J.D. (expected 2001), Stanford Law School

M.A.L.D., The Fletcher School of Law & Diplomacy

B.A., Williams College

3745 24th Street, Apt. 1

San Francisco, CA 94114

415.577.8766

mstrauss@stanford.edu

This paper addresses the problem which process-based environmental trade measures such as the US Marine Mammals Protection Act and the Section 609 of the Endangered Species Act pose to the GATT/WTO legal framework. Drawing an analogy to the policy lessons of US labor regulation cases, then appealing to the methodological approach of law and economics, it goes on to explore a novel potential justification for unilaterlism in the personification of the nation as a utility-maximizing economic actor.

The first section takes on the challenge of presenting and interpreting the principles of neoclassical economics which underlie the argumentation on the debate as it has been posed to date. The initial purpose is to demonstrate that such measures can simultaneously achieve the goal of environmental protection and economic efficiency through an analysis of their capacity to correct a series of pervasive market failures in international trade and its law. In the next section, the essay develops the analogy between the GATT/WTO Article III and Article XX rulings and the US Supreme Court’s Commerce Clause jurisprudence, considering in depth the parallels which have arisen when the court has treated Commerce-based labor regulation. Finally, responding to a recent charge to refine the state-cum-economic-unit analysis in international trade law, the last section will develop the initial law and economics argument (and present and answer some anticipated rebuttals) that individual state actors are prevented from maximizing their utility by a global trade régime which denies them the right to express environmental preferences through unilateral trade measures.

In this context, the conclusion briefly examines alternative and interim solutions in advance of the eventual negotiation of a multilateral environmental framework through the WTO, in particular advocating a resort to a liability rule for nations which would choose to breach.

One of the most controversial matters facing the World Trade Organization ("WTO") in its infancy as the guardian of the General Agreement on Tariffs and Trade ("GATT") rules has been the effort to harmonize its mandate to ensure free trade with the evident need to vouchsafe states’ environmental regulatory polices from nullification by the vagaries of international commerce. It is tempting to characterize the two pursuits as necessarily mutually exclusive, yet a body of work has emerged advancing the proposition that "greening" the GATT is simply a matter of cleverly orchestrating a compromise--either in the interpretation of the GATT itself or in the negotiation of multilateral treaties on the subject. Still others recognize the unique opportunities offered by the existence of an international trading régime towards the achievement of both goals simultaneously.

Situated mainly within the third of these camps, this paper will demonstrate that the WTO’s Appellate Body, through its restrictive definition of a "product" under Article III and its illiberal reading of the Article XX exemptions for environmentally-motivated trade measures, has blunted a potentially salubrious tool for optimizing global gains from trade within the framework of the GATT. In particular, when a WTO Appellate Body ruled in The Tuna-Dolphin Cases that the US Marine Mammals Protection Act ("MMPA") applied an impermissible quantitative restriction against foreign products for the environmental damage resulting from their production rather than their effect once imported, it brought into question a contracting party’s capacity under the GATT rules to impose certain types of environmental trade measures consistent with its domestic environmental policies. The Appellate Body’s more recent application of the Tuna-Dolphin principle in last year’s Shrimp-Turtle ruling served only to intensify concerns that the GATT, as applied, is environmentally unfriendly.

In the aftermath of these rulings, this essay will confront two issues, one abstract and one concrete, both towards the greater purpose of envisaging a coherent and flexible compromise role in the medium-term for environmental trade measures within the international free trade régime. The first is an initial effort to seize Jeffrey Dunoff’s and Joel Trachtman’s recent challenge to administer more comprehensively the methodology of law and economics to international law. As a consequence of its nature and its history, the law of international trade is wracked by a remarkably deep and wide set of market failures, some mutually reinforcing, which have discrete, material policy implications. Following the economic analysis through to one set of its logical conclusions, then, this note will begin with an analytical foray into the law and economics of international environmental trade measures, exploring those market failures which pervade the international legal-economic order so as to identify the environmental consequences and policy imperatives which rationally follow.

The second portion of this paper will delve into the legal-regulatory logic of the process/product distinction which serves at the linch-pin to the GATT Appellate Body’s jurisprudence on recent "external-local" environmental trade measures. The evolution of the same process/product principle in the United States Supreme Court’s ("Supreme Court") federal labor regulation jurisprudence under the Commerce Clause provides perhaps an unexpectedly rich parallel to the GATT’s environmental measures dilemma, consequently offering an instructive body of legal analysis and conclusions on that principle for application to future cases along the lines of the Tuna-Dolphin and Shrimp-Turtle disputes. Against the backdrop of market failure analysis for international trade law as context, the extended analogy demonstrates that the principle of regulating products for the processes of their creation is fundamentally rational within a federal strcuture and, indeed, consistent with the welfare-enhancement goals of the GATT.

The final section will explore in terms of law and economics the instructive (if partially problematic) construction of the state as an economic unit within the global market for trade regulation. The purpose is to demonstrate that environmental trade measures have the potential to yield an allocatively efficient trade result for contracting parties by empowering national units, through individual optimizing behavior, to eliminate otherwise intractable market distortions in international trade. Put differently, this section will demonstrate that, in the words of Howard F. Chang, "to give absolute priority to free trade over the environment [on welfare-promoting grounds] simply ignores an important component of global economic welfare." As other eminent legal scholars have demonstrated in peripheral contexts, such regulations--the foreclosure of certain voluntary consumption choices--can counterintuitively be both welfare- and freedom-expanding, rather than styptic.

The note will conclude with a consideration of some devices which the WTO might implement in the interim--in particular the accomodation of the doctrine of permitting "efficient" breaches--to make a temporary place for (and offer tentative approbation towards) process/based environmental trade measures. In an historical tone, then, it will advance the proposition that the US Commerce Clause regulation analogy, viewed within the political, judicial and economic contexts from which it emerged, illuminates the path which the GATT will eventually need to follow towards the inclusion of a small but essential set of environmental and social trade policies. All this should serve in the interim as a transition towards the negotiation of a multilateral framework on trade and environment explicit and transparent enough to appropriately embody the very solid norms which have developed in recent years on their place and value in international law and policy.

Jeffrey Dunoff and Joel Trachtman issued a challenge earlier this year in the Yale Journal of International Law on the matter of integrating the methodology of law and economics into the analysis of international law. As the authors explicitly acknowledge, the application of an entirely fresh analytical method to a field which, as of yet, has maintained a nearly hermetic innocence of that type of examination summons commentary and dissection from every nook of the newly superimposed discipline. Given the pressing environmental policy dilemmas which have arisen recently under the imprimatur of the GATT and its Appellate Body rulings, a consideration of the existence, pervasiveness and repercussions of the several acuminate market failures in international trade law deserves pride of place among such elaborations.

Arguments for the economic viability of environmental trade regulatory measures, therefore, must begin and end with an exploration of the economic and political-economic principles at work and at stake. The first section briefly describes the core economic principles which justify free trade. The second delineates certain dynamic political-economic observations which underlie some common analyses and recommendations in international trade law, for reference later in the essay. The final group outlines a set of microeconomic market failures which crop up with regularity in the field of international trade and trade law. After these, two additional sections describe legal and policy prerequisites for effective cooperation which, to date, have not been met in the realm of international trade law to the extent necessary to permit the harmonious operation or substantial articulation of effective international environment and free trade policies.

These mainly positive reckonings constitute essential logical building blocks for the analysis throughout the paper. They are organized to guide the reader through the logical steps of thought and research which, cumulatively, tempt him or her to proceed into the type of normative conclusions which policymakers and academics often make using such raw materials. Because they are novel only in their juxtaposition and not alone, the explanations below are minimal sketches which include indicatory references for more profound research.

The oft-cited purpose of the GATT’s various negotiation rounds since World War II has been the recognition that the citizens of the world are materially better off in the aggregate if a state of free trade prevails. The apperception of this condition derives from the interaction between sets of normative principles which follow from several specific theories of microeconomics, developed below.

First, basic to virtually all microeconomic models is the assumption that the costs associated with the transaction itself--getting the good or service from the producer to the consumer--are negligible or, if significant, at least estimable. If transaction costs are high, on the other hand, as they would intuitively seem to be in the early (pre-infrastructural) "state of nature" of international commerce, the rational valuation of the costs and benefits of a transaction is overwhelmed by considerations of the costs of the transaction itself. Barriers to trade burden transactions in just this way. So when barriers to trade fall, distortionary transaction costs are reduced and the market is better able to allocate goods and services according to their true costs and prices. Put simply, efficient allocation is contingent, in part, upon low (or at least accurate) transaction costs economically.

Tariffs and other costs imposed upon the import and export of like products which compete with domestic products impose undue and unnecessary costs upon transactions. Like any tax, these barriers act as wedges between supply and demand. If they are evenly applied, they may be less distortionary in an allocative sense (that is, they are less likely to cause us to choose a different product than we would have chosen in the absence of the tariff). Nevertheless, by construction, they must reduce the number of transactions which would have taken place in their absence, causing a "deadweight loss"--a reduction in aggregate surplus.

In this stylized sense, at least, the case for removing barriers to transactions across borders --i.e., reducing tariff and non-tariff barriers--is straightforward. Support for freer trade is further enhanced by a demonstration of the "gains from trade" principle and the theory of comparative advantage.

