The World Trade Organization 20 Years On: Global Governance by Judiciary

This article presents a narrative about the building of an effective, legitimate judicial system in the World Trade Organization (WTO) through a period of intense diplomatic and political divisiveness and prevailing perception of impasse and malaise in the Organization. At the centre of the narrative is the Appellate Body of the WTO, a standing body of seven jurists charged with deciding appeals of law. The Appellate Body, as will be elaborated, responded to the political conflict and paralysis at the WTO by distancing itself from the Organization and making a number of crucial jurisprudential moves that led to its transformation into an independent court, which has often decided controversial questions in balanced or deferential ways that display, at best, neutrality to the neo-liberal ‘deep integration’ trade agenda reflected in the Uruguay Round of multilateral trade negotiations and many of its results, such as the WTO Agreements on Intellectual Property and on Technical Barriers to Trade, for example. In the early years, the Appellate Body’s deviation from some of the basic tenets of the trade insiders at the WTO led to an open conflict with the trade policy elite, including the delegates of the Members who sit as diplomatic representatives of the membership in Geneva. The end result, however, was the acceptance of the Appellate Body’s authority. The same consensus practice of political and diplomatic decision making at the WTO that made negotiating breakthroughs elusive also made it essentially impossible for the Members to threaten or pressure the Appellate Body effectively since, ultimately, overruling any of its decisions, either through the amendment of a WTO treaty or through an ‘authoritative interpretation’, could not be done absent a consensus of the Members.

1 Introduction

The judicialization of international law through specialized tribunals is an often remarked trend of the last decades. For some, judicialization merely increases anxieties about fragmentation; for others, it inspires hopes that international law, as law, will finally enjoy the institutional thickness that it traditionally lacked when tethered to diplomatic or political arrangements. One would expect judicialization of international law to be a reflection of the enhanced legitimacy and dynamic evolution of substantive norms and the political and diplomatic processes that generate them, which is the story that Ruti Teitel persuasively tells in Humanity’s Law with respect to human rights and the law of war.¹

The World Trade Organization (WTO) presents an alternative and, at first glance, perhaps puzzling counter-narrative. The WTO was conceived at the height of neo-liberalism or the Washington consensus.² But by the time that Uruguay Round was complete and the WTO was born, the atmosphere had changed. The legitimacy of the ‘deep integration’ bargain struck in the Uruguay Round of negotiations and reflected in the WTO treaties came into question almost as soon as the ink was dry, so to speak. It was the riots of Seattle that made the WTO a household name, and it became famous or notorious as a target for the anti-globalization movement.³ But the rioting outside was only part of the story. A legitimacy crisis within the WTO was already brewing with developing countries feeling buyer’s remorse about the result of the Uruguay Round, where, in areas such as trade in services and intellectual property rights, they had made considerable concessions, with (as some developing nations increasingly felt) little concrete action in return. After numerous attempts to conclude a new round of negotiations, which involved the launch of the Doha Round of development in the shadow of the 9/11 attacks, the talks were finally abandoned late in 2015 at the conclusion of the Nairobi WTO Ministerial. According to conventional wisdom and despite agreements on information technology and trade facilitation (customs formalities), the official acceptance of the collapse of the Doha Round constituted almost two decades of political paralysis.

Yet if we turn from the political and diplomatic setting to the dispute settlement system of the WTO, we see a judicial branch in full evolution through this entire period, entertaining hundreds of claims and producing a vast jurisprudential acquis. Despite the deep division among WTO Members about the future of the multilateral trading system, the continuing salience of critiques of economic globalization, the many other events that might be seen as destabilizing international economic order (for instance, the financial crisis of 2007–2011), the sensitive issues often involved in WTO legal disputes (environment, animal welfare, preferences for developing countries, subsidies for renewable energy and the management of scarce natural resources) and the major challenges to binding dispute settlement in other areas of international economic law (investor–state arbitration), the WTO judicial system has been largely spared attacks on its legitimacy.

The question of what makes an international judicial system effective, successful or legitimate goes to the much debated issues about the meaning of ‘compliance’ in international law,⁴ the relationship between empirical or factual legitimacy and normative legitimacy and, ultimately, the interaction of politics and law in international relations.⁵ For now, without prejudging what, or, indeed, whether there are, satisfactory or satisfying answers in these respects at the level of general theory, one can speak in a common sense way of the achievement or success of the WTO judicial system over the last two decades, which has excited admiration and even envy in international legal scholars and practitioners. One does not have to be Donald Trump to see that perceived ‘success’ and legitimacy do have some significant positive relation. Aside from the sheer number of disputes that the states parties (Members) have been prepared to submit to judicialized dispute settlement,⁶ which, increasingly so, is itself some sort of sign at least of empirical legitimacy, one can point to the relative lack of instances where Members have, upon losing a ruling, explicitly chosen not to implement it (ultimately on pain of retaliatory sanctions). While losing parties and sometimes other WTO Members have criticized individual rulings, including by the Appellate Body, these critiques have rarely challenged the overall authority or legitimacy of the WTO judicial mechanism. In the early years of the WTO’s judicial system, some critics, from academia and think tank-type policy institutions,⁷ did question whether in light of apparent judicial activism by the Appellate Body some kind of political or diplomatic control needed to be re-established over judicialized dispute settlement, but these calls never developed lasting traction among WTO Members.

Finally, as already mentioned, in cases involving sensitive issues of policy space, such as trade and environment disputes, the WTO judicial system largely succeeded in avoiding becoming a target of anti-globalization activists or constituencies more generally concerned with non-trade values that could easily be seen to be in conflict with what insiders would regard as the central, liberalizing, if not neo-liberal, mission of the WTO. Only recently has one WTO Member, the USA, launched a persistent attack threatening the Appellate Body’s independence. It has attacked the Appellate Body’s judgments on a seemingly very technical, but sensitive, issue (‘zeroing’), which concerns the application of WTO legal disciplines on a form of unilateral trade action, anti-dumping duties,⁸ even attempting to politicize the process of appointment and reappointment of Appellate Body Members.

This article is aimed at presenting a narrative about the building of a judicial system in the WTO through a period of intense diplomatic and political divisiveness and prevailing perception of impasse and malaise in the Organization. At the centre of the narrative is the Appellate Body of the WTO, a standing body of seven jurists charged with deciding appeals of law.⁹ The Appellate Body, as will be elaborated, has responded to the political conflict and paralysis in the WTO by distancing itself from the Organization and making a number of crucial jurisprudential moves that have led to its transformation into an independent court, which has often decided controversial questions in balanced or deferential ways that indicate neutrality or even caution in regard to the neo-liberal ‘deep integration’ trade agenda reflected in the Uruguay Round of multilateral trade negotiations and many of its results, such as the WTO Agreements on Intellectual Property and on Technical Barriers to Trade (TBT Agreement), for example.¹⁰

In the early years, the Appellate Body’s deviation from some of the basic tenets of the trade insiders at the WTO led to an open conflict with the trade policy elite, including the delegates of Members who sat as diplomatic representatives of the membership in Geneva. The end result, however, was the acceptance of the Appellate Body’s authority. The same consensus practice of political and diplomatic decision making at the WTO that made negotiating breakthroughs elusive also made it essentially im possible for the Members to threaten or pressure the Appellate Body effectively, since overruling any of its decisions, either through amendment of a WTO treaty or through an ‘authoritative interpretation’, ultimately could not be done absent a consensus of the Members. Appellate Body Members were well aware of this situation. Publicly, at least, some Members actually expressed a wish that their rulings could be politically adjusted more easily, implying that the Appellate Body had to accept too much of a burden for the legitimacy of the WTO as a legal system, especially since there were gaps or ambiguities in the legal text.

At the same time, however, it is clear that the Appellate Body was empowered or protected as an independent judiciary because of the obstacles that the consensus decision-making practice, combined with the general context of divisiveness within the Organization, that made it such an effort to change course. The Appellate Body, through case law that may often appear inconsistent – at least where various shifts in approach are inadequately explained¹¹ – has nevertheless developed a number of judicial policies, which have, overall, oriented adjudication towards maintaining a balance between trade liberalization and the right to regulate, i.e., domestic regulatory autonomy. It is very conscious of the legitimacy issues that arise when it passes judgment over domestic policies in sensitive areas of public interest, carefully avoiding the appearance that the Appellate Body is the agent, much less the avant-garde of the neo-liberal project represented by the Uruguay Round, or inspired by the ‘deep liberalization’ telos reflected in agreements such as the TRIPs Agreement.¹² The Appellate Body has taken pains to practise an unobtrusive or light review of the main lines of the policies and has tried to be deferential to the policy objectives that are sought as well as to the domestic choice of the degree of fulfilment of those objectives. The evolution of these elements of jurisprudence is the major substance of the following narrative. Part of the balance, however, has not only been this deference in sensitive cases but also the meaningful, if often procedural, discipline of unilateral trade remedies as well as the careful scrutiny of the fine detail of public policies for arbitrariness or protectionist abuse in implementation. In the Conclusion, this article considers the durability of the edifice constructed by the Appellate Body in the face of the current US attack and whether it is likely to provoke others of a similar kind.

2 The Setting: The Birth of the Appellate Body out of the Troubled GATT-to-WTO Transformation

To understand the judicialization of the WTO, we have to understand not only the dispute settlement procedures of the WTO but also the regime out of which those procedures emerged – the General Agreement on Tariffs and Trade (GATT).¹³ Understanding the regime entails not only an awareness of its main substantive norms but also some awareness of the informal norms, practices and understandings that are not reflected in the legal texts as well as ‘soft law’ declarations or guidelines. Then we have to consider both the way the dispute settlement procedures and the regime were transformed with the creation of the WTO.

The GATT was born from the failure of an ambitious project for a global trade regulatory agency, the International Trade Organization. While it is a one state–one vote international Organization, where decisions including the amendment of the treaty and the creation of new obligations are decided by consensus, the original GATT was dominated by the USA and its post-war partners.The Communist countries were, generally speaking, not GATT members during the Cold War, and with increasing decolonization, developing countries joined the GATT, but by the 1960s, they were increasingly critical of some of the structures of the GATT and were, hence, not viewed as full partners in decision making.

At the start, the GATT was little more than a bare bones structure for progressive negotiated reduction of tariffs on a reciprocal basis among sovereign states, subject to most favoured-nation and national treatment rules. Thus, there was no requirement in the GATT to eliminate tariffs at any given rate or pace. A paramount goal was the avoidance of a protectionist summum malum – the situation where domestic social or economic pressures lead some states to increase or reinstate barriers to trade, thus triggering a ‘tit-for-tat’ response by other states and, eventually, a freefall into discriminatory protectionism that is disastrous for the global economy. This sort of behaviour was widely perceived by the founders of the Bretton Woods system to have led eventually to perilous instability in the interstate system and economic catastrophe in the interwar years, and these phenomena were seen as having contributed to the climate that made fascism, and World War II itself, possible.

In the GATT, allowance was made for a temporary balance of payments-based import restrictions (Articles XII–XV), for safeguards in response to the injury to domestic industries from sudden surges of imports (Article XIX) and for the negotiated rebalancing of concessions (Article XXVIII). The national treatment obligation (Article III),¹⁴ along with Article I on the most favoured nation, was a means of preventing member states from instituting discriminatory domestic policies that would distort competition between domestic and imported products (in other words, cheat on the negotiated bargain), not a mechanism for liberalization per se. Some kinds of domestic policies received explicit, but ambiguous, treatment under the GATT – subsidies were recognized not only as potentially (and illegitimately) trade distorting but also as being in principle not illegal or illegitimate. In response to this studied ambiguity, the GATT explicitly permitted, under certain constraints, self-help in the form of countervailing duties.

Second, the GATT did not, generally, require that the member states constrain private restrictive business practices, but ‘dumping’ (selling abroad at lower prices than in the home market) was disapproved, and the self-help of anti-dumping duties, again under certain constraints, was made the accepted remedy. Furthermore, even discriminatory domestic policies might be permitted if they did not entail arbitrary or unjustified discrimination and could be linked, more or less tightly, to overriding public policy goals such as the protection of human life or health, the conservation of exhaustible natural resources or the protection of public morals (Article XX).

At the same time, the dispute settlement practice evolved out of the general language in Article XXIII of the 1947 GATT into a system that eventually displayed important elements of legalization. The new trade policy elite developed professional working procedures and norms within the GATT, organized the agenda for negotiations and – with very little to go on from the treaty text itself – created and sustained a relatively effective arbitral mechanism for dispute settlement. As persons with the bent of managers and technical specialists, they tended to understand the trade system in terms of the policy science of economics, not in terms of a grand normative political vision. A sense of pride developed that an international regime was being evolved that was not vulnerable to the open-ended normative controversies and conflicts that plagued most international institutions and regimes, most notably, for instance, the United Nations, which is well described by Joseph Weiler:

A dominant feature of the GATT was its self-referential and even communitarian ethos explicable in constructivist terms. The GATT successfully managed a relative insulation from the ‘outside’ world of international relations and established among its practitioners a closely knit environment revolving round a certain set of shared normative values (of free trade) and shared institutional (and personal) ambitions situated in a matrix of long-term first-name contacts and friendly personal relationships. GATT operatives became a classical ‘network’. ... Within this ethos there was an institutional goal to prevent trade disputes from spilling over or, indeed, spilling out into the wider circles of international relations: a trade dispute was an ‘internal’ affair which had, as far as possible, to be resolved (‘settled’) as quickly and smoothly as possible within the organization.¹⁵

Legally oriented dispute settlement in the WTO evolved through referral by the states parties (contracting parties) of the GATT of disputes under the treaty to ad hoc panels (initially called working groups), which were made up of various members of the insider network just described, who were diplomats, retired diplomats, trade officials or academics closely associated with the GATT ‘community’. The ad hoc panel was supported by the GATT bureaucracy, comprising a Secretariat that eventually came to include legal experts (though the panellists themselves were not predominantly trained in law). Above all, their expertise included insider knowledge about the negotiations that had created or modified the legal norms and deliberations inside the GATT institution about their meaning. The point was to produce a solution to a dispute based upon an interpretation of the GATT that was untainted by national interests of the parties, which both parties could somehow accept. To have a legally binding effect, the recommendations of the ad hoc panel had to be adopted by the consensus of the state parties. While a losing party rarely blocked adoption explicitly, panels nevertheless operated in the shadow of the consensus rule and, therefore, would be likely to craft a solution that would make it difficult for a losing party to block a decision, without appearing to be engaged in cheating or avoiding its obligations. The solution reflected not just the best legal analysis of the panellists but, ultimately, the collective wisdom of the institution or, more precisely, its guardians – the insider network.

It was not until the 1970s that the GATT bargain came under sustained stress. The collapse of the gold standard and, with it, the structure for managed macro-economic adjustment foreseen by the Bretton Woods system, combined with the recession of the 1970s and the mounting intellectual and practical (stagflation) challenges to the Keynesian consensus, led to increasing emphasis on micro-economic interventions of various sorts for adjustment purposes as well as to new kinds of trade restrictions – ‘voluntary’ export restraints negotiated under threat of unilateral action – of dubious legality under the GATT. For various reasons, the safety valves for adjustment written explicitly into the GATT did not prove to have the appropriate kind of flexibility to deal with the political economy of adjustment in the 1970s. As for the domestic micro-economic interventions, not only subsidies but also other forms of industrial policy, these challenged the stability of the non-discrimination norm as a means of distinguishing ‘normal’ legitimate domestic policies from ‘cheating’ in the trade liberalization bargain. Differences in approach to the mixed economy were to be tolerated under the embedded liberalism bargain, but under the economic pressures of the 1970s, it was easy to view activist industrial policies as a beggar-thy-neighbour approach to declining industries or declining demand (steel, for instance) – that is, as protectionist cheating on the basic bargain.

Domestic technical regulations gave rise to claims that even facially neutral regulatory requirements constituted disguised protectionism, with regulations creating obstacles to trade by forcing foreign producers to adapt to distinctive requirements of the importing country, which were not obviously justified by non-protectionist regulatory objectives. By the end of the 1970s, it thus became evident that post-war multilateral trade liberalization needed some fine-tuning so as to sustain the embedded liberalism bargain under changed economic and political circumstances. Then came the economic conservative revolution (exemplified by Margaret Thatcher and Ronald Reagan at the level of political leadership) and, with it, a radically different outlook on the problems that ailed the multilateral trading system and their solution. The problem was, at least for the USA, no longer framed in terms of the adequacy of the scope for adjustment under the existing rules of the game. In fact, the normative basis or interventionist adjustment policies were put into question by the moral laissez-faire outlook of the ascendant economic neo-right, aided and abetted by public choice accounts of interventionism as the payment of rents to concentrated, entrenched constituencies. It was natural, then, in defining the US interest in rewriting the rules of the game for multilateral trade to focus on interventionist or otherwise ‘inappropriate’ domestic policies in other countries as barriers to market access for the USA in areas in which it had a competitive disadvantage.

The multilateral rules of the game had enabled Germany and Japan, America’s wartime enemies, to compete successfully in the US market for industrial products. They had also enabled the newly industrializing developing countries to compete successfully in highly labour intensive industries such as textiles. On the other hand, many barriers worldwide hampered America in exploiting its apparent contemporary comparative advantage in knowledge-intensive industries and services. In some, intellectual property was largely unprotected; in most, competition in network services, such as in telecommunications and finance, was severely restricted or limited, while many others still imposed byzantine and archaic regulatory requirements on products, both imported and domestic. In many cases, a business presence in the other country was necessary for the full exploitation of comparative advantage, and here American firms faced severe foreign investment restrictions.

This new agenda, of course, was to become the core of the Uruguay Round agreements, which established the WTO. A common feature was restraint on domestic public policies that extended beyond the non-discrimination obligation of the GATT, pushing in the direction of what Dani Rodrik terms ‘deep integration’. What was required for greater market access was thought, in the predominant economic ideology represented by the Washington consensus, to be also good domestic economic governance: expansive intellectual property protection to spur innovation; de-mono polization and deregulation of network service industries such as telecommunications and finance and scaling down government health and safety and environmental regulation to what could be strictly justified under cost/benefit analysis and by ‘sound’ science. As Rodrik describes, this outlook, variously referred to as the Washington consensus or neo-liberalism, ‘combined excessive optimism about what markets could achieve on their own with a very bleak view of the capacity of governments to act in socially desirable ways. Governments … had to be cut down to size’.¹⁶

The developing countries did sign on formally to the new system. Why did they do so, if it was not unquestionably welfare enhancing?¹⁷ First, due to the debt crisis in the 1980s, many of these countries had been required to engage in unilateral trade and micro-economic policy reform as a condition for International Monetary Fund support for debt re-scheduling. Second, there was the notion that while developing countries might ‘lose’ from some of the agreements, they gained from others, such as commitments to agricultural and textiles trade liberalization. Linkage politics in the Uruguay Round may even have convinced their leaders that the overall package was in their interest, since there was little way to tell. However, perhaps most import antly, the alternative to neo-liberal rules in areas such as intellectual property, food safety regulations and services industries was a further increase in unilateralism in American trade policy – the use of aggressive unilateral remedies under sections 301 and super 301 in attacking what were perceived as being ‘unfair’ trade practices of other WTO Members, whether the lax enforcement of intellectual property rights or the purportedly unnecessary barriers to competition services industries or scientifically unjustified food safety regulations.¹⁸

In the Uruguay Round, the USA agreed to a constraint on unilateralism or self-help in return for new rules and ‘effective’ multilateral enforcement through WTO dispute settlement. Part of this constraint on unilateralism was intensely negotiated disciplines on specific unilateral trade remedies, anti-dumping duties, countervailing duties, and safeguards; because of dissensus as to whether the underlying practices being targeted by such unilateral actions are actually unfair, as well as different views about the extent to which certain interpretations domestic US interpretations of these remedies were opening the door to protectionist manipulation, the disciplines on trade remedies had a messy and incomplete character. The US protectionist lobbies feared having conceded too much; conversely, those concerned about US unilateralism could not be left somewhat dissatisfied with disciplines that often fell short of articulating clear substantive legal standards for ‘self-help’ against ‘unfair’ trade.

For developing countries, who had agreed to neo-liberal rules the substantive legitimacy of which they were questioning throughout, the effectiveness of the new system in holding back US unilateralism (and perhaps that of the European Union (EU) as well) was crucial in making the sacrifice even minimally bearable. In turn, for the USA, or at least the fair trade lobby on Capitol Hill and within the Beltway, multilateral enforcement of the rules had to be effective in order to justify their own sacrifice, precisely, of aggressive unilateralism, while at the same time, this enforcement could not result in overreaching so that the remaining rights to unilateral remedies, which were so zealously negotiated by the USA in, for example, the Anti-Dumping Agreement, were still fully protected.¹⁹ What would be needed to maintain this fine line was, to use a line from a Joni Mitchell song, a ‘strong cat without claws’, a system that could be forceful enough to induce ‘compliance’ through multilateralism and make the constraint of unilateralism meaningful and justified, while not being overly intimidating to domestic agencies handing out trade remedies.

Law-making in the WTO was to remain, as with GATT practice, consensus-based interstate bargaining, and, indeed as Richard Steinberg has pointed out, this became more entrenched and formalized.²⁰ No autonomous or independent law-making or regulating institution was created within the Organization – no equivalent to the UN Security Council or, perhaps more relevant, the European Commission. On the other hand, the Uruguay Round produced a dispute settlement of a judicial sort. The ‘legalizing’ or ‘judicializing’ features of the new system – compulsory jurisdiction and automatically binding dispute settlement reports (through the replacement of the positive consensus with the negative consensus rule), with the ultimate control of dispute settlement outcomes shifting from the membership (political/diplomatic control) to the new Appellate Body – have been repetitively invoked, as Weiler notes earlier, to indicate a rule-of-law revolution or even a constitutional one.

However, in understanding the evolution of the judicial politics of the WTO over the last two decades, it is just as important to recognize how much things did not change or, rather, the extent to which the dispute settlement system remained the same as the one that left the ultimate shape of disputes and their resolution to adjustment by domestic and international trade politics, while in the shadow of the law as it evolved through the jurisprudence of the Appellate Body. First of all, no private right of action was created, nor were WTO Members obliged in a general way, to give direct effect to WTO rules in domestic law and through domestic courts (with some very specific exceptions – the TRIPs Agreement, for example, requiring that certain domestic remedies for violations of TRIPs norms be made available). Thus, the Members retained ultimate control over the filing of the disputes, as well as the dropping of them, and their out-of-court settlement. Second, while much has been made of the possibility of enforcement through the authorized withdrawal of concessions (retaliation/countermeasures), the remedial features of the WTO system make it fall short of a true compliance/enforcement regime.

As has been often noted and, indeed, lamented by free trade hardliners, remedies are only prospective. If after exhausting the appellate process a Member finds itself faced with a definitive ruling of violation against it, then its sole obligation is to alter its measure to bring itself into compliance within a reasonable period of time. There are no damages, or reparations, for the harm caused by the offending measure up to the end of the reasonable period of time or whenever it is modified or withdrawn. In effect, there is a ‘free ride’ to violate WTO obligations for several years, given the length of time the dispute process takes from beginning to end. Second, there is no obligation on the adjudicator to order a losing Member to make particular changes to its laws – indeed, the default understanding is that it is normally up to the losing Member to decide the appropriate legislative or administrative means to address the violations found in the panel and or in the Appellate Body reports. Thus, a Member has the flexibility to attempt to address the concerns of the dispute, ruling in a manner that is least intrusive on its domestic sovereignty. And then there is a further proceeding, under Article 21.5 of the Dispute Settlement Understanding (DSU), if the winning Member deems that what has been done is inadequate to address fully the violations in the original dispute ruling(s).²¹ There can be several rounds of such Article 21.5 proceedings before the nature of the adequate compliance is properly defined, and, moreover, Article 21.5 proceedings, even though in theory they are strictly limited to the question of what was needed to cure the original violation, can lead to a reshaping, and thus a continuation, in a morphed fashion of the dispute well into the future, especially if the measures taken by the losing Member to comply open up a different set of issues about WTO law than those raised by the initial measure. One of the notoriously unclear features of the DSU is at what point the losing Member can simply declare a losing Member to be in non-compliance and ask for retaliation. The most plausible answer is that as long as the issue remains whether what has been done is adequate, Article 21.5 proceedings must run their course.

