instruments cannot be used for environmental protection. In between are numerous proposals to make GATT either more environmentally friendly or less susceptible to eco-protection. Suggestions have also been made for increases in the transfer of technology, financial resources, and property rights to developing countries. If a Green Round commences without greater agreement, it will be a painful negotiation.
Although this article has been critical of the GATT Secretariat's report, it should be noted that much of the report offers a useful message. That is, open trade and environmental protection can be mutually reinforcing goals, especially in democratic, market-based societies.[161] Unfortunately, that message is being drowned out by the authoritarian drumbeats in the report that deny the importance of national environmental leadership and the benefits of competing unilateral standards. Countries are not clones of one another, and need not predicate their environmental protection policies on multilateral approval.
The irony in the "GATT and Environment" debate is not that ecological measures are being evaluated on the basis of a 45-year-old agreement. As we have seen, the authors of the GATT were well aware of the need for unilateral environmental action and allowed for it. The irony is that the GATT Council of the 1990s has been slow to comprehend the connection between trade instruments and environmental protection and the reasons why the GATT is viewed in some quarters as being anti-environment. Thus, the real threat to the future of the GATT is not hordes of Greens trying to ram (or, more accurately, peek) through GATT's gates. The real threat may be the myopia and dogmatism of some of those inside.
5. The Economist, February 15, 1992, p. 78.
13. Antidumping duties are used to offset or "prevent" dumping. See GATT Article VI: 2.
23. UNCTAD, "A New Partnership for Development: The Cartagena Commitment," para. 152, February 1992.
24. UNCED, "Rio Declaration on Environment and Development," Principle 12, of April 2, 1992.
27. U.S. Public Law 92-522, [[section]]108(a).
47. Tuna-Dolphin report, 5.25.
57 See note 51, pp 45-47, 52-53.
60. Genesis 6:19 in The NIV Study Bible (Grand Rapids: Zonedervan, 1985.
61. Tuna-Dolphin report, 5.27.
62. See note 61, 5.25 and 5.32.
63. See "Canada-Administration of the Foreign Investment Review Act," GATT, BISD 30S/140, 5.20.
68. U.S. Department of State, Foreign Relation of the United States, 1933, Volume I, p. 729.
83. Labeling would work if consumers acted "morally," in a Kantian sense.
86. Article XX(d) provides that
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...the protection of patents, trade marks and copyrights, and the prevention of deceptive practices...
Article XX(d) is relevant to the environmental debate because its case law has been used by GATT panels in interpreting Article XX(b).<
99. For a discussion of such an approach, see FTA Salmon and Herring report, 7.07-7.11.
105. GATT Document 1527 (1992), p. 11.
111. GATT Article XX(h) and Ad. Article XX.
113. GATT, Analytical Index, 1989, XX-10-XX-11.
119. GATT Report, p. 26. See GATT Article XXV:5(ii).
120. Tuna-Dolphin report, 6.3.
122. See GATT Article XXX for the requirements to amend the General Agreement.
133. The EC Council has enacted such a ban effective in 1995. Sec Council Regulation No. 3254/91.
137. See Tuna-Dolphin report, 3.14, 3.22, 3.38, 3.51, 4.28, and 4.29.
146. "United States--Measures Affecting Alcoholic and Malt Beverages," 5.79.
153. See Tuna-Dolphin report, 4.2, 5.16, 5.28, and 5.33.
156. See note 84, pp. 146-147.
157. See Max Baucus, "Trade as Environmental Lever," The Journal of Commerce June 3, 1992, at 8A.
ACKNOWLEDGMENTS. Robert F. Housman, Charles C. Pearson, and J. David Richardson. This article expresses the personal views of the author.