The microeconomics of trade have a long pedigree, beginning with David Ricardo’s demonstration that differences in the relative productivity of labor in countries vest in each a "comparative advantage" in the production of some goods (and services) over others. The endowments and marginal productivity of the factors of production in any given nation combine through its citizens’ decisions about how to apply them to and combine them (i.e., how to allocate capital, labor and land) in various available production opportunities. The costs of combining the factors for the production of one good or service may be measured in the foregone possibility (the opportunity cost) of not producing another useful product. In the simplest, two product model in autarky, this trade-off implies a "price" in terms of the quantity of one product which may be exchanged domestically for the other.

At the same time, other nations make similar decisions about trade-offs between the possible uses of productive resources and, with different endowment and factor productivities, come up with different relative "prices". If they choose a different allocation of the factors of production in their jurisdiction, they presumably do so because it is less "costly" (in terms of the lost opportunity to make other products with the same resources).

When two nations which produce the same two goods at different costs and in varied profusion open to trade, they may take advantage of the different relative prices for the identical products to have some more of both goods. This phenomenon, with its accompanying analytic geometry, is known as a "gain from trade" and shows up as a "trade triangle" of additional surplus. Specialization (assuming it leads to price differentials), in other words, pays dividends if you trade.

This illustration of comparative advantage and gains from trade is purposefully cursory, as no more intricate a model is necessary to demonstrate the principles at stake in the trade argument. Moreover, very little turns on any alternative characterization of the gains from trade. As Paul Krugman and Maurice Obstfeld show in their seminal textbook on international economics, most assaults on the perceived deleterious effects of global trade which do attack the theoretical roots are based on shibboleths about the basic meaning of comparative advantage. The underlying normative conclusion that free trade promotes welfare in the aggregate, based on this model of free trade, remains the area of the most robust agreement among economists of all walks.

In spite of these clear aggregate benefits to free trade, it is well known that attendant to the reduction or elimination of protective trade barriers are serious distributive consequences. When the US concluded the North American Free Trade Agreement ("NAFTA") with Canada and Mexico, the decision virtually guaranteed that sectors such as the textile industry would suffer great losses. Because it can provide a cheaper supply of low-skill labor, Mexico has a comparative advantage in the production of textiles. Likewise, the reduction of barriers to trade in steel under the GATT has allowed Russian producers to gain market share at the expense of higher-cost US steel producers. On the other hand, the high-tech firms of California’s "Silicon Valley"--which enjoy a comparative advantage over most computer and software producers in the abundance of high-productivity, high-skill labor--have gained inordinately from free trade. In short, free trade, when embarked upon as a policy, creates stark sets of winners and losers in the redistribution which occurs in its immediate aftermath.

Given the costs of transition (in particular the very heavy and concentrated human costs of unemployment), then, policymakers must query whether or why these redistributions brought about by free trade are justified. Significantly, trade economists acknowledge that the reallocations of wealth which accompany a move to free trade fail one important test of fairness--they are not "Pareto optimal"--that is, the benefits which come to those who gain from trade come at a net cost to others. In effect, in accord with the aphorism, one man’s loss is truly another man’s gain.

In its simplest form, the justification for free trade, then, is therefore not a distributive one--not all gain from free trade and several suffer net losses. But society-wide these redistributions are not zero-sum games. If the gains by one party consistently exceed the losses by the others, free trade will bring benefits in the aggregate. Such an outcome cannot be said to be "fair" in an important legal sense in that it constitutes "a naked attempt . . . to take the private property of A and transfer it to B." And while they may not be defined as a due process clause "taking" strictu sensu and there is seldom a property right in question, these transfers most certainly violate a fundamental sense of fairness in regards to the reliance upon settled expectations. But this redistribution is tolerated and, in fact, preferred as a matter of global policy because it is "Kaldor-Hicks efficient" (or potential Pareto efficient). That is, while some may be worse off for the change, if the welfare outcome sums to more than zero, society has the capacity to compensate A for his loss, leaving him at or above status quo ante and rewarding B the surplus. The worker retraining provisions in NAFTA for losing industries show an example of one such redistributive policy.

The logic that compares these values, in effect, is no more than a society-wide application of a cost-benefit analysis, one not as philosophically unproblematic as it may at first appear. At the risk of shortchanging the gamut of reasonable objections to the use of the Pareto and Kaldor-Hicks models to assuage distributive concerns, this presentation will operate upon the simplifying (and somewhat unwarranted) assumption that the primary actors affected by the implementation of a domestic free trade policy are firms, whose interests can plausibly be classified as utilitarian and whose stores of values can safely be assumed to be completely fungible.

In this reductionist world, at least, this gain and transfer principle is the theoretical fuel for the wealth-generating engine of trade in market economies, and the result is an "expanding pie" for all. A rising tide, when losers are compensated, ought indeed to lift all ships. Whether or not this principle operates in practice--that is to say, whether losers are, in fact, compensated--is a matter of separate inquiry. Presuming for the sake of the construct that practice and principle are one, the benefits of free trade are indisputable.

With the distributive and allocative effects of free trade established above, the desirability of a "default rule" of free trade with a compensation scheme--the absence of trade regulation as the state of nature--should be clear. Being in the best long term interest of all, why, then, has free trade never been the "state of nature"? Why has the still-incomplete effort to establish merely freer trade under the GATT taken more than fifty years and eight negotiating rounds to establish if all are better-off in the long run under free trade? Other microeconomic and political-economic forces have been at work to make cooperation on free trade very difficult. In fact, some profound market failures interact with perverse incentives which inhere in the political-economic structure of liberal democracies to make free trade policies a challenge to establish and maintain.

To begin with, governments which impose tariffs on imports create a flow of valuable revenues which, superficially at least, do not come from the pockets of their citizens. While the costs of tariffs are ultimately borne by both citizens and non-citizens in the form of higher prices for imported and import-competing goods, the actual payments come more often that not from foreigners importing the product, creating the illusion that a tariff is a tax on foreigners. A tariff consequently has an obvious political advantage relative to income, sales or value added taxes, and tends therefore to be a chosen tool of protectionist governments.

In larger part, the need for a political and judicial infrastructure to police the free trade régime flows from individual nations’ seemingly rational propensity to protect their import-competing industries in the short run, according to the game-theoretical outcome of the "prisoners’ dilemma." Briefly (because far too much ink has been spilt recapitulating the prisoner’s dilemma in law journals), this market failure results when a rational choice for an individual to pursue--that is, one which appears to lead to an optimum outcome--leads instead to a sub-optimal outcome when two or more actors play the same game.

It is rational, for example, for any oil exporter to produce and sell as much oil as possible. Consider that revenues equal the product of the quantity sold and the price at which the oil is sold (QxP), and suppose that prices are constant. The more an oil producer sells, the more she makes in revenues. Given the sheer volume of oil sold globally, it is no intellectual leap for a producer to presume that no sale of any given barrel of oil (i.e., the marginal barrel) alone can change the prevailing world price for oil. That is, as to any marginal barrel sold, prices can safely be considered constant. Yet the more producers sell, if they all act with this logic in mind, the more the price of oil will fall. So many marginal barrels sold do indeed flood the market with supply, decreasing everyone’s revenues by sending prices into freefall. So while Q increases greatly, P falls proportionately more.

If they could collude ("cooperate" in the jargon of collective action economics), as the Organization of Petroleum Exporting Countries ("OPEC") were able to do twice in the 1970s and have once again recently attempted to do, the world’s oil producers would reduce their production, prices would rise dramatically and all would capture even higher revenues while producing a much smaller quantity of oil--by far the best outcome for all producers. But this type of cooperation is exceedingly difficult because the one country that chooses not to comply when all others have reaps inordinately high benefits by selling a high volume at a high price. This dynamic tempts each participant to cheat. The perverse result is that the logical and rational action for one actor to pursue--to cheat--ultimately leads to a worse result for all. Any given individual will benefit all the more by cheating when a coordinated policy is established, so maintaining cooperation is extremely difficult.

Similarly (although by removing, not imposing restrictions), as explained supra, a nation benefits greatly from free trade, but its exporters will gain an additional benefit if they may free-ride on other nations’ openness to trade while themselves taxing imports, making their products preferable both at home and abroad. GATT members, then, in a manner similar to the OPEC countries, will be tempted to agree to the terms of free trade in principle, then cheat by establishing protectionist policies to gain the benefit and forego the costs of cooperation. If all "cheat" as it is apparently in their individual interest to do, all are worse off when the régime disintegrates and expensive, distortionary wedges are driven between supply and demand for all imported and exported products.

The savvy economist will recognize that the prisoners’ dilemma story above is incorrect as applied to the individual nation’s free-trade problem. The benefits of free-trade, after all, come to both producers and consumers. As trade policy goes, the apparent benefits to one bring costs to the other. Why would nations eschew free trade for producers’ sake if their consumers would have benefited from the increased quality and lower prices? For example, why would the United States press for dumping penalties or negotiate voluntary export restraints against Russian steel producers or Japanese semiconductor producers who flood the American market with cheap, even below-cost products? Despite domestic producers’ fear that foreign companies will gobble up market share, domestic consumers of these products should be evidently jubilant.