Finally, as for retaliation, it is limited to a withdrawal of concessions of equivalent commercial effect. This gives rise to the possibility of what Alan Sykes refers to as ‘efficient breach’.²² Depending on its domestic political economy, and the social and political sensitivity of the measures that it has been asked to change, the losing Member may well choose to accept the retaliation and maintain its measure. It is paying a ‘price’ but not a high enough price to create decisive incentives to bring itself in conformity with the law as interpreted by the adjudicator.

While the ‘rule-of-law’ features of compulsory jurisdiction, automatically binding dispute reports, judicial oversight of implementation, appellate review and sanctions for non-implementation unquestionably still represent important changes, these changes have also occurred without a fundamental alteration of the nature of the ad hoc panel process, which is the first instance of the WTO. While it is clear from the DSU that the intention was to draw the Appellate Body membership from distinguished respected jurists, the panel system was not professionalized.²³ Panellists are mostly low or mid-level trade officials or retired officials, many are not lawyers and few have trial advocacy experience. The WTO Secretariat remains crucial in orienting the panel reports and motivating them through extensive reasons and citations of authority. There is essentially no distance or independence of the panellists from the WTO insider community; legal advisers from the WTO Secretariat are usually present throughout the panel’s proceedings and deliberations. By establishing appellate review only for error of law and giving appellate review a very tight timeline (60 to 90 days), the Uruguay Round negotiators virtually guaranteed that a factual basis determined by essentially amateur adjudicators and technocrats within the WTO bureaucracy would be decisive in framing the disputes.

All of these features of continuity, and limits on the strict conception of the ‘rule of law’, have to be borne in mind as we consider how the Appellate Body chose to carve out its role, assert its authority and develop the pillars of its jurisprudence.

We now turn to the troubled political setting in which the Appellate Body established itself as a judicial tribunal. As already noted many developing nations had acquired buyer’s remorse almost by the time they had signed the Uruguay Round agreements. At the same time, as previously observed, neo-liberal globalization was already under sustained attack by activists in the USA and Western Europe by 1995. Those features of the Uruguay Round agreements that pointed beyond the traditional GATT non-discrimination norm towards forced harmonization or deregulation in the direction of the neo-liberal model of optimal economic policy for development and growth were understandably the focus of much of the attack, but also important was the apparent cessation of the use of trade sanctions against labour and environmental policies of other countries that threatened human rights and global environmental goals, which occurred through the unadopted Tuna–Dolphin panel reports in the early 1990s and were strongly supported by the insider trade policy elite and the GATT/WTO institution, which largely controlled the panel process, as explained above.²⁴ The neo-liberally oriented trade policy community tended to dismiss the criticisms of the outsider constituencies as ill-informed or as a simple protectionist backlash against progress towards the free trade ideal. Developing countries, on the other hand, were reminded that they had ‘voluntarily’ consented to the deal and that they had much to gain through a system where they could genuinely enforce the rules judicially against more powerful trading nations.

In fact, while the attacks on neo-liberalism multiplied and broadened, the developed country-led trade policy elite, emboldened by the Uruguay Round success but impatient to move further towards the neo-liberal utopia that was permitted in the Uruguay Round (where there was a failure to establish rules on investment beyond the GATT non-discrimination norm, where the TRIPs Agreement still had some exceptions and balancing provisions and where service liberalization commitments were regarded as disappointing), conceived almost immediately of an agenda for new negotiations to push forward where the neo-liberal agenda had been pushed back during the Uruguay Round. A few within the trade policy elite sounded a note of caution²⁵ – perspicacious in retrospect. Now that the Uruguay Round ‘grand bargain’ was done, it might be the appropriate time to consider some rebalancing or adjustments to make the bargain more secure and legitimate in the eyes of those who felt bullied, left behind or worse off, most notably a significant number of developing countries. The Uruguay Round required extensive implementation both by Members and through the creation of new committees and other structures within the WTO. Why the need to push forward immediately with a further ambitious neo-liberal negotiating agenda?

The trade policy insider community surrounding the WTO judged itself, and invited judgment by others, on its success or, as turned out to be the case, on its relative failure to succeed with this new agenda for another large bargain. In fact, through the period in question, the WTO was able to manage two enormously significant accessions – those of China and Russia – to work out a compromise on intellectual property rights and access to medicines, negotiate two agreements on the liberalization of trade in information technology (albeit on a plurilateral basis but involving the players that constitute the vast bulk of these markets) and, more recently, an accord on customs procedures – the Trade Facilitation Agreement (the one item of the Doha Round agenda to be realized).²⁶ In addition, through the device of a waiver, legal security was also provided for in the enforcement of the Kimberly Accord on conflict diamonds, the first time there has been a human rights-related understanding.²⁷ Finally, at the Nairobi Ministerial in late 2015, where the Doha Round was at last buried, an accord on the abolition of agricultural export subsidies was reached, a contentious subject of considerable importance to a number of developing, as well as developed, countries. Why was the WTO judged ineffective or moribund almost throughout this whole period and why would its redemption from this damning verdict have to depend on the eventual achievement of another ‘round’?

Different international organizations have different structures and different needs for renewal or revision of their legal frameworks. The constant negotiation of new packages of multilateral treaty norms is not necessarily the test for the health of an international organization. The mindset of the trade policy elite has been profoundly shaped, however, by the ‘bicycle theory’ of trade liberalization, a notion usually attributed to Fred Bergsten, of the neo-liberal-oriented, Washington, DC, think tank, the Institute for International Economics.²⁸ The theory is that unless one is constantly moving forward with deeper and wider liberalization, the multilateral trading system will collapse just like a bicycle that will fall over if you stop pedalling forward. This hypothesis has never been given any rigorous explanation or justification, either in economics (as Dani Rodrik has pointed out²⁹) or in international relations theory. Yet its influence on the trade policy elite has been enormous. Since tariffs were reduced successfully in repeated rounds of multilateral negotiations, as well as through customs unions and free trade areas and a significant amount of unilateral tariff reduction, the remaining tariffs of significance are largely in ‘sensitive’ sectors such as agriculture, where protection is deeply embedded in domestic political economies, such that making ambitious offers of concessions is extremely difficult politically.

These political economies do change over time, but the change is hard to impel through bargaining at the international level. Thus, the emphasis shifts from tariffs to ‘beyond-the-border’ liberalization, which means subjecting domestic policies to discipline or harmonization beyond the well-established GATT non-discrimination norm (a shift enforced by the neo-liberal ideological tendency and the US use of unilateralism against ‘unfair’ domestic policies, which threatens the GATT equilibrium or bargain, as discussed above). The bicycle theory in this world of limited further gains from negotiated tariff reductions would have to imply that the multilateral trading system can never reach a point of equilibrium where the maximum amount of welfare-enhancing liberalization has been achieved that is consistent with the basic norms of nation-state democracy and reasonable respect for regulatory diversity. Indeed, one may even argue that full implementation of the Uruguay Round would already result in going past this point and, thus, would risk a delegitimizing disequilibrium in favour of liberalization, so to speak.

This is what many constituencies were saying as attempts were made to launch a new round in Singapore, then Seattle, Doha and, finally, Cancun. At the Cancun ministerial meeting in 2003, the EU, in agreeing to take investment and competition (anti-trust) out of the negotiating agenda as a price for unblocking a new round, seemed to be acknowledging that there could be a saturation point beyond which it would make no sense to demand further liberalization/harmonization at the expense of domestic sovereignty/democracy/regulatory diversity. However, if you have not freed yourself from the bicycle theory, then you will simply lack any sense of what equilibrium looks like, let alone that it might well have been reached (at least for some considerable period of time).

As if the bicycle theory were not enough of a distorting lens through which to gauge the needs of the WTO system for revision or renewal, another element in the theology of the trade policy elite has been at work in contributing to the self-condemnation of the WTO ‘institution’ as a failure. This is the notion that to keep the bicycle rolling what is needed is primarily, if not exclusively, new ‘hard’ legal rules, which are themselves difficult to change, rules that ought to be applied to all WTO Members equally and enforced through dispute settlement. This idea is based on the trade policy elite’s superiority complex in relation to the United Nations or other fora that generate softer kinds of cooperation, norm generation and conflict resolution, which, among many in the trade policy elite, one often hears derided as ‘talk shops’.³⁰ Yet these ‘softer’ practices are not simply the result of weak thinking, lack of intestinal fortitude or dreamy utopianism. They reflect the deep reality of political disagreement among sovereign states and the delicacy of the process of narrowing genuine differences in values, perceptions and interests in a large community of member states and, on the basis, at least in principle, of sovereign equality.³¹ Again and again would the then director-general of the WTO hector Members for lack of ‘political will’ to come to agreement in post-Uruguay Round negotiations. Never for a minute was it conceded that after the Uruguay Round there were genuine and serious divergences of perspective and perceived interests among WTO Members about the future of the multilateral trading system, the need to rebalance the Uruguay Round result, the size and nature of gains from a successful new round and, indeed, the most important subject matter priorities for new agreements (which were not necessarily reflected in the formal agenda that was crystallized at the Doha ministerial meeting, a rump of the neo-liberal-oriented Singapore Declaration in 1996).

Instead of the pressure cooker of large-scale negotiations under deadlines (which were in reality always missed), it was arguably a time when further norm creation (if, indeed, it was needed – the point about equilibrium made above) would require long preparation through more open dialogue and deliberation, new economic research and careful thinking about the kinds of structures required to advance liberalization without inordinately burdening developing nations or inappropriately constraining domestic regulatory democracy. It is to the credit of the current director-general of the WTO, Roberto Azevedo, that he quickly dropped the pressure-cooker approach of the French dirigiste bureaucrat Pascal Lamy and, while continuing talks to the extent possible, did so in a new atmosphere of respect for genuine differences that does not malign disagreement as deal breaking or obstructive. This outlook did at least produce the trade facilitation agreement and, most recently, a healthy agreement to disagree – the frank acknowledgement that the differences between WTO Members are such that it is not fruitful to continue the Doha Round exercise. Indeed, the Trade Facilitation Agreement resembled in many ways some of the softer approaches previously held in contempt by the insider trade policy elite, with developing countries having differentiated obligations and the ability to control in significant ways the time frame for the implementation of obligations. Thus, there has been some progress towards a new mentality that is more willing to acknowledge and address openly and respectfully the significance of the diversity among the WTO’s Members, politically and economically.

However, overall, from the time that the Appellate Body was first faced with establishing itself as a legitimate, effective adjudicative body to the present, the WTO ‘institution’ has presented itself and understood itself as being in a state of arrested normative development, to the point that its future relevance and viability could seriously be questioned. I have tried to explain above that this was, in many respects, a self-constructed narrative. When the Appellate Body looked inside the WTO building, it saw an atmosphere of malaise, self-doubt and self-flagellation; when it looked outside the window it saw angry protesters attacking the WTO as an anti-democratic institution threatening social justice, environmental protection and human rights, a symbol of much that was problematic with neo-liberal ideology. We now turn to the legitimacy challenges for the Appellate Body in establishing itself and operating within this frame.

3 The Legitimacy Challenges for the New WTO Judicial System

What is usually referred to as the judicialization of the multilateral trading system – compulsory jurisdiction, automatically binding rulings, sanctions for non-compliance and an Appellate Body – was by its chief architects arguably understood as being part and parcel of the project of deepening and widening economic integration on a neo-liberal model. The features in question would facilitate ‘compliance’ or ‘enforcement’ and, thus, provide a bulwark against vested domestic interests’ pushback against the liberalization agenda. The Appellate Body would provide a safeguard against the occasional aberrent decisions made by the ad hoc panels – the kinds of decisions that under the GATT system were regarded by a consensus of the trade policy elite as being somehow ‘wrong’ and not in line with the way the insider community understood the intent of the law and the evolving practice of the community.³²

A more ambitious way that the new Appellate Body might have seen its role would have been as the ultimate guardian of the new WTO system and its neo-liberal values, adding the rule of law or even, in the more grandiose language of some scholars, ‘constitutionalizing’ the project of economic globalization, orienting its legal interpretations by the norms, practices and professional attitudes of the community that had managed the GATT and successfully concluded the negotiations that created the WTO towards the telos of ever deeper integration through further negotiations (the bicycle theory). The conscious rejection by the Appellate Body of this kind of role is reflected in a statement by one of its founding Members, James Bacchus, some years later that, even though he himself was personally committed to the agenda of deeper integration, it would have been a form of inappropriate judicial activism for the Appellate Body to see its own role as being in aid of that telos:

I am an outspoken advocate for negotiations that would broaden the scope of the WTO treaty to bring within it many more ‘twenty-first century’ global economic and commercial concerns. … To my mind, this must be resolved by negotiation and not by litigation. The responsibilities of the WTO should only be extended at the instigation of the WTO Members themselves. The responsibilities of the WTO should not be extended in the context of discrete disputes as a result of decisions by WTO jurists responding to innovate claims that are beyond the current scope of the WTO treaty.³³

The Appellate Body became activist in a very different sense, creating itself as an independent, semi-autonomous judicial branch of the WTO system, operating at a considerable remove from the political and diplomatic institutions of the WTO. Often the Appellate Body appeared to sympathize with the concerns of typically anti-globalization stakeholders or constituencies, rather than with the neo-liberal sensibility of the insider community (that is, when it did not present itself simply as a treaty interpreter confronting a legal text beyond the fray of the globalization wars). This was a self-conscious judicial policy. Again, founding Appellate Body Member James Bacchus says: ‘I think it is important for those out in the world to know that whatever the failings of the Appellate Body, those failings are not caused by the fact that Members of the Appellate Body are narrow-minded trade gurus who wear blinders and thus cannot see any values other than trade.’³⁴ Thus, the Appellate Body viewed itself as being, in some sense, accountable to those out in the world, stakeholders representing other values and interests than those given primacy by the trade policy insiders. This accountability to outsiders, in terms not only of the manner it balanced values but also in maximizing stakeholder participation to the greatest extent poss ible, given the constraints of the treaties (amicus briefs, open hearings where consented to by the parties), must be set in contrast or juxtaposition to the view within the Appellate Body that its authority as a judicial institution depends not merely upon its independence but also upon its distance from the WTO as an institution. This idea has been most clearly stated by former Appellate Body Member David Unterhalter: ‘[Members of the Appellate Body] do not answer to [the membership] … [since its decisions do not have to be adopted by positive consent of the membership] the Appellate Body is thus removed from political or diplomatic engagement with the membership-an essential pre-requisite for independence.’³⁵

One way of understanding the Appellate Body’s reorientation away from the WTO ‘institution’ is to emphasize the clash between the different sensibilities and professional goals of technocrats and diplomats, on the one hand, and jurists, on the other.³⁶ From this perspective, staffing the Appellate Body with high-level legal professionals almost guaranteed that the jurisprudence of the WTO would not simply be neo-liberal trade diplomacy by other means. Along these lines, George Abi-Saab, an Appellate Body Member who directly succeeded the founding generation, so to speak, has suggested that:

[while] the Appellate Body was initially conceived as an exceptional recourse to harness the odd rogue panel, … once established, institutions evolve according to their inner dynamics … an entity, however ambiguous or lacunary its institutional makeup, once it perceives itself as entrusted with the exercise of the judicial function, evolves according to a legal genetic code towards greater judicialization.³⁷

This logic goes a long way to understanding why the Appellate Body would have taken a judicial, rather than diplomatic, approach to the interpretation of WTO legal norms. However, it cannot fully explain why the Appellate Body distanced itself so dramatically from the GATT/WTO ‘institution’ and its acquis of informal norms, practices and interpretive canons. There is ample room within the interpretive rules of the Vienna Convention on the Law of Treaties (VCLT) (Articles 31–32) for giving a significant place to this acquis – by broadly interpreting ‘subsequent practice’ in Article 31, for example, as some panels did in order to borrow heavily from the GATT acquis.³⁸ As Weiler notes, the distancing of the Appellate Body from the GATT/WTO institution went to the point of ‘sometimes gratuitously scathing [criticism] of panel decisions’.³⁹ As one of the founding Members of the Appellate Body, Claus-Dieter Ehlermann, notes, the insider committee charged with the selection of Appellate Body Members sought deliberately to appoint to the Appellate Body distinguished generalist jurists, not eminent experts in GATT/WTO law.⁴⁰ (Only one insider, the late Julio Lacarte-Muro, was appointed to the founding membership of the Appellate Body, and he was a seasoned diplomat and trade negotiator, not a legal specialist of GATT/WTO law.)

Along the lines of Weiler’s and Abi-Saab’s suggestions, one view would be that by appointing jurists who were not WTO specialists one could only expect them to reconstruct trade jurisprudence using tools and sources from general international law and notions of what is required for a genuinely judicial organ to function properly in the service of the ‘rule of law’. However, another view is also possible. Trade law experts would be more likely to substitute their own view of the law for that of panels, which by reason of the guidance of the Secretariat and the practice of the appointment of panels from the trade policy insider community, represents, in a sense, the accumulated collective wisdom of the ‘institution’. Conversely, since GATT/WTO law is manifestly highly technical and complex, generalist jurists with no prior knowledge of the area would feel it appropriate to correct only manifest legal error, situations where a panel’s reasoning happened to be impossibly unclear or contradictory, while also controlling for non-objectivity situations where a panel might have been influenced by the political interests of a particular Member.

This view is consistent with the notion that the drafters really had the risk of a ‘rogue-panel’ in mind when they created the Appellate Body. And, indeed, often in domestic legal systems, judicial review by general appeals courts of administrative agencies and tribunals in technical areas of regulation such as anti-trust, telecommunications, or food safety has been conceived along the lines of considerable deference to expertise, including leeway for the expert body in legal interpretation. Two practitioner critics of the Appellate Body’s choice for a different approach suggest that:

Appellate Body members in particular are, on the other hand, primarily generalist international jurists who do not share the same common experiences or understandings as their GATT predecessors. … Contrary to the predictions of some long-time observers [footnote omitted] however, the Appellate Body in particular has placed surprisingly little emphasis on drafting history as a means of understanding the intent of WTO agreements.⁴¹

Overall, the mere choice to create in the Appellate Body a quasi-judicial institution staffed with generalist jurists – genuine legalists – cannot itself explain or justify the radical break by which the Appellate Body distanced itself from institutional tradition and practice within the GATT/WTO, which, indeed, as Robert Hudec, Weiler and others have rightly noted, had already moved in a legalistic direction during the last decades of the GATT era.

This declaration of independence, I believe, is supported by other, specific considerations of legitimacy. As described in the previous section, by the time the Appellate Body was delivering its first judgments, the neo-liberal project was already embattled and contested in many countries and the WTO, as represented by the aggressive Uruguay Round agenda, was at the centre of the controversy. How could the Appellate Body not become entangled in the globalization battles, as the ultimate ‘enforcer’ of the WTO neo-liberal agenda, as it was represented in the Uruguay Round agreements, if it were to be explicitly guided in its jurisprudence and institutional orientation by the ‘WTO system’ and the liberalization telos it was understood to represent, both by insiders and enemies?

The negative consensus rule introduced in the Uruguay Round to determine the adoption, or binding character, of rulings by the panels and the Appellate Body was oriented, of course, towards ‘enforcement’ or ‘compliance’. However, it had yet another significance, which is very important to understanding what I call the legitimacy challenge of the WTO judicial system. As former Appellate Body Member David Unterhalter emphasizes, there was no longer any political filter, in effect, for dispute settlement rulings in the Dispute Settlement Body.⁴² These would be adopted, except in the implausible situation where even the winning party voted against it. With the WTO divided against itself, and the political impasse surrounding a new round of negotiations, the ‘institution’ would not have a strong hand to play in disciplining or pressuring the Appellate Body, much less in attempting to control it. Changing the actual WTO rules to override the Appellate Body ruling would itself require positive consensus.

The WTO Agreement did contain a provision that allowed for authoritative interpretations of the WTO treaties to be adopted by a supermajority vote of the membership, but voting was never really practised in the GATT and there was a general aversion that continued into the WTO era to moving away from the consensus decision making.⁴³ In sum, by the combined effect of the negative consensus rule for dispute rulings and the positive consensus requirement (at a time of political dissensus, if not impasse, within the WTO), the insider community of trade diplomats and officials could not plausibly reverse an approach taken by the Appellate Body that was at odds with its (typically neo-liberal) view of the WTO system and its purposes. Explicit statements by various Members of the Appellate Body throughout its history indicate an awareness of the narrative of the WTO as an institution that is blocked or dysfunctional at the political and diplomatic level. In the words of founding Appellate Body Member Mitsuo Matsushita: ‘In national governments, there is a Supreme Court and also there is the legislature. If the Supreme Court decision is unacceptable for legal or political reasons, the legislature may be able to adopt a new law or to reverse it. In the WTO context, the legislature is the ministerial conference. Yet it is not operating so well.’⁴⁴ Another founding Member of the Appellate Body, Julio Lacarte-Muro, summed it up this way: ‘In ten years, the membership of the WTO has never even made a gesture toward interpreting a provision, let alone approved an interpretation, and let alone made an amendment to the WTO Agreements.’⁴⁵

If the Appellate Body were to be seen by neo-liberal constituencies as being too deferential to interventionist government policies, and not sufficiently aggressive as an enforcer of liberalization, it might disappoint certain claimants, particular the USA and the EU, whose corporate lobbies pushed for the Uruguay Round deal. But there were also important anti-globalization constituencies in those Members challenging neo-liberalism. In response to a perception of excessive deference, pro-liberalization constituencies might urge trade officials to take unilateral action in response to the ‘enforcement’ failure. However, such aggressive unilateralism was, as Joost Pauwelyn has pointed out, significantly disciplined in the new WTO system, and unilateralism would create its own legitimacy problems.⁴⁶ While an alternative response might be to shift dispute claims to other forums such as regional trade agreements or reformulate them as investment disputes (easily possible, as is evidenced by a number of instances where WTO disputes also resurfaced as investor–state disputes), such exit is not easy. Article XXIII of the DSU requires that determinations of violations of WTO agreements be made exclusively by the WTO dispute settlement organs. One thus loses any distinctive advantages from WTO law or jurisprudence by shifting to a regional forum. Most regional dispute settlement systems are, generally speaking, rather underdeveloped (MERCOSUR may be an exception), and, as will be discussed later in this article, the WTO Appellate Body, in fact, has adopted a policy that is not accommodating of regional dispute fora operating side by side the WTO dispute settlement system. In sum, for a dissatisfied claimant, exit from the WTO dispute system would be far from costless. On the other hand, a ruling seen by the losing Member and constituencies critical of neo-liberalism or economic globalization as inappropriately intrusive in domestic sovereignty, especially in sensitive areas such as the environment and human health, would likely drag the Appellate Body into the legitimacy woes of economic globalization. The Appellate Body would be embattled, and there would be a possibility that its effectiveness would become questioned due to ‘civil disobedience’ of the losing party, refusing to implement the ruling even on pain of retaliatory sanctions. As the Appellate Body soon found out with the EC–Bananas⁴⁷ and EC–Hormones⁴⁸ disputes, such civil disobedience was a real possibility and even considered by some WTO scholars such as Sykes to be legal – a form of ‘efficient breach’.⁴⁹ Here, the positive consensus rule for political adjustment of dispute rulings would function against the legitimacy of the Appellate Body. It could not easily transfer the responsibility for fixing the systemic risks of ‘civil disobedience’ to a functional political decision-making process.

