Melbourne Journal of International Law
[Competition law and policy has been on the World Trade Organization agenda since 1996. However, progress has been slow. Currently the developing states are not prepared to accept a multilateral competition agreement. This has dealt a blow to those who have championed such an agreement, notably the Competition Directorate-General at the European Commission. Given the present impasse it is useful to revisit the reasons why such an agreement is said to be required. Perhaps the strongest case is that put on behalf of the international trading system. According to this argument the benefits of falling tariffs are being compromised by the growth of private trade barriers. These barriers could be disciplined by the proper application of domestic competition laws, but instead governments are tolerating — if not actually encouraging — them for protectionist reasons. If the situation is permitted to continue, the effectiveness of the multilateral trading system will be put into jeopardy. The solution is said to lie in a multilateral competition agreement. However, constructing such an agreement presents enormous problems.]
The 2003 Ministerial Conference of the World Trade Organization at Cancún in Mexico closed without any agreement. One of the main reasons for this failure was disagreement over the continuing inclusion of the so-called ‘Singapore issues’ in the Doha Round of negotiations — one of which is competition policy. Competition law and policy first appeared on the international agenda at the end of World War II as part of the negotiations that resulted in the introduction of the General Agreement on Tariffs and Trade. Although no provisions relating to competition law were included in the GATT, the issue continued to be discussed in a variety of international fora, albeit as a low level priority. In the last decade the subject has moved closer to centre stage. Pursuant to a decision taken at the Singapore Ministerial Conference in 1996, a WTO committee was set up to look into the issue of the interaction between trade and competition policy. At the Doha Ministerial Conference in 2001 it was agreed that negotiations regarding multilateral cooperation on competition policy would take place ‘after the Fifth Session of the Ministerial Conference on the basis of a decision to be taken, by explicit consensus, at that Session on modalities of negotiations’. Then came Cancún. As a result, the WTO General Council has now agreed that the interaction between trade and competition policy (along with trade and investment) will not form part of the work programme for the Doha Round. Thus, the imminent creation of a competition policy agreement as part of the WTO system has faded.
In light of the failure to move forward on a global competition agreement, it is useful to revisit the reasons that are said to justify such an agreement. In particular, it is useful to examine the connection between international trade and competition law and policy. It was largely due to the concerns of the international trade community that competition law and policy became a topic for international dialogue. Undoubtedly, appearing on the WTO agenda gave competition law and policy a priority that it would have otherwise lacked. The likelihood of a multilateral competition agreement may depend on the strength of this connection.
One of the key connections between trade and competition is the role that private barriers play in obstructing imports. Liberalising market access for foreign goods and services lies at the heart of international trade policy. Therefore, by impeding imports, private market access barriers frustrate that policy. At the same time, preventing anti-competitive private conduct (including market access barriers) is one of the cornerstones of competition law and policy. Thus, there seems to be a natural complementarity between the goals of trade liberalisation and the tools of competition law.
Of course, private barriers to market access are not the only point at which international trade and competition intersect. The role of merger regulation — a significant element in competition law regimes — is an important aspect of the debate over foreign investment. Mergers have the capacity to affect the nature and severity of private access barriers. Anti-dumping measures also impede trade and raise issues related to competition law and policy. Anti‑dumping remedies are viewed with great hostility by developing nations, which see them as deeply and unfairly protectionist. If dismantling private trade barriers to market access is one of the core trade issues for the Western developed states, anti-dumping might be described as one of the core issues for the developing states.
This article will concentrate on the issue of private market access barriers. Following some introductory comments on the distinction between competition law and competition policy, the article considers the nature and importance of market access barriers. In this regard it is critical to understand why the failure to discipline access barriers presents such a problem for trade. The article then looks at the capacity of existing arrangements to solve the trade problem. Given the complexity of competition law and policy, there is only likely to be meaningful progress on a global competition agreement if the trade problem is sufficiently serious and existing arrangements are clearly inadequate. Next, the article investigates some of the conceptual difficulties that confront any attempt to forge an international agreement on competition law. Finally, the article analyses some of the proposals that have been put forward, and ultimately concludes that a multilateral agreement on competition rules is still a long way off.
It is necessary to make some comments about the expressions ‘competition law’ and ‘competition policy’. Unfortunately the extensive literature dealing with the issue of competition law and competition policy in a global context does not use the expressions consistently.
Competition law and competition policy are not synonymous. Competition law is generally taken to refer to the laws that regulate private anti-competitive conduct. Whilst competition laws vary from nation to nation, there are certain core provisions which underpin nearly all competition law regimes. These include: prohibitions on anti-competitive cartel activities (such as price fixing and market sharing by competitors); anti-competitive conduct by dominant firms; and mergers that substantially reduce competition. In the United States, competition law (other than mergers) is invariably referred to as ‘antitrust law’. In Australia and elsewhere, it is sometimes referred to as restrictive trade practices law. The expression ‘competition law’ will be used throughout this article. The precise shape and content of a nation’s competition law — both substantive and administrative — depends to a significant degree on its competition policy.
Competition policy is a much broader and less defined concept. Most often it is used to refer to the rules and policies that determine the conditions of competition within a nation. In this sense competition policy includes not only competition law, but also government policy towards the implementation of the law. A number of states have superficially strong competition laws that are weakly enforced in practice. Sometimes this is due to unavoidable funding constraints and sometimes to a lack of proper understanding. Often however, it is a deliberate policy stance. In many states competition policy tends to be seen as a subset of industry policy. Competition policy also includes government policies on deregulation, privatisation, foreign direct investment, and government procurement practices. When used in its broadest sense to refer to all those factors which influence the nation’s competition conditions, competition policy would also include trade policy and industry policy. Generally however, these areas are treated as discrete (though overlapping) policy domains. Where the expression ‘competition policy’ is used in this article it excludes trade policy and industry policy, unless otherwise indicated. Finally, whether any law or government policy is characterised as part of competition policy is sometimes a matter of perspective. Almost any regulation, if it affects the manner in which business is done, can be characterised as part of competition policy. However, specific regulations often have their origins in considerations far removed from those that underpin competition law. For example, Japan and many European states regulate shop trading hours. In Europe and Japan this is seen as a cultural matter. In the US however, it is seen as a restriction on competition.
It should be noted that the expression ‘competition policy’, when used in the sense just described, is not normative. It makes no assumptions that a particular configuration of markets or industrial organisation is better than another. Furthermore, it makes no assumptions that competition policy is underpinned by any particular object or objects, although there is evidence that a significant number of states view economic efficiency as a key object.
This article is concerned essentially with the internationalisation of competition rules in the context of international trade. It is therefore concerned primarily with competition law. However, because the effectiveness of competition law may be nullified by a policy of non-enforcement or by other legislative provisions inimical to vigorous competition, any discussion on the internationalisation of competition rules must also consider the role of competition policy. Where the concepts of competition law and competition policy are used throughout the article, they carry the meanings discussed above.
Since the introduction of the GATT, international trade has increased substantially, and there is widespread economic support for the notion that it must continue to expand. Further expansion depends in part on the continuing elimination of trade barriers. However, as governments reduce protection at the border in the form of tariffs, market access barriers are being erected behind the border. Alternatively, as tariff barriers come down, existing internal barriers become more important from a trade perspective. It is argued that dismantling internal access barriers will be difficult to achieve without an international agreement to apply appropriate competition rules in protection of trade.
Commentators have divided non-border trade barriers into three classes:
• Government barriers;
• Private barriers; and
• Hybrid barriers.
Government barriers are manifold. They include government monopolies; regulations on entry and exit of firms to and from local markets; government procurement policies; government subsidies to domestic producers; trade facilitation measures (such as port duties, legal fees and foreign exchange commissions); and product and safety standards designed to exclude imports. Government barriers also include environmental and labour laws. Indeed almost any domestic regulation or policy initiative affecting the way business is conducted may be viewed as a barrier to trade. However, although from a broad perspective these are certainly matters that affect the conditions of competition, they are not matters for competition law. While there is much work to be done internationally on the issue of non-tariff government barriers — indeed some would say it is the most critical area — competition law is not a tool that can solve these problems. For example, whilst it is possible in appropriate circumstances to regard government procurement policies that discriminate against foreign suppliers as an abuse of market power and thus technically a competition law issue, the doctrines of sovereign immunity and act of state generally insulate government actions from competition law. Consequently, an agreement on competition laws will not resolve the issue of government barriers.
Private barriers include a whole range of exclusionary tactics that may be employed by private actors to impede the entry of competitors into a market, including horizontal exclusionary practices (such as import cartels, international cartels to allocate markets, joint buying groups, and joint ventures) and vertical exclusionary practices (such as exclusive dealing, requirements contracts, customer and territorial restraints, product bundling, predatory pricing and price discrimination, and refusals to deal in the context of essential facilities). These practices clearly fall within the general ambit of competition law, although whether any given practice is prohibited or not depends on the precise nature of the competition law by which the practice is judged.
‘Hybrid barriers’ is an expression used to describe private barriers that are approved, encouraged or tolerated by the local government. Government involvement in private access barriers may be overt (for example authorising import cartels), or it may be more subtle and less transparent (for example failing to enforce competition laws). Whilst all hybrid barriers are seen as potentially distorting trade, it is the more subtle type of hybrid barrier that presents the greater concern from a global regulatory perspective.
The distinction between hybrid barriers and private barriers should not be overstated. In essence, the two are distinguished by the fact that some private barriers operate without the assistance or toleration of the local government and others operate with such assistance or toleration. In many cases it will be difficult to make this factual distinction. For example, can it be said that a country is tolerating private access barriers simply because it has no competition law? Is a developing country, which routinely fails to enforce its competition laws owing to funding or other capacity constraints, tolerating private access barriers? From an exporter’s perspective these questions may seem unimportant. For the trader the key factor is not that foreign access has been blocked with or without the assistance or toleration of government, but that the exclusionary conduct has not been disciplined. Nevertheless, the distinction does become important when analysing existing and potential disciplinary tools.
Why are private access barriers such a critical issue for the future of international trade? International trade policy is based on the notion of trade liberalisation. However, because of local incentives, that objective can only be achieved through state-negotiated trade agreements. Nations trade access opportunities to their local markets in return for access to foreign markets. In a world of self‑interested states, the liberalisation of international trade can only be achieved if states have confidence in the access deals (the trade agreements) that they make. If deals are not honoured, governments will lose the political support necessary to continue such agreements. In that case, not only will further trade liberalisation be impossible, but existing gains may very well be lost. The existence of private access barriers deprives the exporting nation of part of the deal it negotiated. Failure by the importing state to discipline such activities undermines the state’s international obligations. Consequently, private access barriers are a distinct threat to the future of the international trading system. This is the rationale for linking competition law and policy to international trade and thus to the WTO.
It is not clear how significant the problem of private access barriers is to international trade. Much of the evidence is simply anecdotal. Indeed many of the complaints made by business could just as easily reflect the disappointment of failing to meet inflated export expectations. Additionally, some of the evidence suggests that launching complaints about unfair access barriers may be a useful strategic move by exporters to enlist state aid in facilitating entry into foreign markets. The evidence has led some commentators to argue that the issue of private access barriers has been overstated, and that government barriers are a much more significant threat to trade flows.
Nevertheless, the lack of empirical evidence does not mean that such barriers do not exist. Nations have incentives to engage in protectionism. Competition law provides a convenient vehicle for satisfying those incentives. Firstly, strategic trade theory suggests that under certain conditions, including increasing returns to scale, a nation acts rationally by engaging in protectionist measures. Secondly, public choice theory suggests that government decisions are likely to be made in the interests of special groups (notably, local producers) rather than the public good (consumers). Local producers are likely to pursue local protectionism even where that protectionism harms consumers, for example, through higher prices. The open-textured nature of competition law and the broad discretions given to administrators make competition law ideal for surreptitious protectionism. Theoretically and politically, private-hybrid barriers often make sense.
The reason this issue has caused deep concern — despite the lack of hard empirical evidence — is due to the magnitude of the consequences that would flow from failing to solve the problem. As previously discussed, if confidence is lost in the ability of the international trade system to discipline nations that fail to honour their trade agreements, the international trade system will be in jeopardy. At risk are the enormous gains made under the present multilateral system. There can be no doubt that a significant disruption to international trade would have serious political, social, cultural and economic consequences. For a number of reasons, including welfare, development and the avoidance of trade wars, a bilateral or regional trading system is regarded by many as inferior to a multilateral system. Thus there is considerable pressure on trade policy-makers to find a solution. This explains why trade authorities have been critical of competition authorities for not doing more to solve the problem. Equally, a failure to understand the fundamental nature of the trade dilemma has caused competition authorities to be somewhat dismissive of trade concerns.
Developing states may view the matter in a different light. Although many states have recently adopted competition rules, the recent events at Cancún seem to demonstrate that outside the developed states there still remains considerable apprehension about global competition rules. There are a number of reasons for this. Firstly, developing states are suspicious of the motives of developed states. The push for global rules against private trade barriers is seen by some developing states as an attempt by the developed states to control the growth of significant firms in the less developed economies. Secondly, less developed states are unsure of the effects, and sceptical of the overall benefits, of regulating private import barriers. Implementing and applying an effective and efficient competition law regime is expensive. It also requires a political commitment that many less developed states are unprepared or unable to make. There is also a belief that any global agreement to regulate import barriers may seriously restrict a state’s ability to pursue development strategies based on industrial policy. Thus, the costs of a global competition agreement are perceived by many states to be quite high.
Furthermore, whilst the costs are high, the benefits of a global competition agreement are low. Firstly, the developed states — particularly the US — show no intention of abandoning protectionist-driven anti‑dumping remedies. Secondly, the EU remains resistant to changes in its Common Agricultural Policy. Thirdly, an agreement on market access does little to discipline the power of multinational corporations, a longstanding concern of the less developed states. Developing states would prefer a greater commitment to the principles contained in the Set of Multilaterally Agreed Equitable Principles and Rules for the Control of Restrictive Business Practices negotiated at UNCTAD.
Given the difficulties of achieving multilateral cooperation, it seems reasonable to suggest that, in the absence of extraordinary circumstances, such cooperation is only likely to eventuate if there is a compelling need to solve the problem of private trade barriers and if existing disciplines are incapable of providing that solution. As just discussed, it is doubtful whether there is any widespread acknowledgement of a compelling need. However, assuming that such a need does exist, are existing arrangements, domestic or international, adequate for the task? To answer this, it is necessary to examine unilateral, bilateral and multilateral initiatives.
The most instructive use of unilateral measures comes from the US. Frustrated by a failure to achieve greater market access for its exports, the US developed § 301 of the Trade Act of 1974 (‘§ 301’) to deal with what it determined to be ‘unfair’ trade policies. The impetus for § 301 came from a widespread feeling in the US business community that, having opened its borders to imports, foreign states — particularly Japan and the countries of the European Community — were not reciprocating. Instead it appeared that US exports — not only goods but also services and matters relating to intellectual property — were subject to all manner of protectionist measures. Opinions on § 301 range from ‘aggressive unilateralism’ to a justifiable act of reciprocity.
Many § 301 cases can now expect to finish up within the WTO dispute resolution system. However, this is not necessarily the case with private barriers. It is doubtful whether mere toleration of private access barriers falls within the ambit of the GATT–WTO agreements. Section 301, on the other hand, clearly extends the notion of ‘unreasonable’ actions to government policies and practices that tolerate or encourage private anti-competitive activities that exclude or restrict US exports. Therefore, § 301 remains a possible avenue for US exporters seeking redress against foreign access barriers. The US has deployed § 301 against the Japanese keiretsu system and Japanese distribution practices, but with limited results.
Generally, the use of § 301 has been criticised by other states. This is to be expected given the exercise of raw power that the use of § 301 often entails. Antipathy is most evident where the matter does not involve an issue within the cognisance of the GATT–WTO agreements or some other trade agreement. Where a matter is within its cognisance, the GATT–WTO system has undoubtedly constrained the use of unilateral sanctions by the US.
The objections to § 301 are many. There is no provision in § 301 for the application of comity principles. There are few guidelines on what standards should be applied. Those guidelines that do exist are not necessarily in keeping with US competition law standards. Much less is the USTR explicitly directed to have regard to foreign socio-economic factors. The USTR has no established procedures or guidelines covering hearings, gathering evidence, or making determinations. Remedies involve the suspension of trade concessions or the imposition of duties. These are fairly blunt devices to solve what are essentially privately imposed access restraints.
For these reasons it has to be doubted that § 301 is well designed to deal with complex anti-competitive private activities. Additionally, bilateral arrangements forged under the pressure of unilateral action may tempt the complaining party to seek preferential access treatment for its firms (for example, a guaranteed quota) contrary to the GATT–WTO most favoured nation principle. The most favoured nation principle is one of the keys of the open multilateral trading system. The ability to obtain preferential access treatment provides exporters with incentives to lobby their governments to pursue bilateral agreements at the expense of multilateral arrangements.
In any event, unilateral trade action such as that provided by § 301 is an option realistically open only to relatively powerful nations. Where those states themselves harbour private barriers there is little that the rest of the trading world can do by way of unilateral action.
Even though the source of the complaint may be toleration of private conduct, unilateral trade initiatives such as § 301 are aimed at foreign government conduct. Section 301 is not designed to discipline purely private anti-competitive barriers, although, as previously discussed, the line between purely private barriers and hybrid barriers (based on government toleration) can be very fine indeed. Purely private barriers, however, are the direct focus of competition laws. Where competition law is applied extraterritorially foreign private barriers come very much within its purview. A number of states (principally the US and the states of the EU) apply their competition laws extraterritorially.
There are a number of advantages for a state in applying its law extraterritorially. Firstly, the rules and procedures to be applied already exist. Secondly, where a state is able to impose its law extraterritorially, the outcome is optimal for that state because it has not had to bargain away any elements of its preferred model. Where states can find ways to cooperate in the enforcement of their domestic regimes, extraterritoriality may be the preferred option. The successful pursuit of global hard core cartels in the past decade is an example of extraterritoriality working reasonably well. Cooperation is discussed in the next part of this article.
Extraterritoriality, however, is not a satisfactory solution where it is employed in a hostile environment. In such circumstances serious inter-state conflict is a possibility. As conflict rises, the transaction costs of extraterritoriality also rise. Conflict may arise due to policy differences manifested in substantive law differences. Unlike hard core cartels, there are still substantial policy differences over the regulation of vertical restraints, such as distribution restraints. For example, distribution restraints are treated with greater tolerance in the US than in the EU. Conflict also arises as a result of procedural or systemic differences. Much of the hostility towards the extraterritorial application of US antitrust law may be ascribed to the broad nature of US discovery laws and the ability of US courts to grant treble damages.
Where conflict exists, effective enforcement becomes difficult. Effective enforcement normally depends on access to the defendant’s assets. Where assets lie outside the relevant jurisdiction, effective enforcement depends on the willingness of states to enforce foreign judgments. Not only have states refused to enforce judgments, but attempts by US courts to apply US antitrust law extraterritorially have led in the past to retaliatory legislation. This has included ‘blocking’ laws — both against the collection of evidence and against enforcement — and ‘claw-back’ laws. This leads to under-regulation, and ultimately sets up the possibility of a regulatory race to the bottom.
The effective extraterritorial reach of a state’s competition law is also governed by the deference one state will afford another under the doctrines of foreign sovereign immunity; act of state (and the closely related foreign sovereign compulsion); and the notion of international comity. Thus, private barriers may be insulated against competition law because they result from, or are supported by, government policy.
As with unilateral trade sanctions the main lessons come from the US. In determining the extraterritorial scope of US competition law, it is necessary to have regard to two factors: the so-called ‘effects’ doctrine and the Foreign Trade Antitrust Improvements Act of 1982. The effects doctrine provides that US competition laws apply to any foreign conduct that is intended to affect the US economy, provided the effects are direct, substantial and reasonably foreseeable. International comity could only preclude the exercise of jurisdiction where there was in fact a true conflict between US and foreign law. A true conflict exists only where foreign law requires the defendant to act in the impugned manner. This leaves very little room for the operation of comity. Most competition regimes are proscriptive in nature — they generally do not require firms to act in any particular manner. In the majority of cases US courts will have no discretion to defer to foreign law. A number of states now claim extraterritorial reach similar to that of the US for their competition laws, most notably the EU. Other nations, notably Japan, remain much less aggressive about the extraterritorial reach of their law.
Whereas the effects doctrine applies to inbound commerce, the FTAIA applies to US export trade and commerce. The FTAIA has two important effects. Firstly, it ensures that US export conduct is not affected by US competition law unless it has a direct, substantial and reasonably foreseeable effect on US markets or on US exports. Secondly, the FTAIA ensures that US competition law applies to foreign conduct that affects US exports provided the effect is direct, substantial and reasonably foreseeable. It is the latter effect that is important here. Thus, a boycott of a US exporter (organised by a foreign competitor or group of competitors) would be actionable on the part of the US exporter in the US courts even though there was no adverse impact on US consumer welfare. Arguably the FTAIA leaves little scope for US courts to refuse jurisdiction on comity grounds. As with inbound commerce, only if the foreign boycott was mandated by foreign law would the necessary conflict between US and foreign law exist so that US courts could exercise their discretion to refuse jurisdiction. This suggests that there is ample scope for the application of US competition law beyond US borders. No other state claims such a broad and controversial power to discipline extraterritorial conduct, although this may change if the EU perceives that its efforts to internationalise competition law are not successful.
In the end however, expanding extraterritorialism merely serves to emphasise the importance of the earlier discussion concerning the limitations of extraterritoriality in a hostile environment. For this reason, during the 1990s the US tended to move away from aggressive unilateralism towards bilateral agreements based on comity. As will be discussed in the next section of this article, these bilateral agreements contemplate some use of extraterritorial jurisdiction, but within a cooperative framework. Government tendencies towards bilateralism do not, however, constrain private actions. Thus, unilateralism remains a strong part of private suits. Indeed the attraction of treble damages, class actions, contingency fees and broad discovery rules will ensure that US extraterritorialism continues to be relevant to international competition law.
The EU is the outstanding example of a regional competition regime. No other regional or bilateral arrangement comes anywhere near approximating the substantive and administrative complexity of the European arrangement. By contrast, Australia and New Zealand have unified certain aspects of their respective competition laws. The North American Free Trade Agreement contains limited provisions relating to competition. A number of other regional agreements exist or are in the process of being established, although the competition provisions in each are very limited.
Bilaterally, formal competition cooperation agreements now exist between the US and the EU and between the US and a number of states, including Canada, Australia, Germany, the United Kingdom, Brazil, Israel and Mexico. The EU also has a number of bilateral cooperation arrangements. Australia has cooperation agreements with the US, New Zealand, Korea, Chinese Taipei and Canada. Japan has agreements with both the US and the EU.