The answer lies in the coupling of the prisoners’ dilemma with the now well-elaborated political-economic theory of "public choice." The synergy of the two theories is explained as follows: representatives’ terms are short, insecure, and dependent upon the influx of campaign funds for their longevity. These funds are available only from constituencies which maintain concentrations of discretionary wealth--that is, producers whose interests consequently must be satisfied. Further, workers whose livelihoods depend upon the viability of the local production facility also have a vested interest in protection. Almost inevitably, the consumers (and sometimes producers) who "win" from free trade are elsewhere. There are more of them, but each benefits proportionately less--perhaps even negligibly--certainly not enough to galvanize them as a lobby in favor of a free trade policy. Policymakers will therefore be prone to protect local import-competing industries against liberalized trade for fear that their core constituencies not become severe "losers" from freer trade in the short term. This concentration and immediacy of losses relative to the diffuse and longer-term (but greater in the aggregate) ostensible gains from free trade conspire against a representative’s capacity to support proposals for freer trade, despite its proven long term, general benefit.

In the face of the public choice/prisoners’ dilemma synergy, then, the need to constrain nations’ instinctual protectionist impulses is manifest. If the public choice/prisoners’ dilemma synergy is robust, a supra-national body which invalidates such non-optimal regulations, then, serves the purpose of stanching the otherwise inexorable ebb into a dangerous "race-to-the-bottom" in trade protection. The Commerce Clause in the US Constitution, for example, has been broadly interpreted by the Supreme Court to preclude state or local action which would place undue burdens on interstate commerce. According to the Supreme Court’s construction, by its very existence, the Commerce Clause, even "dormant", prevents states from establishing laws which inhibit the free flow of goods and services to other and between the several states. The Court’s modern treatment of the dormant Commerce Clause was summarized well in Philadelphia v. New Jersey: "The opinions of the Court through the years have reflected an alertness to the evils of economic isolation and protectionism, while at the same time recognizing that incidental burdens on interstate commerce may be unavoidable when a state legislates to safeguard the health and safety of its people."

Presaging the law and economics rationale, the framers justified the strong centralized commerce power of the US Constitution by the need to prevent the several states from descending into pernicious regulatory trade wars. Control over interstate commerce was one of the stalwart prerogatives enjoyed by the central government under the federal Constitution which the Articles of Confederation notably lacked. Fortunately, the commentator schooled in law and economics might say, this textually questionable jurisprudence later defused the potentially destructive market failure of a "race-to-the-bottom" when the US developed into a modern regulatory state with a hierarchy of revenue-collecting mechanisms for each jurisdiction and a basic social welfare apparatus. Often the affliction of a weaker federal structure, this "race-to-the-bottom" is merely a permutation of the prisoners’ dilemma, set forth supra. The payoffs may be, inter alia, revenues, regulatory compliance or protection for import-competing industry. The risk is flight of capital, population, or productive facilities, depending on the context. In short, states hazard the loss of their industrial base if they tax or regulate too aggressively, and as to trade they have the same protectionist impulses as their nation-state counterparts via the public choice dynamic.

The parallel between these federal structures and the GATT rules is robust and attractive to international economists, as it serves as a veritable case-study by transitivity for the virtues of the WTO judicial activism in the invalidation of nations’ exemption for trade-restrictive measures. Like the Supreme Court’s dormant Commerce Clause jurisprudence, the invalidation of measures which burden the free, non-discriminatory flow of goods and services across the world’s borders serves the goal of expanding aggregate welfare through the safeguarding of those significant gains to be had from it. Further, judicial activism in a horizontally federal system may designedly compromise the underlying regulatory régime to a partially desirable degree, ensuring that the market for these goods and services not be encumbered needlessly and expensively.

Unsurprisingly, despite its descriptive appeal, the accuracy of the "race-to-the-bottom" prediction generally is empirically questionable; certainly the effect should vary significantly by the type of runner in the race. An environmental-regulatory race-to-the-bottom is perhaps the most recently controversial. And no matter the vigor of the justification, the judicial body which reflexively invalidates regulations faces the collateral risk that essential regulations, such as those for the protection of health, safety or the environment, will be deemed illegitimate. In order to avoid this result, the Supreme Court has developed over time a set of acceptable aims for regulations of inter-state trade, as well as a series of tests designed to assess whether these regulations were enacted with good faith or with protectionist ulterior motives. As will be discussed in greater detail infra, Article XX of the GATT provides for just such a set of exemptions for national trade measures allegedly motivated by the need to protect the environment, public health and safety, inter alia, but its Appellate Body has consistently refused to find trade-restrictive regulations valid under this article.

A few microeconomic principles are essential in the effort to understand environmental trade regulations as a class and the market failures which plague the realm of international trade. Two market failures in particular--the negative environmental externality and asymmetries of information--acutely afflict international trade, and thus international trade law. Their remediation is the preferred role for environmental policy and is thus an especially legitimate source of justification for environmental trade measures.

Put in general terms, an externality arises when another’s conduct unfairly imposes a cost or confers a benefit on an external party, or "beneficiary", who never bargained to be part of the action or transaction. The external cost or benefit is not always one cognizable in a legal sense as a property or contractual right, and as such, those who suffer from negative externalities do not always have a legal remedy in property, contract or tort for the damage they endure. A man who developed lung cancer after living next to a neighbor who barbecued every night for twenty years, for example, can rarely recover in tort because, among other things, actual causation cannot be proved. Rarely enjoying a legal recourse, those who suffer a negative externality often either suffer in ignorance or silence, or they "bribe" the perpetrator to stop--in effect, paying the polluter.

To apprehend the translation of the externality principle from its microeconomic to its macroeconomic effect, suppose a game in which two factories in separate jurisdictions produce identical radios in a process that also leaves a toxic residual byproduct. Jurisdiction one ("1") imposes elaborate regulations for the safe disposal of the byproduct in order to protect its citizens from environmental harm while jurisdiction two ("2") imposes no such regulations. The factory in 1 pays high costs to properly dispose of the waste while the factory in 2 costlessly dumps its waste into the river, while several families downstream incur great medical and clean-up costs. Free trade in goods between 1 and 2 will prove to greatly enrich (i.e., provide economic surplus to) the producer in 1, whose lower costs allow her to charge lower prices and win market share. It will also greatly enrich consumers in 1 and 2, who pay less for the identical product. Yet the incremental additional surplus had by these three groups because of the lower price that the producer in 2 may offer may be (depending on the shapes of the supply and demand curves for radios and the marginal control and damage cost curves) at least zero-summed by the "negative consumer surplus" (costs) which must be absorbed by the third party "beneficiaries" downstream in 2. In effect, an "unfair" (in an legal-economic sense: characterized by gross externalities) advantage was conferred upon the factory owner in 2 and she thrives as a result of behavior which does not constitute a production efficiency per se, but of a sort of "theft" of surplus, since she causes costly personal harm for which she does not pay.

What is more, this outcome is a single game. When repeated over several periods, the jurisdiction with fewer regulations will continue to gain a price advantage and will re-invest (along with capital brought on from 1) the rents from its previous games in more productive physical plant, choosing, of course, to locate in jurisdictions which have no or close to no marginally-costly regulations and forcing other jurisdiction to rescind their previously burdensome regulations. In this scenario, a "path dependency" result assures exacerbated and longer term entrenchment of distortions and external costs. With interjurisdictional commerce, the endpoint of an ingrained externality is a long-term misallocation of productive resources in violation of the utility of those whose preferences have no palpable value in the market. As Howard Chang explains, "[a]s long as there are no [external] spillovers and governments choose the efficient level of environmental protection (by equating social marginal benefits and costs and thereby maximizing national social welfare), competition in international trade and investment will be efficient. We do not obtain these optimality results, however, in the presence of international environmental externalities."

While it is a subject mainly eccentric to the core goal of this paper, a legal scholar could scant pass by the problem of an externality without commenting on the handful of staple tenets of a just and efficient theory of contracts which are violated in whole or in part by a contract characterized by substantial external effects. Among them are 1) the notable lack of consideration on the part of the third-party beneficiary (especially in the presence of a negative externality, it would be difficult to demonstrate that something of value was exchanged along with the promise, since no promise was made); 2) the absence of an expression of an intention to be legally bound and/or the absence of bargain, and therefore the possibility the contract is one of adhesion; 3) the fact that an incidental beneficiary has no enforcement right against a promisor or promisee, and hence has little legally cognizable control over the transaction; and 4) the possibility as with coercion that many types of external environmental harms that the contract’s assumed risks would be unenforceable on public policy grounds as unconscionable. Without extensive discussion, suffice it here to note that both fairness and efficiency goals are at least partially frustrated when contracts’ effects are in large part external.

Especially important for facilitating the exercise of health and environmental preferences is access to precise information. Accompanying the tendency to globalize markets is an increased temporal decoupling of production and consumption, of place of origin and end-use. If a product "fabriqué en Chine" is consumed in France, essential information which a consumer may need to make his decision whether or not to purchase that product is difficult to ascertain with accuracy as compared to a product produced by a neighbor or sold at a local market. Exacerbating matters are the facts that, first, modern manufacturing and processing technologies often commingle products to the point where control of their characteristics based on their origin (sometimes characteristics which would differentiate the product in the market as to some consumers, sometimes not) is a virtual practicable impossibility.

