Sydney Law Review
BRENDAN J SWEENEY[∗]
Whether there is a need for global competition rules has been a consistent and contested topic of discussion over the last decade and a half. Hopes for a WTO-sponsored global competition agreement received a significant setback at Cancún when competition was removed from the Doha agenda. The lesson from Cancún is that states are only ever likely to agree to global competition rules if there is a compelling international rationale for such rules. Global competition rules must solve problems that other processes — whether unilateral or bilateral, binding or non-binding — cannot solve. The starting point, therefore, is to investigate the nature and intensity of the problems to which global rules may be a solution. If the problems are indeed resistant to other solutions and if the problems are sufficiently intense, global competition rules seem indicated.
In the past decade or so, competition (or antitrust) law and policy has been the subject of vigorous debate in international discussions. There are two main reasons for this. First, there has been a notable shift towards more market-driven forms of political economy. This trend is noticeable not only in Eastern Europe and the former Soviet republics, but also in communist states such as China and Vietnam, in developing democracies such as India, and in the industrialised democracies where there has been a move away from the welfare state to a greater reliance on market-driven solutions. In this new world promoting competition and efficiency has become a central economic principle; and competition law has become a core regulatory mechanism. Secondly, the rapid growth of commercial globalisation has emphasised the tension between conflicting national economic policies.
The widespread move towards more market-driven models of political economy combined with the expanding internationalisation of business and commerce has led to a series of proposals for global competition rules. To date these proposals have been hotly contested. The purpose of this paper is to investigate a possible foundation for pursuing a global competition agreement (‘GCA’). The paper commences with a short history of international competition law and policy. It then describes an analytical structure for examining the need for, and content of, a GCA. This involves developing a typology of competition conduct. The various ‘types’ of competition conduct are then analysed to determine whether they raise international problems and if so, the nature and intensity of the problem. Finally, the paper ventures to suggest some solutions.
The first attempt to forge international competition rules occurred at the end of World War II as part of the negotiations that ultimately resulted in the introduction of the General Agreement on Tariffs and Trade (‘GATT’). Competition rules were included in the Havana Charter, which was the proposed charter for the International Trade Organisation (‘ITO’). The ITO, however, failed to obtain US congressional approval and never came into effect. Considering that at the time only the US had a functioning competition law this seems in hindsight inevitable.
In subsequent years the number of domestic competition regimes grew. In 1958 the European Economic Community was formed and some of its core rules were competition rules. The Allied occupation forces imposed a competition law upon Japan. Although largely ignored in its early years, it was never repealed. Competition rules appeared on the agenda at United Nations Conference on Trade and Development (‘UNCTAD’) and at the Organisation for Economic Co-operation and Development (‘OECD’).
The most significant development, however, in competition policy as a global issue was its inclusion in 1996 on the agenda at the newly created World Trade Organisation (‘WTO’). A WTO working group was formed to examine the interaction between trade and competition policy. The appearance of competition policy at the WTO was very much an initiative of the European Union (‘EU’) with support from Japan and Canada. In contrast, the US was much less enthusiastic and the developing states were wary. Given this lack of universal commitment, it is perhaps not surprising that competition policy was one of the early casualties of the stalled Doha Round. In 2003 the Ministerial Conference at Cancun in Mexico removed competition policy from the Doha agenda.
Competition policy remains an issue at UNCTAD and at the OECD, but the emphasis is on capacity building and improving knowledge about competition issues rather than on creating a global competition agreement (‘GCA’). In keeping with this network approach the International Competition Network (‘ICN’) was formed in 2001. The ICN, a body made up of domestic competition authorities, is essentially a vehicle for disseminating ideas about competition policy with a view to encouraging best practices. It is an initiative of the US and reflects the US preference for avoiding any form of international agreement.
Whether discussion on a GCA will be revived, either at the WTO or some other forum, depends on a variety of inter connected economic and political factors. Those factors have not necessarily lost their cogency simply because of the events at Cancun.