Prior to 1990 very few agreements existed and those that did were essentially concerned with controlling occasions of conflict caused by US extraterritoriality. For example, the 1982 agreement between the US and Australia provides for notification in certain circumstances. Those circumstances are fairly precise. Firstly, where Australia adopts a policy that may have antitrust implications for the US, Australia may notify the US of that policy. Secondly, when the US Department of Justice or Federal Trade Commission decide to undertake an antitrust investigation that may have implications for Australian laws, policies or national interests, the US shall notify Australia of the investigation. There is only one terse provision dealing with cooperation in antitrust enforcement, and that provision is more concerned with controlling the use of evidence-blocking legislation.
In contrast, recent agreements are more concerned with enforcement cooperation than avoidance of conflict. For example, the Agreement between the Government of the United States of America and the Commission of the European Communities regarding the Application of Their Competition Laws begins with the premise that the parties have a common interest in ‘sound and effective enforcement of competition law’, and that sound and effective enforcement is ‘enhanced by cooperation and, in appropriate cases, coordination’. The agreement is based on a framework that involves notification, consultation and cooperation. The parties agree to notify each other when an antitrust matter affecting the other’s interests arises. The parties then undertake to consult with one another, and finally to cooperate and coordinate their enforcement activities to the extent permitted by local law and practice, and the national interest. The agreement also introduces the notion of ‘positive comity’ to international competition enforcement. Positive comity (as understood in the US–EU Agreement) provides that one party may request the other to investigate and, if warranted, remedy anti-competitive activities in accordance with the requested party’s competition law. The request may be made whether or not the alleged anti-competitive activities violate the requesting party’s competition laws, and whether or not the requesting party contemplates taking action under its own competition law. Where a request has been made the requesting party will suspend its investigations subject to certain conditions being satisfied. At least two advantages arise from this type of ‘positive comity’ agreement. The first is that it encourages the state best placed to discipline the relevant anti-competitive activities to take the lead. This restricts the likelihood of conflict arising from extraterritorialism. The second advantage, of particular importance for traders and the disciplining of private trade barriers, is that it encourages the importing state to sanction (at the request of the exporting state) anti-competitive activities whose impact in the requesting state may be limited to a restriction on exports — that is, without any substantial effect on consumer welfare in the requesting state.
Ultimately, however, these agreements are voluntary. The decision whether assistance ought to be granted eventually depends on how the requested state perceives its national interest, including its interest in fostering a climate of reciprocity. Where the request involves issues of market access there may be strong protectionist forces mounted against giving assistance. In fact, the evidence suggests that nations have not shown any great willingness to pursue domestic firms at the request of another state. This reflects the fact that comity requests have the capacity to trigger domestic tensions, for example, between trade representatives and competition authorities. The success of a comity agreement between two states is directly related to the political, economic and social similarity between them. It is also related to the nature of the comity request. Whilst a workable level of cooperation has been achieved in respect of global hard core cartels, the same cannot be said for distribution restraints and domestically focused import cartels. The reason for this is that while there is a shared view concerning the need to discipline global hard core cartels (for example, naked market sharing cartels), there is much less agreement about the nature and effect of distribution restraints that exclude imports. In practice, therefore, bilateral cooperation has concentrated mainly on international hard core cartels.
Additionally, bilateral agreements have not managed to solve all the problems of evidence gathering — in particular, the problems that surround the sharing of confidential information and foreign discovery. Competition law disputes are very fact intensive. Without access to all the relevant information proper decisions cannot be made, yet most bilateral competition agreements expressly exclude the exchange of confidential information. Regulators, no matter how willing to cooperate, cannot supply this information to foreign authorities. Changing this situation will be quite difficult. There are numerous reasons why states protect such information. Firstly, it is important to safeguard the reputation of firms during the investigation process. Secondly, sensitive securities markets need to be protected against the premature release of information. Thirdly, it is necessary to preserve incentives for firms and individuals to volunteer information to competition authorities. Witnesses often request anonymity. Frequently they must be protected against harassment. Lastly, it is vital that sensitive financial and strategic business information does not fall into the hands of the competitors of those firms compelled to provide the information. The last is of particular concern to business. There is a strong suspicion, not necessarily supported by hard evidence, that vital information provided to a foreign regulator will be ‘leaked’ by that authority to its local firms. As the number of bilateral agreements expand, this problem is likely to increase rather than decrease.
There is a considerable amount of informal cooperation between competition authorities. This flows from the informal networks that have been built up over the last decade through participation of competition authorities in international bodies such as the Competition Law and Policy Committee of the OECD, the OECD Global Forum on Competition, the WGTCP at the WTO and the International Competition Network (‘ICN’). Such bodies have been supported and reinforced by various workshops, roundtables and symposia conducted by a variety of institutions, both official (such as the competition workshops run by the Antitrust Division of the US Department of Justice) and unofficial (such as the numerous symposia conducted by US universities). Cooperation at this level may be as simple as sharing experiences, but it is also likely to include informal policy exchanges, and swapping non-confidential information on specific cases and other publicly available information. A priority in recent years has been to build greater understanding and competency in developing states, commonly referred to as ‘capacity building’. There are strong incentives for officials charged with similar responsibilities to seek and provide background information and advice from and to each other. Ultimately, however, this type of informal cooperation between bureaucracies can achieve only so much. Additionally, there may be legitimacy concerns in allowing local policy and practice to be captured by a cabal of unelected and foreign authorities.
There is likely to be continuing use of bilateral and regional trade and competition agreements. These agreements will be subject to positive comity considerations in respect of competition policy, and this may lead to a greater shared understanding of the role of competition law and policy. Furthermore, this trend is likely to be supported by a continuing interaction between domestic competition authorities. However, convergence will not be equal across the range of competition issues. There is likely to be greater convergence on the issue of hard core cartels than on the issue of vertical restraints. Arguably, this will not result in the kind of regulatory framework that trade authorities would like to see. Further, there is no indication that bilateralism has solved the fundamental problems of confidential information and foreign discovery. These remain major obstacles to the successful pursuit of trade barriers. Finally, there are significant costs associated with operating bilaterally as opposed to multilaterally. These costs can only increase as commercial globalisation intensifies and more states develop sophisticated competition regimes.
Firstly, it is necessary to be aware that the GATT–WTO system contains no agreement on competition laws. Secondly, it should be noted that the
GATT–WTO system is not generally able to address purely private conduct. The emphasis therefore falls on the role of government. To determine whether the GATT–WTO system is capable of disciplining private access barriers this article will examine the possibilities under art XXIII of the GATT.
Article XXIII(1)(a) applies where a state violates one of its GATT obligations. This is a common form of dispute. A state would violate its GATT obligations if, for example, it passed a legally binding regulation requiring local firms not to trade in imports, or not to trade beyond a certain level. This could be a violation of the national treatment principle contained in art III(4), or the prohibition on quotas under art XI. However, rarely are such private exclusionary activities mandated by the state. It is much more likely that a state will be accused of having encouraged or tolerated the private trade barriers. While this may still amount to a failure to carry out a GATT obligation within the meaning of art XXIII(1)(a), art XXIII(1)(b), which does not depend on proof of a GATT violation, seems the more promising choice. Article XXIII(1)(c) is largely untested and will not be discussed in this article.
Article XXIII(1)(b) applies where a state can establish that any benefit accruing to it under the GATT is being nullified or impaired or the attainment of any objective under the GATT is being impeded as a result of the application by another state of any measure, whether or not it conflicts with the provisions of the GATT. Potentially, therefore, art XXIII(1)(b) has a specific application (where any benefit accruing to a state under the GATT is nullified or impaired), and a general application (where any objective of the GATT is impeded). The majority of the small number of cases that have arisen have dealt with the first application. It is that application with which this article is mainly concerned.
It is entirely understandable that the GATT contain provisions designed to prevent backsliding through the use of measures not specifically prohibited. This is the purpose of art XXIII(1)(b). However, while the purpose is clear, the scope of art XXIII(1)(b) is not so apparent. Article XXIII(1)(b) is necessarily worded in broad terms. Consequently, on one interpretation it has the capacity to extend beyond occasions of obvious backsliding and to encompass almost all instances where trade obligations are adversely affected. The difficulty is that the circumstances in which trade agreements may be adversely affected by internal policies and activities is probably unlimited. Therefore, art XXIII(1)(b) presents a significant problem of interpretation. The issue is the extent to which art XXIII(1)(b) can and should expand into the undefined space surrounding a state’s defined GATT–WTO obligations in order to secure the performance of GATT–WTO obligations and objects. Having regard to the potential breadth of art XXIII(1)(b), the nature of the WTO dispute resolution system, and the requirements of the Understanding on Rules and Procedures Governing the Settlement of Disputes, it is suggested that the likely and proper approach is one of caution. Recent WTO panels have supported this approach. A non-violation claim under art XXIII(1)(b) is to be regarded as an exceptional remedy requiring detailed justification.
To succeed under art XXIII(1)(b), an exporting state must establish three things. Firstly, the exporting state must establish that the importing state has engaged in a ‘measure’. Secondly, the exporting state must establish the existence of a benefit that has been nullified or impaired by the measure. Thirdly, the exporting state must establish that the measure has had a disproportionately adverse effect on imports compared to local products. This effect is often described as upsetting the competitive relationship between domestic and imported products.
The most striking example of a non-violation claim is where the importing state, following the settling of a tariff agreement, undermines the deal by paying production subsidies to its local producers. However, it is now clear that non‑violation claims are not restricted to subsidies. Therefore, if the exporting state was able to establish that the importing state’s refusal or failure to employ its competition laws against its local firms nullified or impaired the value of the benefits that the exporting state reasonably expected from the importing state’s agreed tariff concessions, the exporting state could arguably maintain a non‑violation claim under art XXIII(1)(b).
The use of art XXIII(1)(b) in this manner would, however, be controversial, as arguably there is a significant difference between curtailing the payment of subsidies to local producers and dictating the implementation of competition law and policy. Nevertheless, there is some precedent for an extended application of art XXIII(1)(b). However, even if art XXIII(1)(b) does extend to competition practices and policies, it will not cover all such practices and policies. Nor will it be easy to establish the elements of art XXIII(1)(b) in any given case.
As already indicated, the exporting state must first prove that the importing state has engaged in a ‘measure’. ‘Measure’ is probably to be interpreted liberally. It is not restricted to government acts that are binding. Non‑binding government acts or policies may produce a course of private action equivalent to that produced by binding measures. The issue is not whether a government act is binding or not, but rather whether in the circumstances there is ‘sufficient government involvement’ to hold the government responsible for the relevant private action. This is very much a question of fact. If the private action effectively depends on the incentives or disincentives (binding or not) created by the governmental act or policy, then that act or policy ought to be regarded as a ‘measure’ for the purposes of art XXIII(1)(b). The more indirect or remote the alleged government involvement, the more difficult becomes the task of establishing a relevant government measure.
The granting of an authorisation or an exemption by a competition authority of private industry practice would undoubtedly amount to a government measure. In Japan — Film, the WTO Panel held that the approval of certain private industry codes by the Japan Fair Trade Commission (‘JFTC’), which exempted code members from scrutiny under the Japanese Antimonopoly Act (Japan’s competition law), had sufficient connection with the Japanese Government to make the act of exempting them a ‘measure’ within the meaning of art XXIII(1)(b). It seems clear that a government announcement that it would not enforce its competition law in a particular case could satisfy the test for a ‘measure’. By the same token, a general announcement that the government would no longer enforce its law against private restraints would satisfy the test for ‘measure’ if, in the circumstances, the existence of the private restraints depended effectively on that government announcement. The Japanese practice of administrative guidance has been held capable of satisfying the requirements of a ‘measure’.
At the opposite end of the scale, where a state has no competition law or no law covering the particular restraint in question, mere failure to enact such a law (in the absence of an explicit promise to rectify the situation) is unlikely to be construed as a measure. The situation in which an importing state has a law covering the private restraint, but fails to act, lies somewhere between these extremes. As art XXIII(1)(b) is regarded as ‘an exceptional instrument of dispute settlement’, it is extremely doubtful that ‘doing nothing’ per se is sufficient to constitute a measure. However, government passivity does not exist in a vacuum. There are always other circumstances which add colour and possibly motive to ‘doing nothing’. It may be possible, therefore, to construct a factual canvas which demonstrates an indirect but sufficient level of government involvement; for example, where the failure to act is inconsistent with previous government intervention. This might be true in theory, but in practice — given the cautious approach of WTO panels to non-violation complaints, and given that the onus of proof lies on the exporting state — the task would not be easy. In any event, establishing a measure is merely the first step.
Having established that the impugned government act, policy or practice is a measure, the exporting state must next establish the existence of a benefit that has been nullified or impaired by the measure. The benefit must have been legitimately expected. In most cases the legitimately expected benefit in question will be improved market access opportunities arising from another state’s tariff concessions. Generally, the benefit will be assumed. However, there can be no nullification or impairment of that benefit by the measure if the measure was reasonably expected. The critical issue, therefore, is whether the exporting state could reasonably have anticipated the measure. If the exporting state could reasonably have expected the measure at the time the tariff concessions were made, then there has been no nullification or impairment of a legitimately expected benefit (being normally the opportunity for improved market access).
Therefore, in practice the exporting state must establish not only that it had no foreknowledge of the likelihood that the benefits of its trade deal would be nullified or impaired by the measure (in this case, the existence of private access barriers or rather the failure properly to discipline them), but also that it could not reasonably have expected such an eventuality. If the importing state had a competition law which proscribed the access barriers in question, the exporting state would have to show that it had a reasonable expectation that the importing state would take action against those barriers in a non-discriminatory fashion. This should be feasible where the importing state traditionally has enforced its competition law vigorously. However, where the importing state has a history of lax enforcement (whether as a result of policy restraints or capacity restraints), it would be difficult for the exporting state to establish that it had the necessary expectation of enforcement. In the absence of any promise by the importing state, an exporting state can have no reasonable expectation that the situation for imports will be made easier than that prevailing at the time the trade deal was made. Rather, the exporting state merely has a legitimate expectation that the situation will not be made worse. Confronted by a history of lax enforcement, the exporting state would probably have to prove that it had a promise of enforcement. Normally, this will be difficult to do. In general, such bilateral competition agreements as exist do not require a state to take action at the request of the other party to the agreement.
Where the importing state has no law prohibiting the relevant private restraints, it is difficult — failing a promise by the importing state to introduce and enforce such a law — to see how the exporting state could argue it had any expectation at all. Only, perhaps, if the importing state had no history of private restraints could the exporting state plausibly argue that it did not expect this kind of nullification or impairment of its trade deal. However, private restraints are generally well known to exist and therefore it is doubtful that this argument could succeed.
Finally, the ‘measure’ must have a disproportionate effect on imports. Article XXIII(1)(b) requires an element of discrimination; the government measure must have the effect of upsetting the competitive relationship between domestic products and imports. Therefore, even if a particular government activity could be characterised as a ‘measure’, and even if it could be established that the measure nullified or impaired the exporting state’s reasonably expected benefit, it would still be necessary to establish that the measure was the cause of domestic production being favoured at the expense of imports — alternatively, that the measure discriminates in favour of one source of imports over another. Discrimination may be de jure or de facto. In practice, most complaints are likely to involve de facto discrimination; that is, a law or policy which is ostensibly neutral being applied in a discriminatory manner. In such cases the exporting state must make ‘a detailed showing of any claimed disproportionate impact on imports’. It will be no simple matter for the complaining state to establish the necessary causal connection between the government measure and the relevant discrimination. The issue of causation has in practice provided the most formidable forensic obstacle.
In sum, if art XXIII(1)(b) has any application against private access barriers, that application is significantly circumscribed. It might apply, for example, where a state which has a history of taking action against exclusionary cartels on a non-discriminatory basis changes its competition policy to permit import cartels. Any trade agreement made prior to the change in policy would presumably have been made with the expectation that import cartels would continue to be proscribed. The change in policy — and its implementation — could therefore amount to a government ‘measure’, and by definition, have a disproportionate impact on import competition. In most cases, however, there is unlikely to be such an overt change in policy. Whilst the exporting state may strongly suspect a change in the landscape, establishing this within the meaning of art XXIII(1)(b) may be very difficult.
Assuming that the exporting state is able to establish a claim under art XXIII, there still remains the fact that the remedies are not well designed to meet the problem. A claim under art XXIII is a state-to-state claim. There is no provision to compensate the private exporters. Further, there is no power in respect of a non-violation claim under art XXIII(1)(b) to require the state to remove the offending ‘measure’. The exporting state must rely on seeking counter‑balancing trade concessions or imposing retaliatory measures.
Article XXIII(1)(b) was argued in Japan — Film by the US against Japan in respect of a dispute between Kodak and the manufacturers of Japanese photographic film. The case is not an exact precedent for the type of situation being discussed in this article because, whilst the substance of Kodak’s complaint was the anti‑competitive nature of distribution practices in the Japanese film market, the US allegations against the Japanese government did not centre on Japan’s failure to apply its competition law. Rather, the US alleged that a raft of Japanese government measures (only some of which were directly related to competition law) enabled Fuji to persist with its anti‑competitive and exclusionary distribution system. Nevertheless, the case has widely been seen as creating a precedent for raising competition policy issues at the WTO.
Kodak argued that distribution practices within the Japanese film market were anti-competitive and effectively excluded Kodak from supplying that market. Kodak further maintained that the anti-competitive distribution practices were made possible by the legislative and administrative practices of the Japanese government. Kodak initiated a § 301 case. The USTR decided in favour of Kodak, and requested a WTO panel to hear the matter. The substance of the US allegation was that Japan, by enacting certain laws and by adopting certain policy and administrative practices, had indirectly protected the Japanese film manufacturers from import competition by enabling those manufacturers to engage in arrangements with their distributors that kept Kodak out of the Japanese film market. According to the US, this was a violation of Japan’s national treatment obligations under art III(4), and a violation of its transparency obligations under art X. It also nullified or impaired the value of the benefits that the US reasonably expected from Japan’s prior tariff commitment — namely, that US firms would have greater access to Japan’s film market, a non‑violation infringement under art XXIII(1)(b). The US failed in its claim. The WTO Panel accepted the argument that a claim under art XXIII(1)(b) could be triggered by government policies or practices other than direct financial subsidies — the normal source of a non-violation claim. The Panel accepted that many of the impugned practices could amount to a ‘measure’ within the meaning of art XXIII(1)(b). However, the Panel then found that none of Japan’s impugned policies or practices had significantly affected Kodak’s competitive position. The US failed to prove that any of the Japanese measures, either singularly or in combination, caused or maintained a system of industrial organisation in the Japanese photographic film industry that nullified or impaired Kodak’s legitimate market access expectations.
Japan — Film opens the possibility that some competition law based claims may be successfully brought to the WTO. However, the case also demonstrates that it will be very difficult to satisfy the necessary requirements. This is not a criticism of art XXIII(1)(b), but simply a reflection of the fact that it was not designed to mediate typical competition disputes. Although the ultimate US object may have been to attack Japan’s Large Scale Retail Stores Law of 1982, on its face Japan — Film looks very much like a dispute between private competitors masquerading as an intergovernmental dispute. As a model for future competition cases before the WTO, it surely poses fundamental questions about the complementarities between the international trade system and competition law and policy. For example, should governments utilise the WTO system to advance the cause of their exporters in disputes with foreign competitors? This issue is discussed in Part V of this article.
On the assumption that private trade barriers are, or will become, a matter of serious concern, this part of the article has explored whether existing arrangements — unilateral, bilateral or multilateral — were capable of providing an effective solution. Unilateral tools — whether special trade measures such as § 301 of the US Trade Act of 1947 or extraterritorial application of domestic competition law — are useful, but there is always the risk that they will provoke hostility. In a hostile environment the evidence suggests that unilateralism is ultimately inefficient and often ineffective. Realistically, it is only available to the more powerful states.
Because of the limitations of unilateralism, there was a distinct shift during the 1990s towards international competition cooperation, particularly at the enforcement level. This resulted in a number of bilateral competition agreements built around the notion of both traditional and positive comity. This web of bilateral state cooperation was supported by cooperation, both formal and informal, between national competition regulators. In organisations such as the ICN, the dialogue has gone beyond enforcement issues to include some substantive matters, particularly in relation to mergers. However, it is necessary to be aware of the limitations in this complex web of cooperation. Firstly, comity by its very nature is discretionary. It works best where the parties share a common view and a level of mutual trust. The evidence suggests that while it has proved useful in regulating global cartels, it has had a weaker effect on distribution restraints and export cartels. Secondly, as commerce continues to globalise, and as more states adopt some form of working competition law, the number of bilateral agreements will grow apace. There is likely to come a time where the cost of negotiating and administering this complex web of bilateral competition agreements will outweigh the benefits. Thirdly, although the number of agreements is likely to grow, many of the less developed states will be left out of this process. Finally, bilateral agreements have not yet solved the problems of confidential information and offshore evidence gathering. As these bilateral agreements increase in number, concern for the protection of vital commercial information is also likely to grow. In other words, there is probably an inverse relationship between expanding enforcement of competition law through bilateral cooperation, and a willingness to exchange vital evidence and information necessary to make the enforcement effective.
If unilateralism and bilateralism have their limitations, what about existing multilateral arrangements? Article XXIII(1)(b) of the GATT was examined to determine whether it had the capacity to handle the problem of private trade barriers. Whilst it seems clear that some relevant competition issues will fall within art XXIII(1)(b), it will not cover all such issues. Even in those cases which arguably come within the scope of art XXIII(1)(b), it will not be easy to establish an infringement. This is not a criticism of art XXIII(1)(b), but rather a recognition that it is not designed to deal with the majority of private anti‑competitive activities that are said to inhibit trade.
If the problem of private access barriers is as pressing as trade authorities have made out (which arguably has not been established), and if existing disciplinary tools are not totally adequate (which arguably has been established), then an international competition agreement of some type is worth exploring. However, as with all policy changes, the shift is only worth the effort if it produces benefits that exceed the cost of implementing the change, and produces better outcomes than existing disciplines. Part V of the article will examine the complementarities and divergences that exist between liberal trade policy and competition law and policy. Part VI of the article will examine some suggested solutions.