Second, the internationalization of trade aggravates asymmetries of information. Producers, suppliers, and retailers have an inherent incentive to obscure or exaggerate any information which may activate preferences. Consumers, in turn, have virtually no control over the provision of that information for products from far-flung places, beyond the regulatory levers their representatives provide them. Just as the expansion of mass production of food and other products ushered in by the industrial revolution eventually spawned the necessity to develop an overarching regulatory body in developed nations, so does an expanding menu of foreign products available domestically, it is arguable, require expanded regulatory activism for health and safety.

The consequence for information provision of this temporal production/consumption bifurcation is that domestic regulators need to be all the more vigilant and, indeed, legally capable of regulating aspects of import (i.e., labeling, etc.), or the import per se of foreign products. By the codification of Article XX, the architects of the GATT must have recognized the need to preserve this type of capability, but actual efforts so far--many of which have admittedly been overzealous in important technical legal aspects--have raised objections from GATT dispute settlement panels and the GATT Appellate Body. Ameliorating the quality and availability of information, like the reduction of burdens to transactions described above, must be an primary goal for those who seek an allocatively efficienct result.

Two prescriptive requirements for efficient allocations of goods and services globally--the allocation of property rights and the enforcement of contracts--achieve the status of primus inter pares when discussing the desirable constitutive architecture of a healthy system of exchange. Even economic libertarians cede some "regulation" as a minimum condition for the efficiency principle of neoclassical economics to function, and the emergence of law and economics analyses in the vein of the Coase Theorem has not attenuated the impulse among legal scholars to at least advocate the pursuit of clarity in defining property rights and assiduousness in the enforcement of contract towards the good of welfare expansion.

Though these rights and their scrupulous protection precede the meaningful development of any principle of welfare, in international law they are notably porously safeguarded on the whole. "The international legal system provides a more graphic illustration of the contingent nature of contract and property than does the domestic system. Compared to the domestic system, the international legal system lacks cognates to the more complete domestic systems of property law (territory and jurisdiction)[ and] contract (treaty and custom)."

Negative environmental externalities, it seems, exert a punctuated effect through international trade. The allocative distortions in systems which project large portions of their effect externally (i.e., outside of privity) should be clear from the above explanation, and affect "states" (both domestic subjurisdictions and nations) in their economic relations with others inordinately. That is, externalities are an especially intractable problem in interjurisdictional (international, interprovincial, interstate, etc.) economic relations. The efficiency with which domestic property law plagued by externality problems is enforced in most developed nations--already spotty--contrasts starkly with the meager set of options available to those who would recover from damage from cross-border pollution internationally. Matters are complicated further by the widely different allocations of property rights among societies according to the variegated social norms which undergird them. The Coase-theoretic logic, were it practically legitimate, would at least assuage our fears of disastrous results in the allocation of resources: allocation itself would remain unaffected by that endowment of property rights. Realistically, states and nations which trade freely with a group of other states are limited by that trade in their capacity to enact an internalization program as a policy, should they so choose, by a trade régime which does not permit them to apply the very minimal structures of sovereign restraint and "coercion" for which they bargained in the "contract" of the General Agreement. Indeed, it is no leap of faith to assert that this very difficulty in sorting out the seemingly relatively uncomplicated "property" rights which are vested in a particular nation under the GATT, coupled with the problems of even defining a "contract" (particularly relative to the perception of the GATT’s allocation of rights and responsibilities upon sovereign nations as public-law--a consequential confusion which will be addressed in greater depth infra) both demonstrate that notions of property and contract in the international sphere are desirous of even basic elucidation and hence, a fortiori the baseline coercive protection we find available to both in well-evolved industrialized economies.

As this essay will demonstrate in its proceeding sections, the above micro- and political-economic models provide intellectual fodder for the general argument, delineated infra, that there are essential social and economic justifications for endowing GATT signatories with a certain tranche of regulatory powers over trade in products produced abroad in an environmentally objectionable manner. Simultaneously, the evolutionary history of domestic environmental policy imposes another set of important constraints upon those policymakers who altruistically hope to reap the proven benefits of free trade while simultaneously safeguarding health, safety, or environmental resources. These constraints are especially onerous when the targeted environmental damage takes place outside of the importing nation.

The GATT Secretariat addressed the matter of the interaction between trade and the environment in 1992, concluding on the whole that, although environmental regulation is a legitimate policy end, trade is not the appropriate means to pursue that policy. In part because the effort to erect the international free trade architecture has been so hard-fought, and in particular out of a skittishness towards any unilateral action among GATT signatories, the GATT Secretariat maintained in the 1992 document that "unilateral restrictions on trade would never be the most efficient instrument for dealing with an environmental problem."

Instead, the GATT Secretariat offered two alternatives to unilateral environmental regulations. The first, of course, was simply to remove the "unilateral" modifier--environmental measures must be multilateral. This condition, given the storied and complex trajectory of the incremental establishment of a global free trade agenda to begin with, must be acknowledged as achievable at best only in the medium term. The second prescription is to remove environmental policy entirely from the international realm. Environmental regulation, says the Secretariat, is mainly a domestic matter and it need not impinge upon trade.

The clean separations are more illusory than they may initially appear. The strategy of "getting prices right" is acknowledged to be one of the most effective and cost-effective policies for the internalization of negative environmental externalities. Yet a domestic policy which itself complies acceptably with the First Welfare Theorem does so by transforming regulations which forbid actions (command and control) into tax and permit systems which simply internalize the previously external social and environmental costs of certain market activities, offering consumers and producers the choice to bear the costs if they will and simultaneously compensating third-party "beneficiaries". The incongruity lies in the fact that foreign products not subject to the same price internalization scheme designed to "get prices right" have the potential to undermine--if not obliterate--the domestic goal of crafting economically efficient environmental regulations. The consequence of the clash of domestic environmental and international trade policies is that many of the former will be either eliminated or blown into the regulatory stone age of command and control remedies, since they no longer can operate by "getting prices right."

Most important of all the policy implications, however, is the stark fact that such a "disconnect" between domestic environmental and trade policies will generate a negative welfare outcome. In effect, as long as we recognize an activity such as unwanted dolphin bycatch and mortality as a public good problem, a GATT panel decision rejecting that prohibition under Article XX is microeconomically inefficient.

The very availability of exemptions for health, safety and natural resource protection in Article XX was presumably premised upon this efficiency principle. It therefore must be asked, given the above demonstration of the need for and potential efficiency of environmental trade measures, why the GATT as interpreted does not permit such an exception? Why, perversely, does it force a return to the clumsy, non-economic (in two senses of the term) domestic environmental policies? With the economic substructure firmly in place, the next task is to examine how the WTO has reacted when nations use trade restrictions to protect external, non-use values.

The superficial explanation for the prevailing restrictive interpretation of Article XX is the recognition, if only sotto voce, that environmental trade measures, no matter their gravitas and value on the whole, are a Pandora’s box for "back-door" protectionism. Much as the US Supreme Court searches vigilantly for an enduring limiting principle when it considers expanding substantive civil rights designations, so too must the WTO/GATT dispute settlement apparatus develop a measure of prophilaxis against the flood of purportedly benign external-local trade measures eligible for Article XX exemption. This section examines briefly the Tuna-Dolphin and Shrimp-Turtle cases before the WTO Appellate Body, then explores close logical parallels within United States Commerce Clause jurisprudence. The critical purpose is to hew just such a workable limiting principle from a review of US labor regulation jurisprudence, dormant Commerce Clause precedents, and several characterizations of the doctrine of "fairness" in trade.

The GATT interpretation of Article XX has unfolded in two recent cases before the Appellate Body in maritime fables of fish, dolphins, turtles and shrimp, worthy of the likes of La Fontaine and Æsop. In the now well-known Tuna-Dolphin Cases, Mexico (joined later by the EU) brought the US MMPA before a GATT dispute resolution panel, claiming that it constituted a numerical restriction on trade--in effect, a quota of zero--impermissible under Article XI of the GATT. In a more recent case, Thailand led the challenge to the US Endangered Species Act Section 609 which controlled the import of shrimp from producers who used "commercial fishing technology which may adversely affect sea turtles"--that is, without using "turtle excluder devices" ("TEDs"). Both cases will be considered in turn.

The US responded to Mexico’s claim by invoking the below provision in Ad Article III:

The MMPA tuna embargo, the US argued, is just such a regulation, applied upon importation, subject to GATT review exclusively under Article III’s strictures. Article XI, it claimed, could therefore not apply.

To the contrary, the Appellate Body determined that the MMPA’s restrictions on imports of tuna caught with a high dolphin mortality were not covered under Ad Article III because the measure did not regulate a product, as Ad Article III and Article III require, but rather the process by which that product was created. Nowhere in Article III, it seems, is there reference to measures required not because of the characteristics of the product itself, but because of the circumstances of its creation. "Article III:4," the Appellate Body stressed, "calls for a comparison of the treatment of imported tuna as a product with that of domestic tuna as a product. Regulations governing the taking of dolphins incidental to the taking of tuna could not possibly affect tuna as a product."