There are a variety of reasons for championing a GCA. Some view a GCA, especially one linked to the WTO, as a useful way of imposing economic reforms on reluctant states. A GCA may also be viewed as a way of overcoming the sub-optimality of domestic regulation; as long as states consider only their national interests (their own consumers and producers), domestic regulation will produce a welfare deficit when compared to that which might be obtained under global regulation. Finally, a GCA may be seen as a way of avoiding the tensions that the extraterritorial application of domestic competition laws often generates. For example, the extraterritorial application of US antitrust law has elicited strong reactions from a number of the US’s main trading partners, including ‘blocking’ laws designed specifically to impede the enforcement of the US laws.
To concentrate on these objectives, however, is to view a GCA from a remote and idealised perspective. In the real world they do not provide sufficient leverage upon which to base negotiations for a GCA. There is no evidence that states are prepared to concede the reallocation of power necessary to achieve these goals.
One of the major conclusions of nearly every analysis done on the issue of international competition policy is that a comprehensive GCA – one that resembles a global competition regime – is most improbable. For a variety of reasons most regard such a regime as undesirable. Therefore, the operative model will remain domestic regulation.
Domestic regulation, however, does not always operate effectively where the relevant conduct is transnational. First, transnational conduct creates incentives for protectionism which domestic regulation is not always able to handle. Protectionism is an international problem where it undermines international trade commitments. Secondly, transnational conduct exposes the problem of concurrent jurisdiction. Concurrent jurisdiction is becoming ever more common in an age of commercial globalisation. While concurrent jurisdiction does not of itself necessitate international competition rules, there are occasions where the consequences do warrant some form of cooperation. The clearest example of this is where one state assumes by default the role of international regulator. This is an international problem because in such cases domestic regulation is rendered meaningless.
It is in these areas, where domestic regulation fails, that international rules are needed most. Given the objections to uniformity and the difficulties in negotiating international rules, some limiting or constraining principle is required. It is suggested that the shape of a GCA should be determined by the minimum requirements necessary to overcome the international problems exposed by domestic regulatory failure. In other words, the shape of a GCA will be informed by what is necessary to overcome protectionism and problems such as the de facto international regulator.
The first task is to categorise competition conduct in a way that facilitates an understanding of the nature of the problems presented by the failure of domestic enforcement. An early meeting of the WTO Working Group on Trade and Competition (‘WGTCP’) commented:
With regard to analytical categories of anti-competitive practices of enterprises and associations that had an impact on international trade, and the discussion of empirical examples of such practices, the representatives of some Members, including the United States and the European Community and its member States, said that the examples discussed by Members and observers in the meeting could be grouped in three broad categories: (i) practices affecting market access for imports; (ii) practices affecting international markets, where different countries were affected in largely the same way; and (iii) practices having a differential impact on the national markets of countries. The observer from the World Bank said that this was a useful analytical framework, while noting that other distinctions could also be relevant. For example, he suggested that it could be useful to distinguish between anti-competitive agreements involving only foreign (exporting) firms, and those which involved both foreign and domestic firms.
This classification is useful, but not entirely satisfactory. The second and third categories depend solely on effects; category (ii) applies where the effects are the same in different states and category (iii) where the effects vary from state to state. This makes no allowance for the type of conduct. The type of conduct is important because often it affects the nature of the international problem to be faced. For example, states tend to agree on the effects of price fixing cartels, but do not agree on the competitive effects of distribution agreements. Therefore, the problems of regulating price fixing cartels generally tend to be quite different to the problems of regulating distribution agreements. Yet, under the WGTCP classification both types of conduct could be in the same category.
Another possibility is to analyse anti-competitive conduct solely according to the nature of the conduct. This has an attraction in that it is a familiar device of competition discourse. Thus, anti-competitive practices could be examined according to whether they involved horizontal agreements, vertical agreements, unilateral conduct or mergers. This classification, however, has problems in that it ignores the possibility that, for example, a domestic horizontal agreement designed to block imports creates different regulatory issues to an international horizontal agreement to divide up world markets.