In sum, non-compliance with its purported binding rulings poses a greater risk on balance to the Appellate Body’s legitimacy than perceptions that it is overly deferential, failing to find violations of WTO rules where it ought. There is a greater downside risk from intrusiveness than there is from deference, generally speaking. Insiders and academic commentators may criticize the latter; but the former is what is likely to lead to headlines and street demonstrations. It follows also that, even where finding a violation, the Appellate Body would be most likely to avoid serious legitimacy consequences, where the finding could be couched in terms that express appropriate deference or sensitivity to non-trade values and institutions as opposed to aggressive enforcement of neo-liberal globalization or ‘deep integration’. Narrow, as opposed to broad, grounds of violation might allow a Member to comply through relatively technical or fine-grained changes to its regulations, leaving in place the main thrust of the public policy, thus lessening the risk of civil disobedience.

At the same time, repeated or chronic failure of the WTO Appellate Body to sanction or discipline discriminatory protectionism, especially abuse of unilateral trade remedies by developed countries, could further exacerbate the legitimacy crisis of the WTO, making the Uruguay Round bargain even more questionable for developing countries and, in the longer term, especially under the pressure of global economic or financial crisis, creating a sense that the constraints are so loose that it is acceptable to revert on a large scale to the kind of protectionism even the original GATT bargain was intended to constrain. Thus, the Appellate Body would need to combine a certain kind of deference in cases involving the dividing line between acceptable domestic policy space and liberalization/integration, while visibly holding the line against the protectionist abuse of permissible policy space and of permitted unilateral trade remedies. This does not mean providing legal security that Members will be held to the strict discipline of the law in every case but, rather, it is a matter of overall confidence in the discipline’s effect over time. The many disputes of a well-functioning adjudicative mechanism will provide sufficient confidence to prevent widespread defection or exit from the system’s restraints and create, as Petros Mavroidis puts it, adopting an expression from Robert Keohane’s theory of international cooperation, ‘diffuse reciprocity’.⁵⁰

4 The Declaration of Independence

The Appellate Body’s initial set of moves to separate itself from, and establish its autonomy in relation to, the WTO as an institution or neo-liberal projects could be collectively described as its ‘declaration of independence’. More apt perhaps is the unforgettable image in the great painting by Jacques-Louis David of Napoleon taking the crown from the pope and crowning himself emperor. By acting like a court and not as part of the enforcement wing of the WTO institution, the Appellate Body created itself as a judicial branch in a distant, even potentially contentious or oppositional, relationship with the WTO institution. This came in a range of decisions over the roughly three-year period in which the anti-globalization movement was refocusing itself on trade leading up to the 1999 Seattle riots and increasingly affecting trade politics in a range of countries including the USA, where Bill Clinton lost fast-track authority in 1997, for example. The following actions were among the ways in which the Appellate Body declared its independence and distance from the WTO ‘institution’:

• employing normative benchmarks and legal standards and sources from outside the domain of GATT/WTO law, unrelated to and sometimes in tension with GATT ‘collective wisdom’, these outside norms including general international law and international environmental law;

• replacing the teleological and functional interpretation characteristic of GATT panels in the service of trade-liberalizing goals with textualism and formalism that abstract from the context of the WTO as an institution and the liberalizing goals of the multilateral trading system and, instead, emphasizing a formal semantic exercise guided by the VCLT,⁵¹ an instrument obviously neutral in terms of the specific values of free trade;

• developing a doctrine of implicit judicial powers, including to fill gaps (the decisions to allow amicus curiae briefs);

• shouting with a megaphone that the Appellate Body will afford no particular deference or even respectful consideration to decisions of the panel under appeal, to the point of what Weiler justly calls ‘gratuitously scathing’ criticism of panel rulings, as noted above;

• rejecting a notion of institutional balance that would require some deference to political/diplomatic rule-making processes of the WTO, even where they are given an explicit role in policing certain institutional norms;

• emphasizing the precedential weight of the Appellate Body’s own decisions relative even to past adopted decisions of GATT panels;

• giving itself a sort of remand authority (completing the analysis) that allows the Appellate Body to illustrate how its correction of the panel’s legal interpretation is to be applied to the facts of the dispute;

• allowing argumentation of cases by private legal counsel unaffiliated with trade officialdom and

• emphasizing consensus rulings by each division of the Appellate Body (avoidance of dissents) and collegiality (all cases discussed among all seven of the members, even if the actual disputes are heard by divisions of three). Thus, the Appellate Body would appear to speak with a single voice in contrast to the political and diplomatic divisiveness within the ‘institution’.

Some of the main ‘articles’ of the Appellate Body’s declaration of independence were articulated in two very early rulings – Japan–Alcohol⁵² and EC–LAN Equipment.⁵³ Neither dispute necessarily raised any important systemic issues. The former case concerned non-discrimination (national treatment) in taxation. Japan had taxed much more heavily classes of alcoholic beverages that were mostly imported and much less heavily other classes mostly produced in Japan. This was the kind of dispute about protective discrimination that had been litigated not infrequently in the GATT era. Not dissimilarly, EC–LAN Equipment was a classic dispute about the meaning of tariff concessions and the interpretation of classifications – namely were certain products to be regarded as computing equipment or telecommunications equipment for the purposes of calculating duties under the EU’s schedule of tariffs? On these kinds of questions, the GATT acquis offered no lack of guidance, with its panel rulings, working parties, practices, normative axioms and other material generated inside the ‘institution’. The panels did not hesitate to draw on this acquis.

Now consider what the Appellate Body did when its jurisdiction was invoked to review these panel reports. First of all, in Japan–Alcohol, while paying lip service to ‘continuity’ between the GATT and the new WTO system, the Appellate Body rejected the panel’s notion that prior GATT reports, even if adopted by the WTO membership, constituted either ‘decisions’ or ‘subsequent practice’ that would somehow be binding or authoritative for the Appellate Body. Adopted GATT panel reports were merely one normative source that should be ‘taken into account when they are relevant’.⁵⁴ The systemic implications of this move by the Appellate Body were blunted or obscured by the fact that the Appellate Body, in its textual interpretation of non-discrimination with respect to taxation, extensively (albeit selectively) cited previous GATT jurisprudence, even though it emphasized that its approach was driven fundamentally by the analysis of the words in the treaty text.

Second, in EC–LAN Equipment, the Appellate Body indicated that the use of the negotiating history as a basis for interpreting commitments under the WTO should be strictly disciplined by Article 32 of the VCLT – that is, that such material should be used solely as a supplementary source of interpretation. In the GATT era, the negotiating history had often been regarded as controlling by the dispute panels, and continuation of this approach was regarded as desirable and likely by such GATT era luminaries as the late John Jackson.

In conspicuously adopting the VCLT as its guiding hermeneutic and strictly and mechanically applying its provisions to each interpretative issue, but in a manner that minimized resort to past institutional practice and negotiating history as sources of interpretation, the Appellate Body risked being taken as the rather unsophisticated judicial body, and numerous critics relished the chance to excoriate the crude literalism of recourse to dictionary definitions and the plodding manner in which the canons in the VCLT were deployed by the Appellate Body in solving interpretative questions. But by acting in this manner, the Appellate Body may in fact have been making a shrewd estimate of the requirements of legitimacy at a time in which the WTO as an institution, and the neo-liberal ethos of the Uruguay Round, were very much in question. The Appellate Body’s mission was defined not by the trade liberalization or ‘deep integration’ telos of the WTO but, rather, by the imperatives governing treaty interpretation in general international law. Moreover, as former Appellate Body Member Abi-Saab would note, ‘despite the oft claimed specificity of international economic law, a thorough look at the jurisprudence (case-law) of the Appellate Body does not reveal any mention of, or reference, to one or more rules of interpretation to this particular field that would come to complement or substitute for’ the general rules of the VCLT.⁵⁵

Third, in EC–LAN Equipment⁵⁶ and also in the India–Patents⁵⁷ case (the first intellectual property dispute in the WTO), the Appellate Body rejected the pro-liberalizing doctrine that WTO commitments should be read in light of the legitimate or reasonable expectations of those seeking the benefit of liberalizing disciplines (exporters or private rights holders in the case of the TRIPS Agreement).⁵⁸ Legitimate expectations were presented by the panels as an interpretative principle that emerged from the GATT acquis. The Appellate Body held that where the claim is one that a treaty provision is violated the expectations are defined by the treaty provision, read semantically, rather than vice versa. An important step was also taken towards the rejection of a liberalizing or neo-liberal telos as a basis for interpretation in the case of EC–Hormones, where the Appellate Body adopted the principle of in dubio mitius, that in case of ambiguity or doubt one should adopt the interpretation least constraining of sovereignty.⁵⁹

In the India–Quantitative Restrictions case, the Appellate Body was faced with arguments that it should display deference to political decision making in the WTO concerning whether delay in a WTO Member removing trade restrictions to protect the balance of payments was justified by considerations of macro-economic policy and development policy as set out in the relevant provisions of the GATT.⁶⁰ In the past, determinations of this kind had been made by a special committee of WTO delegates, the balance of the payments committee. In the committee, India had obtained the agreement of many of its major trading partners on a particular timetable for removing its balance of payments-based restrictions, but the USA was dissatisfied with what it saw as unnecessary delay. The USA’s response was to block consensus in the balance of payments committee and take India to dispute settlement. The situation displayed the intrinsic difficulty with the use of political or diplomatic processes in the WTO to make what might be called mixed judgments of law and policy concerning the way in which Members use flexibilities under the WTO agreements. A single Member can permanently block any decision, where its perceived interests are opposed to the general sense of the membership.

Under the old GATT system, considerable restraint in the use of dispute settlement to deal with these kinds of mixed questions and the disinclination of GATT panels to address them arguably created, or at least complemented, an atmosphere of compromise. In the Uruguay Round, a new WTO instrument was negotiated that tightened up the control on balance-of-payments-based trade restrictions,⁶¹ reflecting a compromise between neo-liberal views and the particular concerns of developing countries to maintain flexibilities. In this instrument, the availability of dispute challenges in the case of ‘application’ of balance of payments-based trade restrictions was explicitly affirmed, while the committee’s process for policing the phasing out of these restrictions was also maintained. With its complaint against India, the USA was bringing to the fore the issue of the equilibrium between these aspects of the Balance of Payments Understanding.⁶²

The Appellate Body held that the expression ‘application’ did not in any way constrain the competence of the dispute settlement organs to review the underlying policy justifications for continuing to maintain balance of payments-based trade restrictions. India’s argument that ‘application’ refers only to issues that arise with the detailed implementation of trade restrictions, not their general policy grounds, was summarily rejected. So was India’s argument that the Appellate Body should limit its competence in such a way as to preserve a meaningful institutional balance between the dispute settlement organs and the balance-of-payments committee. The Appellate Body categorically rejected the notion that institutional balance is a ‘principle of WTO law’. While the Appellate Body suggested that the ‘deliberations and decisions’ of the balance of payments committee should be taken into account, it did not even indicate that were there a consensus in the committee the Appellate Body would be bound to defer, as opposed to coming to its own conclusions about the policy justifications for trade restrictions. Further, responding to India’s argument in the alternative that, even if the dispute settlement organs had competence, institutional balance dictated ‘judicial restraint’, the Appellate Body suggested that, where the dispute settlement organs had competence, ‘judicial restraint’ would be inconsistent with the obligation to exercise this competence when requested to do so by a claimant.

Not long after the India–Quantitative Restrictions ruling, the Appellate Body, in the Turkey–Textiles case,⁶³ had the opportunity to reconsider its rejection of the principle of ‘institutional balance’. Under the GATT, the assessment of regional trade agreements, including customs unions such as the EU, and their consistency with the law and policy of the multilateral trading system, was a function arguably confided to a committee of the delegates, known in the WTO era as the Committee on Regional Trade Agreements (CRTA). As Mavroidis suggests, the original intent of the GATT drafters was that ‘the nature of the multilateral review would come close to that of a merger authority; no [customs union or free trade agreement (FTA)] would be consummated absent multilateral clearance’.⁶⁴ The formal legal conditions for establishing and maintaining a customs union or free trade area are contained in the GATT⁶⁵ and include the liberalization of substantially all trade between the parties to the preferential arrangement and avoidance of greater trade restrictiveness against WTO Members who are non-parties. Ultimately, it is consistency with Article XXIV that allows WTO Members to deviate from the most-favoured nation obligation that is a cornerstone of the multilateral trading system and to treat other parties to the preferential arrangements better than non-party WTO Members.⁶⁶

As preferential trading arrangements proliferated and became a major concern for trade policy scholars such as Jagdish Bhagwati, who were dedicated to the value of non-discriminatory multilateral free trade, it became particularly clear that what Mavroidis calls ‘multilateral clearance’ was largely a failure.⁶⁷ There were few cases where a FTA or customs union was carefully examined ex ante by the CRTA, much less where the committee came to a clear conclusion about the consistency of a particular agreement with the law and policy of the multilateral trading order. In Turkey–Textiles, the issue was whether an otherwise impermissible particular WTO restriction that Turkey had imposed on India could be justified as necessary in order to harmonize Turkey’s external customs regulations in order to fulfil its obligations under a customs union with the EU.⁶⁸ It was not especially controversial as a matter of judicial competence that the dispute settlement organs could review the particular restriction imposed by India and its nexus to the customs union with the EU. However, in Turkey–Textiles, the Appellate Body reinforced its approach in India–Balance of Payments, overturning the view of institutional balance of the panel of first instance as well as the prior GATT practice and finding that judicial competence extended to reviewing the overall or per se consistency of the customs union with the conditions in GATT Article XXIV and not merely the appropriateness of the specific measure.⁶⁹ The Appellate Body held: ‘[W]e would expect a panel, when examining such a meas ure, to require a party to establish [inter alia, whether the customs union fully meets the relevant requirements of Article XXIV]’. The Appellate Body explicitly linked its finding here on the broad scope of judicial review to the rejection of the institutional balance principle in India–Balance of Payments case.⁷⁰ There can be little doubt that in Turkey–Textiles, the Appellate Body was making a pointed statement about the wide extent of its authority over the diplomatic and political organs of the WTO for, as the Appellate Body noted, it did not need to reach the issue of whether it had jurisdiction to review the WTO consistency of a customs union as a whole in order to resolve the appeal.

The Appellate Body rulings in India–Balance of Payments and Turkey–Textiles were subject to forceful criticism by WTO insiders, most explicitly by Frieder Roessler, a former director of the GATT legal secretariat, who represented, it should be noted, India in the former case.⁷¹ Yet there was not a major revolt against the assertion of expansive judicial authority by the Appellate Body. While these decisions reduced the control of the insider trade policy community, they did so with the effect of strengthening pro-free trade disciplines and were thus consistent with the underlying substantive values of the community, even if they were at odds with the taste for diplomacy over hard legal outcomes. The extent to which the Appellate Body was establishing itself as an independent judicial authority, operating at a distance from past GATT practice and from trade policy and trade negotiations, was hardly noticed, save by a rather small group of experts. At the same time, it is hard to imagine that such judicial self-assertion would have been so easily tolerated had the political and diplomatic processes in the WTO been functioning in a robust and effective manner. As is illustrated by the remarks by Appellate Body Members Matushita and Lacarte-Muro cited above, the judges fully understood the vulnerabilities of these processes. They embraced the predominant narrative of impasse and ineffectiveness, and they could frame their activism as a response to a reality they neither created nor could solve. What some criticized as activism was an unpleasant burden imposed by ‘the gap in effectiveness between the WTO’s political bodies and its dispute settlement system’.⁷²

5 The Shrimp–Turtle Rulings: The Watershed

No jurisprudence is more significant than the Shrimp–Turtle dispute for marking the evolution of the Appellate Body as a judicial system independent of, and operating at a distance from, the WTO as an institution and from the ideological and policy orientations that tend to drive it.⁷³ The entire trade/environment debate, with its central importance of turning the attention of the anti-globalization movement to international trade, originated with a GATT case in the early 1990s involving two unadopted GATT panels – the Tuna–Dolphin rulings⁷⁴ – which held that trade restrictions in response to other countries’ environmental policies or practices were per se inconsistent with the GATT. These rulings were without a textual basis in GATT law but based, instead, on some intuitive notion that allowing trade measures to address global environmental externalities was somehow countenancing a slippery slope towards unconstrained green protectionism. In Tuna–Dolphin, the USA had banned tuna products that were caught with methods that led to high levels of dolphin mortality on a non-discriminatory basis – that is, the ban included tuna of US origin. The infamous product/process distinction was invented whereby a country could not defend treating products differently under the GATT based upon their production methods, even where different methods led to different environmental or other harms.

The measures in Shrimp–Turtle were closely analogous – the USA had banned shrimp that was fished with methods that led to high number of deaths of endangered species of sea turtles. Using a different doctrinal conceit than the Tuna–Dolphin panels, the WTO panel in Shrimp–Turtle nevertheless affirmed the overall GATT-era approach that had led to the clash between environmentalists and the multilateral trading system. The panel found that there was a complete incompatibility between non-discriminatory multilateral trade and measures that conditioned imports on the environmental policies or practices of other countries. The Appellate Body had to choose between affirming the orthodox free trader view that prevailed within the trade policy community or acknowledging in some way and attempting to mitigate the harm to the external legitimacy of the system from simply excluding even non-discriminatory trade policies to deal with global environmental problems.

Had the Appellate Body upheld the panel’s approach (or defended the result or a similar result on another doctrinal grounds), it would have clearly been perceived as siding with the ‘institution’ on an issue that sharply divided insiders from import ant outsider constituencies. To use the language some scholars have deployed, the Appellate Body would have chosen ‘internal legitimacy’ over ‘external legitimacy’. The Appellate Body took the opposite course. In Shrimp–Turtle, the Appellate Body revealed an important implication of its carefully crafted independence of, and distance from, the ‘institution’. The Appellate Body was quite capable of giving purchase to constituencies characteristically critical of, if not hostile to, the WTO as a neo-liberal or free trade-driven institution. Yet this revelation was not so easy to see unless one followed carefully the GATT/WTO legal system, because while opening the door in principle to process or production methods (PPM)-type measures, the Appellate Body had found that, under the chapeau of Article XX⁷⁵ – the preambular paragraph – there were elements of discrimination in the manner in which US officials had implemented the legislative ban on turtle-unfriendly shrimp. Thus, the initial reaction of environmental groups was negative, while consummate insiders such as the late John Jackson were muted in criticism of the departure from the PPM theology, unsure just how far the Appellate Body had really intended to go, given that, in the end, it had found several violations in the US scheme.

However, there were other elements in the Shrimp–Turtle ruling that signified a nod to constituencies and values traditionally understood as being external to the WTO and its purposes. One of these was the way that the Appellate Body approached the interpretation of the language ‘conservation of exhaustible natural resources’ in the Article XX(g) exception on which the USA had sought to rely. The claimants, pointing to the negotiating history of the GATT, suggested that ‘exhaustible natural resources’ referred only to non-living resources (petroleum, minerals and so on) and, thus, that the protection of sea turtles was beyond the scope of Article XX(g). Yet case law under the GATT had already established that living species could be deemed ‘exhaustible’ under Article XX(g).

Instead of simply relying on the precedent of an adopted GATT panel report (that the Appellate Body mentioned casually in passing), the judges constructed a complex hermeneutic as if they considered the matter to be one of first impression in the WTO system. Having eschewed teleology in earlier cases when invited to interpret WTO disciplines in light of the purpose of the progressive liberalization of trade, they now endorsed teleology in marking the limits to free trade as articulated in Article XX of the GATT. ‘Exhaustible natural resources’ needed to be read in light of sustainable development, a goal stated in the preamble to the framework agreement establishing the WTO. From this proposition followed the need to bring in the law and policy of biodiversity as it had evolved in recent decades and the broader canon of evolutionary interpretation of WTO norms – interpretation that would necessarily vary depending on how legal regimes outside the WTO themselves changed over time.

The GATT drafting history and the collective memory or wisdom of the ‘institution’ about what the drafters of the GATT meant, and even GATT case law that supported the Appellate Body’s own position, were given short shrift, while external benchmarks from recent and dynamic fields of non-trade international law and policy were elevated to crucial hermeneutic tools. When responding to criticism by the insider trade policy community of its methodology in Shrimp–Turtle, Members of the Appellate Body might have suggested that they were only bringing in non-WTO international law where it was indispensable for solving interpretative controversy about the meaning of the WTO agreements. If so, then doing so in Shrimp–Turtle was only indispensable because of the Appellate Body’s own decision to give so little weight to the GATT acquis on this issue that it needed to open it up as being, to repeat, essentially a question of first impression. A more plausible, but not inconsistent, reading is that the Appellate Body purposely treated the issue of exhaustible natural resources in this way in order to bolster its external legitimacy at a time in which economic globalization was under persistent attack by outsider constituencies, including and especially environmental ones.

This reading is reinforced by the jurisprudential move that led to the most explicit and vehement reaction by the ‘institution’ to the Appellate Body’s new and independent judicial order. The Appellate Body, holding that the panels and, indeed, the Appellate Body itself, had the authority to accept amicus briefs from non-governmental actors, including non-governmental organizations (NGOs) from non-trade constituencies such as environmentalism. The Appellate Body gave a textual justification in the case of the authority of the panels, based upon the right of the panel to seek information in the DSU. However, the Appellate Body also accepted an amicus brief that had been submitted to the Appellate Body itself, without any explanation of its authority to do so. The Appellate Body might have thought that it was exercising an inherent judicial power, and this would be consistent with many of the other moves discussed earlier to create itself as an independent judicial branch of the WTO, such as completing the analysis (acting as a remand authority when one was not provided for in the DSU). Holding that official account might be taken, even in principle or symbolically, of the views of non-state actors on WTO disputes was the culmination of the Appellate Body’s declaration of independence. The Appellate Body, as an autonomous judicial body operating at a distance from the ‘institution’, could enter into a dialogue with outsider constituencies – one unfiltered and unmediated by the political and diplomatic organs of the WTO.

The attacks on the amicus decision multiplied, spreading from trade experts and academic commentators to Member delegates, producing a key test of whether the Appellate Body could withstand sustained political pressure from the ‘institution’. The Appellate Body’s initial response was to provide a grounding for its initial acceptance of an amicus brief in a subsequent case, US–Carbon Steel, which was decided shortly after Shrimp–Turtle:

In considering this matter, we first note that nothing in the DSU or the Working Procedures specifically provides that the Appellate Body may accept and consider submissions or briefs from sources other than the participants and third participants in the appeal. On the other hand, neither the DSU nor the Working Procedures explicitly prohibit acceptance or consideration of such briefs. However, Article 17.9 of the DSU provides [that working procedures are to be drawn up by the Appellate Body]. This provision makes clear that the Appellate Body has broad authority to adopt procedural rules which do not conflict with any rules and procedures in the DSU or the covered agreements [footnote omitted].’⁷⁶

The fuller response came in the context of a different dispute, EC–Asbestos,⁷⁷ where Canada was challenging France’s ban on asbestos, an important public health measure against a toxic substance that had claimed tens of thousands of victims in a number of countries. While the WTO panel had upheld France’s measure under Article XX(b) of the GATT – the human life and health exception – it had also made a ruling that incensed some public health constituencies by finding that, for the purposes of analysing discrimination, asbestos and substitute products were to be regarded as like and in a competitive relationship, even though the substitute products had no record of being lethal to humans. Despite the acceptance of an Article XX defence by the panel, the panel’s likeness analysis signalled a certain obtuseness to the values of human life and health in the assumption that at least at the preliminary stage of the analysis one could be indifferent to the fact that one of the products was killing large numbers of people and the other was not.