The US offered an "even if" argument, claiming that the MMPA was eligible for an exemption from the Article XI violation under both Articles XX(b) and (g), which appear to permit measures discriminating against like-products if they are implemented for the achievement of certain environmental ends. In relevant part, Article XX provides as follows:

In previous Article XX cases, panels have have jealously guarded this series of positive exemptions set forth in the GATT’s Article XX, narrowly construing the Article’s "necessary" and "relating to" clauses. If challenged, for example, the party which imposed the measure under Article XX(b) bears the burden of demonstrating that, inter alia, the particular regulation was "necessary". That is, it must show within reason that no other less trade-restrictive alternative exists for meeting the same end. Consequently, the Appellate Body rejected the prayed-for for an Article XX(b) exemption, stating that the US could not demonstrate that the measure was "necessary" to protect dolphins. The mere possibility that the US and Mexico, in the alternative, could have negotiated a treaty on the matter was sufficient evidence to the Appellate Body that the US had not met this "least trade-restrictive means" test.

Interestingly, the Appellate Body then considered, and rejected the possibility that XX(g) might exempt an extrajurisdictional law, integrating the effect of Article XX’s chapeau in its Article XX(g) analysis. The fact that the MMPA dolphin mortality standard was based on a variable, unpredictable standard, said the panel, went to show that the purpose behind the rule was discriminatory, and thus inconsistent with the chapeau’s "applied in a manner which would constitute a means of arbitrary or unjustifiable discrimination" clause.

The Shrimp-Turtle Case more recently presented a similar fact pattern and therefore received comparable, if more delicate and compromising treatment by the WTO Appellate Body. While it offered several important concessions to those who have argued for a broader interpretation of Article XX’s exceptions, it again mobilized the Article’s chapeau as an amorphous, inscrutable "environmental excluder device."

Briefly, Article 609 of the US ESA required certification of foreign shrimp exporters to ensure that their catch method did not result in excess turtle mortality--a test which generally required that shrimp harvesters use nets with costly and unwieldy TEDs. The regulation applied equally to American shrimp producers under ESA, but was phased in unevenly as to both domestic and foreign producers and as to foreign nations from different regions.

Like Mexico, Thailand argued, inter alia, that this "zero quantitative restriction" violated Article XI. Although neither the Tuna I nor II decisions had been adopted by the WTO, the US did not claim the applicability of Ad Article III after The Tuna-Dolphin Case, and sought sole refuge instead in the Article XX(b) and XX(g) exceptions to Article XI quantitative restriction interdiction. In line with the previous cases, the dispute resolution panel rejected the US’s claim to Article XX exemptions for Section 609. The Panel framed the exemption issue much as the Tuna-Dolphin question had been framed: "whether the scope of Article XX encompasses measures whereby a Member conditions access to its market for a given product on the adoption of certain conservation policies by the exporting countries."

The Panel found that Section 609 would authorize discriminatory treatment to nations which carried out the same practices in harvesting shrimp (because of the varied permissible phase-in times). Because such discrimination was inconsistent with the free-trade-cum-state-of-nature gestalt it read into the GATT as a general matter, the US had violated the chapeau to Article XX. Accordingly, the Panel offered no consideration of Section 609’s compatibility with Article XX’s enumerated exceptions.

Upon US appeal (Shrimp II), this result was upheld, but for different reasons. The mere discriminatory manner in which Section 609 was applied, the Appellate Body found, demonstrated an ulterior protectionist motive which disqualified it for an Article XX exception. Much more interpretively explicit, the Appellate Body rejected the procedural mechanics of the Panel’s interpretation of Article XX because the latter failed to even address the question of whether Section 609 fell within the ambit of Article XX’s enumerated exceptions. Indeed, the Appellate Body asserted that the very meaning of the chapeau could not be apprehended outside of the context of the enumerated exception under which the exemption was claimed.

Assessing the validity of the claimed exception in context, the Appellate Body undertook a four-element test, enunciating what now might be regarded as the set of hurdles which an external-local environmental measure must surmount in the future to receive exemption status under Article XX. First, the Appellate Body explained, Section 609’s target must be an "exhaustible natural resource" consistent with the language of Article XX(g). There was debate as to whether a living, reproducing animal could constitute an exhaustible natural resource, but the Appellate Body found that sea turtles in danger of extinction were indeed such. Second, it was incumbent upon the Appellate Body to determine if the regulation, in the language of Article XX(g), "relat[es] to" the protection of that exhaustible natural resource. It again found in the affirmative, citing the principle elicited in US Reformulated Gasoline that this regulation was not "disproportionately wide in scope and reach in relation to the policy objective." Third, the restriction must have been implemented concomitant with equivalent restrictions upon domestic producers of the same good--a requirement which the US law met, according to the Appellate Body.

Finally, having determined that Section 609 satisfied all of the requirements for an exemption under Article XX(g), the Appellate Body turned to the chapeau. Here, it was not so generous. Despite its initial vigilance in protecting a contextualized meaning of the various exceptions within the language of this preamble, Section 609 was not spared invalidation for its clash with chapeau. The Appellate Body determined that for a measure to be inconsistent with the chapeau’s requirements that it not constitute "arbitrary and unjustifiable discrimination between countries where the same conditions prevail", it must fail an additional conjunctive three-part test. The measure must actually result in discrimination, that discrimination must be arbitrary or unjustifiable and it must occur between countries where the same conditions prevail. The Appellate Body criticized the US for offering a regulation that was unilateral and opaque (given the abstruseness of the certification process) in lieu of exploring the possibility for a multilateral treaty on the matter. It pointed to the complex structure of the régime for certification as proof that the US interest was less to save turtles and more to harmonize foreign regulatory régimes with its own. The US failure to "engage the appellees. . . in serious, across the board negotiations with the objective of concluding bilateral or multilateral agreements for the protection and conservation of sea turtles" was the coup de grâce, in the Appellate Body’s point of view, which demonstrated the true US protectionist ulterior motive lurking behind Section 609’s turtle protection plan. The Appellate Body ultimately found that Section 609 could not pass muster under the Article XX chapeau.

The upshot of the case pertaining to this discussion is, first, that the process/product distinction and the "ten-gallon" chapeau, though not technically precedent after Tuna II, now has effective authority. The WTO Appellate Body is apparently rigid in the jurisprudential position it has staked out that laws which seek to effect trading partners’ behavior in the creation of the product will not enjoy exceptions under Article XX; they are virtual per se violations of the chapeau. Rather (and constituting a thread which runs through the GATT dispute resolution jurisprudence on the environmental exception issue but was only this time aggregated into a reading of the chapeau), a negotiated multilateral treaty on the issue is prescribed as the non-discriminatory alternative available to the offending party. Such a draconian limiting priciple is, of course, mainly impermeable even to bona fide environmentally-motivated trade measures (i.e., those for which the Article XX exceptions were designed), and is thus poorly designed.

It may not be immediately evident why the "product" logic at the core of the tuna and shrimp cases is penny-wise and pound-foolish and, as such, an argument by analogy to the strikingly similar US jurisprudential dilemma of labor regulation should better frame the issue. The US Supreme Court encountered the self-same product/process distinction in its Commerce clause cases between 1918 and 1941. Over the trajectory of these labor regulation cases, the Court progressively ascertained the untenability of distinguishing between process and product, ultimately abandoning this as a material doctrinal distinction altogether.

The process/product confusion, to return to the microeconomics context, begins with the market failure of the negative externality. Of course, products which cause environmental damage are not at all unique in their capacity to impose inordinate negative externalities through the multiple stages of their life-cycles. Many goods and services such as, alcohol, tobacco, pharmaceuticals, or firearms, inter alia, impose severe negative externalities at the point of consumption such that nations often choose to regulate highly the sale, advertisement and use of such products domestically. Another class of products, such as violent or pornographic materials, can cause negative externalities in both its production and consumption. Finally, still other products crafted through the use of unfair, exploitative or inhumane labor conditions impose their externalities exclusively at the point of production, not the point of import, purchase or consumption.

Illustrating this concern, one of the GATT’s enumerated Article XX exemptions--specifically clause (e)--offers contracting parties an exemption for trade-restrictive measures which limit the trade in the products of prison labor. This exemption has an available economic justification (and thus one normatively consistent with the GATT goal of promoting free trade) in that prison labor provides a cost advantage to products that is not derived from an added market "efficiency" in the production process. It is rather the result of a distortion--a subsidy (here the reduced cost of hiring workers who are enjoy fewer freedoms). The degree to which environmental damage associated with the production of a product mirrors this economic characteristic will be discussed infra, however, the Article XX(e) exemption offers an object example of the manner in which labor-motivated regulations resemble environmental ones which aim to eradicate a supposed "evil" attendant to production.

This category of products with high negative externalities in their production should have proved over time most elusive of protective regulations within a régime that limits regulations to the intrinsic characteristics of the product itself. It is this class of products which best illustrates the dilemma faced by the WTO Appellate Panel in Tunas I & II, where the undesirable external environmental damage took place with the production and not the consumption of the product (the former over which the importer has no direct regulatory control).