This paper adopts a system of categorisation that is essentially a hybrid between the WGTCP system and one based solely on the nature of the conduct. Using this categorisation as a vehicle for analysis, it is possible to build a minimum set of rules. The resultant GCA may not be negotiable (at least in the short term), but it does provide a benchmark.
The first category of anti-competitive conduct involves private import barriers. Private import barriers are said to undermine existing and future trade agreements. To the extent that this is true they represent an international problem. A GCA may be a useful tool for solving this problem. The second category involves export cartels. The practice of tolerating export cartels simply because the harm is experienced by foreign consumers is antithetical to the principles of trade liberalisation. A GCA may be essential to overcoming this problem. The third category considers export dumping — that is, selling goods abroad at lower than competitive prices. Export dumping is unique in that it is already the subject of an international agreement (‘the anti-dumping rules’). For this reason it requires separate treatment. There is a strong case that nations misuse the anti-dumping rules to protect local industries from more efficient foreign competition. Continuing use of the anti-dumping rules in this fashion threatens the international trading system because it weakens commitment to the notion of comparative advantage upon which the international trading system is built. This is an international problem. It has been suggested that the anti-dumping rules be replaced with competition rules more attuned to the principles of trade liberalisation. The fourth category is international cartels. There is widespread recognition that hard core international cartels ought to be suppressed. The issue is whether a GCA is necessary to accomplish this. Mergers and monopolistic conduct raise similar international regulatory issues, and together they make up the final category. Mergers and monopolistic conduct excite international attention for a number of reasons, but the most critical regulatory issue centres on those cases where it is not possible to construct different national remedies. Where only a single remedy is possible – that is, an international remedy – domestic regulation becomes an exercise in international regulation.
In the years since it first came into force shortly after the end of the Second World War the GATT has brought about a significant decrease in government tariffs. As a consequence international trade has increased substantially. There is widespread support for the idea that this expansion has been largely beneficial and that further expansion should be encouraged, although many – not just development economists – argue that the emphasis in future must be on freeing up the markets of the industrialised states to the exports of the developing states.
Tariff reductions, however, create incentives to construct alternative forms of protection. One form of protection is for private local firms to erect barriers to foreign entry. Private import barriers include both horizontal exclusionary practices – for example, import cartels, joint buying groups, and joint ventures – and vertical exclusionary practices – exclusive dealing, product bundling, requirements contracts, customer and territorial restraints, predatory pricing and price discrimination.
These exclusionary practices are quintessentially the concern of competition law. Most states now have domestic competition rules. Given, therefore, that states already have the solution in place, why is there a problem? One response is that the problem has been overstated, that it is largely a construct of disappointed exporters. A less benign response is that while states have the domestic competition rules to tackle private import barriers, they also have incentives not to enforce those rules. Under this version states are tolerating import barriers for protectionist reasons. Overcoming the protectionist incentives may require a multilateral initiative.
There are two ways in which private import barriers may be viewed as significant.
First and most obviously, the evidence may demonstrate that private import barriers significantly impede trade flows. Here the evidence is equivocal. Much of it is anecdotal. Such evidence as exists could just as easily reflect the failed aspirations of exporters. It does not point conclusively to a structural trade problem. Of course, lack of empirical evidence does not necessarily mean that such barriers do not exist. Neither private participants – firms engaged in import barriers – nor their governments are likely to be forthcoming on this issue. Nevertheless, in strict quantitative terms doubts remain that the argument for a global agreement has been proven. There is a growing recognition that the failures of trade liberalisation to deliver the level of benefits expected cannot be simply blamed on corrupt and protectionist governments.
Secondly, private import barriers may have a conceptual significance that exceeds their quantitative importance. Because of the allocation of incentives (economic and political), liberal trade policy is built on agreements. States trade access opportunities to their markets in return for access opportunities to foreign markets. 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, existing gains may very well be lost. Failure by governments to discipline private import barriers is perceived as ‘cheating’ on trade deals. The failure of states to discipline private import barriers is undermining the confidence that is necessary to support the multilateral trading system.