POINTS AT WHICH TRADE AND COMPETITION DIVERGE
It is undeniable that there is a philosophical and practical interaction between liberal trade policy and a welfare driven competition policy. Philosophically both policies are derived from the proposition that markets, operating where possible without artificial constraints, have the capacity to deliver significant welfare gains and raise standards of living. In both cases a notion of market access is a key element in realising that object. Practically there are clear points of reciprocal interaction. Restrictive trade practices may frustrate a liberal trade policy, just as a protectionist trade policy can prevent strong domestic competition laws from fully delivering expected welfare benefits. Thus, a domestic cartel may operate to effectively exclude imports from the domestic market even though import barriers are otherwise low. On the other hand, high import barriers, by deterring foreign firms from entering the domestic market, may deprive consumers of many of the benefits they might have expected from a legal regime that otherwise promoted vigorous competition. The creation of the European Common Market in 1957 — and its subsequent success — stands as an example of the dependent relationship of free trade and competition policy.
However, care must be taken not to overstate the synergies between trade and competition policy. There are some critical points of departure. Generally speaking, competition authorities understand competition as a mechanism for enhancing efficiency. Although there are numerous points of disagreement between competition authorities (for example, as to how the competition mechanism operates and how the various types of efficiency should be prioritised), there is general consensus that competition is not pursued as an end in itself. Competition is simply a process. Therefore, the notion of ‘market access’ has a particular meaning to competition authorities. Where a market is already competitive, competition policy is largely neutral on the issue of new entrants. The marginal benefit derived from ensuring access of one new entrant to an already competitive market is not the concern of competition law.
Market access very often means something different to trade authorities. This springs from the fact that whilst trade theory is dominated by the ethic of free trade, trade practice is often profoundly protectionist. In contrast to the process-oriented nature of competition law, trade policy is very outcome‑oriented. Trade authorities broker deals on market access. The touchstone is the deal, not some objective standard such as consumer welfare. Traders and trade authorities, therefore, view market access in terms of competitors’ rights. Whilst theoretically this means the right to an opportunity to enter a foreign market, in practice this often becomes a right to enter a foreign market. Whether that market is already competitive is not germane to the issue of whether a state, by failing to discipline private access barriers, is failing to honour its market access undertakings — that is, honouring its trade deal. From the trade perspective, the importance of access barriers is not the effect they may have on competition, but the effect they have on the integrity of the deal. The trade conception of market access is not necessarily about enhancing welfare. This has led some trade commentators to completely reject the application of competition principles to the trade problem, simply because competition principles will direct attention away from ensuring market access towards inappropriate notions of efficiency.
Equally, for many competition authorities, the danger in allowing the competition law notion of ‘contestable markets’ to become synonymous with the trade notion of market access is that the emphasis will swing from the process of competition and the pursuit of efficiency to competitors’ rights of access. Thus, competition policy has been criticised for ignoring harm to foreign producers. US competition authorities, for whom competition law is very much dominated by the consumer welfare norm, are particularly wary of this attempt to shift the focus of competition analysis. There is great resistance amongst US competition authorities to any process that might change US competition standards.
Furthermore, the structures and processes of international trade regulation and competition law regulation are quite different. Whereas trade has developed an international and unified structure for handling a broad range of disputes, the enforcement of competition rules has remained domestic and diversified. The result is that competition law has as many dispute resolution structures as there are states with a competition law to enforce. Differences abound in procedural requirements, party rules, burdens of proof and appeal processes. Additionally, competition law remedies are more severe, complex and far-reaching than trade remedies. This is due mostly to the fact that competition law (and hence competition remedies) are directed at the private firm, whereas trade law and trade remedies (with limited exceptions such as the anti-dumping remedies) are directed at the state. Because remedies are severe, competition enforcement has traditionally been more judicial in its orientation than trade. However, this may change as WTO dispute resolution mechanisms evolve. In summary, the problems involved in adapting competition tools to trade requirements should not be understated.
Finally, it has to be remembered that not all competition problems are trade problems and vice versa. The regulation of multinational mergers and global cartels are not necessarily matters of direct concern to the international trading system. While the increase in international trade has clearly been a catalyst in the growth of multinational mergers and global cartels, and whilst, in turn, there seems little doubt that multinational mergers and global cartels have the capacity to affect international trade flows, generally multinational mergers and global cartels are not concerned primarily with market access. Equally, the role of environmental and labour standards in building market access barriers is not normally regarded as an issue for competition law. To an even lesser extent are preferential bilateral trade agreements considered to be matters for competition policy. Therefore, the intersection between trade and competition is not the sole dynamic shaping either domain.
When constructing a rule to regulate market access it is necessary to consider not only the shape of the substantive rule but also the shape of any enforcement mechanism. There are three broad options available, namely, unification, harmonisation and unilateralism. These options are by no means discrete. Rather, they describe a continuum of possibilities, bounded at one end by a comprehensive global competition code enforced by a supranational enforcement agency (such as the WTO), and at the other end by prevailing national laws with some cooperation between domestic enforcement agencies (the status quo). It is to be expected, therefore, that many suggested solutions will contain elements drawn from more than one option.
The choice of option is generally informed by a variety of considerations. Thus, the rejection of the unification option may be based on normative considerations (for example, a belief that diversity represents a better policy choice than unification); institutional considerations (for example, a belief that the WTO, the most likely global institution to take responsibility for a comprehensive code, may be inappropriate for the task or may be seriously damaged by taking on that responsibility); and/or political considerations (for example, a belief that the process of negotiating any comprehensive code will entail unacceptable political compromises).
This article will look at some of the more important proposals as they affect the trade problem.
A comprehensive agreement to unify competition rules supported by a supranational dispute resolution mechanism would see market access treated globally as it is presently treated domestically, that is, as part of a unified legal structure dealing with a full range of global competition issues. The pre-eminent example of this kind of unification is the EU. EU competition law is activated whenever trade between Member States is affected. Wholly domestic matters are dealt with by national laws.
However, no one views the EU as a template for global regulation. Indeed, comprehensive unification in any form does not attract much support. The OECD Joint Group on Trade and Competition has dismissed a comprehensive code as undesirable. The EU has recently sought to dispel the notion that it favours comprehensive rule unification. The US has consistently rejected this approach as an option. Many developing states have only just acquired a competition regime. To expect those states to be in a position to negotiate a set of international rules before they have had the opportunity to ‘bed down’ domestic rules is unrealistic. Further, many developing states are very wary of any process to develop a global regime.
The Basic Telecommunications Agreement annexed to the GATS has been suggested as a possible model for a more comprehensive competition agreement. An annex to the Basic Telecommunications Agreement, known as the Reference Paper, contains certain competition provisions. The aim of the Basic Telecommunications Agreement is to liberalise access to central telecommunications networks. In many states such networks are controlled by monopolies, or, at the very least are characterised by the existence of substantial market power. Those states that have committed to the Reference Paper must maintain ‘appropriate measures’ for the ‘purpose of preventing suppliers who, alone or together, are a major supplier from engaging in or continuing anti‑competitive practices’. A ‘major supplier’ is defined. The Reference Paper lists three types of anti-competitive practice:
(a) engaging in anti-competitive cross-subsidization;
(b) using information obtained from competitors with anti-competitive results; and
(c) not making available to other services suppliers on a timely basis technical information about essential facilities and commercially relevant information which are necessary for them to provide services.
This list is not meant to be exhaustive.
To date there exists very little WTO jurisprudence on the Basic Telecommunications Agreement. Recently, a WTO panel examined some of those provisions in respect of Mexico’s regulation of telecommunications services. However, this is the first and only case so far. It may well be that evolving WTO jurisprudence on the Basic Telecommunications Agreement will influence future discussions on global competition. However, for the reasons discussed immediately above, it is suggested that while the Basic Telecommunications Agreement could provide a useful model for developing further sectoral competition obligations, it is unlikely to be seen as a model for a comprehensive agreement.
There is a tension between the content of rules and their enforcement. The more comprehensive the agreement on rule uniformity, the weaker the enforcement mechanism will be. Conversely, where an effective enforcement mechanism is sought, agreement on substantive rules is likely to be minimal. Of these two general pathways the latter is more widely preferred. Consequently, if any rule unification is to occur, it is likely to be at a minimum level. In other words, while it may be possible to unify some competition rules, these rules will be situated within what is essentially an enforcement structure. This weaker form of unification has sometimes been called the ‘convention approach’.
A model that generally fits this description is an agreement along the lines of the TRIPS Agreement. The TRIPS Agreement contains a minimum level of rule unification, a broader range of harmonisation principles and a form of multilateral enforcement mechanism to ensure compliance. The TRIPS Agreement, however, was fashioned against a background quite dissimilar to that which faces competition negotiators. Firstly, there was a long history of treaty cooperation on intellectual property matters, which meant that some of the TRIPS Agreement obligations had already assumed the status of international norms. More importantly, it meant that intellectual property regulation was already accepted as a matter appropriate for international agreement. Secondly, the US and the EU were largely in accord on the need for a global agreement, and Japan was not opposed. In the case of a competition agreement no such accord exists. While the EU supports a competition agreement, the US does not. Thirdly, large business interests had powerful incentives to lobby for the TRIPS Agreement. The same does not apply to a competition agreement. Fourthly, the proponents of the TRIPS Agreement were able to achieve widespread support partly by the threat of US unilateral action, but also by offering certain trade-offs to the developing states (for example, promises to lower tariffs on textiles, clothing and agriculture, as well as promises on capacity building and technical assistance). As many of these trade-offs have failed to eventuate, it is doubtful that those who view a competition agreement with scepticism will be so easily bought this time. Finally, developing states are disillusioned by the manner in which developed states have sought to enforce the TRIPS Agreement for their own benefit to the cost of the developing states.
A more targeted — but in some ways more ambitious — proposal is a dedicated trade rule, in other words, a foreign market access rule.
As complaints about private access barriers include both horizontal and vertical restraints, an effective market access rule must tackle both. Some progress has been made in respect of horizontal restraints, for example, the OECD has adopted a non-binding recommendation against hard core cartels. However, at the margin there is in reality little consensus on precisely what constitutes a hard core cartel. The OECD Recommendation permits individual states to exclude from the rule any cartel they wish. Moreover, the Recommendation makes no attempt to define the policy circumstances in which states may authorise cartel conduct. This remains a matter wholly within the discretion of each state. Thus, the recommendation permits states to exempt export cartels, import cartels, rationalisation cartels, depression cartels and managed trade cartels. It may be questioned whether ultimately the OECD Recommendation has any real substance. As the developed states have failed so far to reach agreement, and as the developing states have even less incentive to agree to a ban, the imminent prospects for a binding global rule are not good.
Vertical restraints are considerably more controversial than horizontal restraints. Only rarely are non-price vertical restraints prohibited per se. They are generally assessed against some standard. The standard is often a complex balance of sometimes competing policies (for example, competition versus cooperation, consumer welfare versus total welfare, economies of scale versus small business protection, allocative efficiency versus innovation). Almost inevitably the standard — in practice if not textually — varies from country to country, and the consequences of its application from case to case. US competition authorities currently adopt the attitude that most non-price vertical restraints are efficiency enhancing and therefore generally not unlawful. EU competition authorities take a broader approach which recognises that a reduction in competitive conditions may outweigh short term efficiency gains. For example, the EC, in pursuit of its obligation to create a common market, has constantly sought to prohibit territorial restraints which enforce discriminatory market segmentation; otherwise powerful firms can extract monopoly profits by preventing the export of their products from a low price destination (one state of the EU) to a high price destination (another state of the EU).
Unifying these divergent standards will require significant compromises to be made. The loss of diversity may be hard to justify given the doubts that exist over the magnitude of the trade barrier problem. There is little evidence that states are ready to compromise on exclusionary vertical restraints. Furthermore, having regard to the differences in the way in which trade and competition authorities view market access, a market access rule along competition law lines may not ultimately satisfy trade authorities. Alternatively, a strong trade-directed, market access rule may in practice disproportionately favour exporters and net exporting states. This may be politically difficult to sell to domestic producers in both developed and developing states. It may have significant adverse implications for developing states and the growth of new industries.
For all these reasons commentators in favour of a market access rule have necessarily searched for a minimalist solution.
A number of suggestions for a dedicated market access rule have centred on the notion of consumer welfare. The rationale for this is that consumer welfare considerations play a role in all competition regimes. Under a consumer welfare proposal, business conduct would amount to an actionable restraint on market access if: (1) it resulted in higher prices (or reduced output) than would otherwise prevail; and (2) there was a provable connection between the conduct and a discriminatory effect on imports. Generally, the first step for a complaining party would be to engage the local legal system — that is, the local competition law. Review would be to a WTO panel, although commentators differ on the standard of review and whether the WTO ruling should be binding. Commentators have devised a range of checks and balances in an attempt to ensure that any rule is not abused. For example, the rule might only apply where imports represented less than a certain percentage of the domestic market.
Aside from the obvious problem of actually achieving consensus on a substantive rule, a rule that rests on a consumer welfare standard measured solely in terms of price competition (a static allocative efficiency model) is normatively suspect. It ignores the dynamic welfare-creating possibilities of innovative efficiency. In many situations it may be more important in welfare terms to permit some sub-optimal pricing in return for more innovation. To the extent that any market access regime forced nations to reject a policy that preferred a trade‑off between innovation and competitive pricing — let alone other factors that local communities may regard as policy imperatives — it seems to be an unwarranted intrusion upon good economics, national sovereignty and proper policy development. In any event, a rule based on consumer welfare is unlikely to satisfy trade commentators. One of the major complaints made by traders against competition law and policy — particularly by US traders against US antitrust law and practice — is that consumer welfare is an inappropriate and ineffective tool where foreign markets are foreclosed by private access barriers. Additionally, if the dispute resolution mechanisms of the WTO have put the whole trading regime under strain, market access disputes to be adjudicated on contentious notions of allocative efficiency are not likely to defuse those tensions. This cannot be good for the WTO.
If a market access rule based on consumer welfare is unlikely to succeed, is it possible to construct another form of market access rule using competition principles? The most obvious solution is to concentrate on the process of competition (maintaining a level of rivalry) rather than on the outcomes. Perhaps the critical substantive difference between US and EU competition policy is that while US policy is firmly fixed on outcomes measured in terms of consumer welfare, EU policy gives greater prominence to the process of maintaining competitive rivalry. This distinction has been evident in US and EC approaches to merger regulation, particularly the Boeing/McDonnell-Douglas merger and the GE/Honeywell merger. The distinction is captured in the oft‑repeated claim by US authorities that US antitrust law and policy protects competition not competitors. In the hands of competition authorities the distinction is important, but hardly reflects an unbridgeable chasm between systems. However, were the EU standard to become the basis for a trade rule, there is a real possibility that a chasm would open. Reflecting the trade view of market access discussed above, the balance that presently exists in EU competition law and policy between rivalry and efficiency would tilt significantly in favour of rivalry. In other words, the trade notion of market access rather than the competition notion would become dominant. It is difficult to see that US antitrust authorities would tolerate a rule that is so opposed to current US competition policy.
A more directly trade-focused approach has been suggested by some US trade commentators. Trade concessions would be accompanied by an ‘explicit warranty’ clarifying that the state will not permit the market access concession to be undermined by ‘private restraints of trade, private collaboration with government bodies, and/or informal administrative guidance’. Examples of market access restraints include ‘price fixing, bid-rigging, monopolising essential port or distribution facilities, and actual or threatened refusals to deal’. Liability would be strict in that it would not depend on any effect the barrier may have on the operation of the market. Breach of this warranty would constitute grounds for seeking compensation for the concession thus impaired.
It is not at all clear what the rule involves, nor how it would work in practice. This suggestion is very imprecise in detailing the nature of those private market access restraints which would trigger the trade remedy. According to the proponents of this rule, private market access restraints are not to be determined by competition law. However, if they are not to be identified by existing competition law and policy principles, how are they to be identified? Ultimately, it may simply be a matter of the exporting state asserting the existence of a barrier. As articulated by its proponents, the rule does not depend on discrimination between local produce and imports. Therefore, the rule goes beyond existing trade remedies such as arts III(4) and XXIII(1)(b) of the GATT. There is absolutely no evidence that states are ready to support a multilateral rule of this type. On the contrary, the rule looks more like a trigger for unilateral action under a mechanism such as § 301 of the US Trade Act of 1974.
Unlike unification, harmonisation does not eschew diversity. The goal of harmonisation is the creation of a relationship of accord or consonance between diverse elements so as to achieve a particular object. What is an appropriate method of harmonising diverse domestic competition rules and practices where the object of the harmonisation is to further the requirements of liberal trade policy? Most commentators who have looked at this issue have stressed the role of competition law enforcement. In other words, the way to advance the object of free trade is to ensure an appropriate level of enforcement of domestic competition law. Appropriate enforcement tends to mean that imports should not be subjected to discriminatory treatment by local authorities. Thus, a non‑discrimination principle, already present in international trade law, provides the baseline substantive rule to which agreed enforcement procedures are added to make a workable harmonisation programme.
One way of mediating the interface between diverse legal regimes to achieve outcomes that are tolerably consistent with the aspirations of free trade is to adapt the existing GATT–WTO rules to ensure that they cover private barriers. For example, art XXIII(1)(b) could be amended so that it explicitly extends to a state’s failure to enforce its competition laws. As previously discussed, there are a number of reasons why art XXIII(1)(b) is not presently an appropriate instrument for regulating private barriers. Firstly, the notion of a government ‘measure’ does not cover all occasions of a state’s failure to discipline private barriers. Secondly, it is difficult for the complaining state to prove that it could not reasonably have expected the importing state’s failure to discipline the relevant private barriers.
A solution, therefore, is an agreement that tackles these problems. Thus, the meaning of ‘measure’ needs to be defined so that it unambiguously extends to government toleration of private access barriers. Furthermore, exporting states might presumptively be entitled to expect importing states to enforce their competition laws according to their text (or perhaps according to practice). Indeed, some commentators have gone further and suggested that the existence of a private access barrier might by itself be viewed as contrary to expectations, whether or not the barrier is contrary to law.
The apparent virtue of this suggestion is that it concentrates on finding a solution within existing instrumental and institutional structures by filling the ‘gaps’ in the GATT–WTO trading system. However, there are problems. Firstly, it cannot be assumed that the ‘gaps’ represent anything less than areas of activity over which states have not yet indicated any willingness to cede sovereignty in return for trade gains. The ‘gap-filling’ exercise therefore goes beyond being a matter of clarification. To extend the definition of ‘measure’ to ensure that it captures instances of toleration will probably become an exercise in constructing a whole new market access rule.
Secondly, introducing a presumption that states will enforce their competition laws according to their text could considerably constrain a state in its valid policy choices. Because competition law is normally complex, and because competition cases are usually very costly, competition authorities often cannot pursue all cases. Deciding which cases to prosecute is itself an important element of competition policy. Given that most competition cases do not involve any foreign party, it is quite possible that cases involving private trade barriers will not be prosecuted, not because of any policy of discrimination against foreign producers, but simply as a matter of making rational policy choices. Further, it is to be expected that a state will need to adjust or realign its enforcement priorities from time to time. For example, a state may quite validly determine to divert some of its limited resources from the pursuit of cartels to monopolistic abuses or merger control. In this sense, a presumption that states will enforce their competition laws according to their text, or according to current practice, does not reflect the reality of competition law enforcement.
Why might this be important? After all, the exporting state will still have to establish that the measure discriminated against imports; Japan — Film demonstrated that this will be difficult to do. Nevertheless, discrimination could occur, and it may occur without any intention to do so. For example, a state may determine, for perfectly rational yet largely domestic reasons, that it should allocate its enforcement resources in the area of cartels. This means that fewer enforcement resources are expended on distribution restraints. The result of this realignment of resources might be that imports are disproportionately affected; not as a direct policy choice, but as a necessary incident of a rational policy choice. In Japan — Film, the WTO Panel held that discrimination for the purposes of art XXIII(1)(b) may be de facto. The Panel also noted that proof of intention to discriminate was not required.
Therefore, to ensure that any rule does not overly restrict a state’s freedom to choose its own domestic policies, it has been suggested that a non-violation claim based on toleration of anti-competitive practices should only lie where the impugned measure (that is, the change in competition law or practice that results in trade barriers not being disciplined) erodes the state’s prior tariff commitment in a manner that is disproportionate to the measure’s purported benefits. In other words, the erosion of the exporting state’s reasonable expectations must now be balanced against the legitimate benefits to the importing state of the measure. This effectively means that WTO panels must make judgments about the relative effects and merits of domestic policies. A standard of review based on such interest-balancing is problematic. In particular, how does a WTO panel determine what weight to give to a domestic policy? A less intrusive test might be to empower WTO panels to assess whether the state could have achieved its legitimate policy objectives with less trade restrictive measures. Although such a standard of review may have an inherent tendency to become trade-centric, it is likely to be more acceptable than an open-ended balancing test.
Finally, there are institutional doubts. Extending the range of non-violation complaints to failures to discipline private restraints could radically increase the number of WTO actions under art XXIII(1)(b), particularly if such actions are viewed as strategically useful in leveraging market access. As competition cases are often complex and lengthy, this will put an increased burden on the WTO adjudication system. Many believe the system may not be ready for that burden. Whilst many states may prefer to concede rather than fight a costly and complex WTO battle, this in itself raises the problem of institutional fairness and may lead to an increase in the level of managed trade. Additionally, there remains the problem of remedies.
Instead of seeking to adapt existing GATT–WTO rules, nations might simply agree on a new procedure for the application, oversight and review of domestic competition laws so as to ensure that they are not applied in a manner that discriminates between local and foreign producers. A possible procedure might be as follows.
Firstly, an exporter claiming to have been harmed by anti-competitive exclusionary activities in the importing state should first seek redress through the importing state’s legal system. To facilitate this, nations must agree to provide non-discriminatory access to their local competition law and policy processes (both administrative and judicial), to provide adequate procedural rules (for example, discovery rules), and to ensure adequate transparency. Where the process is initiated by a request to the importing state to take action to enforce its laws, the importing state must, in keeping with notions of positive comity, give sympathetic consideration to the request. The exporter must be treated in exactly the same manner as local producers.
Secondly, if the importing state refuses or fails to act, the exporting state is entitled to claim personal jurisdiction over the offending firm(s) so that the law of the importing state could be enforced by an action brought in the courts of the exporting state. Alternatively, the refusal or failure to act would entitle the exporting state to an immediate review by some supranational authority, such as a WTO panel.
Thirdly, if conducting an action against the offending firm(s) in the exporting state would be pointless (for example, where no realistic opportunity to enforce a judgment existed), or if a defence of act of state or state compulsion would apply, the exporting state would be entitled to request a hearing by the supranational authority.