The federal structure of the United States as enumerated by the Constitution places Congress in a corresponding quandary over the operation of its power to regulate interstate commerce. Congress is authorized by the Article I, § 8, cl. 3 of the Constitution "To regulate Commerce with foreign Nations and among the several States. . .", a power which was eminently characterized in 1824 by Chief Justice Marshall as "the power to regulate; that is, to prescribe the rule by which commerce is to be governed." The development of the Supreme Court’s jurisprudence articulating the meaning of this clause--which pivoted critically on the matter and manner of labor regulation--proves didactic for those who seek to reform the GATT/WTO to bring its social and economic goals to harmony (to the extent that they are theoretically seperable).

In Hammer v. Dagenhart, the Supreme Court assessed the legitimacy of a 1916 Congressional act interdicting, inter alia, the interstate traffic in the products of child labor. Ronald H. Dagenhart of western North Carolina brought suit on behalf of his three minor sons, all who assembled shoes which the statute had forbidden from entering the stream of interstate commerce. For the five-Justice majority, Justice Day repudiated Congress’s claimed right to regulate labor conditions subsidiarily through the regulation of interstate commerce, distinguishing similar regulations which the Court had previously upheld on the grounds that:

Justice Day went on to conclude that, "[o]ver interstate transportation, or its incidents, the regulatory power of Congress is ample, but the production of articles intended for interstate commerce, is a matter of local regulation."

Justice Holmes dissented forcefully and eloquently (and, ultimately, more memorably) to the proposition that Congress did not enjoy the right to regulate interstate commerce simply because its regulation was not aimed at the commerce itself or that product’s effect while in commerce. Among other persuasive arguments very specific to American constitutional structure, Holmes made the astute point that, "It does not matter whether the supposed evil precedes or follows the transportation. It is enough that in the opinion of Congress the transportation encourages the evil." Holmes cited as further support that the Court had seen fit to permit the Mann Act which regulated the trade in "white slavery" only the previous year in Hoke v. United States, assuredly not because of any inherent harm in the transport.

Arguably because the logic was wholly untenable both textually and logically, Hammer is no longer the law of the land. Though the regulation of labor through commerce power was once again struck down in Carter v. Carter Coal Co. upon the logic that "the effect of the labor provisions [of a Congressional act regulating the wages and hours of coal miners] primarily falls upon production and not commerce," this type of regulation was ultimately held legitimate five years later. Among the Roosevelt Administration’s interventionist New Deal policies was the Fair Labor Standards Act of 1938 which, passed under the Commerce power, prescribed a minimum wage. Like the child labor law, this minimum wage requirement attacked the wage problem as a matter of commerce. Presented with the issue, the Court ruled in United States v. Darby that the distinction between regulations which define or impose an order on interstate commerce and those which prohibit a type of commerce so as to have an effect upon an evil involved in that good’s production was essentially a false one. While this result rested in large part upon a recognition that Congress’s enumerated Commerce power was not textually limited (a result which does not obtain for contracting parties to the GATT), it also emerged out of the steady accumulation of a momentum behind the notion that the regulation of external commerce in products was both an effective and rational way to get at the clear evil of substandard labor conditions--a problem which might otherwise have been exceedingly difficult to reach, if not untouchable.

For present purposes, a more significant logic emerging in Darby was the justification of Commerce-based labor regulations as a means of ensuring "fairness" among the several states. Whereas the principal textual basis for challenges to the federal regulation of labor through Commerce power had previously been the Tenth Amendment, Darby introduced a new, more legitimate reason for Congress’s "local" intervention in labor ("process") issues under the Commerce power--the federal government’s need to put a check on the unfair competition between states in substandard labor. The personification of the federal Commerce power role in regulating labor had metamorphosed from that of the officious intermeddler in 1918 to that of the benevolent referee in 1941. "The [Fair Labor Standards] Act is thus directed at the suppression of a method or kind of competition in interstate commerce which it has in effect condemned as ‘unfair’, as the Clayton Act has condemned other ‘unfair methods of competition’ made effective through interstate commerce."

This justification for the regulation of commerce in those items produced through "unfair" labor practices mirrors the rhetoric associated with the GATT’s Article VI dumping, and subsidy rules. Just as Congress can, under its Commerce power, deem certain productive activities illegal because they are "unfair," the GATT recognizes a series of trade and production processes and activities as injurious to the international trade régime, despite their effect of lowering consumer prices. This is so because, rather than the "level playing field" to which the GATT rules aspire, such practices create instead, if you will, a most ungainly topography. The moniker of "unfair practice" for dumping, subsidies and other activities within the GATT is applied to those practices which derivate from a basic set of acceptable distributive principles of "fairness" in market interactions. The distributive fear is not that there will be winners or losers per se, of course, but rather that the allocation of wins and losses will not be meritocratic--not a function of risks taken or work invested. In the domestic sphere, insider trading and monopoly pricing, for example, "unfairly" rob consumers of surplus, rewarding it to producers who did not take a true risk with their capital to earn it. Though the fairness label is clearly subjective, the designation ought to be clothed in a certain legitimacy or transcendent value since the prohibitions inspired by it are in harmony with the several regulatory conditions that most developed nations regard as prerequisites for the proper functioning of a sustainable market economy in the long term.

Moving from the theoretical to the practical, we may posit two brands of competition. In fair competition, we ought to be entirely willing to countenance long-run industrial restructuring with all its pain and costs if the competitive pressure comes from a more efficient or sophisticated production process. Despite the losses to domestic industry, this is desirable because the alternatives--protecting inefficient industries from these forces--are not. Protection just prolongs the pain, and at a cost. More importantly, such protection blunts the market’s natural tendency to reallocate those now-obsolete resources into more efficient uses. To the extent that capital is fungible, we would prefer to see it allocated in the long run to building Boeing 777s than Wright brother-era planes for commercial transport. In effect, this type of competition results only in pecuniary externalities, and is therefore not to be reproached.

But when the "efficiency" (read here to mean simply an increase in productivity: the bare capacity to provide a product at a lower cost, higher quality, lower price, or some permutation of the three) is the result not of a market enhancement, but instead a practice which violates essential norms of market "fair play" as described above, our comfort with the need to restructure disintegrates. If Hyundai creates a new fuel-cell engine which is twice as energy-efficient and sells for half the price of an internal combustion engine, for example, we are pleased indeed when those companies who produce the obsolete engines either fail, making way for the new fuel-cell-producing companies, or rearrange their productive capacities to compete in the new market. If a large plastics producer in the United States wins a great share of the market because it is secretly employing slave labor, on the other hand, we would be loathe to hope that the reduced prices or costs, or the increased quality of the product that this permits, will cause the "production innovation" to be replicated worldwide. Indeed, if the end result is that a novel efficiency will eventually permeate the industry, its implementation would be horrifying in the case of many "unfair" practices.

Domestically, of course, American law deems a contract for slave labor as unconscionable, no matter the terms of the bargain. But internationally, a foreign trading partner might not maintain or might not enforce such a law, and its trading partners can have little to say to that sovereign nation about its practice or its companies’ practices. This, it seems, is one of several typologies of a worst-case scenario for a trade policy which ignores any objectionable normative component to trade partners’ production practices. The result is not merely that consumers are prevented from exercising their preferences, but that they are made to feel even complicit in the proliferation of the objectionable practice.

Accordingly, the profound interest in regulatory measures for environment and labor standards emerges out of the common desire to ameliorate or eliminate the unfairness concomitant with a set of confirmed market failures in certain types of transactions. Properly understood, these market failures--primarily "external" costs imposed upon unwilling and/or unwitting third parties and asymmetries of information about nations’ willingness to pay for the public good or non-use value of environmental protection--represent a reservoir of dissipated wealth which the international trade régime has, as of yet, been unwilling or unable to capture despite the formidable tools at its disposal.

Because the value-creating mechanism in the free trade régime (quality augmentation aside) is the gradient between prices across jurisdictional borders, most recent advances in environmental policy have involved the imposition of measures to help markets "get prices right" in order to mop up those environmental market failures alluded to infra. It should therefore come as no surprise that the GATT’s interpretation of its environmental role (whatever it may be) relative to its free trade mandate preoccupies the research of would-be international environmental trade policy makers. Precisely because the GATT’s Appellate Body has functionally removed Article XX from the policy tool box, environmentalists have sought "back door" solutions to basic international environmental policy problems, the effort to touch foreigners’ production methods being an emblematic example. The process/product distinction, in other words, conspicuously misses the point: A trade policy has the capacity to let prices do the environmental work by allocating scarce resources more, not less effectively and efficiently.

One intriguing recent proposal to resolve the problem is simply to redefine the word "product" as it appears in Article III and Ad Article III. Christopher Cherry counsels that the very notion of a product ought to comprise its entire history--from "cradle to grave"--in a manner consistent with environmental accounting. The idea is worth a brief recapitulation and further articulation here, as it illustrates well some of the legal-intrepretational challenges of the process/product distinction as enshrined in GATT/WTO Appellate Body jurisprudence.