The Appellate Body thus understandably anticipated the submission of amicus briefs by outsider constituencies in the EC–Asbestos appeal. It decided to enter into what might be called an attempted dialogue with its critics in the institution. While not backing off on its authority to accept amicus briefs, the Appellate Body attempted to address certain criticisms based on considerations of due process and fairness to the parties by promulgating a detailed procedure to be followed to obtain leave to submit an amicus, which would include time and length limits for submissions, and disclosure requirements to address the concern that amicus briefs might be surreptitiously directed or funded by interests connected to the WTO Members parties to the dispute.

While the procedure appeared to make the consideration of amicus briefs a more orderly, open and objective process, it led only to an increased vehemence in the attacks on the Appellate Body by Member delegates. How could the procedure have made things worse since it simply channelled, and in certain ways provided a restraint on, an inherent authority that the Appellate Body had already insisted it possessed? Establishing a procedure ex ante had the appearance of rule making. This was, ultimately, what was intolerable. The Appellate Body’s assertion of competence in India–Balance of Payments and Turkey–Textiles, where rule making through diplomatic and political organs of the WTO was blocked or ineffective, had already given rise to anxieties that the Appellate Body was encroaching on the domain of political and diplomatic decision making in the WTO.

However, these anxieties became significantly more intense and explicit when the Appellate Body began to make rules to govern its relationship to actors outside the ‘institution’. The attacks on the Appellate Body in the Dispute Settlement Body, the deliberative forum of delegates on dispute system issues, became intense, and the chair even communicated to the Appellate Body the dissatisfaction expressed by many delegates. Steve Charnovitz, who provides an excellent account of the whole episode, notes that only the USA defended forthrightly the Appellate Body’s exercise of jurisdiction to set out the procedure for submission of amicus briefs. Developed and developing country delegations alike claimed that the Appellate Body had egregiously exceeded its powers.⁷⁸ Yet by the time of the General Council meeting at which the issue of the amicus procedure was decided and these views had been expressed, the Appellate Body had rejected all of the applications for leave that had been submitted under the procedure. Having taken such an extraordinary set of steps to declare its independence and distance from the WTO ‘institution’ or the membership, what could explain the Appellate Body taking a move that would inevitably be interpreted as being influenced by pressure from the delegates? Indeed, many commentators interpreted this move as an indication that the Appellate Body would reverse its underlying position about its power to accept amicus briefs and not just remove the procedure or never use it again. In fact, the Appellate Body soon indicated in further disputes that the admissibility of amicus briefs remained good law.⁷⁹

However, something happened in the interim. Shortly after the EC–Asbestos amicus controversy, the director of the Appellate Body Secretariat, Debra Steger, a member of the insider trade policy elite and one of Canada’s Uruguay Round negotiators, left her position (it is a secret de polichinelle in WTO circles that this was not an amicable parting of ways). One may speculate whether Steger, who is described by founding Appellate Body Member Ehlermann as particularly strong-willed, had any role, and with what motive, in advising the Appellate Body on the misstep of setting out a procedure and then appearing to cave to pressure by rejecting all of the submissions made under it.

Then, the Appellate Body was given the opportunity to clarify, if not revise, its ruling on the substance of the Shrimp–Turtle dispute. As noted, even though the Appellate Body had found that the overall approach of the US shrimp ban was acceptable under the WTO, some aspects nevertheless remained violations or contrary to the conditions of the chapeau, or preambular paragraph, of Article XX of the GATT, which dealt with the application of measures that a WTO Member is seeking justification for under Article XX. These aspects included inflexibility in the way that the statute was applied to different countries where different conditions prevailed, the failure to negotiate with some countries while negotiating with others a turtle protection agreement as an alternative to an embargo and various shortfalls of due process in the way that decisions were made about import certification. The USA sought to change these aspects of the application of the US law in order to comply with the ruling. Malaysia, one of the original complainants, brought a compliance action under Article 21.5 of the DSU, where it sought to reintroduce arguments about the per se unacceptability of trade measures to target other countries’ environmental policy.

Clearly, there was a belief by some that the Appellate Body, in the face of widespread criticism within the ‘institution’, would back off on this second round and find a way to close the door once again to environmentally based trade restrictions. Instead, the Appellate Body pronounced itself fully satisfied that the USA had addressed its concerns under the chapeau and expressed surprise that Malaysia would, in effect, challenge the authority of the Appellate Body’s original ruling with arguments apparently inconsistent with it. Indeed, the Appellate Body took the occasion to pronounce explicitly on the precedential value of the Appellate Body rulings and the expectation that future panels will follow them.

After the Appellate Body held its ground in the second Shrimp–Turtle ruling, there was no further concerted effort to apply political or diplomatic pressure on the amicus or trade and environment issues. Several years later, the Appellate Body found a basis for opening up its hearings to the public through closed circuit television, by consent of the participants/parties in the dispute (this had already happened at the panel level). For the Appellate Body, this decision was arguably an even more activist move than allowing amicus briefs, as the DSU stated that Appellate Body proceedings are to be confidential. The Appellate Body got around the confidentiality language in the DSU through a notion that WTO proceedings are ‘relational’ and, thus, that parties can agree among themselves to waive aspects of the DSU.⁸⁰ A number of the Members who had objected to acceptance of amicus curiae briefs also vehemently opposed opening the hearings to the public, but this time the criticism was barely noticed.

Part of the reason must surely have been that any Member that was a party to a particular dispute could simply object to open hearings and this would be enough to ensure confidentiality. However, something else was happening around the time of the second Shrimp–Turtle ruling. After failing to agree on the launch of a new round of negotiations at Seattle and then at Cancun, a declaration was achieved at the WTO ministerial in Doha Qatar on the outlines for a new round. One of the elements of this declaration was the stipulation of negotiations on the relationship of multilateral environmental agreements to existing WTO rules. Another was an agreement to negotiate reform of the dispute settlement system. The second Shrimp–Turtle ruling had been circulated just a week before the Doha Declaration. The apparent will at Doha to break through the impasse led to at least a brief hope by some that the criticized aspects of Appellate Body judicial activism might be reversed through treaty amendment as part of the new round. The Appellate Body was given a breathing space from immediate political pressure, though in fact its second ruling in Shrimp–Turtle was more emphatic in the break with the GATT’s past than the first had been, in that an environmentally based trade embargo was found to be without fault whatsoever under the WTO legal system.

However, it soon became apparent that the Doha Declaration had papered over fundamental divisions among the membership about the future orientation of the WTO, including the very meaning of calling the Doha Round a ‘development’ round. The major developed country players sought a further thrust towards ‘deep integration’, only slightly chastened by the broader legitimacy crisis of neo-liberal globalization. In return, developing countries might receive some additional concessions in sensitive areas such as agriculture and the textile and clothing trade. Many developing countries, by contrast, as well as non-governmental constituencies, sought a rebalancing of the result of the Uruguay Round towards greater policy space and more meaningful special and differential treatment of developing countries.⁸¹ Only on issues pertaining to the TRIPs Agreement and access to HIV/AIDS medicines was there some explicit recognition of the need to re-balance the Uruguay Round result away from neo-liberalism.

As the Doha Round negotiations floundered, new issues were emerging such as the relationship of the WTO to climate change mitigation measures with trade dimensions, carbon border adjustment and so on. The WTO director-general insisted inflexibly that new subjects could not be added to the Doha agenda until a successful agreement on the existing items was achieved. This inflexibility and the dirigiste style of Lamy led to considerable frustration and acrimony in Geneva and a crisis of confidence in the WTO as an institution. The Members now had enough grievances with the WTO Secretariat, led by Lamy, and so many renewed or intensified differences among themselves that any sense of a common will to stand up to judicial activism largely atrophied. In this setting, the Appellate Body was able to come to maturity as a judicial body, through twists and turns that moved to solidify and evolve a set of judicial policies that, overall, with some exceptions, have served reasonably well in meeting the legitimacy challenges articulated above.

Once the Appellate Body recovered from its brief misstep in the EC–Asbestos amicus controversy, it became obvious that statements opposing Appellate Body rulings as excess of powers or illegitimate, no matter how numerous or vehement, would not be successful in swaying the Appellate Body to shift course. As Members of the Appellate Body themselves publicly indicated not infrequently, the proper way for the membership to respond to a ruling that was politically or legally unacceptable to it was to ‘legislate’, yet they knew full well what this meant given the practice of positive consensus – the extreme unlikelihood that any decision could be legislatively overruled. The critics of the Appellate Body then began to focus their criticism on institutional or structural ‘imbalance’ between the political or legislative branch and the judicial branch of the WTO. However, this was hardly an argument for the Appellate Body to become a weak court; rather, it was an argument for the ‘legislative’ branch to be strengthened. In sum, by 2003 or so, some level of optimism about Doha and some level of realism about the futility of attacking the Appellate Body itself in the ‘shadow’ of the concensus requirement to change an Appellate Body ruling, allowed the Appellate Body to operate in an atmosphere of relative confidence of its independent authority as a judicial body.

It is at this point that the Appellate Body began to drop the artifice of a mechan ical application of the VCLT, whose first and primary recourse was to the dictionary. Some Appellate Body Members would even publicly admit, more or less, that it was an artifice. According to Abi-Saab, for example, ‘[i]n practice [despite the appearance of strict constructivism] much of the reasoning in interpretation is informed by the object and purpose, either consciously or subconsciously, … even though they may not figure explicitly as such in the analysis’.⁸² Now the dictionary would become at most a beginning step and the VCLT rules were to be applied in a holistic fashion.⁸³ It has become clear, if not always entirely explicit, that the Appellate Body has been applying, if not a judicial philosophy based on an understanding of the appropriate balance or equilibrium of rights and obligations within the WTO in light of the legitimacy challenges of adjudication, at least a set of broad judicial policies that permeate many of its rulings.

The notion that the WTO treaties reflect a kind of fundamental balance or equilibrium between an inherent right to regulate and specific disciplines on its use in the trade context, and that the fundamental task of dispute settlement is to preserve this equilibrium over time, was already nascent in the Appellate Body’s treatment in Shrimp–Turtle of the relationship of the operative provisions of the Article XX exception, which protects policy space, to the conditions in the chapeau or preambular paragraph of Article XX, which prevent discriminatory or protectionist abuse of that policy space. However, it was in the China–Publications case that it became clear that this kind of equilibrium was seen by the Appellate Body as being at the core of the WTO legal system as a whole or at least preserving it was the core of the task of the Appellate Body.⁸⁴ In China–Publications, China sought to invoke the Article XX exception as a defence with respect to obligations in its protocol of accession, even though Article XX is not explicitly incorporated into the protocol. China argued, however, that Article XX was implicitly incorporated through the language ‘right to regulate’ in the protocol. The Appellate Body’s reasoning revealed its philosophy of the WTO legal system in general, as seen from the perspective of the adjudicator’s task:

[W]e see the ‘right to regulate’, in the abstract, as an inherent power enjoyed by a Member’s government, rather than a right bestowed by international treaties such as the WTO Agreement. With respect to trade, the WTO Agreement and its Annexes instead operate to, among other things, discipline the exercise of each Member’s inherent power to regulate by requiring WTO Members to comply with the obligations that they have assumed thereunder. … We observe, in this regard, that WTO Members’ regulatory requirements may be WTO-consistent in one of two ways. First, they may simply not contravene any WTO obligation. Secondly, even if they contravene a WTO obligation, they may be justified under an applicable exception.

As participants in the WTO system, in other words, WTO Members have not subordinated their inherent right to regulate to the telos of deep or ever deepening integration but, rather, have agreed to limit the inherent right to regulate through specific legal disciplines, the contours of which have to be determined by reading the operative provisions and the exceptions as an inseparable whole to ascertain what is permitted and what is not. One cannot presume a broad meaning to an obligation and/or that exceptions are narrow. The kind of equilibrium to be preserved in interpretation is to be ascertained through the holistic view of the interaction of obligations and exceptions in the WTO system, beginning from the notion of the system as a set of fixed, bargained constraints on an inherent plenary power to regulate. Apart from these fixed, bargained limits, there is no ‘constitutional’ requirement in the WTO that the right to regulate be exercised in any particular manner or exercised or not exercised to any particular end. It almost goes without saying that one would not read a human rights treaty or the UN Charter in this way. In the WTO, there is no general relativization of sovereignty to a community normative order held together by a particular shared telos or teloi or vision of right or good. If there is a community ethos, it is a Hobbesian one of the avoidance of the summum malum. In this instance, free fall into beggar-thy-neighbour protective discrimination. A policed equilibrium of rights and obligations sustains enough confidence among Members generally in both the strength and flexibility of the disciplines to hedge effectively against a cascade of defection to unilateralist protectionism.

If the dictionaries, the textualism and the mechanical use of the VCLT all seemed annoying to more elevated legal minds, at least they became eventually reassuring to panels and litigators, who made sure to arm themselves amply with dictionaries and to structure their pleadings around textual readings under all of the elements of the VCLT. The relaxation of this artifice has cumulatively led to a new anxiety. Concern that the Appellate Body is practising a kind of policy-blind formalism has given way to the fear or suspicion that the Appellate Body has gone into free fall towards a kind of legal realist opportunistic decision making case by case, often veering widely on ‘doctrine’. On the other hand, the Appellate Body has been, through twists and turns, working out a number of what I will call judicial policies, which are fairly closely related to the legitimacy challenges described earlier. Analysis and critique of the WTO’s ‘mature’ jurisprudence could usefully be focused by an awareness of these judicial policies and reflection on their significance for the WTO as a whole. The policies have been operating right from the beginning of the Appellate Body’s decision making but have been often somewhat obscured or not brought to full attention, due to the overall focus on a rigid constructivist approach. I now turn to the policies I see most frequently and powerfully at work in the jurisprudence. I do not claim to know to what extent these policies are pursued consciously or unconsciously by the Appellate Body, but they do seem to be at work in the manner in which it comes to its findings in many of the most important and sensitive disputes.

6 Main Judicial Policies of the Appellate Body

A Conditional Deference in the Non-Discrimination Regime

In the Shrimp–Turtle case, the Appellate Body had already signalled that it would examine the consistency of domestic policies with the WTO treaties through a different or broader lens than the prevailing outlook in the WTO’s ‘institution’. That outlook emphasized the progressive liberalization of trade and a suspicion that much regulation is captured by protectionist interests or a pretext for protectionism, if not simply irrational by some economic theory about first best instruments for correcting market failures. It would take many twists and turns in the case law, however, for the Appellate Body to work out the doctrinal edifice for its approach – one that is more sensitive to values that are external to the trade liberalization project (at least in large measure) and more respectful of the choices and constraints of domestic regulators, while sending signals that, at the same time, the protectionist abuse of flexibilities is being effectively identified and constrained.

This doctrinal edifice is what I call the non-discrimination regime. In the non-discrimination regime, the scrutiny of domestic policies is trifurcated. There is, first of all, typically an examination under the national treatment or most-favoured-nation (MFN) non-discrimination norms of whether the policies complained of result in less favourable treatment either of imported products (national treatment) or of imports from some particular WTO Member(s) (MFN). Consideration of regulatory intent or of evidence of purposeful discrimination plays no role in this analysis. The adjudicator makes a determination of whether the products are ‘like’ based upon objective criteria, such as physical characteristics and end uses, while consumer preferences can also be dispositive, and then undertakes a formalistic (not empirical) analysis of whether the regulatory intervention in question has detrimental impact on competitive opportunities for imported like products. In this disparate impact or de facto discrimination analysis, there is no apparent room for consideration of outside values or legitimate regulatory purposes. The approach in effect excuses the WTO litigator from having to make any substantive judgments about the legitimacy or justification of the policies in question.

The second stage of the trifurcated regime is what I call rationality review of policies that have been found to have a detrimental impact on competitive opportunities. Here, the Appellate Body applies a rather deferential standard of review to determine whether, given the impact on trade, the defending Member has acted reasonably in the choice of policy instrument for its chosen objective. Under the GATT, rationality review is undertaken through application of the exceptions in Article XX; under the TBT Agreement,⁸⁵ as is explained further in the next section, rationality review is undertaken through the Appellate Body’s own construct of the notion of a ‘legitimate regulatory distinction’, which it has read into the non-discrimination norms of the TBT Agreement in order to reconcile the approach of the Agreement with that of the GATT. The rationality review is broadly deferential not only to the choice of policy objectives and the level or strictness of regulation but also to the general form of the regulation, the basic choice of the policy instrument.

The third or final stage in the non-discrimination regime is the strict scrutiny of specific or special features of the policies complained of that may lead, again on a formalistic analysis, to a detrimental impact on imports in the way in which the regulatory scheme is applied or operationalized in practice. Here, the Appellate Body may find a flawed procedure, some arcane or anomalous distinction in the fine print of the regulatory scheme, which may often lead to under-inclusiveness (exceptions or limitations on the operation of the scheme that appear to give some advantage to domestic products over imports). This stage of analysis will likely result in the WTO Member being called on to fine-tune its regulatory intervention, without necessarily having to make major changes in the basic choice of policy instrument.

1 Discrimination as Market Disadvantage

The text of the national treatment provisions of the GATT requires that the adjudicator decide whether less favourable treatment is provided for ‘like’ imported products and/or, in the case of taxation measures only, whether dissimilar treatment is provided for directly competitive and substitutable products. From the outset of its jurisprudence, the Appellate Body had veered away from an intent-based, or purpose-based, approach to national treatment, which would examine whether any market disadvantage to imports could ultimately be traced to legitimate public policies. This latter approach had a basis in the GATT acquis – the so-called ‘aims-and-effects’ approach to national treatment. ‘Likeness’ of products was to be determined by seemingly ‘ objective’ market-related considerations, such as physical characteristics, end uses and consumer preferences. If products were alike on the basis of such criteria, then discrimination – a violation of national treatment – occurred where the regulation produced some kind of detrimental impact or market disadvantage for the imported product. This analysis of detrimental impact could be based purely on a formal or hypothetical type analysis. Product A is mostly produced by foreign producers; Product B is largely produced by domestic producers. They are alike based on physical characteristics. Product B is taxed at half the rate of Product A. Thus, there is a detrimental impact on Product A, the like imported product. No econometrical evidence is required, no proof of actual substitution by consumers. National treatment protects equality of competitive opportunities in principle, in the abstract.

The simplicity of this approach – it has an advantage that it requires only minimal analysis of facts, given that it is the panels, which are not made up of professional adjudicators, who make the findings of fact – was challenged in the EC–Asbestos case. In assessing France’s ban on both domestic and imported asbestos, the panel had found that asbestos, a product that had killed or created serious illness in many thousands of people could be considered ‘like’ to substitute products with no record of lethality. While France’s ban was upheld by the same panel under the Article XX(b) health exception, the normative messaging to outsider communities, such as asbestos victims and the public health community, of considering the products ‘like’ and finding that distinguishing them was discrimination was, to say the least, risky from the point of view of the WTO’s legitimacy. This situation lent credence to the views of scholars, such as myself, who warned that there are dangers in impugning as ‘discrimination’ legitimate public policies simply based upon market disadvantage, which are blind to policy considerations, even if the policies could eventually be held to be justified under Article XX.

The Appellate Body pulled a rabbit out of a hat, as it were. First, it held that the panel had not taken into account sufficiently the physical differences between the two products but, more fundamentally, in responding to the outsider constituencies, health should have been taken into account in comparing the two products because the health effects of products influence consumers’ choices. This was a brilliant improvization. The Appellate Body could say health was relevant, not for public policy considerations but, rather, for consumer behaviour considerations and, thus, strictly speaking, not alter its market-based framework of analysis in response to the legitimacy challenge posed in this case. However, for years after EC–Asbestos, the Appellate Body in occasional decisions had been giving hints that, in appropriate cases, it would not find treatment less favourable of imports if the market disadvantage came from some extraneous factor unconnected to the foreign national origin of the imported group of products. In other words, there was an escape valve of sorts from the market disadvantage approach where in a particular case it might pose a legitimacy challenge. With the recent EC–Seal Products decision, the Appellate Body appears to have largely closed these escape valves, apart from the specific one it invented in EC–Asbestos in relation to health, consumer behaviour and likeness.⁸⁶ Significantly, this closure only occurred once the Appellate Body had consolidated a deferential rationality review approach under Article XX to a Member’s overall policy intervention. It is to this jurisprudential development that we now turn.

2 Sliding towards Rationality Review/Deference: The Appellate Body and Article XX of the GATT

Especially towards the end of the GATT era, when a kind of crude economistic ideology and strong deregulatory orientation had come to permeate the insider trade community, it was notoriously almost impossible to justify public policies under Article XX of the GATT. It was possible to defeat a defence under Article XX just by dreaming up a theoretical less trade restrictive alternative policy that could serve the GATT member state’s objective, regardless of its costs, its feasibility and the degree of certainty or uncertainty as to whether, in real world conditions, the policy would actually attain the GATT member state’s objective. Almost always an economist could think of a policy other than trade restrictions that might, in an ideal world, achieve a given policy goal. Defending policies under Article XX under these conditions was essentially a sucker’s game.

The Thai Cigarettes case is a clear illustration.⁸⁷ Thailand banned imported US cigarettes but not domestic cigarettes. Here, Thailand argued that, given the kind of marketing and advertising that were associated with the imported cigarettes, they posed a health risk in terms of attracting young people to cigarette addiction that did not exist in the case of domestic cigarettes. The GATT panel found that Thailand’s import ban could not be justified under Article XX(b) of the GATT on the basis that other, less trade restrictive measures, such as control of advertising, could achieve its public health objective. However, the World Health Organization representative intervening in the panel proceedings noted that ‘[m]ultinational tobacco companies had routinely circumvented national restrictions on advertising through indirect advertising and a variety of other techniques.’⁸⁸ The panel simply ignored this evidence, which suggested an important reason why, for Thailand, advertising regulation might not be a reasonably available less trade restrictive alternative to an import ban. Starting with some of its earliest decisions, the Appellate Body transformed Article XX step by step into an effective means of protecting legitimate policy space under the GATT non-discrimination regime.

In US–Reformulated Gasoline, the Appellate Body considered the precise wording of the ‘exhaustible natural resources’ exception in GATT Article XX(g) to argue for rationality review rather than a strict scrutiny approach.⁸⁹ The requisite connection between the measure and the objective in Article XX(g) was expressed in the wording ‘related to’, which was different from the language ‘necessary’ in the other paragraphs of Article XX, which covered matters such as public morals and human and animal life and health. Then, in the Korea–Beef ruling, the Appellate Body dropped the other shoe, as it were, and reasoned that ‘necessity’ does not necessarily mean necessary.⁹⁰ One meaning of necessary was ‘indispensable’, but that was not the only meaning; a measure could be necessary if it was, on a continuum, significantly closer to being indispensable than to merely making some contribution to the Member’s objective. Necessity in this attenuated sense, however, was to be determined by a holistic judgment based on ‘weighing and balancing’ a series of factors, including the contribution, the importance of the values and interests protected by the law or regulation and the impact of the law or regulation on imports or exports.