As with the model based on adapting the GATT–WTO rules, there is a need to establish discrimination. Again this may prove difficult. There is arguably a significant difference between detecting discrimination in respect of, for example, patent registration or the imposition of governmental protectionist trade measures, and detecting discrimination in the application of competition policy. For example, the Japanese keiretsu system may look like disguised protectionism to a US trade representative, but to a Japanese Government official it may be no more than a natural expression of Japanese socio-economic culture. Should the keiretsu system be challenged as discriminatory? The need to establish discrimination is not a weakness in this suggestion. It is in fact the core element. If discrimination presents substantial forensic difficulties, then that merely indicates the inherent complexity of the whole issue.
Not all private access barriers are discriminatory in character. It is just as likely that they apply equally to locals as well as foreigners. Indeed, it has been argued that most private exclusionary restraints on trade are not discriminatory. Therefore, failure to sanction such barriers is not an exercise in discriminatory domestic enforcement. An agreement based on national treatment principles would have no operation in such cases. Additionally, a non‑discrimination agreement will face the same problem as that discussed in relation to amending art XXIII(1)(b) and which applies equally to the national treatment commitment under art III(4) of the GATT: is it possible to create a test for state discrimination which would extend to tolerating private access barriers without creating a whole new market access rule?
A further problem with this model — and indeed, most others — has been the inability to achieve consensus on the nature and role of a supervisory authority. Indeed, the nature of the review process has been a matter of considerable confusion and controversy, and there is a reluctance to commit an international authority to a review of individual cases. This is understandable, as not only does such a course of action elicit strong political opposition, it also appears to impose an onerous task upon WTO panels. A WTO panel reviewing an individual case would be required to determine not only whether there was a breach of the importing state’s competition law, but also whether the failure to discipline the breach discriminated against imports. However, if the international authority does not review individual cases, what does it do? If its only task is to determine whether a state’s competition law regime is textually and structurally non‑discriminatory, then arguably it will have no operation in the majority of instances where hybrid trade barriers are said to exist. One suggestion is that the international authority have power to examine alleged patterns of failure to enforce competition laws. However, it is difficult to see how a pattern can be detected without looking at individual cases. Additionally, it is difficult to envisage what meaningful order could be made in respect of such a proven pattern. Ultimately, therefore, either the international authority has power to make a case‑by‑case analysis (which most agree is inappropriate), or the whole review process may be ineffective. A possible compromise is an international competition authority with powers to initiate actions under domestic law coupled with an agreement to introduce private rights of action under domestic law. This could significantly reduce administrative bias against foreign interests without unduly impinging on national sovereignty.
Whilst serious doubts exist about the feasibility of multilateral harmonisation, a low level of harmonisation already exists in the form of bilateral cooperation agreements. These agreements, however, are better viewed as a form of cooperative unilateralism. The amount of cooperation is generally determined not by any global considerations (for example, advancing trade liberalisation or global welfare), but by the requirements of effective domestic enforcement. Ultimately, the principal objections to cooperative unilateralism are that the agreements are bilateral, non-binding, and there is no international oversight. Cooperative unilateralism nevertheless remains the preferred option of the US.
Aside from the sovereignty issue, one of the main objections to any form of unification or harmonisation is the likely role of the WTO. A common theme running through the unification and harmonisation proposals is that there must be some international dispute resolution process. The most obvious candidate for this role is the WTO.
The WTO naturally presents itself because of its nearly universal membership, its commercial flavour, and its existing dispute resolution mechanisms. The OECD has a severely restricted membership of developed states only, and no comparable history of dispute resolution. UNCTAD is perceived by the developed states as being dominated by the less developed states and for that reason has been sidelined, although the recent failure at Cancún may reinvigorate the role of UNCTAD. The International Court of Justice lacks any enforcement mechanism. A new dedicated competition forum with a binding dispute mechanism would be difficult to achieve. Suggestions that new international institutions be formed are often met with extreme hostility. This seems particularly apt considering recent popular sentiments towards the WTO.
Although the WTO presents as a convenient forum, there is considerable debate as to whether its dispute mechanisms are well adapted to handling competition matters. As previously mentioned, there is a perception in some quarters that the decision-making processes of the WTO are likely to favour trade norms at the expense of other goals. Even if there is no trade bias, there is concern that WTO panels will be ill-equipped to make proper assessments about national laws and the performance of national competition authorities. For example, some have doubted that a WTO panel would be competent to judge whether a discretion has been appropriately exercised by a domestic regulator. Some doubt that the WTO dispute resolution system is able to handle factually intensive and complex competition cases. There is also apprehension that creating a WTO competition-based agreement would simply result in a new form of WTO-sanctioned trade regulation. There are a number of other concerns that have been expressed in response to the suggestion that the WTO be given responsibility for global competition matters. A full analysis of this issue, however, is beyond the scope of this article. Rather, the purpose of the discussion has been simply to point out that while the existence of the WTO has made speculation about competition rule unification or harmonisation a realistic undertaking, many questions remain to be answered about the role of the WTO before it can be said that a unification or harmonisation choice is free of institutional issues.
There is evidence to support the claim that private market access barriers are restricting the growth of international trade. This problem is expected to increase as business becomes more globalised and as tariffs continue to be reduced. Some of the most effective private access barriers — both horizontal and vertical — are said to be anti-competitive restraints. Existing unilateral, bilateral and multilateral devices have had only limited success against these barriers.
Domestically, anti-competitive conduct is regulated by competition law. The number of states with a competition law has grown significantly in the last decade. Consequently, trade commentators view competition law as an appropriate tool for disciplining private access barriers. However, competition law, in the eyes of many trade authorities, is not being properly enforced against such barriers. On the contrary, so it is argued, governments have encouraged or tolerated market blockage for protectionist reasons. However, there may be other reasons why governments have failed to act. Firstly, a number of states simply do not have the requisite capacity to run a sophisticated competition law regime. Secondly, and more importantly, competition law does not have precisely the same objectives as trade policy. Consequently, its principles and processes are not the same as trade. Although both disciplines use the language of market contestability, they do not necessarily have the same goals in mind. Even where a state has the capacity and the will to use its competition law against a trade barrier, the barrier may not be anti-competitive. It may not be unlawful. Thus, competition authorities are very wary of attempts to employ competition rules to achieve market access for imports.
There have been many suggestions for solving the apparent trade problem. Unilateral tools are useful, but there is always the risk that they will provoke hostility. In a hostile environment unilateralism is ultimately inefficient and politically dangerous. However, unilateral tools can be improved through the creation of an environment of cooperation, and there are a number of bilateral agreements built around the notion of comity that seek to foster that cooperation. Administrative authorities, perhaps because they are unable to offer anything more, have tended to praise, and even champion, the role of these comity agreements. However, comity is as much an aspiration as a concrete structure. It works only as long as the parties share a common view. Consequently, it has proved useful in regulating global cartels, but it has had little effect on distribution restraints and export cartels. Further, those agreements that exist tend to be between states of similar economic and social background. Thus, while bilateral agreements will continue to do much to aid the effective operation of domestic competition law regimes, ultimately there needs to be something more.
Most commentators have now backed away from the ideal of a uniform world competition law operated by a world competition authority. Many concerned with the trade problem have investigated the possibility of a trade‑related rule — either a dedicated market access rule or a less ambitious non-discrimination rule. However, these lesser forms of multilateralism have also struggled for acceptance. The non-discrimination approach is essentially a minimal harmonisation scheme: in its various versions it seeks to find some way of mediating domestic competition regimes by utilising a principle of non‑discrimination between locals and foreigners. However, de jure discrimination is probably rare and de facto discrimination is hard to prove, particularly where it is based on government inaction. Moreover, the national treatment commitments in existing WTO agreements already cover most cases of discrimination. In addition, these schemes have tended to founder on the requirement that some international authority be empowered to enforce the non‑discrimination rule. The only institution that could realistically take on that role is the WTO, which brings institutional challenges to the problem that make it even more intractable. The developing states remain suspicious of the whole competition debate and the US remains opposed to a multilateral solution. These two factors make it difficult to envisage a multilateral agreement in the near future.
[*] BCom, LLB (Melbourne); Barrister and Solicitor of the Supreme Court of Victoria, Senior Lecturer, Department of Business Law and Taxation, Faculty of Business and Economics, Monash University.
 The WTO was established by the Marrakesh Agreement Establishing the World Trade Organization, opened for signature 15 April 1994, 1867 UNTS 3 (entered into force 1 January 1995) (‘Marrakesh Agreement’). On the outcome from the WTO Ministerial Conference at Cancún, see WTO, Briefing Note, Day 5: Conference Ends Without Consensus (2003) <http://www.wto.org/english/thewto_e/minist_e/min03_e/min03_14sept_
e.htm> at 1 October 2004.
 Marrakesh Agreement, above n 1, annex 1A (General Agreement on Tariffs and Trade) 1867 UNTS 190 (‘GATT’). Competition law provisions were included in the Havana Charter for an International Trade Organization (‘Havana Charter’), the proposed charter for the International Trade Organization (‘ITO’). See art 46 of the Havana Charter, in United Nations Conference on Trade and Employment, Final Act and Related Documents, UN Doc E/CONF.2/78 (1948) (the conditions for the entry into force of the Havana Charter, set forth in art 103, were not fulfilled within the prescribed time limit. The ITO failed to obtain US congressional approval and never came into effect). See American Bar Association Sections of Antitrust Law and International Law and Practice, Report on the Internationalization of Competition Law Rules: Coordination and Convergence (1999) American Bar Association, 4 <http://www.abanet.org/antitrust/comments/2000/convexe.html> at 1 October 2004 (‘ABA Internationalization of Competition Law Rules’). See John Braithwaite and Peter Drahos, Global Business Regulation (2000) 177. See also Spencer Weber Waller, ‘Neo-Realism and the International Harmonization of Law: Lessons from Antitrust’ (1994) 42 University of Kansas Law Review 557, 558–9.
 Competition law and policy has been debated at various international fora including the Organisation for Economic Co-operation and Development, the United Nations Conference on Trade and Development (‘UNCTAD’) and the Asia-Pacific Economic Cooperation (‘APEC’).
 Singapore Ministerial Declaration, WTO Doc WT/MIN(96)/DEC (1996) .
 Doha Ministerial Declaration, WTO Doc WT/MIN(01)/DEC/1 (2001) .
 Doha Work Programme, WTO Doc WT/GC/W/535 (2004) [1(g)] (Draft General Council Decision). The 2002 Annual Report of the Working Group on the Interaction between Trade and Competition Policy (‘WGTCP’) at the WTO revealed considerable disagreement on nearly all the Doha issues: See Report (2002) of the Working Group on the Interaction between Trade and Competition Policy to the General Council, WTO Doc WT/WGTCP/6 (2002).
 The reasons most often proffered as justifying a global competition agreement include: (1) the interaction between competition law and policy and international trade liberalisation; (2) the inability of domestic regimes to regulate many commercial activities that take place at a multi-state level; (3) the friction caused by the clash of different competition systems arising from the application of domestic laws extraterritorially; (4) the duplication of costs arising from compliance with different systems, particularly but not solely in respect of merger reporting; and (5) the global welfare deficit that results from domestic regulation (assuming that maximisation of social welfare is the regulatory goal, domestic regulation is sub-optimal when compared to global regulation). See generally Sir Leon Brittan and Karel Van Miert, ‘Towards an International Framework of Competition Rules — Communication to the Council’, Doc No COM(96)284; Daniel Tarullo, ‘Norms and Institutions in Global Competition Policy’ (2000) 94 American Journal of International Law 478; Eleanor Fox, ‘Toward World Antitrust and Market Access’ (1997) 91 American Journal of International Law 1; Andrew Guzman, ‘Is International Antitrust Possible?’ (1998) 73 New York University Law Review 1501; Andrew Guzman, ‘Antitrust and International Regulatory Federalism’ (2001) 76 New York University Law Review 1142.
 Export cartels also distort patterns of trade. A discussion of export cartels is beyond the scope of this paper.
 The concern with anti-dumping is not limited to the developing states. Japan has also argued that anti-dumping rules are a core issue: See International Competition Policy Advisory Committee to the Attorney General and the Assistant Attorney General for Antitrust, United States of America Congress, Final Report (2000) 266 (‘ICPAC Report’).
 Any discussion on the role of mergers in facilitating the existence and effectiveness of private trade barriers is beyond the scope of this article.
 See ABA Internationalization of Competition Law Rules, above n 2, 17. See also Eleanor Fox, ‘What is Harm to Competition? Exclusionary Practices and Anticompetitive Effect’ (2002) 70 Antitrust Law Journal 371.
 See WGTCP, Communication From the European Community and Its Member States,
WTO Doc WT/WGTCP/W/115 (1999) 4 (Submission from the European Community and its Member States to the WGTCP).
 A nation’s competition law also depends on judicial factors. This is particularly true of the US, but applies wherever the judiciary is the final interpreter of legislative provisions.
 See Bernard Hoekman and Peter Holmes, ‘Competition Policy, Developing Countries and the World Trade Organization’ (Policy Research Working Paper No WPS2211, World Bank, 1999). See also I M Destler, ‘US Approach to International Competition Policy’  Brookings Trade Forum 395, 396–8; Bernard Hoekman and Petros Mavroidis, ‘Economic Development, Competition Policy, and the World Trade Organization’ (Policy Research Working Paper No WPS 2917, World Bank, 2002). The nature of competition policy is discussed in the Report (1998) of the Working Group on the Interaction between Trade and Competition Policy to the General Council, WTO Doc WT/WGTCP/2 (1998) .
 See Dianne Wood, ‘Cooperation and Convergence in International Antitrust: Why the Light is Still Yellow’ (Speech delivered at the The New Antitrust Paradox: Policy Proliferation in the Global Economy Conference, American Enterprise Institute for Public Policy Research, Washington DC, 22 April 2003) <http://www.aei.org/news/newsID.17274/news_detail.asp> at 1 October 2004. Arguably, Japan has always treated competition policy as an aspect of industry policy: See Harry First, ‘Antitrust Enforcement in Japan’ (1995) 64 Antitrust Law Journal 137, 142–5, arguing that Japanese authorities give a higher priority to protecting national economic interests than to protecting consumers. Competition policy can easily become part of industry policy where competition law enforcement is left in the hands of bureaucracies. Where a competition law regime has an unfettered private right of action (for example in the US), it is much more difficult for special interests to capture competition policy.
 This is particularly true of the US and some other Western developed states. However, even in these states there are a variety of other objectives, including fair trade, wealth distribution, social and environmental policy, and protection of national industry.
 See, eg, Joseph Stiglitz, Two Principles for the Next Round: Or, How to Bring Developing Countries in from the Cold (1999) 6–13 <http://www.worldbank.org/knowledge/chiefecon/articles/geneva.pdf> at 1 October 2004. See also Robert Hudec, ‘A WTO Perspective on Private Anti-Competitive Behavior in World Markets’ (1999) 34 New England Law Review 79, 82. This is based on the argument that the liberalisation of trade will increase the wealth of nations. However there are caveats, particularly in respect of the needs of the developing world. See Joseph Stiglitz, ‘Development Policies in a World of Globalization’ (2002) 4 <http://www-1.gsb.columbia.edu/faculty/jstiglitz/download/DevelopmentGlobalization.pdf> at 1 October 2004, warning that trade liberalisation has tended to be asymmetric with the developed states benefiting at the expense of the developing states.
 For example, Since World War II and the introduction of the GATT, the average (trade‑weighted) most favoured nation tariff rates have dropped from 40 per cent to 4 per cent: OECD, The Development Dimensions of Trade (2001) 10.
 See, eg, American Bar Association Sections of Antitrust Law and International Law and Practice, Report concerning Private Anticompetitive Practices as Market Access Barriers (1999) American Bar Association, 7 <http://www.abanet.org/antitrust/comments/2000/ma_rpt.doc> at 1 October 2004 (‘ABA Report on Market Access’). See also ICPAC Report, above n 9, 203–9.
 Procurement policies, and trade facilitation measures form part of the ‘Singapore issues’, which contributed to the failure of the Ministerial at Cancún: See WTO, Briefing Note, Day 5, above n 1.
 Where product and safety standards are mandated by government, they are better viewed as a government barrier. Where the standards are set by industry and the government’s role is one of encouragement or toleration, the barrier is better defined as a hybrid barrier.
 See, eg, the list of laws and regulations complained about by Kodak in its dispute with Fuji and the Japanese Government: Japan — Measures Affecting Consumer Photographic Film and Paper, WTO Doc WT/DS44/R (1998) (Report of the Panel) (‘Japan — Film’). This case is discussed at the text accompanying below n 172. It has been a regular complaint of the US and the EU that Japan uses a web of legislative provisions and bureaucratic guidance to nullify the effect of its competition law and permit private barriers to flourish. See generally James Southwick, ‘Addressing Market Access Barriers in Japan through the WTO: A Survey of Typical Japan Market Access Issues and the Possibility to Address Them through WTO Dispute Resolution Procedures’ (2000) 31 Law and Policy in International Business 923, 928–9.
 For a discussion on government barriers, see Shanker Singham and D Daniel Sokol, ‘Public Sector Restraints: Behind-the-Border Trade Barriers’ (2004) 39 Texas International Law Journal 625.
 See Report (1998) of the Working Group on the Interaction between Trade and Competition Policy to the General Council, above n 14, –. See also ABA Report on Market Access, above n 19, 8; ICPAC Report, above n 9, 203–5.
 See ABA Report on Market Access, above n 19, 7.
 Competition regimes contain a wide range of authorisations, exemptions and exclusions that directly or indirectly affect trade. For example, many countries have traditionally exempted certain sectors (including defence related matters, energy, transport and communications) from foreign competition. Authorisations, exemptions and exclusions vary from state to state. See generally Jürgen Basedow and Stefan Pankoke, ‘General Report’ in Jürgen Basedow (ed), Limits and Control of Competition with a View to International Harmonisation (2002) 1–62; OECD, Antitrust and Market Access: The Scope and Coverage of Competition Laws and Implications for Trade (1996) 7, 17–20; WGTCP, Overview of Members’ National Competition Legislation, WTO Doc WT/WGTCP/W/128/Rev.2 (2001) (Note by the Secretariat — Revision); Exceptions, Exemptions and Exclusions contained in Members’ National Competition Legislation, WTO Doc WT/WGTCP/W/172 (2001) (Note by the Secretariat).
 See generally ABA Report on Market Access, above n 19, 15–19; ICPAC Report, above n 9, 204–5.
 See Stiglitz, Two Principles, above n 17.
 See Hudec, A WTO Perspective, above n 17, 82–3.
 Ibid. See also Julian Epstein, ‘The Other Side of Harmony: Can Trade and Competition Laws Work Together in the International Marketplace?’ (2002) 17 American University International Law Review 343, 364–5.
 See Jagdish Bhagwati, ‘Afterword: The Question of Linkage’ (2002) 96 American Journal of International Law 126, 130. Failure to discipline private exclusive practices will force states into unilateral action. This is the very thing that the GATT–WTO system was designed to prevent. It is clear that market access issues were at the heart of the EU’s move to bring competition policy into the WTO. For a discussion on this issue, see Philip Marsden,
A Competition Policy for the WTO (2003) 161–8. Market access issues have also been the key factor driving calls by US trade interests for better US antitrust enforcement: See, eg, Alan Wolff, ‘The (Notionally) Bridgeable Chasm between Antitrust and Trade Policy’ (2003) 47 New York Law School Law Review 167. One of the most prolific academic commentators on trade and competition has been Eleanor Fox. Fox also sees market access (market blockage) as the issue that brings trade and competition together: See, eg, Fox, ‘Toward World Antitrust and Market Access’, above n 7.
 See OECD, Joint Group on Trade and Competition, Trade and Competition Policies: Options for a Greater Coherence (2001) 15–16 (‘OECD Report on Trade and Competition Policies’) commenting that the evidence of access barriers was contested. See also ICPAC Report, above n 9, 224–5, determining that the evidence ‘while uneven, is sufficient to show that private, governmental, and mixed public–private restraints that inhibit market access are a problem’. See also ABA Report on Market Access, above n 19, 24. See Tarullo, above n 7, 484–5, who argues that non-enforcement of competition laws has not yet been established as a significant barrier to trade, although it may become more important as more government monopolies are privatised. See also William Barringer, ‘Competition Policy and Cross Border Dispute Resolution: Lessons Learned from the US — Japan Film Dispute’ (1998) 6 George Mason Law Review 459, 474. But see Epstein, above n 30, 362 citing a 1996 OECD report, Antitrust and Market Access: The Scope and Coverage of Competition Laws and Implications for Trade (1996) as evidence that private barriers are a matter of concern.
 See ICPAC Report, above n 9, 211–19, discussing specific instances brought to the attention of the Advisory Committee by US business interests, business associations and economists concerning access barriers in Japan, the EU, Latin America and elsewhere. See also WGTCP, Annual Reports to the General Council, WT/WGTCP/1–WT/WGTCP/7
(1998–2003). For example, US and European interests have long held the opinion that Japan’s huge trade surpluses must result from substantial trade barriers having regard to its low average tariff rates and low quantitative import restrictions: See Southwick,
above n 22, 923. Indeed, US exporters have long held the view that there are deep cultural forces at work in Japan to exclude foreign imports: See Sara Dillon, ‘Fuji — Kodak, the WTO, and the Death of Domestic Political Constituencies’ (1999) 8(1) Minnesota Journal of Global Trade 197, 202.
 See generally, ABA Report on Market Access, above n 19, 6–19. Many complaints follow a similar pattern, often with Japan as the subject of the complaint. The exporter complains that despite having a competitive product at a competitive price, it has been unable to make any inroads into the Japanese market. The exporter then points to some circumstantial evidence that seems to implicate local producers or local buyers and local government in anti‑competitive conduct. The exporter then draws the conclusion that the anti-competitive conduct must have occurred and that it must be the cause of the exporter’s failure to make sales.
 Ibid 67.
 See Michael Trebilcock, ‘Competition Policy and Trade Policy: Mediating the Interface’ (1996) 30(4) Journal of World Trade 71, 106; Kevin Kennedy, ‘Foreign Direct Investment and Competition Policy at the World Trade Organization’ (2001) 33 George Washington International Law Review 585, 613–15; Friedl Weiss, ‘From World Trade Law to World Competition Law’ (2000) 23 Fordham International Law Journal 250, 269–70.
 Strategic trade theory challenges the notion that the increase in international trade is due to trade liberalisation. Strategic or ‘new’ trade theory holds that an increasing proportion of trade can be explained by technological innovation rather than countries’ comparative advantage (natural resources). The implications of this include the proposition that national welfare can be increased by government promotion of national champions and by strategic use of protectionist policies. However, international trade economists have since somewhat backed away from this position. The gains from government protectionist intervention are modest and difficult to predict. Policy retaliation by other states could remove any gains. Recent studies suggest that while a reduction in trade barriers is generally beneficial there is probably no single trade strategy which will optimise the economic performance of every state: See Edward Buffie, Trade Policy in Developing Countries (2001) 1–7.