There are at least two attractions for local governments in tolerating private import barriers. First, some of the adverse, local industry effects of tariff cuts are ameliorated. This is politically attractive to governments concerned not to alienate powerful industry lobby groups. Secondly, the fact that the import barriers are private removes the government from direct involvement. Therefore, it is more difficult for critics to make a successful case that the government is failing to honour its international commitments.
In summary, under this conceptual analysis the importance of private import barriers lies (at least for the time being) not so much in their quantitative significance as in their capacity to undermine the trust upon which the trading system is built. Quantitative considerations are likely to become more important with the passing of time as commercial globalisation continues to grow and tariffs continue to fall.
A GCA is not necessary if there are existing mechanisms for achieving a workable solution. In this respect there are three broad possibilities. First, exporting states could pursue a solution through the extraterritorial application of their competition laws. Even though there has been a growth in the number of states willing to apply their competition laws extraterritorially, this type of unilateral solution is realistically open to only a relatively small number of the most powerful nations. The reason for this is that the importing state has incentives to block any extraterritorialism. Only the larger, more powerful states can overcome these blocking tactics.
A more promising avenue is through the use of existing WTO agreements including the GATT. Although a WTO panel has accepted the notion that the GATT remedial provisions may be activated by the application or non-application of domestic competition laws to conduct that excludes imports, in practice this would be difficult to establish where the allegation against the importing state is essentially one of passivity (mere toleration). The US-Japan film case demonstrates the forensic hurdles facing the complaining state. The US was unable to establish the necessary link between Kodak’s failure to achieve expected levels of market penetration of the Japanese film market and Japanese government conduct.
The third possibility lies in the application of positive comity. The 1998 EU-US positive comity agreement provides that one party at the request of the other party will give consideration to investigating and, if warranted, remedying anti-competitive activities in accordance with the requested party’s competition law. The requested state must agree to devote adequate resources to the investigation, to use its best endeavours to pursue all relevant information, to complete the investigation within six months and to keep the requesting state informed of progress.
Positive comity makes sense as a weapon against private import barriers because it recognizes that the appropriate jurisdiction is the importing state. The problem is that the obligation is non-binding. This is cause for concern because as previously discussed states (when confronted with private import barriers) have incentives to tolerate breaches of their competition law. This suggests that there is a need for states to commit to non-discriminatory enforcement of their domestic competition laws, and to reinforce that commitment with some mechanism to ensure compliance. Some form of supranational oversight seems indicated.
The most obvious candidate for international oversight is the WTO. No other international forum is as well qualified, and to create a new international forum would be a very difficult task. However, there are powerful objections to the WTO adopting an oversight role in respect of a GCA. First, there is concern that WTO panels will be ill-equipped to make proper assessments about competition laws, competition cases and the performance of national competition authorities. According to one trade law scholar:
WTO panels do not have the legal tools to conduct the massive fact-finding expeditions that occur in [US antitrust] cases, especially if the private enterprise in question is allowed to conduct its defense, and if confidential business information is involved.
Secondly, there is concern that the WTO already has a heavy workload dealing with traditional market access issues. A competition agreement would simply be a further distraction. Extending the range of WTO complaints to include failures to discipline private import restraints could radically increase the number of WTO actions, 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 is not ready for that burden.
Thirdly, as with any international agency, the WTO suffers from perceptions of capture and illegitimacy. These perceptions may be stronger for the WTO than for other institutions such as the ICN because of the WTO’s much more extensive powers.