Clearly, the older GATT jurisprudence with its dependence on the least trade restrictive alternative test had assumed that a measure must be indispensable to be ‘necessary’ in the sense that no other measure other than a trade restrictive one (at least trade restrictive to that extent) could achieve the same contribution to attaining the Member’s objective. The notion of ‘weighing and balancing’ undid the doctrinal purity of least trade restrictive alternative. To some, it implied an alternative proportionality analysis of measures that were asserted to be ‘necessary’, though not ‘indispensable’. That what the Appellate Body was doing was really shifting to rationality review would only become apparent in a series of later cases, US–Gambling (which dealt with public morals under the General Agreement on Trade in Services provision that is the functional equivalent of GATT Article XX),⁹¹Brazil–Retreaded Tyres⁹² and EC–Seal Products.⁹³

In order to make a prima facie case that its measure is justified as necessary under Article XX, a WTO Member would only need to show that its measure made a ‘material contribution’ to the objective – a material contribution would have to be of a significant nature, the Appellate Body seemed to be saying, given the level of trade restrictiveness of the measure. However, the Appellate Body also held that there is no need to quantify or measure the extent of the contribution. That a measure of the kind would address the problem in a meaningful way was a matter of common sense reasoning, not empirical proof. To put it bluntly, if you do not quantify or measure a ‘contribution’, it is really impossible to say if it is significantly closer to being indispensable than to making any including a trivial contribution to the attainment of the Member’s objective. Once a Member makes a prima facie case of its measure making a sufficient – ‘material’ contribution – then the complainant might raise ‘reasonably available’ less trade restrictive alternatives. As long as the defending Member, however, can provide a reasonable explanation of why it did not adopt such an alternative, its justification would stand. Such a reasonable explanation could include: ‘[The alternative measure] is merely theoretical … the responding Member is not capable of taking it, or where the measure imposes an undue burden on that Member, such as prohibitive costs or substantial technical difficulties.’⁹⁴ In other words, an Article XX review largely boils down to assessing the overall reasonableness of the Member choosing the measure that it uses to achieve its objective, given the trade restrictiveness of that measure.

That this is the logical outcome of the various twists and turns of the Appellate Body on ‘necessity’ under Article XX becomes very clear in the EC–Seal Products ruling. In this case, the Appellate Body found that the EU’s ban on seal products was necessary for the protection of public morals, which were understood in terms of animal welfare or countering cruelty to animals. The Appellate Body clarified: ‘We therefore do not see that the Appellate Body’s approach in Brazil–Retreaded Tyres sets out a generally applicable standard requiring the use of a pre-determined threshold of contribution in analysing the necessity of a measure under Article XX of the GATT 1994.’⁹⁵ The Appellate Body further held: ‘[I]n order to qualify as a “genuine alternative”, the proposed measure must be not only less trade restrictive than the original measure at issue, but should also “preserve for the responding Member its right to achieve its desired level of protection with respect to the objective pursued.” The complaining Member bears the burden of identifying possible alternatives to the measure at issue that the responding Member could have taken.’⁹⁶ Finally, the Appellate Body clarified that any substantiated reason why the alternative measure is not ‘reasonably available’ (not just the reasons listed in Brazil– Retreaded Tyres) may be sufficient to rebut the complainant’s invocation of hypothetical less trade restrictive alternatives:

[The formulation in Brazil–Retreaded Tyres] does not foreclose the possibility that there may be other indications that the alternative measure is ‘merely theoretical in nature’. As we see it, if there are reasons why the prospect of imposing an alternative measure faces significant, even prohibitive, obstacles, it may be that such a measure cannot be considered ‘reasonably available’.⁹⁷

3 Chapeau or Chapeau-like Strict Scrutiny

If the Appellate Body needed to make assurances of policy space to establish and enhance its legitimacy in an era where neo-liberal globalization is highly contested, it certainly also needed to show it could maintain meaningful constraints on protectionist abuse of public policies that undermined the value or integrity of the basic GATT-like commitments on border measures. From its earliest jurisprudence under Article XX of the GATT, the Appellate Body underlined the importance of the chapeau of Article XX in allowing such a balancing act. As explained in the next section, the Appellate Body has assimilated the norms of the Uruguay Round’s TBT Agreement to the GATT-based anti-discrimination regime (Articles I, III and XX). Thus, chapeau-type strict scrutiny also occurs with the Appellate Body’s discrimination termination under Article 2.1 of the TBT Agreement, which includes both national treatment and MFN obligations.

What typifies chapeau or chapeau-like strict scrutiny (in the case of TBT) is a focus on the fine print of the regulatory scheme and particularly on those features that raise concerns that the way the scheme is applied in practice may entail elements of discrimination – a detrimental impact on like imported products or (in the case of MFN) like imported products of particular WTO Members. Rationality review displays deference not only to the collective preferences of the society in question but also to the general form of intervention: a ban, a mandatory labelling scheme, a PPM, no matter. However, under strict scrutiny of the fine detail, any distinction or classification that could give rise to a discriminatory impact in the actual operation of the scheme is suspect and must be explained as being, essentially, indispensable to the regulatory objective of the scheme as a whole. As with the preliminary analysis of discrimination under the national treatment or MFN operative provisions (Articles I and III), there is no need for the complainant to show actual deleterious impact but only a formal analysis that the distinction or classification could operate in such a way as to disfavour imports or in the case of MFN imports from certain WTO Members.

A dramatic illustration of this scenario is the recent EC–Seal Products case. The EU’s ban on seal products was accompanied by an exception for indigenous subsistence seal hunts. Since Greenland had a large indigenous seal hunt, discrimination was found on the basis that Canada’s and Norway’s commercial sealing industries could have suffered a deleterious impact. The theory would have to be that demand would be shifted from Canadian and Norwegian seal products (banned) to Greenland products (EC origin) that are permitted under the indigenous exception. However, after the EU ban on non-indigenous hunts, Greenland’s seal industry in fact exhibited a tremendous downturn in sales. There was certainly no empirical evidence that consumers were responding to a ban on commercially hunted seal products by buying more indigenous products.

There are a number of ways in which reserving strict scrutiny for the fine print blunts or mitigates the intrusiveness of putting a Member’s domestic regulations under a microscope. First of all, in many instances, the ‘fine print’ has been a matter of regulations or administrative practices, which can be altered without the need to alter the legislative scheme itself. This was conveniently the case with the problems that the Appellate Body identified under the chapeau in Shrimp–Turtle. Even if some legislative change is required, tweaking the details of a complex regulatory scheme may not raise the kinds of sensitive political problems involved in attempting a major overhaul.

Perhaps most significantly in case after case where the Appellate Body has found discrimination at the stage of chapeau or chapeau-like strict scrutiny, the problem has been under-inclusiveness of one sort or another: in US–Clove Cigarettes, the problem (under Article 2.1 of the TBT Agreement was that the USA had banned clove cigarettes but not all other flavoured cigarettes that appeared to raise the same public health concerns (most notably, menthol);⁹⁸ in EC–Seal Products, as just mentioned, the indigenous exception as well as certain other kinds of very limited exceptions; in Tuna–Dolphin II, strict monitoring and verification of the ‘dolphin-safe’ label had been extended to the fisheries that applied to the complainant Mexico but not to certain other Tuna fisheries and in US–COOL, the country-of-origin labelling scheme of the USA imposed burdensome tracing and record-keeping requirements on operators who were processing partly foreign origin meat, but the actual information given to consumers about national origin was less than that generated by the considerable regulatory burden.⁹⁹ In each instance, it would clearly be possible to respond to the finding of chapeau or chapeau-like discrimination by making the regulatory scheme stricter or more watertight, which is likely to better serve, as opposed to undermining, the main interests and values behind the regulation.

When the EU closed or narrowed some of the exceptions in the seal ban that had been found by the Appellate Body to not meet the conditions of the chapeau, the animal welfare activists who pushed for the ban in the first place understandably cheered. Their cause had actually benefited from strict scrutiny under the chapeau. Of course, it is always possible – and this is what the complainants count on perhaps – that the kinds of adjustments required by the Appellate Body would be unacceptable, given the alignment of domestic interest groups around the particular regulatory scheme, pro and con. In the US–COOL case, for example, the USA sensibly responded to the Appellate Body ruling by requiring that the full information that operators were required to collect be provided to consumers. However, this led to some additional issues for the Appellate Body upon review of US compliance. In this instance, given the powerful US domestic interests that opposed country-of-origin labelling (though it is enormously popular with consumers) and the attitude of the Republican-dominated Congress, mandatory country-of-origin labelling may be abolished altogether, the existing law has been repealed. In this case, the WTO complainants may have succeeded in their objective, even though it was open in principle to the USA to respond to the chapeau problems identified by the Appellate Body by making the scheme tighter or stricter.

But consider the incentive effects of focusing chapeau strict scrutiny on features that are under-inclusive or can be modified by making the scheme tighter or stricter. A potential claimant runs the risk that it will spend millions of dollars and several years and receive a positive result from the Appellate Body, only to find out that it is a Pyrrhic victory or, worse, that the increasing strictness of the new scheme makes it even harder to achieve the market access that the claimant is seeking.

Another feature of the focus of chapeau or chapeau-strict scrutiny is what I would call a kind of reverse ‘regulatory chill’ effect. The ‘fine print’ features that are impugned under the chapeau or chapeau-like strict scrutiny may well be idiosyncratic features of a particular Member’s regulatory scheme, which another Member seeking to regulate the same kind of problem using the same sort of general policy instrument would not need to duplicate or include in their own regulation. To take the EC–Seal Products example, the concerns of the Appellate Body about the EU indigenous exception would obviously be irrelevant where a Member had no indigenous population and thus no need for this kind of exception. By limiting strict scrutiny to regulatory design or operational features that are likely to be specific to a particular Member’s situation, the general signal that the Appellate Body sends is a positive one of policy space, but with a warning not to abuse this policy space for protectionist ends.

B Assimilation of ‘Post-Discriminatory’ Uruguay Round Norms with Non-Discrimination

As just discussed at some length, in the presence of a normative dissensus about the future direction of the WTO and under conditions of continuing contestation over the neo-liberal approach to trade liberalization, the Appellate Body has understandably shifted the focus of its scrutiny of domestic regulations from second-guessing substantive domestic policy choices to an emphasis on the prevention of protectionist abuse or the arbitrariness of domestic regulations, on process norms and, above all, on the examination of discriminatory elements in the detailed legal, regulatory and administrative provisions that operationalize the substantive policy choices. However, the Appellate Body, of course, cannot simply wish away the existence of the Uruguay Round WTO agreements, which, to no small extent, are informed by a neo-liberal or Washington consensus outlook, and seek to go beyond the discrimination norm in disciplining purportedly irrational or inefficient domestic regulations that are viewed as barriers to market access while not necessarily having any features that discriminate against imports.

The Agreement on Sanitary and Phytosanitary Measures (SPS Agreement)¹⁰⁰ and the TBT Agreement were clearly seen by their architects as moving well beyond the basic non-discrimination norms of the GATT – national treatment and MFN – towards spurring market-friendly regulatory reform or deregulation as well as global regulatory harmonization through the use of international standards. One of the most remarkable aspects of the judicial activism of the Appellate Body has been to read (with some zigzagging in its early years) what Hudec calls the ‘post-discriminatory’ provisions in these agreements¹⁰¹ in such a way as to give little opportunity to claimants to challenge measures that either would not violate the non-discrimination provisions of the GATT or would be upheld under the Article XX exceptions.

In the case of the TBT and SPS Agreements, this hermeneutic strategy has been most explicit. In the US–Clove Cigarettes ruling, the Appellate Body pointed to language in the preamble of the TBT Agreement that resembled or reiterated elements of the GATT non-discrimination regime (including Article XX exceptions) and held:

The balance set out in the preamble of the TBT Agreement between, on the one hand, the desire to avoid creating unnecessary obstacles to international trade and, on the other hand, the recognition of Members’ right to regulate, is not, in principle, different from the balance set out in the GATT 1994, where obligations such as national treatment in Article III are qualified by the general exceptions provision of Article XX.¹⁰²

This statement is remarkable. It suggests that the WTO’s judiciary must not read provisions of the TBT Agreement that are different from, and additional to, those in the GATT in such a way as to create a ‘balance’ that is more liberalizing or more restrictive of domestic regulatory autonomy than the GATT itself. In the case of the SPS Agreement, the Appellate Body, in its very first decision, made the remarkable teleological pronouncement that the ‘post-discriminatory’ harmonizing features of the SPS Agreement were merely a means to disciplining discrimination and disguised restrictions of international trade (as opposed to exhibiting a neo-liberal ‘deep integration’ objective):

The ultimate goal of the harmonization of SPS measures is to prevent the use of such measures for arbitrary or unjustifiable discrimination between Members or as a disguised restriction on international trade, without preventing Members from adopting or enforcing measures which are both ‘necessary to protect’ human life or health and ‘based on scientific principles’, and without requiring them to change their appropriate level of protection.¹⁰³

The text and structure of the TBT Agreement, taken on their own, suggest a greater limitation on policy space and more second guessing of domestic policy choices than in the case of the GATT. (This is understandable because the spur for the TBT Agreement was the sense that the GATT was inadequate to address regulatory barriers to trade that were in the form of regulations that might not be discriminatory but were overly cumbersome or inefficient.) First of all, the national treatment and MFN provisions of the TBT Agreement (Article 2.1) are not made subject to an Article XX-type public policy exception. Second, the provision of the TBT Agreement that at first glance resembles Article XX of the GATT – Article 2.2 – actually requires justification for all of the technical regulations that do not create unnecessary obstacles to trade, regardless of whether they violate the non-discrimination or any other provisions of the TBT Agreement or GATT. Third, Article 2.4 of the TBT Agreement requires WTO Members to use international standards as a basis for their technical regulations where available and relevant and appropriate. This kind of regulatory harmonization is nowhere to be found in the GATT.

Now let us see how the Appellate Body has managed claims under these provisions so as to ensure that the balance between trade liberalization and regulatory autonomy remains unaltered from the GATT. First of all, the Appellate Body simply read into Article 2.1 of the TBT Agreement a kind of Article XX exception. A feature of a technical regulation found to provide less favourable treatment either to a like imported product (national treatment) or of some imported products (MFN) could nevertheless be found not to violate Article 2.1 if this feature stemmed exclusively from a ‘legitimate regulatory distinction’. With respect to Article 2.2 of the TBT Agreement, the Appellate Body has placed much more emphasis on the language in the second paragraph that refers to a measure being required to be no more trade restrictive than necessary rather than the ‘unnecessary obstacle’ terminology of the first. In articulating what trade restrictive means, the Appellate Body refers to a previous ruling that interprets restrictions on trade in the context of Article XI of the GATT, which makes illegal quantitative restrictions that tend to be inherently discriminatory against imports.

The Appellate Body has given no indication that it can conceive of a measure as being trade restrictive unless it has the kinds of effects on conditions of competition that would lead in any case to a finding of a violation of non-discrimination norms in the GATT. At the same time, overturning findings of violation of Article 2.2 of the TBT Agreement by the panels in US–Clove Cigarettes and Tuna–Dolphin II, the Appellate Body has emphasized that applying Article 2.2 involves the same kind of weighing and balancing exercise as GATT Article XX, with a considerable margin of appreciation in examining possible alternative measures that a Member might take that are less restrictive, and that there is no need to quantify the contribution of the measure or possible alternatives to the Member’s objectives. The Appellate Body has also underlined elements in Article 2.2 that point to deference or margins of appreciation – for example, unlike Article XX of the GATT the initial burden of proof is on the complainant to show excessive trade restrictiveness, and Article 2.2 requires that in determining whether a measure is more trade restrictive than necessary the nature of the risks that would occur if its objective were not to be fulfilled must be taken into account. The overall emphasis is usually on a holistic exercise to determine the reasonableness of the Member’s policy choice. In light of this inquiry, it could be asked what other instruments might have been available that could make an equal contribution to this objective, while being less trade restrictive. Given these jurisprudential moves, it seems hard to conceive of an instance where a Member could make a successful claim under Article 2.2 against a measure that did not violate the GATT.

Now let us turn to Article 2.4 of the TBT Agreement, namely international standards. In an early case under the TBT Agreement, EC–Sardines, the Appellate Body appeared to lurch in a neo-liberal direction, implying that Article 2.4 implied a large measure of regulatory harmonization – a very close fit or relationship between any technical regulation and the international standard, providing very little flexibility for regulatory diversity.¹⁰⁴ In Tuna–Dolphin II, which addressed the lack of any definition of an international standard in the TBT Agreement, the Appellate Body – in an unusual reliance on the work of a WTO committee – outlined a set of criteria that, cumulatively, very few international standardization initiatives are likely to meet at present. These criteria include ‘[a]ll relevant bodies of WTO Members should be provided with meaningful opportunities to contribute to the elaboration of an international standard so that the standard development process will not give privilege to, or favour the interests of, a particular supplier/s, country/ies or region/s’ as well as complete transparency at the drafting stage. There is ample scope for the Appellate Body to resist demands for regulatory harmonization through Article 2.4 due to the burden the complainant faces in proving that the above-mentioned criteria have been met fully with regard to the standards in question.

In sum, through its readings of Articles 2.1, 2.2 and 2.4 of the TBT Agreement, the Appellate Body has made it effectively impossible, or at least very unlikely, to succeed with a claim under the TBT Agreement that would not also succeed under the GATT. At the very outset, the Appellate Body had established that the TBT and SPS Agreements were to be applied in parallel with, and not to the exclusion of, the GATT (although both of the newer agreements should typically be considered first). The incentives are now considerable for the parties to frame disputes about technical regulations as being, essentially, GATT disputes about discrimination and/or policy justifications through the general exceptions. After all, as the Appellate Body itself pronounced, the balance between trade liberalization and regulatory autonomy established by the GATT is to be maintained.

The route to this result under the SPS Agreement has been more tortuous. Arguably, the SPS Agreement goes farther than the GATT towards a ‘post-discriminatory’ order that judges the rationality of public policies since it appears to demand that the SPS measures be sustained on the basis of scientific rationality. Article 2.2 of the SPS Agreement requires that SPS measures be based on scientific principles and supported by ‘sufficient’ scientific evidence. Article 5.1 stipulates that measures be based on a scientific assessment of risk. In the first case under the SPS Agreement, EC–Hormones, the Appellate Body rejected a proceduralist approach that would have interpreted the science requirements as obligatory inputs into the process of deliberation and decision, not substantive rationality standards against which the WTO judiciary would judge SPS regulations. This would have been a very effective and direct route to neutralizing or taming the most legitimacy-threatening neo-liberal ‘post-discriminatory’ elements of the SPS Agreement. However, the Appellate Body would have had to deal with the fact that there is no phase-in period for SPS obligations, and they clearly applied to existing, not just new, measures.

Thus, interpreting the requirements in a procedural way would have produced a sort of retroactivity problem. It would have also highlighted a legitimacy problem of a different nature with the SPS Agreement’s reliance on science. Few developing countries have the capacity or resources to conduct scientific risk assessments of the kind seemingly intended by the SPS Agreement, and, thus, they would be penalized if the requirement of science were a requirement imposed on the domestic regulatory process, as opposed to an objective standard that could be fulfilled by pointing to a risk assessment conducted elsewhere or by an international body, which never was in fact part of the actual regulatory or political process that produced the measure.

Hence, the Appellate Body found more subtle and indirect ways of avoiding the WTO judiciary being turned into a science court for domestic regulations. First of all, the Appellate Body held that science did not necessarily mean mainstream or majority scientific opinion. It was a sovereign right of a WTO Member to choose among differing scientific opinions, as long as the scientists were competent and a methodology corresponding to the scientific method in the broadest sense was adopted. Nor did risk need to be quantified. Second, in a moment of (relatively rare) eloquence, the Appellate Body held that the risk to be considered was not only the kind of risk that could be assessed in a laboratory ‘but also risk in human societies as they actually exist, in other words, the actual potential for adverse effects on human health in the real world where people live and work and die’.¹⁰⁵ Third, the Appellate Body insisted on the sovereign right of a WTO Member to determine its own level of protection. Thus, one took a society’s risk preferences as one found them, and it would not be unacceptably unscientific for a Member to regulate so as to seek to attain a level of risk approaching zero, no matter what degree of trade restrictiveness of the regulation. If a Member’s chosen level of protection was high enough, even a risk assessment that suggested the risk was very small would be sufficient to meet the requirements of the SPS Agreement. Fourth, the Appellate Body largely gutted the regulatory harmonization provision of the SPS Agreement by holding that the requirement that ‘[m]embers shall base their sanitary or phytosanitary measures on international standards’ was of an aspirational, soft law character, despite the use of the word ‘shall’. It did so on the basis that the purpose of the provision was stated as ‘harmoniz[ing] … on as wide a basis as possible’ and that this had an aspirational ring to it. Fifth, even though the EU had not invoked a provision of Article 5.7 of the SPS Agreement that allowed provisional regulations in the absence of sufficient scientific evidence on a precautionary basis, the Appellate Body held that, while it was not persuaded that there is a precautionary principle in international law with a normative force that would override SPS treaty provisions, nevertheless as an interpretive matter:

a panel charged with determining, for instance, whether ‘sufficient scientific evidence’ exists to warrant the maintenance by a Member of a particular SPS measure may, of course, and should, bear in mind that responsible, representative governments commonly act from perspectives of prudence and precaution where risks of irreversible, e.g., life-terminating, damage to human health are concerned.¹⁰⁶

Finally, the Appellate Body had to reckon with a provision of the SPS Agreement that appears to be very intrusive in democratic decision making about risk. Article 5.5 requires WTO Members to avoid ‘arbitrary or unjustifiable distinctions’ in the levels of protection it seeks in different situations … ‘if such distinctions result in discrimination or a disguised restriction on international trade’. Reversing the panel, which had taken the view that the existence of ‘arbitrary or unjustifiable distinctions’ raised a presumption that there was ‘discrimination or a disguised restriction on international trade’, the Appellate Body held that in fact the core of the Article 5.5 obligation was to avoid discrimination or disguised restrictions on international trade. The existence of ‘arbitrary or unjustifiable distinctions’ in the levels of protection in different situations merely functioned as one ‘warning’ signal of whether the measure was discriminatory or a disguised restriction.¹⁰⁷ Thus, the latter issue, essentially non-discrimination, becomes the crucial one for applying Article 5.5. Indeed, as noted at the outset, the Appellate Body made a revealing statement that harmonization under the SPS Agreement is merely a means to achieving the goal of avoiding discrimination or disguised restrictions on international trade, conceptually distancing itself from a neo-liberal post-discriminatory agenda of imposing a notion of regulatory rationality or efficiency through the trading system.

Despite all of these elements in reading down the ‘post-discriminatory’ provisions of the SPS Agreement into something approaching the GATT non-discrimination regime, the Appellate Body found the EU to be in violation of the SPS Agreement. The key point was that the EU had more or less admitted the lack of ‘laboratory’ scientific evidence to support its banning of synthetic growth hormones in meat products but then argued that the science did not take into account the possibility that doses of growth hormones were being administered that were much higher than what was indicated by good veterinary practice. As noted, the Appellate Body was open to real world risk, not just ‘laboratory’ risk, as a scientific basis for SPS measures, but the EU had failed to produce any study that showed the presence of this real world risk – that is, abuses in the administration of hormones to animals, which threatened humans with much higher exposures to residues than would be indicated by laboratory studies, supposing proper veterinary practices.