 Public choice theory holds that policy makers act in a rational or self-interested manner. Therefore, decision-making is subject to capture by the interest group most capable of providing the policy makers with political, financial or other private benefits. Normally, this will be the group with the lowest organisational costs. In most cases local producer interests would be best placed to capture those who determine competition law and policy.
 Hudec, A WTO Perspective, above n 17, 83; Patricia Isela Hansen, ‘Antitrust in the Global Market: Rethinking “Reasonable Expectations”’ (1999) 72 Southern California Law Review 1601, 1640–1.
 See Paul Stephan, ‘Sheriff or Prisoner? The United States and the World Trade Organization’ (2000) 1 Chicago Journal of International Law 49, 67.
 See Hudec, A WTO Perspective, above n 17, 84; Wolff, ‘The (Notionally) Bridgeable Chasm’, above n 31, 180–1, citing a number of instances where US antitrust law failed the needs of trade, including the failure of US antitrust authorities to contribute anything meaningful to the resolution of Kodak’s complaint against the alleged closure of the Japanese photographic market (the Japan — Film case). For a discussion of this case, see text accompanying footnote below n 172.
 For a discussion on the respective views and approaches of trade and competition authorities, see generally Douglas Rosenthal, ‘Jurisdiction and Enforcement: Equipping the Multilateral Trading System with a Style and Principles to Increase Market Access’ (1998)
6 George Mason Law Review 543.
 See ICPAC Report, above n 9, 267, citing responses of Kenya to the WTO.
 See, eg, the concerns expressed by India in the WGTCP: Report on the Meeting of
23–24 April 2002, WTO Doc WT/WGTCP/M/17 (2002) .
 See, eg, Trade Policy Review Body, Trade Policy Review — United States,
WTO Doc WT/TPR/S/126 (2003) ix (Report by the Secretariat), detailing various protectionist aspects of US anti-dumping and countervailing duty policy; Gregory Husisian, ‘When a New Sheriff Comes to Town: The Impending Showdown between the US Trade Courts and the World Trade Organization’ (2003) 17 St John’s Journal of Legal Commentary 457, 464, arguing that US anti-dumping laws are more concerned with protectionism than with righting the supposed trade distorting impact of foreign behaviour. There is broad consensus that the anti-dumping rules are applied in a protectionist manner. However, there is less consensus over the question as to whether they should be applied in such a manner.
 See Set of Multilaterally Agreed Equitable Principles and Rules for the Control of Restrictive Business Practices, GA Res 35/63, UN GAOR, 35th sess, 83rd plen mtg,
UN Doc Res/35/63 (1980) (‘UNCTAD Principles’). The UNCTAD Principles were recently reaffirmed: See UNCTAD, Review of All Aspects of the Set of Multilaterally Agreed Equitable Principles and Rules for the Control of Restrictive Business Practices,
UN Doc TD/RBP/CONF.5/15 (2000). Although these provisions may not be very effective, their mere presence has been used by some states (for example India and Malaysia) to criticise proposals for WTO competition rules made by developed states, particularly the EU, at the WGTCP: See Report on the Meeting of 1–2 July 2002, WTO Doc WT/WGTCP/M/18 (2002) –. At least the UNCTAD Principles has some development provisions designed to protect less developed states against the economic power of multinationals. Cf OECD Committee on International Investment and Multinational Enterprises, The OECD Guidelines for Multinational Enterprises, annex 1,
7–9, DAFFE/IME(2000)20 (2000) <http://www.olis.oecd.org/olis/2000doc.nsf/LinkTo/Daffe-
ime(2000)20> at 1 October 2004; OECD, The OECD Guidelines for Multinational Enterprises: A Key Corporate Responsibility Instrument (2003) OECD Policy Brief, <http://www.oecd.org/dataoecd/52/38/2958609.pdf> at 1 October 2004.
 Trade Act of 1974 § 301, as amended 19 USC § 2411 (1996) authorises the US Trade Representative (‘USTR’) to initiate negotiations with a foreign state whenever the USTR determines that any US rights under a trade agreement have been violated or US benefits denied, or whenever the USTR determines that any act, policy or practice of a foreign government is ‘unjustifiable’, or ‘unreasonable or discriminatory’ and burdens or restricts US commerce. To be ‘unjustifiable’ an act, policy or practice probably has to be illegal under international law. To be ‘unreasonable’ an act, policy or practice does not have to violate any international legal rule; it is sufficient that it is ‘otherwise unfair and inequitable’. Any practice that ‘denies fair and equitable ... market opportunities’ is deemed unreasonable, including government ‘toleration’ of ‘systematic anticompetitive activities’ by private enterprises in its territory. In making the determination of reasonableness the USTR is to take into account the ‘reciprocal opportunities in the US for foreign nationals and firms’. If the offending nation fails to negotiate a mutually satisfactory agreement with the US, the USTR may impose economic sanctions against it.
 The US had strategic reasons for unilaterally opening its borders. It saw this as a necessary step in rebuilding the economies of Europe and Japan as a bulwark against communism, particularly Soviet communism. As these economies recovered and the balance of trade moved significantly against the US, the strategic concerns became less important and were replaced by a strong desire to seek reciprocity. The history and nature of § 301 is discussed in Alan Swan, ‘“Fairness” and “Reciprocity” in International Trade: Section 301 and the Rule of Law’ (1999) 16 Arizona Journal of International and Comparative Law 37; William Barringer and James Durling, ‘Out of Focus: The Use of Section 301 to Address Anticompetitive Practices in Foreign Markets’ (1996) 1 UCLA Journal of International Law and Foreign Affairs 99, 101–8. See also Jagdish Bhagwati and Hugh Patrick (eds) Aggressive Unilateralism: America’s 301 Trade Policy and the World Trading System (1990); Braithwaite and Drahos, above n 2, 178–9; ABA Report on Market Access,
above n 19, 84–5; Hansen, above n 39, 1614–17, discussing the US policy of ‘aggressive unilateralism’. See generally Harvey Applebaum, ‘The Interface of the Trade Laws and the Antitrust Laws’ (1998) 6 George Mason Law Review 479, 485–6.
 See Jagdish Bhagwati, ‘Aggressive Unilateralism: An Overview’ in Bhagwati and Patrick (eds), above n 48, 1–45.
 See Swan, above n 48, 37. See also Hansen, above n 39, 1617.
 See Marrakesh Agreement, above n 1, annex 2 (Understanding on Rules and Procedures Governing the Settlement of Disputes) 1869 UNTS 401, art 23 (‘DSU’), which requires that in respect of any violation or non-violation claim under the GATT–WTO agreements, countries have recourse to, and abide by, the rules and procedures set out in the DSU. Countries should not act unilaterally in determining whether an infringement of the
GATT–WTO system has occurred. Trade Act of 1974 § 303(a)(2), 19 USC § 2413(a)(2) requires that where an international dispute resolution system applies to a § 301 case, the matter shall be submitted to that system. With the creation of the Marrakesh Agreement, above n 1, annex 1B (General Agreement on Trade in Services) 1869 UNTS 183 (‘GATS’) and the Marrakesh Agreement, above n 1, annex 1C (Agreement on Trade-Related Aspects of Intellectual Property Rights) 1869 UNTS 299 (‘TRIPS Agreement’), many disputes that previously fell outside the GATT are now covered. However, the DSU will not prevent unilateral action. The US, for largely domestic political reasons, has made it clear that it will continue to pursue solutions under § 301. For a discussion of § 301 and its relationship to the GATT–WTO system see United States — Sections 301–310 of the Trade Act of 1974, WTO Doc WT/DS152/R (1999) (Report of the Panel) (‘US — 301’).
 See discussion on the applicability of the GATT–WTO system to private barriers at text accompanying below n 120.
 Trade Act of 1974, 19 USC § 2411(d)(3)(B).
 The keiretsu system is a form of industrial organisation involving relationships, or a combination of relationships, between closely tied (but independently incorporated) corporations. The relationships are characterised by mutual share-holdings, interlocking directorships and similar relationships: See Mitsuo Matsushita, ‘The Structural Impediments Initiative: An Example of Bilateral Trade Negotiation’ (1991) 12 Michigan Journal of International Law 436, 440. Tamura identifies three distinct keiretsu structures, namely horizontal groups, production keiretsu, and distribution keiretsu: Jiro Tamura, ‘US Extraterritorial Application of Antitrust Law to Japanese Keiretsu’ (1993) 25 New York University Journal of International Law and Politics 385, 399.
 This at least is the opinion of the former head of the USTR’s Japan office: See Southwick, above n 22, 929. Section 301 has also been deployed against Korea, Taiwan, China, the countries of the EU, India, Brazil and others. For a discussion on the use of § 301 see ABA Report on Market Access, above n 19, 82–7; ICPAC Report, above n 9, 213–18.
 See generally US — 301, above n 51. The Panel was requested by the EU. Japan, Korea and other regular targets of § 301 participated as third parties.
 See Southwick, above n 22, 937. This is probably not an exercise of comity but a realist appreciation of US interests. Where the matter is within the cognisance of the WTO, the state upon which trade sanctions are imposed under § 301 may challenge that imposition at the WTO as not complying with the GATT–WTO agreements: See, eg, United States — Imposition of Import Duties on Automobiles from Japan Under Sections 301 and 304 of the Trade Act of 1974, WTO Doc WT/DS6/1 (1995) (Request for Consultations by Japan). If the US imposed § 301 sanctions that were inconsistent with the GATT–WTO agreements it would risk undermining the whole WTO system: See Braithwaite and Drahos, above n 2, 179.
 The origins of comity lie in the emergence of the notion of state sovereignty. Comity was employed to explain why states applied foreign law in certain circumstances. The state did so not as a matter of obligation, but as a mark of courtesy afforded to an equal sovereign. Where public policy required the application of local law, comity had no operation. These notions (of something less than an international obligation, of something discretionary but nevertheless imbued with custom and expectations of reciprocity) still underpin the modern uses of comity. The uses of comity however, are not always consistent, and its meaning remains elusive. Maier has described it as ‘an amorphous never-never land whose borders are marked by fuzzy lines of politics, courtesy, and good faith’: See Harold Maier, ‘Extraterritorial Jurisdiction at a Crossroads: An Intersection between Public and Private International Law’ (1982) 76 American Journal of International Law 280, 281. Nevertheless, despite its apparent indeterminacy, it is a useful doctrine (at least diplomatically, if not judicially) that encompasses all the reasons why one state or legal system might defer to another state or legal system. For a discussion on the origins and uses of comity, see Joel Paul, ‘Comity in International Law’ (1991) 32 Harvard International Law Journal 1; see also Brian Pearce, ‘The Comity Doctrine as a Barrier to Judicial Jurisdiction: A US–EU Comparison’ (1994) 30 Stanford Journal of International Law 525; ‘Predictability and Comity: Toward Common Principles of Extraterritorial Jurisdiction’ (1985) 98 Harvard Law Review 1310.
 See Barringer, ‘Competition Policy and Cross-Border Dispute Resolution’, above n 32, 470, who argues that there are ‘no established procedures for the submission of evidence, there is no legal standard for evaluating the sufficiency of the evidence, and there is no meaningful opportunity for a public hearing. There is also only limited guidance on what constitutes a restrictive practice’; See also Epstein, above n 30, 354; Spencer Weber Waller, ‘Can US Antitrust Laws Open International Markets?’ (2000) 20 Northwestern Journal of International Law and Business 207, 227–8.
 For example, the USTR is directed to have regard to reciprocity of access opportunities. US competition law concentrates on consumer welfare. See ICPAC Report, above n 9, 251, advising against expanding unilateral trade tools such as § 301 to cover anti-competitive activities because of the strong possibility that trade authorities would employ standards different to those applying under US antitrust law.
 See Hiroyuki Fujii, ‘The Kodak–Fuji Dispute: A Spectrum of Divergent Colors and a Blueprint for a New WTO Procedure for Disputes Involving Government Toleration of Anti-Competitive Practices’ (1997) 2 UCLA Journal of International Law and Foreign Affairs 317, 332.
 Trade Act of 1974, 19 USC § 2411(c).
 See Barringer, ‘Competition Policy and Cross-Border Dispute Resolution’, above n 32, 474. It should be noted that Barringer represented Fuji in its dispute with Kodak and the USTR. See also Epstein, above n 30, 354; ICPAC, above n 9, 252.
 See Japan — Trade in Semi-Conductors, GATT BISD, 35th Supp, 116, GATT Doc L/6309 (1988) (Report of the Panel, adopted on 4 May 1988) (‘Japan — Semi-Conductors’). See also Eleanor Fox, ‘International Antitrust and the Doha Dome’ (2003) 43 Virginia Journal of International Law 911, 920.
 See GATT, above n 2, art 1. In Japan — Semi-Conductors, GATT BISD, 35th Supp, 116, GATT Doc L/6309 (1988) (Report of the Panel, adopted on 4 May 1988) the European Economic Community and others unsuccessfully alleged that Japan had provided preferential access treatment to US producers as a result of a § 301 case initiated by the US. In 1991 however, the US managed to obtain from Japan an agreement for a market share target of 20 per cent. See Southwick, above n 22, 942.
 See Anne Krueger, ‘Challenges Facing the Multilateral Trading System: An Overview’ in Lee-Jay Cho and Yoon Hyung Kim (eds), The Multilateral Trading System in a Globalizing World (2000) 3, 10.
 As a trading bloc the EU has considerable power but because of the diverse interests of its member states, it is difficult to forge the necessary consensus to apply a § 301-like instrument effectively and consistently. See Braithwaite and Drahos, above n 2, 179.
 One of the basic reasons for the strong US reaction against a global competition agreement has been the fear that any agreement must derogate from the US optimum model: See ABA, Report on Market Access, above n 19, 25–39, 79–82, 99–100.
 Where an agreement between competitors (the cartel) exists solely to increase the welfare of the cartel members at the expense of both consumer welfare and total welfare, it is commonly referred to as a hardcore cartel. See OECD, Recommendation of the Council concerning Effective Action against Hard Core Cartels (1998) C(98)35/FINAL, 3 <http://www.oecd.org/dataoecd/39/4/2350130.pdf> at 1 October 2004 (‘OECD Recommendation’).
 See generally OECD, Report on Hard Core Cartels (2000); ICPAC Report, above n 9,
163–200, annex 4-A; John Connor, Global Price Fixing: Our Customers are the Enemy (2001); Simon Evenett, Margaret Levenstein and Valerie Suslow, ‘International Cartel Enforcement: Lessons from the 1990s’ (2001) 24 World Economy 1221, 1237–8.
 See generally Fox, ‘What is Harm to Competition?’, above n 11, 371. Ironically, exclusive distribution arrangements have been one of the major complaints voiced by US exporters against Japanese producers: See, eg, Japan — Film, WTO Doc WT/DS44/R (1998) (Report of the Panel) discussed at text accompanying below n 172. There is a very real possibility that, in many instances, US exporters and trade authorities have been complaining about Japanese activities that would be perfectly legitimate in the US. One of the reasons the Panel in Japan — Film decided that the US had failed in its bid to establish the necessary causal connection between Japanese government activities and discriminatory effects against US film imports was that the Japanese film market largely conformed to the normal film market structure around the world, including the US market. Each market tended to be dominated by a single brand distribution structure. This was precisely the market structure that Kodak was complaining about: at [10.173]. See also Barringer, ‘Competition Policy and Cross-Border Dispute Resolution’, above n 32, 465–7, who argues that the US trade authorities applied different standards when determining the legitimacy of Japanese film manufacturers’ distribution restraints than are applied to such restraints under US antitrust law; Daniel Gifford, ‘Antitrust and Trade Issues: Similarities, Differences, and Relationships’ (1995) 44 DePaul Law Review 1049, 1062–3, who argues that US criticisms of Japanese keiretsu relationships remain illegitimate so long as the US fails to employ the same principles as those applied under US antitrust law.
 See Jürgen Basedow, ‘International Antitrust: From Extraterritorial Application to Harmonization’ (2000) 60 Louisiana Law Review 1037, 1039, who claims that extraterritorial application of competition law has proved inadequate in a global business context because of the rules on application and enforcement, rather than because of those on substantive provisions. See also Sharon Foster, ‘While America Slept: The Harmonization of Competition Laws Based Upon the European Union Model’ (2001) 15 Emory International Law Review 467, 507–8; Vishnu Sharma, ‘Approaches to the Issue of Extraterritorial Jurisdiction’ (1995) 5 Australian Journal of Corporate Law 45.
 See William Dodge, ‘The Structural Rules of Transnational Law’ (2003) American Society of International Law: Proceedings of the 97th Annual Meeting 317.
 See, eg, Re Uranium Antitrust Litigation  USCA7 143; 617 F 2d 1248 (7th Cir, 1980) (‘Westinghouse Case’). The case is fully discussed in Warren Pengilley, ‘Extraterritorial Effects of United States Commercial and Antitrust Legislation: A View from “Down Under”’ (1983)
16 Vanderbilt Journal of Transnational Law 833.
 In the wake of the Westinghouse Case, evidence-blocking legislation was passed by Australia, Britain, Canada, New Zealand and South Africa. The legislation was designed to prevent citizens from giving evidence or supplying documents relevant to a case in a foreign jurisdiction exercising extraterritorial jurisdiction where national security or the national economic interest was involved. Additionally, some legislatures empowered their governments to order that certain foreign judgments not be enforced: See, eg, divs 2 and 3 of the Foreign Proceedings (Excess of Jurisdiction) Act 1984 (Australia), repealing Foreign Proceedings (Prohibition of Certain Evidence) Act 1976 (Australia) and Foreign Antitrust Judgments (Restriction of Enforcement) Act 1979 (Australia). Japan has legislation which can be used to block foreign judgments against Japanese defendants that are ‘contrary to the public order or good morals in Japan’ — it has been argued that this could be used against a treble damages claim: Geralyn Trujillo, ‘Mutual Assistance under the International Antitrust Enforcement Assistance Act: Obstacles to a United States–Japanese Agreement’ (1998)
33 Texas International Law Journal 613, 628.
 Claw-back legislation provides that any excessive damages (that is, damages in excess of those actually suffered) paid pursuant to an award in a foreign jurisdiction can be clawed back in an action against the recipient of those excessive damages: See, eg, Foreign Proceedings (Excess of Jurisdiction) Act 1984 (Australia) s 10; Protection of Trading Interests Act 1980 (UK) c 11, s 6. See generally Deborah Senz and Hilary Charlesworth, ‘Building Blocks: Australia’s Response to Foreign Extraterritorial Legislation’ (2001)
 MelbJlIntLaw 3; 2 Melbourne Journal of International Law 69.
 Extraterritoriality largely cancels out the incentives firms might otherwise have to locate anti-competitive conduct in jurisdictions with relaxed or non-existent competition laws.
 Foreign sovereign immunity is generally not extended to the commercial acts of foreign sovereigns: See, eg, Foreign States Immunities Act 1985 (Australia) s 11; Foreign Sovereign Immunities Act of 1976, 28 USC § 1605(a)(2) (2002); State Immunity Act 1978 (UK) c 33, s 3.
 The act of state doctrine refers to the non-justiciability or non-enforceability of ‘public act[s] of [a] foreign sovereign within its territorial jurisdiction on matters pertaining to its governmental sovereignty’: See US Department of Justice, Antitrust Division, Antitrust Enforcement Guidelines for International Operations (2nd ed, 1995) [3.33] <http://www.usdoj.gov/atr/public/guidelines/internat.htm> at 1 October 2004 (‘US International Guidelines’). See also Restatement (Third) of Foreign Relations Law (1987)
s 443(1) (‘Restatement’); Banco Nacional de Cuba v Sabbatino,  USSC 48; 376 US 398 (1964). The principle was applied to antitrust cases in Occidental of Umm al Qaywayn Inc v A Certain Cargo of Petroleum Laden Aboard the Tanker Dauntless Coloctronis,  USCA5 1190; 577 F 2d 1196
(5th Cir, 1978), certiorari denied 442 US 928 and Occidental Petroleum Corp v Buttes Gas and Oil, Co 331 F Supp 92 (Cal, 1971) affirmed United States of America v Jesus
Alvarez-Franco,  USCA9 509; 461 F 2d 1261 (9th Cir, 1972), certiorari denied 409 US 950. It was also discussed in Mannington Mills, Inc v Congoleum Corp,  USCA3 315; 595 F 2d 1287, 1292 (3rd Cir, 1979). In Australia see Petrotimor Companhia de Petroleos SARL v Commonwealth of Australia  FCAFC 83; (2003) 128 FCR 507. In the UK, see Buttes Gas and Oil Co v Hammer (No 3)  AC 888.
 Act of state and foreign sovereign compulsion are treated as related but separate doctrines in the US. In the US foreign sovereign compulsion applies (both as an issue going to jurisdiction and as a defence) where (1) the impugned private act is mandated by the foreign state; (2) there is a threat of tangible sanctions; and (3) the defendant has made a good faith attempt to comply with US law: Restatement s 441. Mere approval or even involvement by a foreign sovereign is not sufficient; United States v Watchmakers of Switzerland Information Center  Trade Cases 70,600; Hartford Fire Insurance Company v California,
 USSC 100; 509 US 764 (1993); Mannington Mills, Inc v Congoleum Corp,  USCA3 315; 595 F 2d 1287, 1293 (3rd Cir, 1979). Foreign sovereign compulsion was applied to deny antitrust jurisdiction where the Venezuelan Government specifically ordered the defendant not to trade with the plaintiff oil company: Interamerican Refining Corp v Texaco Maracaibo, Inc, 307 F Supp 1291, 1298 (D Del, 1970). See also Trugman-Nash, Inc v New Zealand Dairy Bd, Milk Products Holdings (North America) Inc, 954 F Supp 733 (D NZ, 1997) (NZ Dairy Board compelled by statute to refuse plaintiff’s application for licence to export dairy products to the US). See generally Don Wallace Jr and Joseph Griffin, ‘The Restatement and Foreign Sovereign Compulsion: A Plea for Due Process’ (1989) 23 International Lawyer 593.
 For a description of comity see above n 59.
 Foreign Trade Antitrust Improvements Act of 1982, 15 USC § 6a (1994), amending the Sherman Act (‘FTAIA’). FTAIA is a convoluted piece of legislation, the precise meaning and scope of which is still contentious.