To begin with, the GATT’s definition of "product" is nowhere specifically enumerated, of course. Rather, it has come to be construed as simply that which falls physically within the four corners of the object itself--an exclusively physical signification--not as a result of any formal preambulatory definition, but instead out of the GATT’s Appellate Body rulings in Tunas I & II. Rather like the old saw about Constitutional law, the GATT definition of the term "product" is simply what the GATT Dispute Resolution body says it is. This construction of the term, of course, is entirely rational, and its meaning is certainly a matter of law over which reasonable people could differ. But while a reclassification would be no less legitimate, it would be a natural subject for critique and gainsaying. Obstinence in the face of this interpretive reclassification is, if the cross-topical overstretch can be forgiven, the "separate but equal" of trade and environment. An alternative construction of "product" would be more fair, make more sense, and solve an entire class of problems. While not exactly a legal fiction, its persistence entrenches the WTO Appellate Body’s interpretive end-run around those guarantees which were seemingly bargained for formally under GATT Article XX.

At the same time, Cherry’s proposition that a product carries the process of its production with it, first of all, is not a mere sleight of hand. It has reputable legal roots, though he does not detail them. German law, for instance, has required for several years that manufacturers take responsibility for the cost associated with the disposal of their products, a charge which many firms have implemented through recovery programs which reuse valuable components. While perhaps not explicitly, this policy begs the consumer and the producer to reconsider their modern conception of the market as a place where transitory, impersonal transactions take place. Rather, there is a certain history to a transaction for a good which we all recognize if we are willing to abstract a bit.

A consumer also shares a degree of responsibility for a product’s effects in the world--past and future (if foreseeable)--throughout its life cycle. If it is constructed by forced labor, we indisputably endorse and encourage that activity just by our market choice even if we oppose that activity on moral grounds. If we knowlingly purchase cheaper petroleum from a firm which dumps its hazardous waste in the rivers around it, we are complicit in the harm which that waste causes to the plant’s neighboring families, though nothing within the measure of the product itself gives us this indication.

The difference between the three dimensional, physical product and, to coin the term, the "essential" product, is striking. Adding the fourth dimension, as it were, to the product’s length, width and height (i.e., keeping track of the character and the consequences of its change over time) may strike some as too "metaphysical". All the same, it has certain consequential philosophical underpinnings:

Aristotle defines the form of a thing such as a home (which can only come into existence once a house has been transformed by a change over time) to exist only through the inclusion within that thing’s nature of the change which has operated upon it.

We are entirely willing to understand animate objects in terms of their fourth-dimensional development; to wit: a person is defined by his experiences; take them away when conceiving of someone and you have robbed him of much of his personality and soul. Still, Aristotle’s "essential" definition of a thing’s nature may seem slightly foreign when applied to inanimate objects. But when we acknowledge that products have "life cycles", we are not trying to transform them from the inanimate to the animate. Rather, we are recognizing, almost as if in a play by Chekhov or Sartre, that the interaction with objects and instrumentalities (here the consumption of products) forms an interface between living things surrounding goods in a way which has external, diachronic import (that is, it has repercussions in space and over time, beyond those in the atomistic, fleeting "instant" of the transaction). The "essential" product, in the metaphysical sense, gains its currency from the fact that those operations which have been brought to bear upon a thing, as much as the physical items which constitute it, have social and economic meaning because both the production and consumption of that product tangibly effect peoples’ lives at the point of production, the point of consumption and between. This explains in part our willingness to abstract away from the physical shoes that the Dagenhart boys were producing to feel a certain comfort with a regulation on commerce in those shoes; the product, in effect, is tainted, because our purchase of them will lead to more hardship and forfeited youths.

In one arena--environmental economics--this philosophical proposal is not even novel; because environmental harm is so frequently the result of gross externalities, the "history" of a product, including what happens during its production, transport and consumption, is a paramount consideration. A mercury battery, once incarnate, ussually does you no harm. But does that simple fact dismiss its effect even if its production in a given plant causes forty serious birth defects per year in the neighboring community from mercury poisoning? No matter one’s perspective on the allocation of legal norms and property entitlements, this product cannot be divorced from the process which creates it, and its meaning must be made descriptively thicker by a consideration of its history.

Environmental economists see this health and safety hazard as an input, with associated costs which must be accounted for, like labor, capital and land. They see the history of two otherwise identical products as clearly differentiating them in an economically meaningful way. That is why information is so important to the would-be green consumer; the fact that increasingly globalized trade reduces consumers’ access to reliable information as to the actual social and environmental impact--including external costs--a product has had during its life cycle is thus all the more problematic.

In a practical sense, Cherry’s product redefinition argument is compelling on several grounds. First, of course, it is no strain to conclude that the Article XX architecture is imperfect in its current state given the high-profile clash between free-trade and environmental policy goals and the flurry of scholarly work it has engendered. Either the text or the interpretation needs to be changed if Article XX is to have any purpose. And because the text is much more difficult to change than an interpretation which has no binding precedential value, the insertion of a broadened definition of product is an appealing stroke of economy. Third, there is an historical argument of changed circumstances since Article XX came into force in 1947; since a notion of international environmental policy did not even exist at the time of its inception, reading the GATT as textually blind to such a typical source of environmental harm as a product’s production history, in the contemporary context, seems foolhardy. It would be tantamount to reading the Constitution as blind to the regulation of automobile traffic because the word "car" is not mentioned in the text.

The case for the interpretive revision becomes more compelling when juxtaposed with other, related options as well. Just one clearly impracticable example would be the proposal that environmental externalities be codified and permitted as acceptable classificatory bases for the discrimination between otherwise like-products. Still within the vanguard of the "essential" product school, this innovation would operate through the mechanism of Article III, rather than require a favorable finding (with its conjoined, albatross-like burden of proof) for an Article XX exception. Product classification has never been a clean, neat science within international trade law, and here the notion is to capitalize upon the inherent malleability of the art to codify those externalities worthy of differentiation. For example, highly-sulfurous coal, given the type of like-products determinations utilized in the Japanese Schochu case, could be classified simply as "coal" for the purposes of Article III. But an externality-based differentiation principle would clearly make the clean and dirty coal not like products. Such an argument was proposed explicitly for a unilateral US regulation in US Reformulated Gasoline, but was rejected by the GATT Panel as an illegitimate basis for defining products as not "like."

The unmistakable difficulties with an externality-based classification abound. Briefly, first, this would require a textual change to Article III, which is totally unfeasible. Second, as explained supra, externalities are, to a crucial extent, the products of variegated, often conflicting norms. Although it is likely that there would be a fair degree of universality in the development of a multilateral consensus on a few acceptable externalities, many choices would clash and still others would be claimed only for their ancillary protectionist value. In short, any solution which requires an incorporation and definition of externalities for product classification under Article III would require the negotiation of a multilateral treaty on the subject. An interpretive change to recognize the "product" language in Article III as "essential" would be much preferred, but also unlikely in a short time frame. Embracing the notion of the "essential" product is a long run remedy, in the same vein and category of as the development of a multilateral framework on acceptable environmental trade measures.

Perhaps the literature on the "essential" product reading has been muted simply because the proposal is not very realistic as a medium-term measure. Rejecting the more formal categorization of product differentiation standards in Article III based on external effects leaves the international environmental policymaker with even fewer formally realizable solutions, particularly because externalities are so rampant that any effort to recognize a product’s history is bound to be both incomplete and unwieldy. In the face of the very heterogeneity of norms which necessarily underlie any system of externalities, the task is daunting. For the GATT’s very practical purposes, in fact, even the "essential" product interpretation is unlikely to prevail soon, and so we move on below to more pragmatic proposals for etching out a viable limiting principle for acceptable Article XX environmental exceptions in the interim.

While the US labor-environment jurisprudential analogy above is apt for the sake of bringing out complications with the naked process/product distinction, an obvious weakness is embedded in the US Constitution-GATT analogy. In the GATT example of Tunas I & II and Sprimps I & II, a decision to permit a regulation would have conferred unilateral regulatory power upon an individual political subdivision; it would have permitted unilateral trade measures by a member nation. On the other hand, a Supreme Court ruling to permit a Commerce power regulation applies to Congress--the supranational body which coordinates multilateral acts among all subsidiary states. The equivalent institution to Congress would be the WTO itself.

Importantly, first, this structural/organizational flaw does not seem to contradict the speciousness of the process/product split logic implied by the analogy, nor does it mitigate the apparent gravity of the Appellate Body’s recent functional eviscerations of the Article XX enumerated exceptions. It does, however, seem to lend support to the compromise notion that any environmental regulatory measures, to be deemed acceptable in the long term, ought to orbit around the centralized infrastructure of the WTO itself, or at least must be multilaterally negotiated. Here, even in its weakness, the analogy has strength for exposing the implications of that structural difference; individual, subsidiary states, unlike Congress, have an incentive to cheat if they make their own unilateral regulations in the pattern described above as a "race-to-the-bottom". The question of abuse will be addressed infra, but suffice it to say here that this is the most prominent justification for vesting in the Appellate Body a great deal of power and discretion to review claimed unilateral exemptions.

Many commentators have taken to defending those unilateral environmental trade measures which the GATT/WTO dispute resolution arm has invalidated, motivated by the seeming compatibility of these free-trade exceptions with the plain language of Article XX. If ever there were a reason to permit unilateral trade measures, they point out, the protection of a nation’s exhaustible natural resources or its health and safety standards should come first to mind. Nevertheless, the fear of trade wars and the abuse of the few textual grants of permissible unilateral exceptions within the GATT give pause lest they "eat up the rule under the guise of an exception." Let there be environmental exceptions to the GATT Article III rules, we might say, but let them be limited to those on which all can agree, administered centrally and multilaterally.