After many years of non-compliance with the Appellate Body ruling, the EU finally produced some new scientific studies that indeed showed directly the risk from synthetic growth hormones. It is in revisiting the dispute in this context that the Appellate Body has been able to clarify and reinforce the elements of deference built into its approach to the SPS Agreement. This clarification was particularly necessary since in two cases in the intervening years the Appellate Body had appeared to accept intrusive approaches by panels that put risk assessments under a microscope – as if the panels themselves were equipped to determine what kinds of scientific inquiry or methodology were adequate to ascertain the nature or extent of the risk the defending Member was seeking to regulate (Australia–Salmon; Japan–Apples).¹⁰⁸ In the case of Japan–Apples, the Appellate Body even threatened to undermine the fundamental pillar of respect for the regulating Member’s chosen level of protection. In that case, the Appellate Body failed to overturn a finding of the panel, which read a proportionality requirement into Article 2.2 of the SPS Agreement and held there was a violation because the measure was ‘clearly disproportionate’ to the risk identified in the scientific studies. (At the same time, the Appellate Body seemed to protect its prior approach by reading the panel reference to ‘disproportionate’ as a way of expressing the notion that there was no rational relationship between the risk as determined by science and the regulation actually adopted, as opposed to a fully blown proportionality analysis.)

In its new ruling, US–Hormones Suspension,¹⁰⁹ the Appellate Body reiterated many of the deferential elements in its original ruling which had been ignored by the panel in this new phase of the dispute (the acceptability of non-majority, non-mainstream scientific opinion and the right to regulate based on ‘real world’ risk such as abusive veterinary practices). However, most importantly, the Appellate Body made a clear statement that the SPS Agreement does not invite the WTO judicial system to make its own judgment on the scientific justification of the defending Member’s substantive regulation. The role of science in the adjudication of SPS claims is much more limited, and apart from the requirement that the defending Member invoke scientific evidence, essentially the same as a rationality review under the non-discrimination regime. (Even under the GATT, a defending Member would likely present some kind of scientific evidence to support its regulation.) Thus, according to the Appellate Body:

it is the WTO Members task to perform the risk assessment. The panel’s task is to review that assessment. Where a panel goes beyond this limited mandate and acts as a risk assessor it would be substituting its own scientific judgement for that of the risk assessor … and, consequently, would exceed its functions under Article 11 of the DSU. Therefore the review power of a panel is not to determine whether a risk assessment is correct, but rather to determine whether that risk assessment is supported by coherent reasoning and respectable scientific evidence.¹¹⁰

Anyone familiar with complex tort litigation, or even anti-trust litigation or investor–state dispute settlement, will know the phenomenon of each party being able to produce an expert or experts with a stellar curriculum vitae, a teaching appointment at a leading university and so on, who offers a carefully reasoned study in support of that party’s position on the scientific facts. The Appellate Body said clearly in EC–Hormones that all that a Member must do under the SPS Agreement is provide a report by some competent, respectable scientist (who may well be expressing a minority or idiosyncratic point of view) that is coherently reasoned and that supports the existence of the risk against which the Member is regulating. The Member is then free to take measures against that risk that corresponds to the level of protection that it has a sovereign right to determine. This does not solve all of the legitimacy concerns with the science requirement in the SPS Agreement, to be sure, because smaller developing countries may well have less capacity to ‘SPS-proof’ their regulations through eliciting or producing such studies. However, in principle and formally, the Appellate Body has really minimized – indeed, close to eliminated – any real difference in scrutiny between the GATT non-discrimination regime and the ‘post-discriminatory’ features of the SPS Agreement.

The Appellate Body has perhaps been fortunate that it has not often been required to decide on sensitive issues of policy space on other Uruguay Round agreements that reflect a ‘Northern’ neo-liberal view of trade and domestic policy – in particular, the TRIPs Agreement and the Agreement on Subsidies and Countervailing Measures (SCM Agreement).¹¹¹ Even in these relatively rare cases, the Appellate Body has found techniques to afford policy space while disciplining the specifically discriminatory element of the policies. The one case where the Appellate Body was faced with applying the substantive standards of the TRIPs Agreement, US–Havana Club¹¹² raised the issue of whether the USA could deny trade market protection where the mark was originally owned by a business that had been confiscated during the Cuban revolution. The mark was then acquired from the Castro regime by a major Europe-based drinks conglomerate, Pernod, but in the USA was used by the Bacardi group. The Appellate Body rejected the very broad reading of trademark protection urged by the EU and, significantly, held that not only the rights that existed under the Paris Convention for the Protection of Industrial Property (a treaty under the World Intellectual Property Organization (WIPO) that long predated the TRIPs Agreement) were incorporated into the TRIPs Agreement but also all of the flexibilities and exceptions.¹¹³

At the same time, the Appellate Body pointed to a gap in the TRIPs Agreement itself. Nowhere does this agreement specify the domestic regulation of who may own intellectual property; it only addresses the nature and scope of the rights that must be protected. In theory, governments could achieve very broad policy goals by placing conditions or restrictions on the ownership of intellectual property (subject to discrimination disciplines). The Appellate Body also engaged in chapeau-type strict scrutiny, finding a violation of the non-discrimination norms in the TRIPs Agreement because an arcane provision of the US trademark regime appeared to impose on foreign persons some kind of additional burden or regulatory step that did not apply to US persons. The USA had argued that in practice this could not lead to discrimination because US persons would not in any case have a route available to them for protection under the provisions in question, due to other aspects of US law. However, exhibiting a typical formalism, the Appellate Body found a violation because this anomaly at least created an apparent hypothetical possibility of less favourable treatment. This being said, the general signal sent by US–Havana Club was that the Appellate Body would read the substantive standards of the TRIPs Agreement narrowly, assuming no greater degree of harmonization of intellectual property protection than is strictly indicated by the text of the TRIPS Agreement or the WIPO treaties regime that pre-dates TRIPS and the neo-liberal intellectual property agenda.

Perhaps this is one reason why pro-intellectual property interests have not pushed the USA or the EU to challenge the use of TRIPs flexibilities by developing countries in WTO dispute settlement, instead preferring pressures tactics such as threats of removal of the general system of preferences (GSP) (one case was brought against Brazil concerning compulsory licensing and then withdrawn). One of the few instances where WTO Members have succeeded through negotiation in re-adjusting the Uruguay Round result towards a more pro-South or less neo-liberal direction is with respect to issue of access to medicine, where patent rights under the TRIPs Agreement have been asserted to prevent licensing to provide low-cost medication to poor people in the South. The accord to adjust the TRIPs Agreement on this issue, which entails considerable administrative obstacles to a developing country compulsorily licensing an essential drug that is produced in a different country, may have been acceptable to the USA and the pharmaceutical lobbies in the shadow of the risk that the Appellate Body might have found even broader flexibilities in the existing text of the TRIPs Agreement or in the gaps in that text.

C Respect for Collective Preferences

As Mavroidis has rightly observed, the Appellate Body’s rejection of the product/process distinction in Shrimp–Turtle signalled an approach of deference or respect for the goals or objectives of regulation adopted by the WTO Members.¹¹⁴ The Appellate Body will simply not make a judgment on the preferences of a given society in regarding what it regulates. This deference extends to the intensity or strictness of regulation. Thus, as noted above, beginning with the EC–Hormones case, the Appellate Body affirmed the right of a WTO Member to determine its own level of protection against a given harm. In principle, a government could seek in its regulation to achieve a risk of zero. The implication of this level of respect for collective preferences is the rejection of the notion of proportionality in the evaluation of the relationship between means and ends, discussed in the last section on the rationality review. As explained, ‘weighing and balancing’ as practised by the Appellate Body leads to an overall assessment of whether there are comprehensible reasons behind the policy choices of a WTO Member, given its own objectives and its obligation under WTO law to avoid unnecessary trade restrictiveness. Proportionality, at least in its strict version, could lead to the invalidation of a policy instrument that makes a contribution towards the achievement of a given level of protection, where that contribution appears small relative to the level of trade restrictiveness in question. In other words, with proportionality, the WTO Member is required to make some sacrifice of the achievement of its chosen level of protection in order to avoid trade restrictiveness. This kind of trade off is not consistent with the respect for collective preferences, as the Appellate Body understands it. (As discussed in the previous section, in one case under the SPS Agreement, Japan–Apples, the Appellate Body deviated from its otherwise consistent rejection of proportionality.)

Respect for collective preferences also goes to the relative weight or priority that a given society attaches to different regulatory objectives. Here, the Appellate Body has on occasion deviated from its full respect for collective preferences but, notably, only in order to justify affording an additional margin of deference under rationality review in cases where human life or health is at stake (EC–Hormones and EC–Asbestos). What the Appellate Body has never done is to apply explicitly a relatively higher level of scrutiny in a judgment in which the regulatory objective was less vital, or should be less vital, for a given society.

An apparent textual difficulty in operating the respect for collective preferences in a consistent manner is that, with respect to the exceptions in Article XX of the GATT, for example, only some regulatory objectives are listed but not all. One way of handling this issue, which is recommended by some scholars including myself, would be to build collective preferences into the determination of whether a measure is discriminatory and, thus, required to be justified in the first place under Article XX. Yet as explained above, in the discussion of the Appellate Body’s non-discrimination regime, while the Appellate Body did go in this direction in at least one case – EC–Asbestos – where consumer preferences about health were considered in determining whether the treatment of different products could properly be compared for the purposes of an analysis of discrimination, overall (after the twists and turns in the case law), the Appellate Body has moved towards an approach where discrimination is found on the basis of an impact on competitive relationships alone, without regard for the basis of the distinctions in policy choices or collective preferences.

It is perhaps no accident, however, that in the very same case where the Appellate Body completed or fully articulated its orientation towards an exclusively competition-based approach to discrimination – EC–Seal Products – the Appellate Body also reinforced its resources to ensure the respect for collective preferences under Article XX of the GATT by reaffirming a broad reading of the meaning of ‘the protection of public morals’ in Article XX.¹¹⁵ This broad reading, already endorsed in US–Gambling¹¹⁶ and China–Publications,¹¹⁷ appears to cross-cut all substantive fields of regulation, focusing on whether the measure in question is deemed by a given society to be derived from the fundamental beliefs or values of that society.

The significance of such an approach for collective preferences was perhaps not fully grasped with US–Gambling and China–Publications because in those cases the kinds of restrictions at issue, including controls on betting and the censorship of films, could be seen to be rather pervasive, traditional or conventional forms of ‘moral’ regulation. In EC–Seal Products, what was at issue was animal welfare, namely the prevention of cruelty to seals. While in fact, as Langille, Sykes and I have shown, regulations addressing animal cruelty have been an element of public morality for some time in many societies, in general, the tendency in the trade policy community has been to make light of the notion of opposing the seal hunt as a genuine moral matter (as opposed to a sentimental fad stoked by fanatical NGOs and celebrities hungry for more publicity). For the panel, which grasped properly the respect for collective preferences already evident in the Appellate Body jurisprudence, and for the Appellate Body itself, there was no place for an objective inquiry into whether concern for animal welfare generally, or, indeed, for the suffering of seals in particular, could or should be a matter of the fundamental beliefs or values of Europeans. This inquiry was essentially a matter of a declaration or assertion by the EU, speaking for its citizens, who were subject ultimately to an implicit condition of good faith – that is, that the declaration not be a sham or pretext for the protection of domestic commercial interests.

Many expected a kind of anti-hypocrisy condition to be put on the invocation of public morals – if you act against cruelty to seals, to show you are serious that animal cruelty is a serious moral matter, then you have to demonstrate that you are as concerned for the suffering of foxes, chickens or pigs. The rejection of this kind of argument (strongly urged by the claimants Canada and Norway) illustrates the consistency of the respect for collective preferences by the WTO judicial system. Caring more about some animals than others, or prioritizing some animal welfare causes over others, is not a question of rationality or irrationality nor does it necessarily raise the spectre of hypocrisy; it is simply a function of the collective preferences of a particular society, which the WTO adjudicator has no business second-guessing. Overall, the approach to public morals in EC–Seal Products should allow the Appellate Body to be consistent in its respect for collective preferences, providing the possibility of justifying measures for objectives that are not explicitly stated in the other exceptions in Article XX.

However, there are questions about the respect for collective preferences that are raised by other aspects of the Appellate Body’s jurisprudence that have been far from fully answered. The SCM Agreement, a product of the overall neo-liberal orientation of the Uruguay Round, disciplines subsidies that have certain competition-distorting effects, but without an exceptions provision such as Article XX of the GATT.¹¹⁸ The disciplines are essentially indifferent to collective preferences and are a product – in some significant measure – of the anti-industrial policy/anti-‘picking winners’ bias of the thinking on regulation and its reform that dominated in the West in the 1980s and early 1990s, when it began to be seriously challenged. Some have suggested reading Article XX to the SCM Agreement on the grounds that it is a lex specialis to the GATT that elaborates but incorporates the basic approach in the GATT to policy space. In Canada–Renewable Energy, the Appellate Body was faced with an important challenge with legitimacy implications.¹¹⁹ All renewable energy markets have historically been premised on government support; when one does not take into account negative environmental externalities, the cost of generating renewable energy has been higher than in the case of fossil fuels. In a competitive marketplace where consumers only care about the lowest price for a given amount of electricity, no renewable energy would be generated.

Yet the SCM Agreement is indifferent to policy objectives, even those that are apparently imperative or vital as to mitigate climate change. In this context, the Appellate Body had to be extremely creative in finding a way of bringing in respect for collective preferences. It did so through the concept of a ‘benefit’, as interpreted in the jurisprudence of the SCM Agreement. The subsidies disciplines depend on a notion that a benefit has been provided, which is an advantage over the situation of a normal competitive market. The Appellate Body hypothesized that a market could itself be the product of collective preferences. In Ontario, the government had constructed a renewable energy market and structured it to achieve certain policy objectives. One had to take this framework as a given and ask not whether the price provided to renewable energy producers provided a benefit in relation to providers of fossil fuel energy but, rather, whether there were competition-distorting subsidies within the renewable energy market as created and structured by the Ontario government, based upon Ontario’s collective preferences. One may well ask whether this approach may also have implications for the Appellate Body’s choice of a purely competition-based approach to determining the existence of discrimination. If the government has created different markets for two products based upon collective preferences, is it proper to postulate a competitive relationship between the products as if there were a single market?

Many observers were surprised by the result in Canada–Renewable Energy. However, Antonia Eliason and I had suggested years before the complexity of determining a market benchmark in subsidies disputes related to support for renewable energy.¹²⁰ Consider, for example, the Appellate Body’s ruling in the US–Foreign Sales Corporations dispute.¹²¹ This dispute concerned the sensitive issue of how the USA approaches the tax liability of corporations. The USA typically taxes on the basis of nationality rather than residency, and the result can be that US corporations are faced with double taxation, where their operations abroad are such that they attract tax liability in other jurisdictions, which typically tax on the basis of residency. The US foreign sales corporations (FSC) scheme was intended to address in part this problem, by exempting certain US entities with particular structures from US tax liability when operating offshore. In effect, the FSC scheme could be viewed as allowing the USA to preserve its sovereign choice for a different approach to tax liability over other countries, without penalizing its own industries and traders. However, it is unquestionable that Congress had managed to build various tax ‘loopholes’ in the scheme that provided certain competitive advantages to particular kinds of US entities operating abroad, going beyond simply adjusting or compensating to prevent double taxation. The EU challenged the scheme as an illegal subsidy under the SCM Agreement.

The essential issue became whether the FSC scheme resulted in the USA foregoing revenue ‘otherwise due’. One view of this expression ‘otherwise due’ is that it indicates a comparison of the particular measure being challenged as a subsidy against the benchmark of a ‘normal’ tax system, one that achieves the revenue collection goal without altering competitive relationships, favouring particular industries and so forth – a Washington consensus view of how a tax system should operate. The Appellate Body, however, held that the SCM Agreement permitted a Member to adopt any tax system it wished. Therefore, the comparison in determining whether revenue was ‘otherwise due’ had to be internal to the Member’s chosen tax system – that is, the benchmark would be from some general or default taxation rule from which the Member was alleged to have deviated to provide a particular competitive advantage. Conveniently, the Appellate Body was able to find such a rule in the case of the USA. However, there is no necessity to structure a tax system such that differential treatment takes the form of a deviation from a default rule. To see how respect for collective preferences works, let us return to the example of renewable energy. A WTO Member can have a policy that taxes renewable energy industries much less than fossil fuels, which can be expressed simply as two tax rules, one for the renewables sector and a different one for the fossil fuels sector. There is no revenue foregone that is otherwise due because the rule for fossil fuels is simply a different rule for a different sector, not some kind of default or general taxation rule that sets the norm for what is ‘due’.

A different kind of question about respect for collective preferences is raised by the US–COOL case, where the Appellate Body considered under the TBT Agreement US regulations mandating that information be provided to consumers through labelling about the national origin of certain meat products. The Appellate Body avoided an inquiry into the policy objectives in providing consumers with the information in question. Consumers might view national origin as a surrogate for the quality or safety of meat (which could be rational or could be based on prejudice or misinformation or some of both). Or they might have preferences against certain countries. Could facilitating the latter preferences be intrinsically inconsistent with the concept of non-discrimination in WTO law?

A third question emerges from the way in which the Appellate Body has been operating its strict scrutiny of discrimination. As noted, the Appellate Body operates strict scrutiny of discrimination under the chapeau of Article XX by requiring a tight justification of any distinctions or exceptions in the way that a regulatory scheme is designed to be applied in practice against the stated objective. The EC–Seal Products judgment displays the difficulty the Appellate Body has had in imagining a situation where distinctions or exceptions reflect different objectives – each legitimate – that need to be traded off to some extent. It was difficult for the Appellate Body to see the indigenous people’s exception as a reflection of the strength of the collective preferences for the protection of the traditional way of life of indigenous peoples that could legitimately limit the fulfilment of collective preferences with respect to the protection against animal cruelty. The Appellate Body did not exclude the legitimacy of trading off these different goals through an exception but required a kind of harmonization, such that attaining the indigenous objective detracted to the minimum extent necessary from the ‘main’ or ‘principal’ objective of addressing animal cruelty. This seems to limit the extent to which the regulator can take into account the relative strength or intensity of different sets of collective preferences in determining how to make trade-offs within a given regulatory scheme.

D Selective Judicial Minimalism

Where a substantive norm is ambiguous, seemingly not coherent or based on a delicate but elusive historical compromise, the Appellate Body has favoured, selectively, judicial minimalism. Minimalism can consist of leaving open the meaning of the norm itself while emphasizing procedural or justificatory steps that Members must take in order to show they have had the norm somehow in their consideration. The Appellate Body may weaken the norm into a guidepost, allowing that a Member may satisfy the obligation if it takes an approach different to the Appellate Body’s guidance as to what is required – a limited deference to the Member’s own reading of how the obligation applies in certain situations. As already noted, in the Shrimp–Turtle case, there was a strongly proceduralist focus in the Appellate Body’s chapeau strict scrutiny. The Appellate Body faulted the USA for having negotiated with some WTO Members and not others about an agreement that would forestall a trade embargo, for rigid application of statutory criteria that did not take account of conditions in different countries and for the lack of reasons for decisions on individual importation applications. Since the general focus of the chapeau analysis, also discussed earlier, is how the measure is, or will be, actually applied or operationalized, there is a large role for proceduralist judicial minimalism in that analysis.¹²² Judicial minimalism as proceduralism, however, is not without its risks and has been subject to justified criticism.

Strictness with respect to procedures seems legitimate where the defending Member has a highly developed legal system, a regulatory democracy that is similar to the US or EU type of model.¹²³ In the case of less developed countries, proceduralism can result in obligation overload, which is perhaps why, as already noted, the Appellate Body has rejected a proceduralist turn in some instances and rejected, for example, an interpretation of the risk assessment requirement in the SPS Agreement that would have required that the actual WTO Member taking the measure itself conduct a scientific risk assessment. Overall, selective judicial minimalism is a dispute or controversy avoidance technique that may have positive legitimacy effects in a situation where there are deep divisions between the membership, even with respect to the WTO’s aims and future direction. At the limit, though, it can also produce a sense of legal incoherence that may erode trust in the system and its ability to provide clear guidance as to the meaning of legal rules. The question is whether the Appellate Body’s jurido-political intuitions and judgment are up to the requisite selectivity.

An excellent example of the strengths and perhaps also of the risks of selective judicial minimalism as a means of avoiding controversy concerning the meaning of legal norms is the GSP dispute.¹²⁴ One of the most divisive issues between developing and developed countries that pertains to what it means for the Doha Round to be a ‘development round’ was that of special and differential treatment of developing countries. One such kind of treatment has been the provision of tariff preferences to developing countries – that is, lower rates of tariff than the MFN rate for developing country exports to developed country WTO Members.¹²⁵ Such non-reciprocal preferences represented a partial victory for developing countries in the struggle over a new international economic order in the 1960s and 1970s. A framework was created to allow for an MFN exception for these preferences and to encourage individual developed countries to grant them, but the developing nations failed in their demand that these preferences were binding. By the time of the Doha Round, many of these preferences had their special value – or ‘preferentiality’ – eroded by the reduction of tariffs on many products on a MFN basis as well as by the proliferation of preferential trade agreements, including between developed countries that had eliminated tariffs to zero or close.

Furthermore, the preferences, having been granted voluntarily, were increasingly encumbered by conditions (ranging from anti-terrorism, to the protection of intellectual property rights, to human rights and environmental protection) as well as by other forms of unilateral decision making, such as whether a country or product had ‘graduated’ from the GSP – that is, had become competitive enough so as to no longer justify this form of special and differential treatment. The legal instrument under the GATT and the WTO that allowed a deviation from the MFN in order to operate the GSP – the Enabling Clause¹²⁶ – incorporated a number of criteria or desiderata from pre-existing GATT practices and documents, including that the preferences be operated in a non-discriminatory manner. However, the drafting of this instrument made it far from clear whether the criteria were of a legally binding nature or whether they reflected the ultimate aspiration for what would be achieved through a voluntary preference scheme. In March 2002, notably a few months after the launch of the Doha Development Round, India launched a dispute at the WTO challenging conditions related to labour rights, environmental performance and drug enforcement practices that had be incorporated in the EU’s GSP scheme. Meeting these conditions would allow a developing country to receive the highest margin of preferentiality for its imports into the EU. India claimed that the conditions violated the non-discrimination provision of the Enabling Clause, which India argued was ‘hard law’ and prohibited any distinctions between different developing countries. A successful ruling for India might have given India and other developing countries a boost to negotiate greater legal security for this kind of special and differential treatment in the Doha Round, and this was perhaps the gambit. Yet striking down conditionality on GSP preferences would have undermined a very strong understanding, especially among legislators in the EU and the USA, that GSP is a voluntarily conferred benefit to which strings can be attached. Being prohibited from attaching those strings might well have led to a loss of interest from legislators in supporting the GSP at all.

As Greg Shaffer and Yvonne Apea summarize, ‘the GSP case represents a lawyer’s paradise of ambiguous legal provisions interpreted by judicial bodies in a case having significant political and institutional implications’.¹²⁷ The judicial minimalism of the Appellate Body was enabled by India’s own decision to limit its challenge to the drug enforcement conditions, dropping the claims against labour and environmental conditionality. Given that the Appellate Body had recently affirmed in its second ruling in Shrimp–Turtle the legitimacy in principle of trade measures that were in response to other countries’ environmental policies and had apparently empowered in some measure some of the constituencies critical of neo-liberal trade liberalization, dropping environmental and labour conditions was an understandable choice by India to reduce the risk of its gambit. However, a feature of the drug enforcement conditionalities was that they were not accompanied by any specific criteria to determine a country’s entitlement to obtain the preferences. Listing a country was an act of essentially unfettered bureaucratic discretion, with no requirement to give reasons or an explanation. This feature made the drug preferences fundamentally different from those related to environmental and labour conditionalities, which were rather precisely defined.