 See Hartford Fire Insurance Company v California,  USSC 100; 509 US 764 (1993) (‘Hartford’). The effects doctrine has now been applied to a criminal prosecution: See United States v Nippon Paper Industries Co, 944 F Supp 55 (Mass, 1996), reversed  USCA1 97; 109 F 3d 1 (1st Cir, 1997), certiorari denied 118 S Ct 685 (1998).
 Hartford,  USSC 100; 509 US 764, 799 (1993). Attempts to temper the jurisdictional expansionism of the effects doctrine by recognising an extended form of comity analysis appear to have failed as a result of Hartford: See Timberlane Lumber Co v Bank of America,  USCA9 255; 549 F 2d 597 (9th Cir, 1976). But see Spencer Weber Waller, ‘The United States as Antitrust Courtroom to the World: Jurisdiction and Standing Issues in Transnational Litigation’ (Working Paper, Chicago School of Law, Loyola University, 2002) 8 <http://www.luc.edu/law/academics/special/center/antitrust/freeborn.pdf> at 1 October 2004, who argues that the decision in Hartford is often simply ignored. US Department of Justice Antitrust Division, Antitrust Enforcement Guidelines for International Operations (2nd ed, 1995) [3.2] <http://www.usdoj.gov/atr/public/guidelines/internat.htm> at 1 October 2004, also implies that interest balancing on comity principles is still available to US courts. In fact, the decision in Hartford, which was by a bare majority, remains controversial. Both Souter J (for the majority) and Scalia J (in dissent) applied the Restatement. It has been claimed that both applied the incorrectly, thus leaving this area of the law unclear. See Russell Weintraub, ‘Globalization’s Effect on Antitrust Law’ (1999) 34 New England Law Review 27, 29; Andreas Lowenfeld, ‘Conflict, Balancing of Interests, and the Exercise of Jurisdiction to Prescribe: Reflections on the Insurance Antitrust Case’ (1995) 89 American Journal of International Law 42, and the reply by Larry Kramer, ‘Extraterritorial Application of American Law after the Insurance Antitrust Case: A Reply to Professors Lowenfeld and Trimble’ (1995) 89 American Journal of International Law 750.
 Of course, this may not be a bad thing. There are serious doubts that comity has the capacity to be an appropriate principle of jurisprudence: See William Dodge, ‘Extraterritoriality and Conflict-of-Laws Theory: An Argument for Judicial Unilateralism’ (1998) 39 Harvard International Law Journal 101. See also Spencer Weber Waller, ‘The Twilight of Comity’ (2000) 38 Columbia Journal of Transnational Law 563, 570 who argues that US courts have proved incapable of applying a comity test in a consistent and principled manner.
 Aholstrom Osakeyhitio v Commission of the European Communities (C89/85) 
ECR I-5193. On the extraterritorial application of competition laws, see generally Basedow and Pankoke, above n 26, 27–9.
 See generally Takaaki Kojima, ‘International Conflicts over the Extraterritorial Application of Competition Law in a Borderless Economy’ (Fellowship Paper, Weatherhead Center for International Affairs, Harvard University, 2001–02) 6, 36 <http://www.wcfia.harvard.edu/fellows/papers01-02/kojima.pdf> at 1 October 2004 who argues that Japan applies a more traditional notion of the territorial principle in determining the scope of its Act concerning Prohibition of Private Monopolization and Maintenance of Fair Trade 1947 (Japan) (‘Antimonopoly Act’). Kojima does claim that even Japan is shifting to a broader interpretation, particularly in respect of mergers. However, given Japan’s traditional employment of horizontal and vertical linkages between commercial entities, it has to be doubted that Japan will adopt an ‘effects doctrine’ approach to distribution restraints any time in the near future. The extraterritorial application of Australian competition law is governed by s 5(1) of the Trade Practices Act 1974 (Australia). This is narrower in scope than the US effects doctrine. It provides that the competition law provisions of the Trade Practices Act 1974 (Australia) only apply to extraterritorial conduct by firms which are incorporated in Australia or which carry on business in Australia, or to Australian citizens or persons ordinarily resident in Australia.
 See US International Guidelines, above n 79, [3.122]. Originally, and despite the FTAIA, the Department of Justice had indicated its intention not to act unless US consumers were affected. This was referred to as Footnote 159. It was deleted in the 1992 version of the guidelines and has not reappeared: See Epstein, above n 30, 351.
 US International Guidelines, above n 79, [3.2], argue that if the US competition authorities have determined to bring suit the US courts have no role to play in comity analysis. This has not been legally tested. The US International Guidelines do not, of course, affect a private right of action.
 The fact that subject matter jurisdiction exists does not necessarily mean that personal jurisdiction exists. See US International Guidelines, above n 79, [3.11]. The evidence suggests that there has been little cooperation with US outbound extraterritorialism: See Merit Janow, ‘Public, Private, and Hybrid Public/Private Restraints of Trade: What Role for the WTO?’ (2000) 31 Law and Policy in International Business 977, 978.
 See Waller, ‘The United States as Antitrust Courtroom to the World’, above n 84, 14.
 See Treaty of Amsterdam Amending the Treaty on European Union, the Treaties Establishing the European Communities and Certain Related Acts, opened for signature
2 October 1997  OJ C 340 (entered into force 1 May 1999) arts 81–6. The EU competition regime applies only to cross-border activities. Otherwise all competition law and policy is a matter for the Member States. The EU made no attempt to unify competition law or policy within the States.
 See Protocol to the Australia New Zealand Closer Economic Relations — Trade Agreement on Acceleration of Free Trade in Goods,  ATS 18, art 4 (signed and entered into force 18 August 1988) (‘ANZCERTA’). Article 4 relates to anti-dumping measures. For a discussion of ANZCERTA, see Rex Ahdar, ‘The Role of Antitrust Policy in the Development of Australia–NZ Free Trade’ (1991) 12 Northwestern Journal of International Law and Business 317.
 Opened for signature 17 December 1992, 32 ILM 289 (entered into force 1 January 1994) (‘NAFTA’).
 NAFTA art 1501 requires treaty members to proscribe anti-competitive business conduct and to take appropriate action. However, by virtue of art 1501(3), breaches of art 1501 are not subject to the NAFTA dispute resolution procedures. For a discussion of the competition aspects of NAFTA, see Spencer Weber Waller, ‘The Internationalization of Antitrust Enforcement’ (1997) 77 Boston University Law Review 343, 356–60.
 See generally Report (2001) of the Working Group on the Interaction between Trade and Competition Policy to the General Council, WTO Doc WT/WGTCP/5 (2001) .
 For a list of the relevant US bilateral agreements, see ICPAC Report, above n 9, annex 1-C.
 The EU has arrangements with the US and Canada amongst others. For a list of the relevant EU bilateral agreements, see Competition Directorate General, International — Bilateral Relations, European Union (EUROPA) (1998) <http://europa.eu.int/comm/competition/international/bilateral> at 1 October 2004.
 See, eg, Cooperation Agreement between the Commissioner of Competition (Canada), The Australian Competition and Consumer Commission and the New Zealand Commerce Commission regarding the Application of their Competition and Consumer Laws (signed and entered into force 14 October 2000) Australian Competition and Consumer Commission <http://www.accc.gov.au/content/item.phtml?itemId=381432&nodeId=file41
106de89acfe&fn=ACCC%20‑%20New%20Zealand%20Commerce%20Commission%20(NZCC)%20-%20commissioner%20of%20competition%20(Canada).pdf> at 1 October 2004.
 Agreement concerning Cooperation on Anticompetitive Activities, Japan–USA, 4 Trade Reg Rep CCH ¶ 13 507 (signed and entered into force 7 October 1999); Agreement between the European Community and the Government of Japan concerning Cooperation on Anticompetitive Activities, signed 10 July 2003,  OJ L 183/12 (entered into force
9 August 2003).
 Charles Stark, ‘Jurisdiction and Enforcement: International Cooperation in the Pursuit of Cartels’ (1998) 6 George Mason Law Review 533, 535. An exception may be the agreement between the US and Germany: Agreement relating to Mutual Cooperation regarding Restrictive Business Practices, signed 23 June 1976, USA–Federal Republic of Germany, 27 UST 1956 (entered into force 11 September 1976). According to the ICPAC Report, this agreement concerned antitrust enforcement and reflected the ‘strong post-World War II German antitrust enforcement tradition’: ICPAC Report, above n 9, annex 1-C, v. The signing of this agreement, however, was an isolated incident. Enforcement cooperation did not really begin in earnest until the 1990s.
 Agreement relating to Cooperation on Antitrust Matters, Australia–USA, 1369 UNTS 43 (signed and entered into force 29 June 1982).
 Ibid arts 1(1), 1(2). For a discussion of the Agreement relating to Cooperation on Antitrust Matters, see Stephen Ramsey, ‘The United States–Australian Antitrust Cooperation Agreement: A Step in the Right Direction’ (1983) 24 Virginia Journal of International Law 127. See also the Memorandum of Understanding between the Government of the United States of America and the Government of Canada as to Notification, Consultation and Cooperation with respect to the Application of National AntiTrust Laws, 23 ILM 275 (signed and entered into force 9 March 1984). For a comment on the Memorandum of Understanding, see Jean Gabriel Castel, Extraterritoriality in International Trade: Canada and United States of America Practices Compared (1988).
 See discussion on blocking legislation at above n 75.
  OJ L 95/47 (signed and entered into force 23 September 1991); corrected at  OJ L 131/38 (‘US–EU Agreement’); see also Stark, above n 101, 535–6.
 This agreement and others that follow the same structure trace their origins to OECD recommendations: See OECD, Revised Recommendation of the Council concerning
Co-operation between Member Countries on Anticompetitive Practices Affecting International Trade (1995) C(95)130/FINAL <http://www.oecd.org/dataoecd/60/42/21570317.pdf> at 1 October 2004. See also Edward Swaine, ‘The Local Law of Global Antitrust’ (2001) 43 William and Mary Law Review 627; ICPAC Report, above n 9,
annex 1-C, v.
 US–EU Agreement, above n 105. Supplemented by Agreement between The Government of the United States of America and the European Communities on the Application of Positive Comity Principles in the Enforcement of Their Competition Laws  OJ 173/28 (signed and entered into force 4 June 1998).
 See Waller, ‘Can US Antitrust Laws Open International Markets?’, above n 59, 230, who argues that few competition systems have shown impartiality between their own citizens and foreigners.
 ICPAC Report, above n 9, 23.
 Ibid: Positive comity is only truly effective where there is ‘correspondence between the parties’ antitrust laws and enforcement commitment’.
 Ibid 207, 210; Epstein, above n 30, 364–5; Fox, ‘International Antitrust and the Doha Dome’, above n 64, 925. See also Marsden, A Competition Policy for the WTO, above n 31, 83, who claims that ‘a patchwork quilt of individual agreements among countries cannot constitute a web that would be broad enough, or strong enough, to capture all international cartels, let alone other forms of anti-competitive activity’.
 OECD, Report on Hard Core Cartels, above n 70, 29; Report (1999) of the Working Group on on the Interaction between Trade and Competition Policy to the General Council,
WTO Doc WT/WGTCP/3 (1999) .
 The only dedicated competition agreement allowing for the exchange of confidential information is that between the US and Australia: See Agreement on Mutual Antitrust Enforcement Assistance, signed 27 April 1999, Australia–USA,  ATS 22 (entered into force 5 November 1999). There are a number of mutual legal assistance treaties (‘MLATs’) providing for mutual legal assistance in criminal matters. However, MLATs are only relevant in competition regulation where the treaty partners have made competition infringement a crime. To date this still tends to be the exception rather than the norm. However, the situation is changing: See, eg, Enterprise Act 2002 (UK) c 40, ss 188, 189. The UK has recently introduced criminal penalties for cartel offences. The criminal cartel offence operates alongside the civil sanction provisions of the Competition Act 1998 (UK) c 41. For a list of sanctions by state see OECD, Fighting Hard Core Cartels: Harm, Effective Sanctions and Leniency Programmes (2002) annex B; see also Basedow and Pankoke, above n 26, 25–7.
 See OECD, Report on Hard Core Cartels, above n 70, 7. The Report makes the point that most OECD states actually ban the sharing not only of confidential information but also of much information that cannot in any sense be regarded as confidential. Section 155AA of the Trade Practices Act 1974 (Australia) is probably to be read in this way. On the US position see Trujillo, above n 75, 627. US antitrust provisions prohibit the sharing of information received during an enforcement investigation unless a mutual assistance agreement exists or the person who supplied the information consents to its dissemination: See Clayton Act of 1914 § 7A(h), 15 USC § 18a(h) (2000); Federal Trade Commission Act,
15 USC § 57a(j) (1994). US Department of Justice authorities must comply with the secrecy rules applicable to a criminal grand jury investigation. Japanese law has similar provisions: See discussion in Trujillo, above n 75, 628. For a discussion generally on the issues relating to information sharing between national regulators, see OECD, Global Forum on Competition: Information Sharing in Cartel Investigations (2002) CCNM/GF/COMP(2002) <http://www.oecd.org/dataoecd/57/40/1820435.pdf> at 1 October 2004; see also Nina Hachigian, ‘Essential Mutual Assistance in International Antitrust Enforcement’ (1995)
29 International Lawyer 117. Additionally, the law of confidentiality goes beyond specific competition law restrictions. Most states have general laws protecting the secrecy of relevant business information.
 See Hachigian, above n 114, 127.
 See ICPAC Report, above n 9, 191–2; Ibid 127; John Parisi, ‘Enforcement Co-operation among Antitrust Authorities’ (1999) 20 European Competition Law Review 133, 139.
 The first Global Competition Forum was conducted by the OECD in October 2001. The WGTCP was established by a decision of the WTO Ministerial Conference held in Singapore in December 1996: Singapore Ministerial Declaration, above n 4, .
 The ICN was formed in 2001 and was very much a US initiative. It derives its creative inspiration from the findings of ICPAC Report, above n 9. The ICN is described by US authorities as a ‘virtual network’. It has no headquarters. The US still supports this structure despite growing pressure to develop a small but permanent bureaucracy: See R Hewitt Pate, ‘The DOJ International Antitrust Program — Maintaining Momentum’ (Speech delivered at the American Bar Association, Section of Antitrust Law, 2003 Forum on International Competition Law, New York, 6 February 2003).
 For example, the Fordham Corporate Law Institute, Annual Conference on International Antitrust Law & Policy, New York.
 See Charles James, ‘International Antitrust in the 21st Century: Cooperation and Convergence’ (Speech delivered at the OECD Global Forum on Competition, Paris,
17 October 2001).
 See Singapore Ministerial Declaration, above n 4, . See also Report (2002) of the Working Group on the Interaction between Trade and Competition Policy to the General Council, above n 6, –.
 Provisions in the GATT, above n 2, which affect conditions of competition include art II(4) (dealing with import monopolies); arts III(2) and III(4) (maintaining competitive conditions between local products and imports); art XI (dealing with quantitative restrictions); and art XVII (dealing with state trading enterprises). A prohibition on voluntary export restraints and orderly marketing arrangements (whether in respect of exports or imports) is contained in the Marrakesh Agreement, above n 1, annex 1A (Agreement on Safeguards),
1869 UNTS 154, art 11(1)(b). Competition provisions are also found in: TRIPS Agreement, above n 52, art 40; GATS, above n 52, arts VI, VIII, IX and XXIX (Annex on Telecommunications); Marrakesh Agreement, above n 1, annex 1A (Agreement on Technical Barriers to Trade), 1868 UNTS 120, art 2. The most comprehensive competition provisions are to be found in the Fourth Protocol to the General Agreement on Trade in Services, opened for signature 15 February 1997  ATS 9 (entered into force 5 February 1998) (‘Basic Telecommunications Agreement’), particularly in an annex to this Protocol (‘Reference Paper’). The competition provisions contained in the Reference Paper apply to major suppliers of basic telecommunications services, which, in practice, largely restricts its operation to present or former state-owned telecommunications monopolies. See generally, Kevin Kennedy, Competition Law and the World Trade Organisation: The Limits of Multilateralism (2001). On the role of the competition-like provisions in the GATT–WTO agreements see generally Ernst-Ulrich Petersmann, ‘International Competition Rules for Governments and for Private Business: A “Trade Law Approach” for Linking Trade and Competition Rules in the WTO’ (1996) 72 Chicago-Kent Law Review 545, 561–5. See also, particularly in relation to services, Mark Warner, ‘After Seattle: Is There a Future for Trade and Competition Policy Rule-Making?’ (2000) 26 Brooklyn Journal of International Law 307. In relation to the competition provisions in the Reference Paper see
Mexico — Measures Affecting Telecommunications Services, WTO Doc WT/DS204/R (2004) (Report of the Panel) (‘Telmex’). This is the first WTO telecommunications case and provides some insights into how WTO panels might approach the interpretation of competition issues. The significance of this case extends beyond the telecommunications industry because the Reference Paper has been suggested as a template for other sectoral agreements. On the Reference Paper see also Hudec, A WTO Perspective, above n 17, 91–6; Laura Sherman, ‘“Wildly Enthusiastic” About the First Multilateral Agreement on Trade in Telecommunications Services’ (1998) 51 Federal Communications Law Journal 61. On the background to the Telmex case, see Rachel Rosenthal, ‘United States v Mexico: The First Telecommunications Challenge Confronting the World Trade Organization’ (2002)
10 CommLaw Conspectus 315. For a critique of the WTO decision in Telmex, see Philip Marsden, ‘Trade and Competition — WTO Decides First Competition Case — With Disappointing Results’ (2004) 16 Competition Law Insight 3.
 Article XXIII(1) provides that an exporting state may seek a remedy
[i]f [it] should consider that any benefit accruing to it directly or indirectly under this Agreement is being nullified or impaired or that the attainment of any objective of the Agreement is being impeded as the result of (a) the failure of another contracting party to carry out its obligations under this Agreement, or (b) the application by another contracting party of any measure, whether or not it conflicts with the provisions of this Agreement, or (c) the existence of any other situation[.]
 For an analysis of the application of art III(4) to private trade barriers see Claus-Dieter Ehlermann and Lothar Ehring, ‘WTO Dispute Settlement and Competition Law: Views from the Perspective of the Appellate Body’s Experience’ (2003) 26 Fordham International Law Journal 1505. For a discussion on art III(4) and its relationship to the trade and competition debate, see Marsden, A Competition Policy for the WTO, above n 31, 125–32.
 Thus, for example, a state may violate its GATT national treatment obligation if, through the use of non-binding measures, it creates a situation where private actors are induced to engage in exclusionary conduct that discriminates against imports. See Japan — Semi‑Conductors, GATT BISD, 35th Supp, 116, GATT Doc L/6309 (1988), 155 (Report of the Panel, adopted on 4 May 1988). The government measures must do more than merely facilitate the private conduct. The effectiveness of the private conduct must be ‘essentially dependent’ on the government’s non-binding measures: See Hudec, A WTO Perspective, above n 17, 97. See also Marsden, A Competition Policy for the WTO, above n 31, 144.
 There is no reason why the one act, practice or policy may not at the same time give rise to a cause of action under art XXIII(1)(a) and art XXIII(1)(b). See European Communities — Measures Affecting Asbestos and Asbestos-Containing Products, WTO Doc WT/DS135/AB/R, AB-2000-11 (2001)  (Report of the Appellate Body) (‘EC — Asbestos’).
 Article XXIII(1)(c) applies to ‘any other situation’ causing nullification or impairment of benefits. The scope of art XXIII(1)(c) is not clear. Article XXIII(1)(c) was raised by the EEC in its request for a working party to investigate the ‘imperviousness of the Japanese market’, allegedly caused by a range of barriers including cartels and distribution restraints. The matter, however, proceeded no further. See Japan — Nullification or Impairment of the Benefits Accruing to the EEC under the General Agreement and Impediment to the Attainment of GATT Objectives, GATT Doc L/5479, C/M/167 (1983) (Request for a Working Party). Article XXIII(1)(c) was introduced to deal with global upheavals such as world depressions. Hudec argues that art XXIII(1)(c) could be applied, albeit controversially, to the non-enforcement of competition law: See Hudec, A WTO Perspective,
above n 17, 99–100. However, a decision under art XXIII(1)(c) is not binding unless adopted by consensus. It is difficult to see why any state would want such a precedent.
 A claim under art XXIII(1)(b) is often referred to as a non-violation claim. In fact, art XXIII(1)(b) covers both violation and non-violation claims. Article XXIII also applies to disputes under TRIPS Agreement, above n 51. A non-violation claim is included in GATS, above n 51, art XXIII(3) and in the Agreement on Government Procurement, opened for signature 15 April 1994, 1915 UNTS 105, art 22 (entered into force 1 January 1996).
 See European Economic Community — Payments and Subsidies Paid to Processors and Producers of Oilseeds and Related Animal-Feed Proteins, GATT BISD, 37th Supp, 86, GATT Doc L/6627 (1989)  (Report of the Panel, adopted on 25 January 1990)
(‘EEC — Oilseeds’):
The idea underlying [the provisions of art XXIII] is that the improved competitive opportunities that can legitimately be expected from a tariff concession can be frustrated not only by measures proscribed by the General Agreement but also by measures consistent with that Agreement. In order to encourage contracting parties to make tariff concessions they must therefore be given a right of redress when a reciprocal concession is impaired by another contracting party as a result of the application of any measure, whether or not it conflicts with the General Agreement.
A subsequent panel has described this as an example of pacta sunt servanda, which by virtue of customary international law applies to the GATT–WTO agreements to the extent that it is not expressly or impliedly excluded: Korea — Measures Affecting Government Procurement, WTO Doc WT/DS163/R (2000) [7.93]–[7.96] (Report of the Panel)
(‘Korea — Procurement’). Pacta sunt servanda requires that treaties be performed in good faith: See Vienna Convention on the Law of Treaties, opened for signature 23 May 1969, 1155 UNTS 331, art 26 (entered into force 27 January 1980).
 To some extent the issue resembles the concerns attending the development of an implied term of good faith in relational contracts.
 Formerly the GATT dispute resolution system resembled a diplomatic process. This was because any WTO panel decision could be vetoed by any state (including the state against whom the complaint was brought). The WTO dispute resolution system has moved from a diplomatic-oriented process to a more rule-based one. A panel decision may only be rejected by unanimous vote (including the vote of the successful state).
 DSU, above n 51, art 3(2) provides that ‘[r]ecommendations and rulings of the DSB [Dispute Settlement Body] cannot add to or diminish the rights and obligations provided in the covered agreements’; but see Ehlermann and Ehring, above n 124, 1557.