Finally, there is a danger that subjecting competition principles to adjudication by a trade body will lead to a corruption of those principles. Take the fundamental notion of market access for instance. Generally speaking, market access is understood in theoretical terms by both competition and trade theory as part of a mechanism for enhancing efficiencies. In theoretical terms, provided a market is contestable, it is irrelevant whether the players in the market are local or foreign. It is the process of competition, not the competitor, that is important. This broadly describes the manner in which national competition agencies act (at least in the industrialised states). It does not describe how national trade authorities act. Trade authorities act largely on behalf of exporters. Their role is to seek and ensure access to foreign markets. This is the operational nature of international trade. Trade authorities view market access in terms of the competitor (that is, the exporter), not the process. The danger in allowing the competition law notion of market access to become synonymous with the trade notion of market access, in a context where trade authorities are likely to have the role of interpreting texts, is that the emphasis will swing from the process of competition and the pursuit of efficiency to competitors’ rights of access. This criticism has been particularly strong among US competition authorities where competition law tends to rely heavily on an efficiency norm. There is great resistance amongst US competition authorities to any process that might ‘water-down’ that norm.
A major impediment to any GCA on private import barriers is the attitude of the developing states. The developing states demonstrated at Cancún that competition issues are far from the top of their list of priorities. They are suspicious of the motives of developed states. For example, 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. Alternatively, it is just another way to prise open foreign markets. According to an UNCTAD Report ‘a key concern of developing countries [in removing competition policy from the Doha Work Program] was that a multilateral framework at the WTO might be used as a means of securing increased access to developing-country markets’.
Developing states are wary of the costs and sceptical of the overall benefits of regulating private import barriers. Maintaining a competition regime, for instance, is an expensive business. Developing states would benefit far more if the industrialised states honoured their promises to liberalise key areas of trade and industry policy – for example, US protection of its cotton and sugar industries and the EU’s resistance to changes in its Common Agricultural Policy. Developing states would also benefit greatly from a radical overhaul of the anti-dumping rules.
Not only did Cancún demonstrate that a GCA is not a priority for the developing states, it also demonstrated that until a critical number of developing states are enlisted to the cause a GCA will remain a distant prospect.
Although empirical evidence for the damage done to trade by private import barriers is somewhat inconclusive, it is a problem that is unlikely to diminish with time. In any event, the case for regulating private import barriers does not ultimately rest on a quantitative argument. Rather, it rests on the damage that the perception of private import barriers can do to the multilateral trading system. This discussion has suggested that there is a case for a GCA based on an agreement to enforce domestic competition laws in a non-discriminatory fashion. Because of incentives to engage in protectionism some form of supranational oversight seems warranted. However, it is one thing to make a case for international rules; it is an entirely different thing to construct a consensus. While the developing states and the US remain opposed, progress is unlikely.
Export cartels are agreements between exporters to act collusively in respect of some aspects of their export activity. They are often referred to as export associations. Collusion may range from joint production and/or joint marketing to naked price fixing and market allocation.
Export cartels may reduce or enhance consumer and aggregate welfare depending on the circumstances. Where the cartel is comprised of small to medium-sized businesses and its aim is to increase the value of exports by reducing costs or improving products, the cartel is likely to be welfare-enhancing in both the exporting and the importing state. These cartels present no global competition problem.
Where the cartel is between firms with a large share of the foreign market the cartel has the capacity to extract supra-competitive benefits in the importing state. The importing state suffers a loss of consumer welfare without any off-setting increases in producer welfare. If the cartel were local it is likely that it would be prohibited. The exporting state, however, may experience an increase in aggregate welfare because, even if it suffers some loss of consumer welfare, it will also experience an increase in producer welfare (due to the cartel’s ability to capture monopoly rents in the importing state). In these circumstances, the exporting state has incentives to encourage such cartels. This type of export cartel represents an international problem.
The importing state, which has incentives to discipline export cartels, does not have territorial jurisdiction. The exporting state, which has clear territorial jurisdiction, has few incentives to discipline export cartels. Consequently, to restrain export cartels, the importing state is forced to act extraterritorially. Because the exporting state has incentives to ensure that the cartel is not disciplined, the exporting state may try to block the importing state’s attempts at extraterritorialism. This conflict in incentives means that export cartels are not optimally regulated. No state has an incentive to change its law unilaterally, unless it is to exempt export cartels if it does not already do so. To overcome the incentives requires some global rule.