The panel below held for India that any conditionality in the granting of GSP preferences was incompatible with non-discrimination in the Enabling Clause. The Appellate Body’s approach was quite different. First of all, it agreed with India that the non-discrimination requirement in the Enabling Clause was a binding hard law commitment. Not to have done so, arguably, would have sent a very negative message about legal security in special and differential treatment and would have made the atmosphere of the Doha Round even more tense in terms of divisions between developing and developed countries.

However, having said that non-discrimination applied, the Appellate Body rejected the panel’s view that any distinction would constitute discrimination. Instead, the Appellate Body emphasized the importance of transparency and due process to the non-discrimination norm and relied heavily on the lack of objective criteria in the drug preferences. Drawing on the language in the Enabling Clause, the Appellate Body did impose one substantive discipline: to be non-discriminatory, the conditionality had to make a positive contribution to the development needs of the country concerned. Yet it was very unclear to what extent the Appellate Body would engage in real scrutiny of the relation between the conditionality and the development need (as long as tariff preferences in principle could make a contribution and the development need was related to an ‘objective standard’ such as some multilateral treaty deploying a concept of development). In any event, the Appellate Body made it rather clear that it would not be inclined to make its own judgment about the meaning of ‘development’ or development needs.

According to Shaffer and Apea, many developing countries have been deeply disappointed by the Appellate Body ruling.¹²⁸ In effect, only through future dispute cases could they test how deferential the Appellate Body intended to be in determinations of preference-giving developed countries concerning the meaning of the key substantive norm of positive contribution to development needs. Shaffer and Apea themselves take the view that the Appellate Body was in fact placing a high hurdle in front of a preference-giving Member and regrets the unwillingness of developing country members to test this in further litigation. But by leaving the key substantive norm undertermined, the Appellate Body made any such effort inherently risky, for it could lead to a legal baseline unfavourable to efforts to secure greater legal security for preferences, which is the key objective of many developing countries.

Selective judicial minimalism has loomed large in another area of WTO law that has presented something of a political minefield – the constraints on unilateral trade measures that have been traditionally permitted under the multilateral trading order, including anti-dumping duties and countervailing duties in response to purportedly unfair trade practices of other Members and safeguard or emergency action in response to a crisis in a domestic industry due to a sudden increase in imports. Negotiations in the Uruguay Round to constrain these forms of unilateralism were intense and difficult – not only the USA but also the EU placed a considerable weight on being able to retain the ability to use these instruments, including in response (implicitly) to demands of protectionist domestic constituencies. Other countries saw these kinds of unilateralism, especially anti-dumping duties, as being among the most damaging forms of protectionism remaining, especially given the ease with which definitions of ‘dumping’ and other legal standards could be manipulated by ‘captured’ domestic agencies. Economists have generally seen the legal standards for imposition of unilateral fair trade remedies as having no sound economic basis (though they can be explained in positive political economic terms of the demand of domestic interest groups for protectionism). There is no good economic theory to determine if ‘dumping’ – selling in export markets at a higher price than in domestic markets – is ‘unfair’ or to determine whether a particular subsidy is ‘unfair’ such that it merits a unilateral response (the neo-liberal orientation towards multilateral discipline of subsidies was incorporated in the Uruguay Round’s SCM Agreement, as noted above, but the same neo-liberal outlook was sceptical of using subsidization as a pretext for a unilateral protectionist response).

With the substantive standards for disciplining unilateral trade remedies lacking coherence, and the product of a rather brutal power-based negotiation, it is not surprising that the Appellate Body has tended to resort to selective judicial minimalism in these cases. In the early cases on safeguards – US–Lamb Meat¹²⁹ and US–Steel Products¹³⁰ – the Appellate Body tackled the substantive norms that required that to apply safeguards the sudden increases in imports must be due to unforeseen developments and that the increases in imports must have a causal relation to the injury in the domestic industry. Giving an economics-based meaning to these norms is well nigh impossible, so the Appellate Body essentially faulted a lack of reasoning or inadequate consideration of the evidence or a failure to take into account all factors that might be contributing to the injury. It impugned the quality of the process by which the domestic agency imposed the safeguards while giving little guidance as to what the substantive norms that the agency was to apply actually do mean. At some point, the quantity and quality of reasons given for the decision, and evidence considered, would be sufficient, but to know at what point this was would require arguably some understanding of the substantive norm. The safeguards cases strained the limits of judicial minimalism, although one can understand the logic of the Appellate Body in resorting to it.¹³¹

Another instance of judicial minimalism is a countervailing duties case, US–Softwood Lumber.¹³² At issue was a long-standing dispute between the USA and Canada about the price at which the Canadian government sold timber from government lands. The US lumber industry argued that this price was lower than that which would result from a market mechanism to determine prices (for example, auctions of the kind often used in the USA). The industry’s claim was that the ‘artificially’ lower prices for access to the trees were in fact a subsidy to Canada’s lumber industry, lowering the prices of inputs. The WTO SCM Agreement provides that the benchmark for determining whether the price at which the government sells a good or service is below market and, therefore, potentially a subsidy is the market condition in the defending Member – in this case, Canada. However, the USA argued in the WTO dispute that since the government, in fact, dominated sales of timber rights in Canada, even private sales would be greatly influenced by the government’s administered prices. Thus, there was no genuine market benchmark in the Canada, and the US agency should therefore be able to use auction prices in the USA as the benchmark. The Appellate Body agreed with the USA that the private Canadian prices were inadequate as a market benchmark, but the alternative benchmark proposed by the USA was one that was not contemplated by the SCM Agreement. So the Appellate Body simply said to the USA that it could use another benchmark, if it could justify that benchmark, without indicating what would or would not be acceptable. In effect, the Appellate Body softened the treaty benchmark into a form of indicative guidance or a presumptive standard, from which the Member’s agency could deviate where adequately justified.

In the more recent ‘zeroing’ cases, where the issue is whether the amount of an anti-dumping duty should be lower to give the dumping firm a ‘credit’ for negative dumping – that is, for those sales where its price is higher in the export market than in the domestic market – the Appellate Body has strayed from judicial minimalism with rather ominous results.¹³³ ‘Zeroing’, which is the failure to count in negative dumping transactions in determining the anti-dumping duty (so as to lower it, and it is so named because these transactions are simply given zeros) is neither explicitly permitted nor prohibited under the Anti-Dumping Agreement. The Appellate Body began with something like a judicial minimalist approach, avoiding a hard per se rule against zeroing in all circumstances and addressing the practice in a case-by-case, contextual manner. Eventually, pressed by further complaints, the Appellate Body found that the practice of ‘zeroing’ was incompatible with the norms of the Anti-Dumping Agreement, including the requirement to make a fair comparison of prices and to determine duties based upon comparing aggregate domestic and export transactions. Relentless pressure was being applied to the Appellate Body from the US trade representative (USTR) to give zeroing a green light. While the USA did implement gradually the Appellate Body’s rulings in some fashion, it sent a note of criticism to the Appellate Body itself in two cases. In other cases, panels of first instance refused to follow the Appellate Body’s approach to the zeroing issue (at the time, the head of the legal secretariat serving the panels of first instance was an American, Bruce Wilson, with a Washington, DC, insider background). The departure of US Appellate Body Member Merit Janow from the Appellate Body without seeking renewal for a second term and the USA’s failure to allow her successor Jennifer Hillman to serve a second term may well be related to the zeroing controversy.¹³⁴ Finally, the USTR apparently resolved to block any appointment of a new Appellate Body Member who is likely to be independent of trade insider circles, especially any academic.

The politicization of the appointments process, observed by political scientist Mark Pollack in an important recent article, could be seen as an inevitable outcome of the Appellate Body’s assertions of independence and authority in a treaty community that likes to see itself as ‘membership driven’. However, when the USA blocked the appointment of James Gathii as an Appellate Body Member, most of the community at first stood up to US pressure, resulting in the failure of the first appointment process, a deadlock where the Appellate Body was not fully staffed for a period of time, while Kenya finally had little choice but to withdraw Gathii’s candidacy. Significantly, the individual ultimately chosen was a consummate WTO insider. While the narrative of this article has emphasized how generally the consensus rule has protected the Appellate Body against effective political interference or pressure, the USTR under the Obama administration found a way of using the consensus practice to produce politicization by holding out until a candidate who was seen to be amenable to deciding cases with sensitivity to Member views was finally chosen over one who had the appearance of an independently minded jurist.

Why could the Appellate Body have not backed off to a judicial minimalist stance over zeroing, simply stating that where an agency zeroes it has to ensure, or perhaps give reasons to show, that the practice is ‘fair’ in all things considered? As we have elaborated, these kinds of minimalist moves have saved the Appellate Body from becoming immersed in political controversy over other heavily negotiated, but unprincipled, trade-offs in the constraints on unilateral trade remedies. The aggressiveness with which the USA applied pressure and the relative openness (to show protectionist domestic constituencies that it was playing tough) probably forced the Appellate Body to stand completely firm. Even if jurisprudentially defensible (after all, as noted, ‘zeroing’ was not explicitly banned and different Members had different views in the negotiations on its compatibility with the overall normative approach of the Anti-Dumping Agreement), softening its approach would have made the Appellate Body appear to have been subject to pressure, not from the ‘institution’ or the membership in general but, rather, from one Member, the USA. It would have undermined the consideration that, ultimately, probably allowed many other Members to put up with decisions that were not only independent but also anomalous from a trade insider perspective. Namely, that, overall, having a judicial organ that can counter to some extent the power-based nature of the WTO system, with the USA as its most powerful Member, is worth it. Fortunately, one may question whether the kind of pressure applied in the zeroing controversy is much more than a reflection of the style of senior members of the current USTR. More is to be said about this in the conclusion.

E The Autonomy of the WTO Judicial System from Other Trade Fora

As noted above, in its ‘declaration of independence’, the Appellate Body pronounced that it would interpret the WTO treaties in light of non-WTO public international law, where relevant, and even in preference to the accumulated collective wisdom of the GATT/WTO ‘institution’. In contrast to this openness and resistance to ‘fragmentation’, the Appellate Body has declined to engage in what Teitel and I call ‘cross-judging’. It has not engaged in dialogue, much less taken a posture of comity or complementarity, in relation to other fora for the settlement of trade disputes. Instead, the Appellate Body has tried to maintain ‘clinical isolation’ from these other trade fora, to use its own words, for what it is not doing in relation to international law in general. Article 23 of the DSU stipulates that the WTO dispute settlement system shall be the exclusive forum for determining violations of the WTO agreements. Article 23 is an important element of what was discussed above and an important, if not crucial, element of the Uruguay Round. The USA, in particular, has accepted constraints on aggressive unilateralism, which includes the option of self-help through trade sanctions or economic threats where another Member was determined by US authorities to have violated its WTO obligations, pursuant to section 301-type trade law.

However, Article 23 also implies that WTO Members cannot simply exit the WTO dispute settlement procedure by migrating their dispute to a non-WTO forum. With the self-constructed narrative of WTO political failure, the spectre of the WTO simply becoming irrelevant as Members have more success in moving forward with regional negotiations has been often raised. The final acknowledgement of the failure of the Doha Round was preceded by the apparent success of the Trans-Pacific Partnership negotiations. In cases where aspects of disputes spill over between a regional trade forum and the WTO dispute settlement system, the Appellate Body has made it clear that the WTO dispute settlement organs are completely autonomous and have no obligation to facilitate the role of regional fora. The message might be said to be that if there is a dispute that relates to WTO legal rules, better to bring it to the WTO in the first place, as the exclusive forum that deals with these rules.

As discussed above, in the Brazil–Retreaded Tyres case, the Appellate Body took a deferential view of Brazil’s overall scheme of banning imports of retreaded tyres, considering it to be a rational choice in furthering environmental and health objectives. However, when the Appellate Body moved to scrutiny under the chapeau of Article XX, things took a different turn. Brazil had also been sued in the MERCOSUR regional forum for its tyre scheme and had lost in that forum, possibly because it did not make the same kind of defence that it had done in the WTO. As a consequence, Brazil had been obligated under MERCOSUR’s dispute settlement system not to apply the import ban to MERCOSUR member states. The Appellate Body held that this was unjustified discrimination because it was unrelated to the environmental and health objective of the main measure and, therefore, that Brazil’s non-application of its measure to the MERCOSUR members failed the chapeau. One might think that this whole issue ought to have been resolved under Article XXIV of the GATT, which, as discussed above, concerns an exemption from the MFN obligation where it is necessary to form, or maintain, a customs union or free trade area.

Yet as we saw in the Turkey–Textiles case, the Appellate Body took a strict or narrow view of the exception, holding that it should be subject to strict judicial scrutiny, not merely diplomatic oversight, in the relevant WTO committee. Formalistically speaking, Brazil could comply with both its MERCOSUR and its WTO obligations by withdrawing its measure as opposed to simply not applying it to non-MERCOSUR members. Thus, arguably, based on the Appellate Body’s strict approach, Article XXIV would not be available. But the Appellate Body showed no interest in facilitating Brazil’s compliance with both sets of obligations and with the maintenance of a domestic regulatory scheme that had already passed rationality review. Had the Appellate Body taken a positive view of the co-existence of regional fora with the WTO dispute settlement system, it could easily have found that Brazil’s discrimination in favour of MERCOSUR members was not ‘unjustifiable’ – there is no definition of ‘unjustifiable’ in Article XX and no textual basis for confining acceptable justifications to those that pertain to the main purposes of the general regulatory scheme itself.

In the Mexico–Soft Drinks case,¹³⁵ Mexico sought justification under Article XX(d) of the GATT, which refers to measures necessary to enforce laws, regulations and requirements, for sanctions against the USA to block the formation of a panel under the North American Free Trade Agreement (NAFTA) to resolve a claim that Mexico had against the USA under NAFTA.¹³⁶ The Appellate Body held that ‘laws, regulations, and requirements’ referred exclusively to domestic law and could not include international agreements such as NAFTA. Since many international norms are implemented in domestic law and in some legal systems with automaticity, the Appellate Body itself had to acknowledge that this distinction could not easily apply if international norms were part of a domestic legal system. However, the clear message was that the WTO dispute settlement system was not available to address gaps or ineffectiveness in regional dispute arrangements. The Appellate Body opined:

Mexico’s interpretation would imply that, in order to resolve the case, WTO panels and the Appellate Body would have to assume that there is a violation of the relevant international agreement (such as the NAFTA) by the complaining party, or they would have to assess whether the relevant international agreement has been violated. WTO panels and the Appellate Body would thus become adjudicators of non-WTO disputes. … This is not the function of panels and the Appellate Body as intended by the DSU.¹³⁷

In other cases, the Appellate Body has held that it could make determinations about the legal requirements of other legal systems and whether they are met, where needed, in order to apply a WTO legal rule (in the EC–Bananas dispute,¹³⁸ the Lomé Convention in India–Patents¹³⁹ and India’s domestic administrative and constitutional law). In Mexico–Soft Drinks, the legal rule was Article XX(d) and the issue of the US ‘compliance’ with NAFTA was a question simply subordinate to whether Mexico’s measures were required to secure compliance. So the Appellate Body, following earlier case law, could simply have said that the operative provision being applied was Article XX(d) of the GATT and finding with respect to NAFTA compliance was merely ancillary to determining whether Mexico’s measures were consistent with the WTO-covered agreements. So it is not hard to see that there is an implicit judicial policy of ‘clinical isolation’ from regionalism. There may also be a concern that regional fora must respect the exclusivity of WTO dispute settlement to determine WTO violations, as required by Article 23 of the DSU – that is, to favour ‘clinical isolation’ from the other side, as it were, and offer a reciprocal standoffishness.

Finally, in Peru–Agricultural Products,¹⁴⁰ Peru argued that Guatemala should be barred from making a claim against it for WTO violations where the measures in question had been clearly agreed to be permissible under a FTA that had been concluded between Peru and Guatemala but not yet entered into force. Did the provision in the FTA constitute, at a minimum, some kind of representation that Guatemala would not pursue a claim against Peru for maintaining this sort of measure? The Appellate Body held that the provisions of the VCLT on inter se agreements (Article 41) did not apply to allow a sub-set of WTO Members to agree to permit among themselves otherwise WTO-inconsistent measures. The WTO’s own rules that related to contracting out of, suspending or limiting WTO obligations, including rules on waivers and Article XXIV of the GATT, trumped the VCLT. In effect, the WTO agreements constituted an exhaustive lex specialis concerning deviations from WTO obligations applicable among some, but not all, WTO Members. Again, this ruling has to be understood hand in hand with the strict or narrow view of the exception for FTAs and customs unions in GATT Article XXIV.

In sum, common to all of these cases is the reinforcement of the autonomy, if not a certain kind of supremacy or at least primacy, of the WTO dispute settlement system, at a time when regional and bilateral agreements and negotiations proliferate, partially in response to the supposed blockage or failure of the political and diplomatic processes of the WTO. On balance, the Appellate Body makes exit to regional dispute settlement harder and certainly shows no interest in treaty interpretations that could accommodate or facilitate harmonious co-existence with regional regimes as opposed to two or multiple solitudes.

7 Conclusion

While the future of the WTO as an institution is in question or at least in flux, with no consensus now about what it is able to do or what it should do in the future, the WTO’s Appellate Body is a formidable engine of global economic governance, probably the most active and productive of all international courts not only in the number and range of its decisions but also in the number of disputes that its jurisprudential guidance has helped to settle, often out of the courtroom. The Appellate Body operates under many constraints: it is required to decide appeals in a 60-to-90-day time period; it must ultimately rely on a factual record and findings of first instance from panels that are not professional adjudicators and are resource-constrained; it cannot award damages; it must take every appeal brought to it. On the other hand, what is appealed and the scope of the appeal is entirely in the hands of states who are the parties to the disputes. In the presence of backlash against the Uruguay Round result that created the WTO and more generally the intense contestation of neo-liberal globalization or ‘deep integration’, the Appellate Body sought to discern in the corpus of WTO treaties an equilibrium between domestic regulatory autonomy and trade liberalization very much inspired by, or anchored in, the original GATT – a respect for regulatory diversity and flexibility towards domestic policy interventions that characterized the GATT in the period when it enjoyed the greatest legitimacy or acceptance (post-war embedded liberalism).

The Appellate Body has attempted to make sense of, and in a way soften or blunt, the Uruguay Round treaties that tend to veer in a ‘deep integration’ direction, using the kind of non- discrimination prism that was central to the GATT founders’ view of the dividing line between legitimate domestic policies and those that represent or threaten cheating on bargained limitations on border measures such as tariffs. In general, the Appellate Body has looked not primarily to the intuitions of the GATT/WTO insider community and its traditions, understanding that there are legitimate public policies acceptable within the treaty framework, but, rather, to the outside. Thus, it has been open to curbs on trade that facilitate objectives such as the protection of animal welfare and climate change mitigation, which have simply not been brought into the mandate of the WTO as an institution. The equilibrium between domestic regulatory autonomy and trade liberalization discerned by the Appellate Body is very much a construction – one that is normatively stabilizing at a time when there are few agreed answers about the costs and benefits of globalization or the ideal shape of global economic governance in relationship to differing domestic policy paths. For the contestants in these debates on either side, this normative stabilization cannot but seem to have an element of the arbitrary and artificial to it. Yet it may well have contributed to a sense that, while the WTO appears to be stalled in its negotiating functions during this period, there has been some basic durability to the given legal framework and its enforceability, helping to resist a major reversion to beggar-thy-neighbour during the financial and economic crisis of 2007–2011. The judicial policies that the Appellate Body has deployed to navigate along the equilibrium it has constructed deserve to be articulated explicitly and debated. As I have suggested in this article, they have sometimes been deployed deftly; in other cases, less so or counter-productively, from the point of view of legitimacy.

The concerns raised by Manfred Elsig and Pollack, among others, concerning the recent apparent politicization in the Appellate Body are a reflection, more than anything else, of its power. Constituencies try to influence who is on the US Supreme Court, for example, for the same reason. I believe the Appellate Body will be able to withstand such efforts, which probably will backfire in the case of tactics as heavy handed as those used by the Obama administration US Trade Representative (USTR) in the last few years. There are few exits from the WTO dispute settlement system, and blocking or threatening to block appointments until the Appellate Body is stacked with those who will do the bidding of the USTR risks, in the end, jeopardizing the stature and effective functioning of a system that the USA itself needs for dispute settlement, given its legal commitment to not revert to unilateralism and given the relatively underdeveloped regional fora. Recently, US Appellate Body Member Thomas Graham was re-appointed without difficulty. It is notable also that in the Trans-Pacific Partnership, the USA has accepted that the dispute settlement institutions for that mega-regional organization must take into account the jurisprudence of the WTO. Larger than the current life of the WTO ‘institution’, the Appellate Body, as this move suggests, may well have come of age as a true court of world trade.

1

R. Teitel, Humanity’s Law (2011). Teitel shows how the International Criminal Tribunal for the Former Yugoslavia adopted a broad teleological interpretation of its powers, based upon the compelling nature of the aims and values that it was purportedly designed to serve and the need progressively to realize these. See also Weiler, ‘The Transformation of Europe’, 100 Yale Law Journal (YLJ) (1991) 2403, on the relationship of the European court as a central actor in the transformation of Europe to the political process.

2

See D. Rodrik, Has Globalization Gone Too Far? (1997).

3

See S.A. Aaronson, Taking to the Streets: The Lost History of Public Efforts to Shape Globalization (2001), ch. 6, at 7.

4

Howse and Teitel, ‘Beyond Compliance: Rethinking Why International Law Really Matters’, 1 Global Policy (2010).

5

A synoptic treatment that remains one of the most insightful is Helfer and Slaughter, ‘Toward a Theory of Effective Supranational Adjudication’, 107 YLJ (1997) 273.

6

P.C. Mavroidis, Dispute Settlement in the WTO: Mind over Matter (2016) (manuscript on file at European University Institute (EUI), Florence). Mavroidis notes that 500 disputes had been submitted to the World Trade Organization (WTO) dispute settlement system between the creation of the WTO and November 2015.

7

C. Barfield, Free Trade, Democracy, Sovereignty (2001), most notably.

8

See C. Bown and T.J. Prusa, U.S. Antidumping: Much Ado about Zeroing, Working Paper, World Bank Policy Research (2010).

9

Members of the Appellate Body are appointed for a four-year term that is renewable once. They are expected to spend half of their time on Appellate Body business and are compensated with a salary on that basis. They may participate in other professional activities as long as they make themselves available when necessary to decide appeals and there is no conflict of interest that arises from the other activities. Candidates are put forward by their home countries and considered by a Selection Committee of WTO insiders, the director-general and member state delegates with senior positions in the various political and diplomatic councils of the WTO. However, the ultimate decision about appointments is by the consensus of the membership.

10

Agreement on Trade-Related Aspects of Intellectual Property Rights 1994, 1869 UNTS 299; Agreement on Technical Barriers to Trade (TBT Agreement) 1994, 1868 UNTS 120.