 See Japan — Film, WTO Doc WT/DS44/R (1998) [10.36] (Report of the Panel). This was also the view of the only Appellate Body to have examined a non-violation complaint:
EC — Asbestos, WTO Doc WT/DS135/AB/R, AB-2000-11 (2001)  (Report of the Appellate Body). As the Panel in Japan — Film stated at [10.36], ‘[t]he reason for this caution is straightforward. Members negotiate the rules that they agree to follow and only exceptionally would expect to be challenged for actions not in contravention of those rules’.
 See Japan — Film, WTO Doc WT/DS44/R (1998) [10.30] (Report of the Panel) citing GATT jurisprudence and art 26.1 of the DSU, above n 51; see also EC — Asbestos,
WTO Doc WT/DS135/AB/R, AB-2000-11 (2001)  (Report of the Appellate Body).
 In Japan — Film, WTO Doc WT/DS44/R (1998) [10.41] (Report of the Panel), the Panel said that the text of art XXIII(1)(b) required the complaining state to establish three elements: ‘(1) application of a measure by a WTO Member; (2) a benefit accruing under the relevant agreement; and (3) nullification or impairment of the benefit as the result of the application of the measure’. The test was applied by the Panel in EC — Asbestos,
WTO Doc WT/DS135/R, (2000) [8.283] (Report of the Panel). This part of the Panel’s report was not reviewed by the Appellate Body in EC — Asbestos, WTO Doc WT/DS135/AB/R, AB-2000-11 (2001) (Report of the Appellate Body).
 See, eg, The Australian Subsidy on Ammonium Sulphate, GATT BISD, vol II, 188, GATT Doc GATT/CP.4/39 (1950) 193 (Report of the Working Party, adopted by the Contracting Parties on 3 April 1950) (‘Australian Subsidy’); Treatment by Germany of Imports of Sardines, GATT BISD, 1st Supp, 53, GATT Doc G/26 (1952) 58 (Report of the Panel, adopted by the Contracting Parties on 31 October 1952); EEC — Oilseeds, GATT BISD, 37th Supp, 86, GATT Doc L/6627 (1989) – (Report of the Panel, adopted on
25 January 1990); Japan — Film, WTO Doc WT/DS44/R (1998) [10.82] (Report of the Panel).
 Japan — Film, WTO Doc WT/DS44/R (1998) [10.38] (Report of the Panel). See, eg, Australian Subsidy, GATT BISD, vol II, 188, GATT Doc GATT/CP.4/39 (1950) (Report of the Working Party, adopted by the Contracting Parties on 3 April 1950); Follow‑Up on the Panel Report ‘European Economic Community — Payments and Subsidies Paid to Processors and Producers of Oilseeds and Related Animal-Feed Proteins’, GATT BISD, 39th Supp, 91, GATT Doc DS28/R (1992) (Report of the Members of the Original Oilseeds Panel). See discussion in Hudec, A WTO Perspective, above n 17, 90; Hansen, above n 39, 1629–30.
 Japan — Film, WTO Doc WT/DS44/R (1998) [10.38] (Report of the Panel).
Trade-distorting subsidies are now expressly covered by the Marrakesh Agreement, above n 1, annex 1A (Agreement on Subsidies and Countervailing Measures), 1869 UNTS 14. An unlawful subsidy would now amount to a violation.
 See, eg, James Rill, ‘Comments and Discussion’ in  Brookings Trade Forum 378, 382; Frieder Roessler, ‘The Concept of Nullification and Impairment in the Legal System of the World Trade Organization’ in Ernst-Ulrich Petersmann (ed), International Trade Law and the GATT/WTO Dispute Settlement System (1997) 123, 134, who argues that given ‘the broad extension of the multilateral trade order into the area of domestic policies, the field of potential application of [art XXIII(1)(b)] has diminished significantly’; John Jackson, ‘The Role and Effectiveness of the WTO Dispute Settlement Mechanism’  Brookings Trade Forum 179, 204. See Barringer, ‘Competition Policy and Cross Border Dispute Resolution’, above n 32, 477, who argues that the use of WTO panels to discipline private anti‑competitive activities extends the jurisdiction of the WTO well beyond that which most states believe it has. See also John Linarelli ‘The Role of Dispute Settlement in World Trade Law: Some Lessons from the Kodak–Fuji Dispute’ (2000) 31 Law and Policy in International Business 263, 334, who argues that art XXIII ought to be interpreted and applied in an expansive manner that furthers the object of trade liberalisation, and Robert Howse and Michael Trebilcock, ‘The Fair Trade–Free Trade Debate: Trade, Labor, and the Environment’ (1996) 16 International Review of Law and Economics 61, 78, who discuss the use of art XXIII(1)(b) in relation to environmental and labour concerns as barriers to trade. Contra Dillon, above n 33, 219–48, who argues that any expanded use of art XXIII(1)(b) is democratically illegitimate. In 1958 a GATT expert group advised against the application of art XXIII(1)(b) to restrictive business practices: Restrictive Business Practices: Arrangements for Consultations, GATT BISD, 9th Supp, 170, GATT Doc L/1015 (1960) (Report of Experts, adopted 2 June 1960). See also Hansen, above n 39; Tracy Abels, ‘The World Trade Organization’s First Test: The United States–Japan Auto Dispute’ (1996) 44 UCLA Law Review 467; see generally Ehlermann and Ehring, above n 124, 1555–6.
 See generally Japan — Film, WTO Doc WT/DS44/R (1998) (Report of the Panel). But see Sung-Joon Cho, ‘GATT Non-Violation Issues in the WTO Framework: Are They the Achilles’ Heel of the Dispute Settlement Process?’ (1998) 39 Harvard International Law Journal 311, 316–20, who argues that while the interpretation of art XXIII(1)(b) ranges between a minimalist or restrained approach and an expansive or activist approach, the weight of precedent favours restraint.
 The difficulties of establishing a claim under art XXIII(1)(b) in the context of competition disputes are discussed in Hudec, A WTO Perspective, above n 17, 98–101. See also Sung‑Joon Cho, above n 140, 316–20. See generally Bernard Hoekman, ‘Competition Policy and the Global Trading System: A Developing Country Perspective’ (Policy Research Working Paper No WPS1735, World Bank, 1997).
 Japan — Film, WTO Doc WT/DS44/R (1998) [10.49] (Report of the Panel). ‘Measure’ is probably to be treated as having a broader meaning than ‘requirement’ under art III(4), although in practice the difference may not be very great. ‘Requirement’ has been held to extend to non-binding government acts, practices and policies: at [10.51].
 Ibid [10.49].
 Ibid [10.56].
 GATT and WTO panels have held that Japanese administrative guidance programs may amount to government measures: See, eg, Japan — Semi-Conductors, GATT BISD,
35th Supp, 116, GATT Doc L/6309 (1988), 154–5 (Report of the Panel, adopted on
4 May 1988); Japan — Film, WTO Doc WT/DS44/R (1998) [10.45] (Report of the Panel).
 Japan — Film, WTO Doc WT/DS44/R (1998) [10.45] (Report of the Panel).
 This is a difficult issue which in part merges with the issue of causation. In Japan — Film, for example, the Panel was of the opinion that certain advisory reports and survey reports issued by the Japanese Ministry of International Trade and Industry (‘MITI’) (since 2000, the Ministry of Economy, Trade and Industry) were not by themselves ‘measures’ in that they did not offer incentives or disincentives for private parties to act in a manner similar to that which would result from a legally binding measure: Ibid [10.121]–[10.122],
[10.135]–[10.136], [10.147]–[10.149]. On the other hand, MITI’s ‘Guidelines for Rationalizing Terms of Trade for Photographic Film’ — which were accompanied by a plan of implementation — were held to be a measure: at [10.156]–[10.161].
 Most competition laws make provision for a process of authorisation or exemption by a competition authority. A notable exception is the US. The US has a limited scheme of guidance but the main body of US competition law leaves it to firms to make their own assessments as to whether their conduct is lawful or not. The competition authorities do publish guidelines, but these are not legally binding. As a part of a modernisation program EU competition law will, in the future, rely more on self assessment: See Council of the European Union, Council Regulation (EC) No 1/2003 of 16 December 2002 on the Implementation of the Rules on Competition Laid Down in Articles 81 and 82 of the Treaty, opened for signature 16 December 2002,  OJ L 1/1 (entered into force 24 January 2003).
 See Mitsuo Matsushita, International Trade and Competition Law in Japan (1993) 74–169; Alex Seita and Jiro Tamura, ‘The Historical Background of Japan’s Antimonopoly Law’ (1994) University of Illinois Law Review 115; Robert Stack, ‘Western Law in Japan: the Antimonopoly Law and Other Legal Transplants’ (2000) 27 Manitoba Law Journal 391, 391.
 Japan — Film, WTO Doc WT/DS44/R (1998) [10.327]–[10.328] (Report of the Panel).
 See ibid [10.44]–[10.46]. A set of self-regulating rules, designed to ensure ‘fair trade’ in industry, promulgated by the Fair Trade Promotion Council (‘the Council’) (a grouping of private sector trade associations) was held to be a government measure. The Council was set up and operated under the guidance of the JFTC. The JFTC also played a significant role in determining the content of the self-regulating rules. There was sufficient likelihood that private parties would act in conformity with the rules to consider them administrative guidance attributable to the Japanese Government and thus a measure within the meaning of art XXIII(1)(b): at [10.293]–[10.299], [10.311]–[10.314]. On the issue of administrative guidance, see also Japan — Semi-Conductors, GATT BISD, 35th Supp, 116, GATT Doc L/6309 (1988) 153–8 (Report of the Panel, adopted on 4 May 1988). In this case the Panel’s comments were directed at art XI(1), but are clearly applicable to art XXIII(1)(b):
Japan — Film, WTO Doc WT/DS44/R (1998) [10.45] (Report of the Panel).
 Japan — Film, WTO Doc WT/DS44/R (1998) [10.36] (Report of the Panel). See also Hudec, A WTO Perspective, above n 17, 99; Ehlermann and Ehring, above n 124, 1557.
 See, eg, Ehlermann and Ehring, above n 124, 1557.
 See, eg, Japan — Film, WTO Doc WT/DS44/R (1998) [10.118]–[10.1222], [10.133]–[10.136] (Report of the Panel).
 Ibid [10.61].
 Ibid. This derives from the nature of tariff agreements: See EEC — Oilseeds, GATT BISD, 37th Supp, 86, GATT Doc L/6627 (1989),  (Report of the Panel, adopted on
25 January 1990) where the Panel considered that the main value of a tariff concession is that it provides an assurance of better market access through improved price competition. Contracting parties negotiate tariff concessions primarily to obtain that advantage. Article XXIII does not protect trade flows: at ; See also Japan — Film, WTO Doc WT/DS44/R (1998) [10.82] (Report of the Panel). The degree of legitimately expected access will depend on the measure in question. See EC — Asbestos, WTO Doc WT/DS135/R (2000) [8.285] (Report of the Panel).
 See Japan — Film, WTO Doc WT/DS44/R (1998) (Report of the Panel). But see
EC — Asbestos, WTO Doc WT/DS135/R (2000) [8.285]–[8.286] (Report of the Panel), warning that the assumption may not be legitimate where the measure in question does not relate primarily to economic conditions but to some other matter such as public health and safety.
 See Japan — Film, WTO Doc WT/DS44/R (1998) [10.72]–[10.81] (Report of the Panel). The Panel in Japan — Film at [10.79]–[10.80] set out tentative guidelines for determining whether a measure ought to have been reasonably expected.
 Where the importing state has a history of lax enforcement (and there has been no indication that this is likely to change) the exporting state would be regarded as having taken this into account in settling the terms of its trade agreement with the importing state. The exporting state will be presumed to know the importing state’s laws. See Korea — Procurement,
WTO Doc WT/DS163/R (2000) [7.119] (Report of the Panel). It is also suggested that it will be presumed to know that state’s history of enforcement. It is up to the exporting state to rebut this presumption.
 Japan — Film, WTO Doc WT/DS44/R (1998) [10.232] (Report of the Panel).
 Normally there is no binding obligation on a state to act at the request of another state. Although a number of bilateral arrangements require a state to whom such a request is made to actively consider that request (a positive comity request), there is generally no obligation to accede to the request. See discussion accompanying above n 106.
 Japan — Film, WTO Doc WT/DS44/R (1998) [10.82] (Report of the Panel).
 See authorities cited in above n 136.
 See Japan — Film, WTO Doc WT/DS44/R (1998) [10.86] (Report of the Panel).
 See ibid [10.85]–[10.86].
 Ibid [10.85]. The Panel clearly envisaged a high level of proof in the case of de facto discrimination. It should be noted that the Panel accepted that in the case of proving de facto discrimination, it is permissible to lead evidence of intention: at [10.87].
 See the discussion by the Panel in ibid [10.83]–[10.88]. Not only does the issue of causation present difficult conceptual problems, but its analysis in any given circumstance is often very fact-intensive.
 See DSU, above n 51, art 26(1)(b).
 Hudec, A WTO Perspective, above n 17, 90; Hoekman, Competition Policy and the Global Trading System, above n 141, 10–11.
 See Japan — Film, WTO Doc WT/DS44/R (1998) (Report of the Panel). For a comprehensive account of the Kodak–Fuji dispute see Linarelli, above n 139. See generally Hansen, above n 39, 1623; Barringer, ‘Competition Policy and Cross-Border Dispute Resolution’, above n 32, for a discussion of the issues involved in the case from the perspective of the legal representatives from Fuji.
 See Hudec, A WTO Perspective, above n 17, 97.
 But see Weiss, above n 36, 273, arguing that the Kodak–Fuji case was atypical of
GATT–WTO practice. According to Weiss this was ‘a “bad case” which would probably not normally have been brought, but for enterprising attorneys’.
 60 Fed Reg 35 447 (1995). The failure to exhaust local remedies before going to the WTO has been severely criticised as inconsistent with international and US custom: See Barringer and Durling, above n 48, 111.
 61 Fed Reg 30 929 (1996).
 61 Fed Reg 52 835 (1996).
 See Japan — Film, WTO Doc WT/DS44/R (1998) [10.22] (Report of the Panel). The alleged measures fell into three categories:
(1) distribution “measures”, which allegedly encourage and facilitate the creation of a market structure for photographic film and paper in which imports are excluded from traditional distribution channels; (2) restrictions on large retail stores, which allegedly restrict the growth of an alternative distribution channel for imported film; and (3) promotion “measures”, which allegedly disadvantage imports by restricting the use of sales promotion techniques.
The laws complained about included, inter alia, the Large Scale Retail Stores Law of 1982 (Japan) (which limited the size of retail stores) and the Law against Unjustifiable Premiums and Misleading Representations 1962 (Japan) (which limited the amount that manufacturers could offer to wholesalers and/or retailers by way of rebate or premium to attract custom). The policies complained about included, inter alia, Cabinet decisions relating to restrictions on foreign investment. The administrative measures included a host of guidelines and reports promulgated by MITI relating to the distribution of photographic film: at [10.23].
 See 61 Fed Reg 52 835 (1996).
 Article X of the GATT requires states to promptly publish all measures affecting international trade and to administer these measures in a ‘uniform, impartial and reasonable’ manner.
 At least one commentator, however, regards the US as having achieved a victory of sorts: Dillon, above n 33, 220, argues that the US strategy was to force Japan to amend its Large Scale Retail Stores Law of 1982. In fact, this occurred. Dillon argues that in return for a repeal of the law the US withdrew its complaint under GATS.
 Japan — Film, WTO Doc WT/DS44/R (1998) [10.38] (Report of the Panel).
 Ibid [10.90]–[10.367].
 Ibid. See Hansen, above n 39, 1626. The claim under art III(4) failed for much the same reasons as the claim under art XXIII(1)(b) failed: Japan — Film, WTO Doc WT/DS44/R (1998) [10.368]–[10.382] (Report of the Panel). For a discussion on the claim under art III(4) see Linarelli, above n 139, 324–7. The US also failed in its claim under art X(1) (alleging a lack of transparency): Japan — Film, WTO Doc WT/DS44/R (1998)
[10.383]–[10.401] (Report of the Panel).
 See Dillon, above n 33, 220.
 See above n 58 and accompanying text.
 Some commentators have doubted that a web of bilateral arrangements could prove effective, let alone efficient: See above n 108 and accompanying text.
 Kennedy, ‘Foreign Direct Investment and Competition Policy’, above n 36, 587–8; Allan Fels, ‘Competition and Globalisation’ (Paper presented at the Conference on Trade and Competition with the EU in the New Millenium, The University of Melbourne,
14–16 December 2000) 5 <http://www.accc.gov.au/content/item.phtml?itemId=255536 & no eId=file3eb0831b9d927 & fn=COMPETITION%20 & amp;amp;%20GLOBALISATION.pdf> at 1 October 2004.
 Treaty Establishing the European Economic Community, opened for signature 25 March 1957, 298 UNTS 11 (entered into force 1 January 1958).
 See Frederic Scherer, ‘International Competition Policy and Economic Development’ in Lee-Jay Cho and Yoon Hyung Kim (eds), The Multilateral Trading System in a Globalizing World (2000) 67–8.
 See generally OECD, ‘Consistencies and Inconsistencies between Trade and Competition Policies’ in OECD, Trade and Competition Policies for Tomorrow (1999) 21; Tarullo, above n 7; Marsden, A Competition Policy for the WTO, above n 31, 125–9; Weiss, above n 36. See also Douglas Rosenthal, above n 42, 547–51; Applebaum, above n 48, 480–5, comparing US trade law and antitrust law. For a discussion of the historical tensions between the objects of trade and competition policy, see Trebilcock, ‘Competition Policy and Trade Policy’, above n 36, 72–4.
 This is not to say that efficiency is the only goal of each and every competition regime. This is clearly not the case. Rather, it simply means that efficiency enhancement is one of the major goals of all regimes.
 See ICPAC Report, above n 9, 210; Epstein, above n 30, 361–2. This issue is discussed in Marsden, A Competition Policy for the WTO, above n 31, 125–8. See also Philip Marsden, ‘A WTO “Rule of Reason”?’ (1998) 19 European Competition Law Review 530; but see Thomas Schoenbaum, ‘The Concept of Market Contestability and the New Agenda of the Multilateral Trading System’ (1996) American Society of International Law Insights <http://www.asil.org/insights/insight2.htm> at 1 October 2004, who claims that many at the WTO and the OECD have argued that the objective of WTO trade policy should be ‘international contestability of markets’, based on the principles advanced by William J Baumol. However, it is doubtful that most national trade authorities pursue this vision. First, international trade exhibits characteristics of a prisoner’s dilemma. Although all states would be better off with reduced access barriers, generally speaking, trade authorities act rationally by demanding the maximum possible market access in return for minimal dismantling of trade barriers. Secondly, it cannot be assumed that trade authorities generally act in the public interest. It is just as likely that they act for local producers. This reinforces the first point, that the object of national trade authorities is a bargain-driven notion of market access.
 See Barringer, ‘Competition Policy and Cross Border Dispute Resolution’, above n 32, 462. Barringer was the US legal representative for Fuji in the Kodak–Fuji dispute. Barringer, has distinguished trade and competition authorities on the same basis: at 460. See also Douglas Rosenthal, above n 42, 548–9; Alan Wolff, Thomas Howell and John Magnus, ‘Trade and Competition Policy: A Suggested US Strategy’ (Paper submitted to the ICPAC, 4 November 1998) <http://www.dbtrade.com/publications/icpac.pdf> at 1 October 2004.
 See Hudec, A WTO Perspective, above n 17, 82–3. While some trade authorities may refer to the goal of contestable markets, trade authorities are influenced by exporters whose objectives are less likely to be driven by the principles of welfare economics.
 It has been suggested that US trade policy has shifted from a focus on national treatment to a focus on reciprocity of treatment. See generally Michael Young, ‘Lessons from the Battlefront: US–Japan Trade Wars and their Impact on the Multilateral Trading System’ (2001) 33 George Washington International Law Review 753.
 Hudec, A WTO Perspective, above n 17, 83.
 Hoekman, Competition Policy and the Global Trading System, above n 141, 8.
 See, eg, Wolff, ‘The (Notionally) Bridgeable Chasm’, above n 31.
 See Hoekman, above n 141, 19. See also Imelda Maher, ‘Competition Law in the International Domain: Networks as a New Form of Governance’ (2002) 29 Journal of Law and Society 111, 126–7, who argues that the WTO WGTCP Policy quickly reframed and narrowed its trade and competition agenda to fit trade concerns. See generally David Leebron, ‘Linkages’ (2002) 96 American Journal of International Law 5, 26, who discusses the possibility of institutional bias where diverse areas of regulation are linked in a single institutional structure; see also Tarullo, above n 7.
 For a discussion on this point see Marsden, A Competition Policy for the WTO, above n 31, 159–99.
 Robert Litan and Carl Shapiro, ‘Antitrust Policy During the Clinton Administration’ (2001)
56 <http://repositories.cdlib.org/iber/cpc/CPC01 – 022> at 1 October 2004.
 Competition law remedies also vary significantly from state to state. For example, the US permits successful private applicants to claim treble damages. Some states do not even allow private suits. In some states cartel activities attract criminal sanctions, but not in others. On remedies available in various states, see generally OECD, Joint Group on Trade and Competition, Remedies Available to Private Parties Under Competition Laws (2000) COM/DAFFE/CLP/TD(2000)24/FINAL <http://www.olis.oecd.org/olis/2000doc.nsf/Link To/COM-DAFFE-CLP-TD(2000)24-FINAL> at 1 October 2004.
 Epstein, above n 30, 345.
 See, eg, Karl Meessen, ‘Competition of Competition Laws’ (1989) 10 Northwestern Journal of International Law and Business 17; William Dodge, ‘An Economic Defense of Concurrent Antitrust Jurisdiction’ (2003) 38 Texas International Law Journal 27.
 See, eg, ICPAC Report, above n 9, 273; Tarullo, above n 7, 491–4.
 A number of proposals are discussed at length in Marsden, A Competition Policy for the WTO, above n 31, 159–287. See also Douglas Rosenthal, above n 42.
 See, eg, Frederic Jenny, ‘Competition Law and Policy: Global Governance Issues’ (2003) 26 World Competition 609, 620. The strongest supporters of the unification approach are to be found amongst those who view the maximisation of global welfare as the most significant issue. This is a normative stance and even its proponents are sceptical about its prospects: See, eg, Guzman, ‘Is International Antitrust Possible?’, above n 7, 1548.
 OECD, Joint Group on Trade and Competition, International Options to Improve the Coherence between Trade and Competition Policies (2000) COM/TD/DAFFE/CLP(99)102/FINAL, 3 <http://www.olis.oecd.org/olis/1999doc.nsf/link to/com-td-daffe-clp(99)102-final> at 1 October 2004.