It is not clear just how significant export cartels are. There is no consensus between states on this issue. Some of the industrialised states, including the EU and Japan, have argued that export cartels distort trade and should be banned. The US, on the other hand, takes the view that most export cartels involve firms without market power. Such cartels are likely to be pro-competitive and even efficiency-enhancing and should not be condemned. This echoes sentiments expressed at the time of the introduction of US export exemptions in 1918.
Developing states, including India, China and Indonesia, contend that anti-competitive export cartels are a problem, but a problem restricted to the industrialised states. According to the developing states, their export cartels – in addition to being justified in development terms – normally consist of small firms without market power. Should there be any move to ban export cartels, developing states have indicated that their export cartels should be exempt.
Although the empirical evidence against export cartels is lean, commentators have pointed to a number of negative indirect effects that are said to justify some concerted action. First, exempting export cartels provides foreign buyers with incentives to form buyers’ cartels in retaliation. Secondly, where an export cartel comes into existence, producers in the export country who are not members of the cartel have strong incentives to seek membership of the cartel. This strengthens and perpetuates the cartel. Thirdly, export cartels promote international cartels (made up of a number of export cartels). An international cartel formed from a number of export cartels is likely to be more stable than other international cartels which do not have the advantage of receiving state support. Fourthly, because the importing state may seek to use its competition law extraterritorially against the export cartel and because the exporter’s government has incentives to protect the export cartel and because the export cartel will normally only be able to argue the doctrine of state compulsion if its own government has compelled, not just encouraged, the cartel, the exporting government has an incentive to move towards compulsion (not just toleration) of export cartels. This increases the danger of trade conflict.
In summary, the case against export cartels is uncertain. Perhaps the strongest argument in favour of some action is symbolic. Promoting or tolerating export cartels designed solely to increase exporter welfare at the expense of foreign consumers (hard core export cartels) is a form of neo-mercantilism. It is a ‘beggar-thy-neighbour’ policy which harms the principle of trade liberalisation.
The major obstacle to a GCA on export cartels is the lack of consensus about their competitive significance. But it is not the only obstacle. How would an export cartel rule be structured? Different states deal with export cartels in different ways. In some states – for example, the EU and most of its Member States – export cartels are generally not unlawful because they have no local, anti-competitive effects. There is no need for any special exemption. In other states, where hard core cartels are prohibited per se (that is, without any competitive effects test), exemptions are available for export cartels that have no, or no appreciable, adverse local effects. This is the case, for example with the US, Australia and India. The different approaches rule out any simple solution. An agreement not to exempt export cartels from local law or not to discriminate between local and export cartels would have no effect in those states where cartels are judged under a competitive effects test.
Therefore, the first step in constructing a rule will be to determine whether primary authority for judging export cartels should be given to the importing state, the exporting state or some supranational body. There is no support for a supranational body taking on this kind of primary responsibility. To hand responsibility to the importing state will require a significant sovereignty sacrifice by the exporting state. It will also place a heavy burden on developing states attempting to curtail export cartels emanating from the developed states.
The most attractive suggestion is to leave primary responsibility with the exporting state, but with the proviso that it must take into account foreign effects. No exemption would be available simply because the cartel is an export cartel. This means that all states will be required to introduce an export cartel rule based on international effects, but otherwise the interference with state sovereignty is quite limited. The rule could be limited by defining export cartels to include only hard core cartels where the effects are experienced predominantly off-shore. The advantage over the importing state alternative is that the problems of jurisdiction and foreign evidence-gathering are minimised. Evidentiary problems are minimised because it is in interests of the importing state – where the relevant evidence is likely to be located – to co-operate in the collection and presentation of that evidence. The protectionist urge implicit in export cartels means that some form of supranational oversight is probably necessary. Again the logical forum is the WTO. As with import cartels this presents a significant obstacle to any GCA.