The Supreme Court enjoys the ability under the Commerce Clause to invalidate any state regulations which unduly burden commerce between the states themselves or the states and foreign nations. Unsurprisingly given the relative youth of the WTO’s dispute resolution Appellate Body (and, a fortiori, its poorly-elaborated jurisprudence on nation’s claims to legitimate exceptions to the free-trade mandate), some commentators have turned to the US Commerce Clause to locate a limiting principle for prescriptive jurisdiction where potentially legitimate trade-blocking excpetions emerge. To square the analogy, if North Carolina, and not Congress had chosen to ban the import of shoes constructed by child labor, the statute might have been constitutionally preempted, running afoul of the "dormant" Commerce Clause as an excessive burden on free trade among the states in a manner more purely parallel to the GATT Appellate Body’s invalidation of the US MMPA. After Gibbons v. Ogden in 1824, Congress reserved for itself the exclusive competence to place obstacles in the way of free trade between states. But with no specific textual basis, the Court’s construction of the Commerce Clause’s grant to Congress of the "Power [to] regulate Commerce [among] the several States," includes the further, judicial imperative to negate encumbrances upon commerce between the states even when Congress has not acted on the matter at hand. Free trade among the states, the Court has multiply declared, is the state of nature which the framers intended by the Commerce Clause.

Importantly, the Court’s approach is manifestly and advantageously (if only technically slightly) different from the GATT/WTO Appellate Body’s test for Article XX’s chapeau, because it 1) applies a different quantum-level test for state actions which are enacted with a demonstrable protectionist motive; 2) works backwards from purpose to result, rather than the reverse (as the Appellate Body does by seeking first the validity of the claimed end under an Article XX enumerated exception, then proceeding to the chapeau to assess intent); and, most importantly, 3) applies a cost-benefit balancing test when a regulation cannot be shown to have protectionist roots. To the contrary, the Appellate Body has yet to consider at all in environment cases the claimed (non-economic) regulatory benefit to the domestic jurisdiction as a counterweight to the dangers of walling off a certain channel of commerce, perhaps because it sees no welfare-expanding purpose for regulation at all.

Despite the apparent appeal, the US Commerce Clause approach has been multiply criticized as unworkable, primarily out of concern that the WTO Appellate Body simply lacks the resources, cerdibility and powers required to effectuate the integration of a US Supreme Court-style balancing test. Importantly, the Appellate Body does not have the capacity to find facts (dormant Commerce Clause balancing tests tend to be extremely fact-intensive), it does not sit within a broader federal structure of explicitly separated powers and it has, as of yet, little political legitimacy. Because the viability of a balancing test for Article XX which operated like a Commerce Clause test would be contingent upon the immediate and sucessful development of the WTO Appellate Body’s jurisprudential capacities, the analogy to US commerce regulation will end here.

A thread which continues outside of the analogy to the Commerce Clause, however, is the theoretical possibility that nations’ external-local environmental trade preferences might be legitimately safeguarded through unilateral trade measures. Should the WTO be concerned in the abstract about permitting unilateralism?

A host of arguments have been marshaled against permitting unilateral trade regulations under the WTO/GATT régime, many of which extend outside the realm of specifically environmental measures. In a general sense, since the GATT is a multilateral framework over trade intended to provide a bulwark against the development of barriers to trade--be they unilateral, bilateral, plurilateral or multilateral--unilateralism above all seems an awkward instrument to be countenanced under its auspices. After all, the great innovations of the GATT were the principles of "most favored nation" status ("MFN") and Article III’s nondiscrimination provision, both Herculean efforts at multilateralizing previously perniciously unilateral actions or plurilateral interactions. As Daniel Farber comments, "[i]t is quite possible to favor either free trade or environmental protection as purely unilateral local policies. Anyone who favors multilateral action on either one, however, should probably favor multilateral action on the other as well." First it is claimed that unilateral measures will always tend to favor domestic interests over external ones, violating by nature the GATT’s core Article III (and Article XX chapeau) non-discrimination principle. Such policies, if they are indeed the inevitable outgrowth of permissible unilateralism, not only needlessly distort international trade but also breed cynicism among those who would otherwise support the colorable purpose behind the measure. Second, of course, they risk inciting a race-to-the-bottom or an insidious retaliatory trade war, with all of its associated economic and political baggage. Third, they fly in the face of the central notion of pacta sunt severanda--that laws between sovereign nations are mutually binding because they represent a consensus to be bound. Fourth, as Robert Hudec points out, a system which countenances the use of unilateral measures invites less-than-scrupulous trading partners to develop double standards in their trade policies. Lastly, and related to the previous points, unilateral measures are most useful for the precise reason that they are most dangerous; they provide outlets only for those who fundamentally disagree with most other nations. Arguments for their necessity depend upon a discontinuity between the party imposing the measure’s legal/economic norms and those of most participants in the world trading system. Certainly if we were all to have the sanction to impose our idiosyncratic will upon the world trading system only to satisfy our esoteric whim, trade would suffer and resentments would pile up; "[t]he General Agreement. . . would provide legal security only in respect of trade between a limited number of contracting parties with identical internal regulations."

The case against unilateralism is strong, but not airtight. As John Jackson has remarked, "[g]iven the imperfections of the international system, . . . environmental policy experts can legitimately argue that there must be some room for unilateral nation state actions designed to support the world environment." Similarly, Edith Brown Weiss laments, "surely there must be instances where countries are at the forefront of identifying risks to areas of common concern and should not be forced as a principle of international law to continue to contribute to environmental degradation." Finally, as Naomi Roht-Arriaza remarks, "[o]ften the example and moral pressure of unilateral action or the threat of unilateral sanctions accelerate [sic] the process of international lawmaking." Following are three reasonable justifications for medium-term unilateralism in the face of a régime which prohibits a nation’s preferred trade activity.

Some see unilateral measures as obligatory given their aura of "civil disobedience" under a prohibitionist régime. This point of view is especially in evidence where there is a conflict of norms on the trade issue in question (as in the example of trade and the environment). Unlike the justifications below, an act in "civil disobedience" is not aimed at developing an eventual multilateral agreement per se. Instead, it is most useful when time is of the essence and relief is not available on the immediate horizon (e.g., rejecting the import of a product which a nation perceives as dangerous but for which the GATT Appellate Body has not granted an exemption). It is this characteristic of unilateral measures--"the ability of unilateral measures to operate quickly and precisely to improve the environment [and] the relative ease with which they may be enacted" which makes them attractive.

It is fascinating, in this context, first, to note the sense of public law which the rhetoric of "civil disobedience" invokes; if we see the GATT as simply a series of contracts between nations on a set of bound tariffs, the talk of civil disobedience seems a bit melodramatic, if not downright artificial. The issue is entirely bound up in the overall problem of the dual perception of the GATT as simultaneously public and private law, which renders the task of assessing the appropriate way to register of dissent or grievance exceedingly difficult. Jagdish Bhagwati, for example, invites us to consider the indecorousness of unilateral environmental trade measures by appealing to "Mahatma Ghandi’s idea of nonviolence," which, he reminds, "spread far and wide, not because India had economic power to force it on others. . . [but] simply because of its inherent and powerful moral attractiveness." Impliedly, here a nation’s decision not to consume (ironically, a peaceable boycott) is tantamount to a Bismarckienne realpolitik power game; at least where the civil disobedience strand motivates a GATT-inconsistent trade measure. With this perspective a reference to Ghandi seems only to reinforce the point it was intended to refute. Polemics aside, if nothing more, civil disobedience unilateralism is fascinating for the raw nerves and conflicting undercurrents of international economic law it exposes. Other justifications, below, focus on the more designedly constructive benefits of a unilateral measure.

Outside of emergencies and the perceived imperative to act on principle, nations may rally behind an environmental trade measure in order to capture the positive externality of the "bandwagon effect". Similar to the very process by which customary international law is developed, unilateral measures initially supported by a small minority may enkindle a consensus-building process where, had a party simply complied, one might otherwise not have developed; here, the unilateral measure is a sort of "seed crystal". If civil disobedience is a tool of the relatively powerless, those with the power of suasion are the agents provocateurs. While perhaps in the minority where subjective norms about trade regulations are concerned, a clever, persuasive nation might nonetheless have the capacity to bring about this type of regulation in the face of an amorphous lack of international consensus (i.e., ultimately, a multilateral framework may develop on the type of regulation initially imposed unilaterally). Because this type of unilateral action is not backed by power, but rather by intellect and craft, it is perhaps the most interesting, yet likely to be the most infrequent in an international arena still dominated by power politics. The domestic political analog would be the revolutionary.

The temptation is to pass over this role lightly, and certainly few examples of multilateral trade agreements with such "persuasive" roots come easily to mind; still, "on the contrary," as Abraham and Antonia Handler Chayes attest, "the fundamental instrument for maintaining compliance with treaties at an acceptable level is an iterative process of discourse among the parties the treaty organization, and the wider public." Not surprisingly, the Chayeses go on to describe the revelation in their research that this type of "sovereignty" of discourse is as important for maintaining (or "managing") a treaty and compliance with it, as it is for encouraging adaptations and adjustments.