 See, eg, Mario Monti, ‘The EU Views on Global Competition Forum’ (Speech delivered at ABA meeting, Washington, 29 March 2001) <http://europa.eu.int/rapid/press ReleasesAction.do?reference=SPEECH/01/147 & format=HTML & aged=0 & language=EN & guiLanguage=en> at 1 October 2004, stressing that the EC favours neither comprehensive rule unification nor case by case global dispute resolution. For a discussion on the EU see Marsden, A Competition Policy for the WTO, above n 31, 161.
 See ICPAC Report, above n 9, 263. The history of the US approach to international competition agreements has been detailed in a number of articles. See, eg, Waller, ‘The Internationalization of Antitrust Enforcement’, above n 95.
 See, eg, Manisha Sheth, ‘Formulating Antitrust Policy in Emerging Economies’ (1997)
86 Georgetown Law Journal 451.
 See OECD Report on Trade and Competition Policies, above n 32, 14.
 See, eg, the results of the WTO Ministerial Conference at Cancún, where one of the reasons for failure to reach an agreement was the attitude of the developing states to the inclusion of competition policy: See WTO, Briefing Note, Day 5, above n 1.
 Basic Telecommunications Agreement, above n 122, [1.1].
 Reference Paper, above n 122, [1.1].
 Ibid [1.2].
 See Telmex, WTO Doc WT/DS204/R (2004) (Report of the Panel).
 An analysis of this case is beyond the scope of this article. For a criticism of the decision see Marsden, ‘Trade and Competition’, above n 122.
 See Warner, above n 122, 339–48.
 See, eg, Wolfgang Fikentscher, ‘Antitrust, Market Conceptualization and the World Trade Organization — the Convention Approach’ in Roger Zäch (ed), Towards WTO Competition Rules: Key Issues and Comments on the WTO Report (1998) on Trade and Competition (1999) 73.
 TRIPS Agreement, above n 51, sets out minimum standards for the protection of intellectual property.
 See Josef Drexl, ‘Trade-Related Restraints of Competition — The Competition Policy Approach’ in Zäch (ed), above n 220. See generally Guzman, ‘Is International Antitrust Possible?’, above n 7, who favours such an agreement; Andrew Guzman, ‘Introduction — International Regulatory Harmonization’ (2002) 3 Chicago Journal of International Law 271, 273. But see Brian Hindley, ‘Competition Law and the WTO: Alternative Structures for Agreement’ in Jagdish Bhagwati and Robert Hudec (eds), Fair Trade and Harmonization: Prerequisites for Free Trade? (1996) vol 2, 333, 333–48, who discusses, and generally dismisses, a TRIPS-type competition agreement. See also Hudec, A WTO Perspective, above n 17, 86–9, who questions both the efficacy and the likelihood of a TRIPS-type agreement.
 For example, the principle of national treatment appeared in the Paris Convention for the Protection of Industrial Property, opened for signature as revised 14 July 1967,
828 UNTS 306, art 10bis (entered into force 26 April 1970 insofar as concerns arts 13–30) and the Berne Convention for the Protection of Literary and Artistic Works, opened for signature as revised 14 July 1967, 1 UNTS 217 (entered into force 29 January 1970 insofar as concerns arts 22–38(1)). Intellectual property regulation also spawned an international organisation, first the United International Bureaux for the Protection of Intellectual Property and subsequently the World Intellectual Property Organization.
 See Braithwaite and Drahos, above n 2, 63.
 The role of US business interests in the creation of the TRIPS Agreement is outlined in ibid 61–3.
 In addition, there has been a distinct reluctance on the part of the US to bring its anti‑dumping laws into compliance with GATT–WTO requirements: See Trade Policy Review Body, Trade Policy Review — United States, above n 45, 47–51.
 For example, developing states have been disillusioned by the manner in which the developed states have sought to interpret the exceptions to patentable protection contained in arts 27(3)(b), 30 and 31. See, eg, Review of the Provisions of Article 27.3(b): Summary of Issues Raised and Points Made, WTO Doc IP/C/W/369 (2002) (Note by Secretariat), which discusses the conflicting interpretations of art 27(3)(b) made by developed and developing states.
 See generally ABA Report on Market Access, above n 19, 80, 95; see also ICPAC Report, above n 9, 201–80. It is difficult from the available evidence to determine whether horizontal or vertical restraints present the more significant problem. Some commentators however, feel that vertical restraints are more important: See Barringer, ‘Competition Policy and Cross Border Dispute Resolution’, above n 32, 469 who argues that most complaints about private access barriers involve vertical restraints; see also Jason Kearns, ‘International Competition Policy and the GATS: A Proposal to Address Market Access Limitations in the Distribution Services Sector’ (2001) 22 University of Pennsylvania Journal of International Economic Law 285, 311.
 See OECD, OECD Recommendation, above n 69.
 The OECD Recommendation does, however, provide that states must be transparent in their exemptions and prepared to review the exemption from time to time.
 See Dianne Wood, ‘Cooperation and Convergence in International Antitrust: Why the Light is Still Yellow’ (Speech delivered at the The New Antitrust Paradox: Policy Proliferation in the Global Economy Conference, American Enterprise Institute for Public Policy Research, Washington DC, 22 April 2003) <http://www.aei.org/news/newsID.17274/news_detail.asp> at 1 October 2004, arguing that a close inspection of the OECD Recommendation leads to the conclusion that states have not agreed to very much at all: See also Warner, above n 122, 309–11.
 For example, there is no evidence to suggest that nations are prepared to base an agreement on per se provisions of the US Sherman Antitrust Act of 1890, 15 USC § 1. This solution was suggested by former Chief Democratic Counsel for the US House Judiciary Committee, Julian Epstein. See Epstein, above n 30, 365–6. Again, the solution faces the problem that there is little consensus on import cartels which arguably represent the most trade-inhibiting type of horizontal restraint.
 See, eg, Trebilcock, ‘Competition Policy and Trade Policy’, above n 36, 105; see also Marsden, A Competition Policy for the WTO, above n 31, 108–12.
 In the US this is referred to as the ‘rule of reason’ approach.
 For a comparison of US and EU approaches to exclusionary practices see Fox, ‘What is Harm to Competition?’, above n 11: See also Marsden, A Competition Policy for the WTO, above n 31, 109. On the attitude of the European Commission to vertical restraints see generally Commission Regulation (EC) No 2790/1999 of 22 December 1999 on the Application of Article 81(3) of the Treaty to Categories of Vertical Agreements and Concerted Practices  OJ L 336/21 (entered into force 1 January 2000) and Commission Notice — Guidelines on Vertical Restraint  OJ C 291/1. See also European Commission, Competition Policy in Europe: The Competition Rules for Supply and Distribution Agreements (2002); Philip Lowe, ‘An Overview of EU Competition Law and Policy on Commercial Practices’ (Speech delivered at ABA Meeting, Brussels,
16 October 2003) <http://europa.eu.int/comm/competition/speeches/text/sp2003_038_
en.pdf> at 1 October 2004.
 See Commission Decision of 30 October 2002 relating to a Proceeding pursuant to Article 81 of the EC Treaty and Article 53 of the EEA Agreement  OJ L 255/33; Commission Decision of 28 January 1998 relating to a proceeding pursuant to Article 85 of the EC Treaty  OJ L 124/60. See also Scherer, above n 189, 69.
 See discussion at text accompanying above n 33. The preservation of rule diversity has been seen as an important factor against unification: See Meessen, above n 204; Dodge, ‘An Economic Defense of Concurrent Antitrust Jurisdiction’, above n 204.
 See, eg, Marsden, A Competition Policy for the WTO, above n 31, 131.
 Eleanor Fox has also proposed a market blockage rule. See Fox, ‘Toward World Antitrust and Market Access’, above n 7. However, given that the design of the rule is left entirely to each state, Fox’s proposal more closely resembles a non-discrimination scheme than a unification scheme.
 See Kearns, above n 228, 311–13: See also Charles Gastle, ‘The Convergence of International Trade and Competition Law through a WTO Market Access Code’ (1999)
8 Currents International Trade Law Journal 3, 13–14. Epstein, above n 30, 367, has argued that a global market access rule for vertical restraints should be based on the US essential facilities doctrine. This is a more restricted rule than one based on consumer welfare.
 See Kearns, above n 228, 313, who refers to the requirement that the action or inaction by the country’s competition authority ‘significantly distort trade in the market access context’; see also Gastle, above n 238, 14, who argues that the complaining party should be required to show a causal connection between the impugned conduct and its lack of success in entering the market. This would include evidence that the complainant had the product and the capacity (financial and technical) to enter the foreign market successfully.
 See, eg, Gastle, above n 238, 14. This is a fairly standard requirement of those who recommend WTO involvement.
 See Kearns, above n 228, 312, who argues that the Panel’s decision should not be binding. It should be noted that Gastle, above n 238, 14, would limit the right of review to situations where the complaining party was able to establish that it would have been treated differently under its own law.
 For example, Gastle, above n 238, 14, would only apply the market access rule if the import market share of domestic consumption is less than 15 per cent. Gastle would also provide an exemption for any public conduct that is currently exempt by the GATT–WTO agreements: at 15–16; for example, the exemptions under art XX of the GATT, above n 2.
 See, eg, Charles Weller, ‘Harmonizing Antitrust Worldwide by Evolving to Michael Porter’s Dynamic Productivity Growth Analysis’ (2001) 46 Antitrust Bulletin 879.
 See, eg, Alan Wolff, ‘WTO Dispute Settlement and the Particular Problem of Trade and Competition Policy’ (Speech delivered at the Conference on Dispute Resolution in the World Trade Organization, Washington DC, 24 June 1998); see also ICPAC Report,
above n 9, 210, 249–50.
 The WTO is discussed below at text accompanying below n 290.
 For a discussion of possible rules based on maintaining rivalry see Marsden, A Competition Policy for the WTO, above n 31.
 See William Kolasky, ‘What is Competition?’ (Speech delivered at the Seminar on Convergence, The Hague, 28 October 2002) expressing the view of the Department of Justice’s Antitrust Division that efficiencies are often more important than concentration.
 See Eleanor Fox, ‘Global Markets, National Law, and the Regulation of Business: A View from the Top’ (2001) St John’s Law Review 383, 394–5. Fox summarises the difference between US and EU merger policies:
United States authorities normally approve a merger if it cannot be proven to raise prices. The European Commission normally disapproves a merger or imposes regulatory conditions if the merger significantly enhances the market share of a dominant firm, creates joint dominance, or seriously distorts the playing field for competitors.
 In the GE–Honeywell merger the US Department of Justice determined that the efficiencies generated by the merger would put the merged entity in a position to offer better products to consumers at cheaper prices. Consequently, it favoured approving the merger. See Charles James, ‘International Antitrust in the 21st Century: Cooperation and Convergence’ (Speech delivered at the OECD Global Forum on Competition, Paris, 17 October 2001). The EC, on the other hand, relied on what has been termed ‘portfolio theory’ to conclude that the merged entity would stifle competition. The Commission opposed the merger: See Initiation of Proceedings (Case Comp/M.2220 — General Electric–Honeywell)  OJ C 74/6. See generally Alec Burnside, ‘GE, Honey, I Sunk the Merger’ (2002) 23 European Competition Law Review 107. In the Boeing–McDonell‑Douglas merger the US Federal Trade Commission approved the merger, but the EC rejected it. For a discussion on this merger, see Daniel Gifford and E Thomas Sullivan, ‘Can International Antitrust be Saved for the Post-Boeing Merger World? A Proposal to Minimize International Conflict and to Rescue Antitrust from Misuse’ (2000) 45 Antitrust Bulletin 55.
 The US believed that the EC had rejected the GE–Honeywell merger because the merged entity would be too good a competitor. For US officials this was antitrust heresy. It amounted to protecting the competitor, not the process of competition: See James,
above n 251; see also William Kolasky, ‘US and EU Competition Policy: Cartels, Mergers, and Beyond’ (Speech delivered at the Council for the United States and Italy Bi-Annual Conference, New York, 25 January 2002).
 See discussion at text accompanying above n 192.
 Many European commentators also have objections to the EU approach to merger analysis: See, eg, Matthias Pflanz and Cristina Caffarra, ‘The Economics of GE/Honeywell’ (2002) European Competition Law Review 115, who criticise the EC’s analysis of portfolio effects in the GE/Honeywell case; see also Burnside, above n 251, 108, who claims that the EC’s substantive analysis was speculative and economically ‘novel’. See also discussion in Marsden, A Competition Policy for the WTO, above n 31, 214–27.
 See ICPAC Report, above n 9, 249–50; Wolff, ‘The (Notionally) Bridgeable Chasm’,
above n 31, 187–9; Wolff, ‘WTO Dispute Settlement and the Particular Problem of Trade and Competition Policy’, above n 246.
 Wolff, ‘WTO Dispute Settlement and the Particular Problem of Trade and Competition Policy’, above n 246.
 Wolff, Howell and Magnus, above n 193, 8.
 See discussion in Marsden, A Competition Policy for the WTO, above n 31, 187–92.
 For a range of criticisms of this type of market access rule, see ibid 241–7.
 See Wolff, Howell and Magnus, above n 193, 7–8.
 On § 301 of the Trade Act, see discussion at text accompanying above n 47.
 Although the expression ‘harmonisation’ is used widely throughout literature dealing with trade and competition law, this use is not consistent. This article adopts the meaning ascribed to ‘harmonisation’ by Martin Boodman, ‘The Myth of Harmonization of Laws’ (1991) 39 American Journal of Comparative Law 699, 702, who states that harmonisation is a process in which diverse elements are combined or adapted to each other so as to form a coherent whole while retaining their individuality. In its relative sense, harmonisation is the creation of a relationship between diverse things. Its absolute and most common meaning, however, implies the creation of a relationship of accord or consonance.
 See Waller, ‘Neo-Realism and the International Harmonization of Law’, above n 2, who discusses the harmonisation of competition laws.
 A considerable amount of work has been done at both the OECD and the WTO on this issue. See, eg, OECD, Joint Group on Trade and Competition, Outline of (A) Core Principles, Common Approaches and Common Standards and (B) Bilateral and Multilateral Approaches (1999) COM/TD/DAFFE/CLP(98)97/FINAL <http://www.olis.oecd.org/olis/1998doc.nsf/linkto/com-td-daffe-clp(98)97-final> at 1 October 2004; OECD, Joint Group on Trade and Competition, International Options to Improve the Coherence between Trade and Competition Policies (2000) COM/TD/DAFFE/CLP(99)102/FINAL <http://www.olis.
oecd.org/olis/1999doc.nsf/linkto/com-td-daffe-clp(99)102-final> at 1 October 2004; and OECD, Joint Group on Trade and Competition, The Role of ‘Special and Differential Treatment’ at the Trade, Competition and Development Interface (2001) COM/TD/DAFFE/CLP(2001)21/FINAL <http://www.oecd.org/dataoecd/49/37/2072318.pdf> at 1 October 2004. See also WGTCP, The Fundamental Principles of Competition Policy, WTO Doc WT/WGTCP/W/127 (1999) (Background Note by the Secretariat).
 See discussion at text accompanying above n 142.
 See Hansen, above n 39, 1642–9. The expanded meaning of ‘measure’ could, of course, go beyond competition law and include any government action that had the relevant effect. For example, the Japanese Government measures relied upon by the US in the Japan — Film dispute included a raft of legislative and administrative provisions, only some of which were directly related to Japan’s competition law. See discussion at text accompanying above n 171.
 See, eg, Hudec, A WTO Perspective, above n 17, 101.
 For example, many competition authorities set and publish enforcement priorities.
 See Japan — Film, WTO Doc WT/DS44/R (1998) [10.86] (Report of the Panel), which noted that the principle of de facto discrimination has been consistently applied by GATT and WTO panels in respect of arts I (Most Favoured Nation) and III (National Treatment) of the GATT.
 See ibid [10.88]. The Panel, however, commented that intention may not be entirely irrelevant. Evidence of intent to discriminate may have probative value.
 See Hansen, above n 39, 1642–3.
 Ibid 1643. Hansen, who suggested the balancing test, was herself wary of giving WTO panels the power to make such value judgments. Consequently, she recommended that ‘WTO tribunals should give greater weight to the importing nation’s own laws, as well as emerging international norms and the requirements of transparency’.
 See Trebilcock, ‘Competition Policy and Trade Policy’, above n 36, 99, who argues that a least trade restrictive means test has often been applied by GATT–WTO panels when making determinations concerning the exceptions in art XX of the GATT.
 See, eg, Hudec, A WTO Perspective, above n 17, 100.
 See discussion at text accompanying above n 167.
 See, eg, John McGinnis, ‘The Political Economy of International Antitrust Harmonization’ (2003) 45 William and Mary Law Review 549. Some proposals adopt non-discriminatory enforcement as a baseline, but go further. For example, the EU has proposed an additional rule against cartels and an agreement on core competition principles: See, eg, Mario Monti, ‘The EU Views on Global Competition Forum’ (Speech delivered at ABA meeting, Washington, 29 March 2001) <http://europa.eu.int/rapid/pressReleasesAction.do?reference=SPEECH/01/147 & form at=HTML & aged=0 & language=EN & guiLanguage=en> at 1 October 2004.
 See Fox, ‘Toward World Antitrust and Market Access’, above n 7, 24. The allocation of prescriptive jurisdiction by agreement to the importing state is a form of conflict of laws rule. Such a rule is necessary to prevent system clashes caused by the extraterritorial application of domestic competition laws. The expansive jurisdiction claimed by US courts (‘the effects doctrine’) and increasingly by other jurisdictions, make the rule necessary (extraterritoriality is discussed at Part IV(A)(2) of this article). There are a number of possible conflict of law rules, but most would be considerably more problematic (normatively and politically) than the suggested rule. For example, one possibility is to allocate jurisdiction on the basis of some form of interest-balancing test. This kind of test owes much to notions of comity. However, there are doubts that it is a useful juridical rule. Leaving aside the case by case difficulties of applying such a rule, there is first the problem of obtaining agreement on a list of factors that are to be weighed, and secondly, the problem of forum bias. It should be noted that the optimal choice of law rule may be different where the competition issue is not market access but, for example, an international merger or an international cartel.
 Determining the precise form that due process and transparency ought to take raises complex questions which are beyond the scope of this article; see generally Jeffrey Waincymer, ‘Transparency of Dispute Settlement within the World Trade Organization’  MelbULawRw 32; (2000) 24 Melbourne University Law Review 797.
 See generally Fox, ‘Toward World Antitrust and Market Access’, above n 7, 24.
 This is discussed by Trebilcock, ‘Competition Policy and Trade Policy’, above n 36, 99. Trebilcock is of the opinion that the keiretsu system should not be regarded as discriminatory given its central role in Japanese industrial and commercial organisation. It is unlikely that a Japanese court (much less a Japanese regulator) would declare the keiretsu organisational structure an unlawful discrimination.
 See Marsden, A Competition Policy for the WTO, above n 31, 148–50.
 See Part VI(C)(2) of this article.
 This criticism has been made about the power of WTO panels in respect of the TRIPS Agreement.
 See Warner, above n 122, 317–18.
 This has been a constant criticism made by US officials: See Tarullo, above n 7, 490–2.
 An international competition authority was suggested by the International Antitrust Code Working Group set up in 1993: See Draft International Antitrust Code as a GATT-MTO-Plurilateral Trade Agreement (International Antitrust Code Working Group Proposed Draft 1993), published and released 10 July 1993, 65 Bureau of National Affairs Antitrust and Trade Regulation Report 1628 (Special Supp 19 August 1993) (‘Draft Code’). For a discussion of the Draft Code, see generally Daniel Gifford, ‘The Draft International Antitrust Code Proposed at Munich: Good Intentions Gone Awry’ (1997) 6 Minnesota Journal of Global Trade 1. The Draft Code approach to an international competition authority is favoured by Josef Drexl, ‘Do We Need “Courage” for International Antitrust Law? Choosing between Supranational and International Law Principles of Enforcement’ in Josef Drexl (ed), The Future of Transnational Antitrust — From Comparative to Common Competition Law (2003) 311.
 See discussion at text accompanying above n 97.
 See Joel Klein, ‘A Note of Caution with respect to a WTO Agenda on Competition Policy’ (Speech delivered at the Royal Institute of International Affairs, London,
18 November 1996); see also ICPAC Report, above n 9, 279. This view has been maintained by the Bush Administration: See Charles James, ‘Antitrust in the Early 21st Century: Core Values and Convergence’ (Speech delivered at the Program on Antitrust Policy in the 21st Century, Brussels, 15 May 2002); Ehlermann and Ehring, above n 124, 1514.
 Most commentators see the WTO as the logical repository for any market access rule:
See Fox, ‘International Antitrust and the Doha Dome’, above n 64, 926–7; Guzman, ‘Is International Antitrust Possible?’, above n 7, 1501.
 Braithwaite and Drahos, above n 2, 195, argue that UNCTAD has become largely irrelevant.
 See John Jackson, ‘The Linkage Problem — Comments on Five Texts’ (2002) 96 American Journal of International Law 118, 125.
 See the discussion in Ehlermann and Ehring, above n 124.
 See Tarullo, above n 7, 489–92, who argues it is necessary to be aware of the institutional limitations of the WTO both as a forum for negotiating a global competition agreement and as a forum for providing dispute resolution services; cf Ehlermann and Ehring, above n 124, 1532, who reject the probability of trade bias.
 ICPAC Report, above n 9, 27–8. See also Wolff, ‘The (Notionally) Bridgeable Chasm’, above n 31, 184. It has been suggested that this problem could be overcome by adding competition experts to the WTO personnel. Alternatively, WTO panels could call upon competition experts as the need arises.
 See, eg, Kennedy, ‘Foreign Direct Investment and Competition Policy’, above n 36, 623.
 See, eg, Wolff, ‘The (Notionally) Bridgeable Chasm’, above n 31, 184–5, which argues that the WTO Secretariat demonstrated its inability to handle complex, factually-intensive cases in Japan — Film.
 See, eg, ibid 185.
 See generally Robert Hudec, ‘The New WTO Dispute Settlement Procedure: An Overview of the First Three Years’ (1999) 8 Minnesota Journal of Global Trade 1; Jackson, ‘The Role and Effectiveness of the WTO Dispute Settlement Mechanism’, above n 139, 179; see also Jan Aart Scholte, Robert O’Brien and Marc Williams, ‘The WTO and Civil Society’ (1999) 33 Journal of World Trade 107.