To conclude, the evidence does not yet warrant a GCA to solve the problem of export cartels. However, if the problem of welfare-reducing, export cartels grows, some form of GCA along the lines described appears desirable.
A matter consistently raised in the competition-trade debate is the issue of anti-dumping measures. The GATT-WTO system permits states in certain circumstances to take action against low-priced imports (that is, dumping by foreign producers). A state can impose anti-dumping duties when it is determined that foreign producers are selling an export below the price at which the same product is supplied to the exporter’s domestic market (the product’s ‘normal value’). As a result, many states have introduced domestic laws authorising anti-dumping measures.
The use of anti-dumping measures has been a source of conflict. Developing nations and some developed ones (including Japan) have argued in the past that anti-dumping remedies are nothing more than a form of protectionism. Economists generally agree. In fact, many regard the expanding use of anti-dumping remedies as a major threat to the multilateral trading system. Those opposed to the current anti-dumping practices have suggested that a rule based on competition principles should replace the existing anti-dumping rules.
On the domestic producer side there are equally strong convictions that anti-dumping measures are necessary to counter ‘unfair’ trade. The politically successful catch-cry is that anti-dumping rules are essential for the maintenance of a level playing field.
There is an economic rationale for protection against dumping. Dumping is economically inefficient if it enables the foreign producer to obtain monopoly power in the market by eliminating rivals. Once in possession of monopoly power the producer is able to extract supra-competitive profits. Consumers suffer a welfare loss, which is not even partially offset by local producer gains (as it would be in an autarky). This type of conduct is referred to as predatory pricing and is generally regarded as anti-competitive. Most states have the power to deal with such conduct under domestic competition law.
Anti-dumping laws, however, are not limited to the situation just discussed. Typically anti-dumping laws may be applied in circumstances where there is no predation. It is not necessary to prove that the exporter or exporters had substantial power in the relevant market or that individually or collectively they are trying to drive competitors out of that market. These are normally the minimum requirements to establish predatory pricing. According to welfare-based competition theory, this form of price discrimination without predation imposes no costs on the target market (the importing nation). There is no competitive injury. The costs (if any) are borne by the consumers of the exporting nation. Consequently, any anti-dumping law to discipline foreign price discrimination without predation is simply protectionism and not justified. Even more broadly-based competition regimes are not likely to regard price discrimination as unlawful unless accompanied by significant anti-competitive effects. Thus, injury to competitors without evidence of injury to the competitive process itself is unlikely to be proscribed by competition law.
Competition advocates have further complaints against antidumping laws. Where an anti-dumping claim is levelled at a group of exporters, the response of the exporters is often to enter into a voluntary restraint agreement under which they consent to restricting output or, what amounts to the same thing, raising prices. This leads to the exporters forming an export cartel to manage the reduction in their collective export output. Alternatively, their government is forced to manage the cartel. In either event, the outcome is contrary to good competition policy.
Finally, it has been shown that the mere presence of anti-dumping laws is a sufficient incentive for firms to alter their market behaviour. Thus, market outcomes may be distorted even when no anti-dumping duty is actually imposed.
If low priced imports are judged solely through the prism of predatory pricing, the case for competition rules appears quite strong. However, there are objections to characterising low-priced imports in this narrow, economic-centric manner. Consequently, there are objections to scrapping the anti-dumping rules in favour of a predatory pricing rule.
The primary object of anti-dumping law, so it is argued, is to protect domestic industry and the domestic economy against injurious, low-priced imports. Many of the injuries that may be inflicted on domestic producers by low-priced imports are not recognised under competition rules (at least not without additional evidence that the process of competition has been substantially harmed). These injuries cause sufficient economic and social harm to warrant regulation. A non-exclusive list of such injuries might include loss or reduction in expected investor profits perhaps leading to a reduction in future capital investment, reduction in employment resulting in lower national income and tax revenues, waste of human skills that may have required a significant investment in time and capital to acquire, reduction in research and development. Recently, anti-dumping remedies have also been linked to safety concerns. In other words, anti-dumping laws are said to be justified on policy grounds that look beyond traditional competition policy and even economic policy.