11

For criticisms along these lines, see F. Roessler, Changes in the Jurisprudence of the WTO Appellate Body during the Past Twenty Years, Working Paper no. RSCAS 2015/72 (2015) (on file at Robert Schuman Centre for Advanced Studies, Global Governance Programme, EUI).

12

Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS Agreement) 1994, 1869 UNTS 299.

13

General Agreement on Tariffs and Trade (GATT) 1994, 55 UNTS 194. The following borrows freely from earlier work, especially Howse, ‘From Politics to Technocracy and Back Again: The Fate of the Multilateral Trading System’, 96 American Journal of International Law (AJIL) (2002) 94; Howse and Nicolaidis, ‘Enhancing WTO Legitimacy: Constitutionalization or Global Subsidiarity?’, 16 Governance (2003) 73. See also A. Lang, World Trade Law after Neoliberalism: Reimagining the Global Economic Order (2011).

14

The main provisions of Article III read as follows:

Article III: National Treatment on Internal Taxation and Regulation

1. The contracting parties recognize that internal taxes and other internal charges, and laws, regulations and requirements affecting the internal sale, offering for sale, purchase, transportation, distribution or use of products, and internal quantitative regulations requiring the mixture, processing or use of products in specified amounts or proportions, should not be applied to imported or domestic products so as to afford protection to domestic production.

2. The products of the territory of any contracting party imported into the territory of any other contracting party shall not be subject, directly or indirectly, to internal taxes or other internal charges of any kind in excess of those applied, directly or indirectly, to like domestic products. Moreover, no contracting party shall otherwise apply internal taxes or other internal charges to imported or domestic products in a manner contrary to the principles set forth in paragraph 1 …

4. The products of the territory of any contracting party imported into the territory of any other contracting party shall be accorded treatment no less favourable than that accorded to like products of national origin in respect of all laws, regulations and requirements affecting their internal sale, offering for sale, purchase, transportation, distribution or use. The provisions of this paragraph shall not prevent the application of differential internal transportation charges which are based exclusively on the economic operation of the means of transport and not on the nationality of the product.

15

Weiler, ‘The Rule of Lawyers and the Ethos of Diplomats: Reflection on WTO Dispute Settlement’, in R. Porter et al. (eds), Efficiency, Equity, Legitimacy: The Multilateral Trading System at the Millennium (2001) 334, at 334–336.

16

D. Rodrik, The Globalization Paradox (2011), at 77.

17

Silvia Ostry, an important Uruguay Round negotiator and Canadian trade official, bluntly describes the Uruguay Round bargain in the following terms: ‘The essence of the South side of the deal – the inclusion of the new issues and the creation of the new institution – was to transform the multilateral trading system. … the most significant feature … was the shift in policy focus from the border barriers of the GATT to domestic regulatory and legal systems – the institutional infrastructure of the economy. The barriers to access for service providers stem from laws, regulations, administrative actions which impede cross-border trade and factor flows. … In the case of intellectual property the move to positive regulation is more dramatic since the negotiations covered not only standards for domestic laws but also detailed provisions for enforcement procedures to enforce individual (corporation) property rights. … And, lest we forget, all this in return for minimal liberalization in agriculture and textiles and clothing.’ S. Ostry, The Uruguay Round North-South ‘Grand Bargain’: Implications for Future Trade Negotiations (2000), available at http://sites.utoronto.ca/cis/ostry/docs_pdf/Minnesota.pdf (last visited 22 February 2016).

18

This is very cogently discussed in Mavroidis, supra note 6.

19

Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade (Anti-Dumping Agreement) 1994, 1868 UNTS 201.

20

Steinberg, ‘In the Shadow of Law or Power? Consensus-Based Bargaining and Outcomes in the GATT/WTO’, 56 International Organization (2002) 339.

21

Understanding on Rules and Procedures Governing the Settlement of Disputes 1994, 1869 UNTS 401.

22

Sykes, ‘The Remedy for Breach of Obligations under the WTO Dispute Settlement Understanding: Damages or Specific Performance?’, in M. Bronckers and R. Quick (eds), New Directions in International Economic Law (2000) 351.

23

The European Union (EU) made a proposal in the stalled post-Uruguay Round dispute settlement negotiations for the professionalization of the panels, but it was not well received by other WTO members, including the USA. WTO, Dispute Settlement Body Special Session Document, Contribution of the European Communities and Its Member States to the Improvement of the WTO Dispute Settlement Understanding, Doc. TN/DS/W/1, 12 March 2002.

24

WTO, United States – Measures Concerning the Importation, Marketing and Sale of Tuna and Tuna Products – Report of the Appellate Body (Tuna–Dolphin), 13 June 2012, WT/DS381/AB/R.

25

E.g., Ostry, supra note 17.

26

See Eliason, ‘The Trade Facilitation Agreement: A New Hope for the World Trade Organization’, 14 World Trade Review (2015) 643. Agreement on Trade Facilitation, WTO Doc. WT/L/931 (2014).

27

Waiver Concerning Kimberley Process Certification Scheme for Rough Diamonds, Doc. G/C/W/432/ Rev. 1, 24 February 2003.

28

F. Bergsten, Toward a New International Economic Order (1979); J. Bhagwati, Protectionism (1988).

29

D. Rodrik, Trade Policy as Riding Bicycles (2007), available at http://rodrik.typepad.com/dani_rodriks_weblog/2007/07/trade-policy-as.html (last visited 22 February 2016).

30

See, e.g., R. Sally, New Frontiers in World Trade: Globalization’s Future and Asia’s Rising Role (2008).

31

On the ethos required to reorient the WTO to such a process, see Howse and Nicolaidis, ‘Towards a Global Ethics of Trade Governance’, Law and Contemporary Problems (2016) (forthcoming). Mavroidis has rightly noted: ‘[T]he WTO has a very important mandate anyway which is independent of the success/failure of rounds: discussions in the various committees manage to produce better communication across trading nations, and resolve many disputes as well.’ P. Mavroidis, Right Back to Where We Started from (or Are We?) (2011) (unpublished manuscript on file at EUI, Florence)

32

The view that the negotiators considered the Appellate Body as a kind of ‘afterthought’, a kind of safety valve against a rare panel ruling that was seriously anomalous in terms of the general institutional understanding of the legal norms at issue, is well established in the literature. See, e.g., van den Bossche, ‘From Afterthought to Centrepiece: The Appellate Body and Its Rise to Prominence in the World Trading System’, in G. Sacerdoti, A. Yanovich and J. Bohanes (eds), The WTO at Ten: The Contribution of the Dispute Settlement System (2006) 289.

33

Bacchus, ‘Not in Clinical Isolation’, in G. Marceau (ed.), A History of Law and Lawyers in the GATT/WTO: The Development of the Rule of Law in the Multilateral Trading System (2015) 507, at 515–516.

34

American Society of International Law, ‘WTO Appellate Body Roundtable’, in Proceedings (2005), at 182 (emphasis added).

35

Unterhalter, ‘The Authority of an Institution: The Appellate Body under Review’, in Marceau, supra note 32, 466, at 469.

36

See the seminal essay of Weiler, ‘The Rule of Lawyers and the Ethos of Diplomats: Reflection on WTO Dispute Settlement’, in Porter et al., supra note 15, 334.

37

Abi-Saab, ‘The Appellate Body and Treaty Interpretation’, in G. Sacerdoti, A. Yanovich and J. Bohanes (eds), The WTO at Ten: The Contribution of the Dispute Settlement System (2006) 453.

38

Vienna Convention on the Law of Treaties (VCLT) 1969, 1155 UNTS 331.

39

Weiler, supra note 36, at 146.

40

C.-D. Ehlermann, ‘Revisiting the Appellate Body: The First Six Years’, in Marceau, supra note 32, 487.

41

D. Wilson and L. Starchuck, Judicial Activism in the WTO: Implications for the Doha Negotiations, September 2003, available at http://jpkc.zzu.edu.cn/esjmyzzf/ebook/lw/Judicial_Activism_in_the_WTO_-_Implications_for_the_DOHA_Negoti.pdf (last visited 22 February 2016).

42

Ibid., at 11.

43

Agreement Establishing the World Trade Agreement (WTO Agreement) 1994, 1867 UNTS 154.

44

American Society of International Law, supra note 34, at 181.

45

Ibid., at 177.

46

Pauwelyn, ‘The Transformation of World Trade’, 104 Michigan Law Review (2005) 1.

47

WTO, European Communities – Regime for the Importation, Sale and Distribution of Bananas – Report of the Appellate Body (EC–Bananas), 25 September 1997, WT/DS27/AB/R.

48

WTO, EC Measures Concerning Meat and Meat Products (Hormones) – Report of the Appellate Body (EC–Hormones), 13 February 1998, WT/DS26/AB/R, WT/DS48/AB/R.

49

See Sykes, ‘The Remedy for Breach of Obligations under the WTO Dispute Settlement Understanding: Damages or Specific Performance?’ in M. Bronckers and T. Cottier (eds), New Directions in International Economic Law (2000) 347; EC–Bananas, supra note 47; EC–Hormones, supra note 48.

50

Mavroidis, supra note 6, at 4.

51

VCLT, supra note 38.

52

WTO, Japan – Taxes on Alcoholic Beverages – Report of the Appellate Body, 1 November 1996, WT/DS8/AB/R, WT/DS10/AB/R, WT/DS11/AB/R.

53

WTO, European Communities – Customs Classification of Certain Computer Equipment – Report of the Appellate Body, 22 June 1998, WT/DS62/AB/R, WT/DS67/AB/R, WT/DS68/AB/R.

54

WTO, Japan – Customs Duties, Taxes and Labelling Practices on Imported Wines and Alcoholic Beverages – Report of the GATT Panel, L/6216, 10 November 1987, BISD 34S/83, at 13.

55

Abi-Saab, supra note 37, at 460.

56

Ibid., at 15.

57

WTO, India – Patent Protection for Pharmaceutical and Agricultural Chemical Products – Report of the Appellate Body, 16 January 1998, WT/DS50/AB/R.

58

TRIPs Agreement, supra note 12.

59

EC–Hormones, supra note 48, at 12.

60

WTO, India – Quantitative Restrictions on Imports of Agricultural, Textile and Industrial Products – Report of the Appellate Body, 22 September 1999, WT/DS90/AB/R.

61

Understanding on the Balance-of-Payments Provisions of the General Agreement on Tariffs and Trade 1994, (WTO, 1995).

62

Ibid.

63

WTO, Turkey – Restrictions on Imports of Textile and Clothing Products – Report of the Appellate Body (Turkey–Textiles), 19 November 1999, WT/DS34/AB/R.

64

P. Mavroidis, The Regulation of International Trade (2016) at 292.

65

GATT, supra note 13, Art. XXIV.

66

Ibid.

67

Bhagwati, supra note 28.

68

Turkey–Textiles, supra note 63, at 22

69

WTO, India – Quantitative Restrictions on Imports of Agricultural, Textile and Industrial Products – Report of the Appellate Body (India–Balance of Payments), 22 September 1999, WT/DS90/AB/R.

70

Ibid., para. 60.

71

Roessler, ‘Are the Judicial Organs of the World Trade Organization Overburdened?’, in Porter et al., supra note 15, 13.

72

Ehlermann and Ehring, ‘The Authoritative Interpretation under Article IX: 2 of the Agreement Establishing the World Trade Organization: Current Law, Practice and Possible Improvements’, 8 Journal of International Economics and Law (2005) 803, at 813.

73

WTO, United States – Import Prohibition of Certain Shrimp and Shrimp Products – Report of the Appellate Body, 6 November 1998, WT/DS58/AB/R.

74

WTO, United States – Restrictions on Imports of Tuna – Report of the GATT Panel, 3 September 1991, DS21/R, DS21/R (unadopted); WTO, United States – Restrictions on Imports of Tuna – Report of the GATT Panel, 16 June 1994, DS29/R (unadopted).

75

GATT, supra note 13, Art. XX. The text of Article XX reads as follows: ‘Subject to the requirement that such measures are not applied in a manner which would constitute a means of arbitrary or unjustifiable discrimination between countries where the same conditions prevail, or a disguised restriction on international trade, nothing in this Agreement shall be construed to prevent the adoption or enforcement by any contracting party of measures:

(a) necessary to protect public morals;

(b) necessary to protect human, animal or plant life or health;

(c) relating to the importations or exportations of gold or silver;

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

(e) relating to the products of prison labour;

(f) imposed for the protection of national treasures of artistic, historic or archaeological value;

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

(h) undertaken in pursuance of obligations under any intergovernmental commodity agreement which conforms to criteria submitted to the CONTRACTING PARTIES and not disapproved by them or which is itself so submitted and not so disapproved;

(i) involving restrictions on exports of domestic materials necessary to ensure essential quantities of such materials to a domestic processing industry during periods when the domestic price of such materials is held below the world price as part of a governmental stabilization plan; Provided that such restrictions shall not operate to increase the exports of or the protection afforded to such domestic industry, and shall not depart from the provisions of this Agreement relating to non-discrimination;

(j) essential to the acquisition or distribution of products in general or local short supply;

Provided that any such measures shall be consistent with the principle that all contracting parties are entitled to an equitable share of the international supply of such products, and that any such measures, which are inconsistent with the other provisions of the Agreement shall be discontinued as soon as the conditions giving rise to them have ceased to exist. The CONTRACTING PARTIES shall review the need for this sub-paragraph not later than 30 June 1960.’

76

WTO, United States – Countervailing Duties on Certain Corrosion-Resistant Carbon Steel Flat Products from Germany – Report of the Appellate Body, 19 December 2002, WT/DS213/AB/R, para 39.

77

WTO, European Communities – Measures Affecting Asbestos and Asbestos-Containing Products – Report of the Appellate Body, 5 April 2001, WT/DS135/AB/R.

78

Charnovitz, ‘Judicial Independence in the World Trade Organization’, in L. Boisson de Chazournes, C.P.R. Romano and R. Mackenzie (eds), International Organizations and International Dispute Settlement (2002) 219.

79

WTO, European Communities – Trade Description of Sardines – Report of the Appellate Body (EC–Sardines), 23 October 2002, WT/DS231/AB/R.

80

WTO, United States – Continued Suspension of Obligations in the EC – Hormones Dispute – Report of the Appellate Body, 16 October 2008, WT/DS320/AB/R.

81

See the excellent analysis of Grainger-Jones and Primo Braga, ‘The Multilateral Trading System: Mid-Flight Turbulence or Systems Failure?’ in R. Newfarmer (ed.), Trade, Doha, and Development: A Window into the Issues (2006) 27.

82

Abi-Saab, supra note 37.

83

See, e.g., WTO, European Communities – Customs Classification of Frozen Boneless Chicken Cuts – Report of the Appellate Body, 27 September 2005, WT/DS269/AB/R, WT/DS286/AB/R.

84

WTO, China – Measures Affecting Trading Rights and Distribution Services for Certain Publications and Audiovisual Entertainment Products – Report of the Appellate Body (China–Publications), 19 January 2010, WT/DS363/AB/R.

85

Agreement, supra note 10.

86

WTO, European Communities – Measures Prohibiting the Importation and Marketing of Seal Products – Report of the Appellate Body (EC–Seal Products), 18 June 2014, WT/DS400/AB/R / WT/DS401/AB/R.

87

WTO, Thailand – Customs and Fiscal Measures on Cigarettes from the Philippines – Report of the Appellate Body, 15 July 2011, WT/DS371/AB/R.

88

Ibid., 16.

89

WTO, United States – Standards for Reformulated and Conventional Gasoline – Report of the Appellate Body, 20 May 1996, WT/DS2/AB/R.

90

WTO, Korea – Measures Affecting Imports of Fresh, Chilled and Frozen Beef – Report of the Appellate Body, 10 January 2001, WT/DS161/AB/R, WT/DS169/AB/R.

91

WTO, United States – Measures Affecting the Cross-Border Supply of Gambling and Betting Services, Report of the Appellate Body (US–Gambling), 20 April 2005, WT/DS285/AB/R. General Agreement on Trade in Services 1994, 1869 UNTS 183.

92

WTO, Brazil – Measures Affecting Imports of Retreaded Tyres – Report of the Appellate Body (Brazil–Retreaded Tyres), 17 December 2007, WT/DS332/AB/R.

93

EC–Seal Products, supra note 86.

94

Brazil–Retreaded Tyres, supra note 92, para. 156.

95

Ibid., para. 5.213.

96

Ibid., at para. 5.261 (footnote omitted).

97

Ibid., at para. 5.277.

98

WTO, United States – Measures Affecting the Production and Sale of Clove Cigarettes – Report of the Appellate Body (US–Clove Cigarettes), 24 April 2012, WT/DS406/AB/R.

99

WTO, United States – Certain Country of Origin Labelling (COOL) Requirements – Report of the Appellate Body, 23 July 2012, WT/DS384/AB/R / WT/DS386/AB/R.

100

WTO Agreement on the Application of Sanitary and Phytosanitary Measures 1994, 1867 UNTS 493.

101

Hudec, ‘Science and “Post-Discriminatory” WTO Law’, 26 Boston College International and Comparative Law Review (2003) 185.

102

US–Clove Cigarettes, supra note 98, para. 96.

103

Ibid., para. 177.

104

EC–Sardines, supra note 79.

105

EC–Hormones, supra note 48, para. 178.

106

Ibid., para. 124.

107

Ibid., para. 213.

108

WTO, Australia – Measures Affecting Importation of Salmon – Report of the Appellate Body, 6 November 1998, WT/DS18/AB/R; WTO, Japan – Measures Affecting the Importation of Apples – Report of the Appellate Body, 10 December 2003, WT/DS245/AB/R.

109

WTO, United States – Continued Suspension of Obligations in the EC – Hormones Dispute – Report of the Appellate Body, 14 November 2008, WT/DS320/AB/R.

110

Ibid., at 3507.

111

Agreement on Subsidies and Countervailing Measures (SCM Agreement) 1994, 1867 UNTS 14.

112

WTO, United States – Section 211 Omnibus Appropriations Act of 1998 – Report of the Appellate Body, 1 February 2002, WT/DS176/AB/R, at 589.

113

Paris Convention for the Protection of Industrial Property 1967, 828 UNTS 305.

114

Mavroidis, supra note 64. This section has been much influenced by my work with Joanna Langille on ‘pluralism’ in the WTO. See Howse and Langille, ‘Permitting Pluralism: The Seal Products Dispute and Why the WTO Should Accept Trade Restrictions Justified by Non-instrumental Moral Values’, 37 Yale Journal of International Law (2012) 368. See also Howse, Langille and Sykes, ‘Pluralism in Practice: Moral Legislation and the Law of the WTO after Seal Products’, 48 George Washington International Law Review (2015) 81.

115

EC–Seal Products, supra note 86.

116

US–Gambling, supra note 91.

117

China–Publications, supra note 84.

118

SCM Agreement, supra note 111.

119

WTO, Canada – Certain Measures Affecting the Renewable Energy Generation Sector/Canada – Measures Relating to the Feed-in Tariff Program – Report of the Appellate Body, 24 May 2013, WT/DS412/AB/R, WT/DS426/AB/R.

120

Eliason and Howse, ‘Domestic and International Strategies to Address Climate Change: An Overview of the WTO Issues’, in T. Cottier, O. Nartova and S.Z. Bigdeli (eds), International Trade Regulation and the Mitigation of Climate Change (2009) 48.

121

WTO, United States – Tax Treatment for ‘Foreign Sales Corporations’ – Report of the Appellate Body, 20 March 2000, WT/DS108/AB/R.

122

Michael Ioannadis has written thoughtfully and significantly on procedural approaches to adjudicative legitimacy in the WTO. See Ioannadis, ‘Deference Criteria in WTO Law and the Case for a Procedural Approach’, in L. Gruszczynski and W. Werner (eds), Deference in International Courts and Tribunals: Standard of Review and Margin of Appreciation (2014) 91.

123

This is a notion that has been developed by my colleagues Kevin Davis and Benedict Kingsbury.

124

WTO, European Communities – Conditions for the Granting of Tariff Preferences to Developing Countries – Report of the Appellate Body, 20 April 2004, WT/DS246/AB/R, at 925

125

The following draws from my previous scholarship on this dispute. See Howse, ‘Back to Court after Shrimp–Turtle? Almost But Not Quite Yet: India’s Short Lived Challenge to Labor and Environmental Exceptions to the European Union’s Generalized System of Preferences’, 18 American University International Law Review (2003) 1333; Howse, ‘Reconciling Political Sanctions with Globalization and Free Trade: India’s WTO Challenge to Drug Enforcement Conditions in the European Community Generalized System of Preferences: A Little Known Case with Major Repercussions for “Political” Conditionality in US Trade Policy’, 4 Chicago Journal of International Law (2003) 385; Howse, ‘The Death of GSP? The Panel Ruling in the India–EC Dispute over Preferences for Drug Enforcement’, 1 Bridges (ICTSD) (2004) 7; Howse, ‘Appellate Body Ruling Saves the GSP, at Least for Now’, 4 Bridges (ICTSD) (2004) 4.

126

Decision on Differential and More Favorable Treatment,Reciprocity and Fuller Participation of Developing Countries, 28 November 1979, GATT BISD (26th Supp.) (1980).

127

G. Shaffer and Y. Apea, Institutional Choice in the General System of Preferences Case: Who Decides the Conditions for Trade Preferences? The Law and Politics of Rights (2005), available at http://www.ictsd.org/downloads/2013/02/institutional-choice-in-the-general-system-of-preferences-case.pdf (last visited 22 February 2016).

128

Shaffer and Apea, supra note 126, at 32.

129

WTO, United States – Safeguard Measures on Imports of Fresh, Chilled or Frozen Lamb Meat from New Zealand and Australia – Report of the Appellate Body, 16 May 2001, WT/DS177/AB/R, WT/DS178/AB/R.

130

WTO, United States – Definitive Safeguard Measures on Imports of Certain Steel Products – Report of the Appellate Body, 10 December 2003, WT/DS248/AB/R, WT/DS249/AB/R, WT/DS251/AB/R, WT/DS252/AB/R, WT/DS253/AB/R, WT/DS254/AB/R, WT/DS258/AB/R, WT/DS259/AB/R.

131

See the critique of A. Sykes, The Safeguards Mess: A Critique of WTO Jurisprudence, Working Paper, University of Chicago (2003).

132

WTO, United States – Final Countervailing Duty Determination with Respect to Certain Softwood Lumber from Canada – Report of the Appellate Body, 17 February 2004, WT/DS257/AB/R.

133

See the excellent account in C. Bown and T. Prusa, US Antidumping: Much Ado about Zeroing, Working Paper, World Bank Policy Research (2010).

134

See Elsig and Pollack, ‘Agents, Trustees and International Courts: The Politics of Judicial Appointment at the World Trade Organization’, 14 European Journal of International Relations (2012) 1.

135

WTO, Mexico – Tax Measures on Soft Drinks and Other Beverages – Report of the Appellate Body (Mexico–Soft Drinks), 24 March 2006, WT/DS308/AB/R.

136

North American Free Trade Agreement 1992, 32 ILM 289, 605 (1993).

137

Mexico–Soft Drinks, supra note 134, para. 76.

138

WTO, European Communities – Regime for the Importation, Sale and Distribution of Bananas – Report of the Appellate Body, 25 September 1997, WT/DS27/AB/R, at 591.

139

WTO, India – Patent Protection for Pharmaceutical and Agricultural Chemical Products – Report of the Appellate Body, 16 January 1998, WT/DS50/AB/R, at 9.

140

WTO, Peru – Additional Duty on Imports of Certain Agricultural Products – Report of the Appellate Body, 31 July 2015, WT/DS457/AB/R.