Economists and competition authorities see little benefit and much danger in the present anti-dumping laws. It has been suggested that the minimal solution is an international agreement to apply domestic competition laws to imports in a non-discriminatory manner. Dumping by foreign firms should be treated as would price discrimination by a local firm.
However, just as the benefits of greater market access have provided exporters with incentives to apply political pressure to lower tariffs, so have falling tariff barriers (coupled with the core trade principle of non-discrimination, particularly most favoured nation status) created the incentives for local producers to harness their political power to meet the threat from foreign competition by insisting on anti-dumping measures. It has to be recognised that anti-dumping laws have considerable political support. It is unlikely that nations will unilaterally repeal these provisions. In fact, the trend is very much in the opposite direction with the less developed states, for many years the principal critics of anti-dumping duties, now applying their own anti-dumping measures.
If there is to be any GCA ridding the GATT-WTO system of these measures, it will be necessary to offset the political support that anti-dumping laws attract. This is highly unlikely. The states most likely to have a hand in fashioning any GCA, the US and the EU, are also the states most prone to employing anti-dumping laws. To date the US has remained implacably opposed to any negotiations on anti-dumping, particularly as part of a WTO agenda. Even if the political forces for retention of anti-dumping can be offset, consensus will be difficult to achieve because states do not have a consistent approach to regulating predatory pricing.
It is now widely accepted that some forms of collusion between competitors should be suppressed, including agreements on naked price fixing, output restrictions, market sharing and bid-rigging. Broadly speaking, collusion of this type is designed to increase producer welfare (the welfare of the cartel members) at the expense of both consumer welfare and aggregate welfare. Agreements of this nature are often called hard core cartels.
A recent OECD report described the harmful effects of hard core cartels:
Cartels harm consumers and have pernicious effects on economic efficiency. A successful cartel raises price above the competitive level and reduces output. Consumers (which include businesses and governments) choose either not to pay the higher price for some or all of the cartelised product that they desire, thus forgoing the product, or they pay the cartel price and thereby unknowingly transfer wealth to the cartel operators. Further, a cartel shelters its members from full exposure to market forces, reducing pressures on them to control costs and to innovate. All of these effects harm efficiency in a market economy.
Where the cartel traverses international borders it is an international cartel.
Hard-core cartels need to be distinguished from agreements formed for more legitimate purposes. Production joint ventures and research and development consortia are commonly encouraged because they have the capacity to add to competition and to produce significant efficiencies and innovation gains; in other words, consumer and aggregate welfare are likely to grow as well as producer welfare.
While it is possible to distinguish between hard core cartels and other collusion in the abstract, it is not always so easy in practice. This is partly a question of identification; in practice, the task of determining the motivation and welfare effects of cartels is not always straightforward. It is also partly a question of the criteria used; not all states apply an unadulterated welfare standard. Some states will assess the legality of a cartel on additional criteria such as industry policy (including export enhancement). Nearly all states allow for exemptions or grant authorisations. Therefore, what may sometimes seem a hard core cartel to one state or group of states could be viewed by another state as a beneficial arrangement. Consequently, when states proclaim that they are against hard core cartels, it should not be assumed that they are necessarily referring to precisely the same thing or that they will not disagree over the status of a particular cartel. In some respects, the expression ‘hard core cartel’, particularly when used in an international context, is a convenient device for reaching an agreement in principle without having to sort out all the details. Nevertheless, these difficulties of definition should not be overstated. Many private international cartels are blatantly hard core and universally recognised as such.
Although governments will always remain susceptible to political lobbying for protectionism (even to sanctioning international cartels), most states view the protection of international hard core cartels as poor policy. Consequently, the international regulatory issues tend to be different than those posed by import and export cartels (where protectionism remains a key concern). Curbing international hard core cartels raises problems of global detection, prosecution and enforcement. Each of these exposes gaps in domestic regulation. Finding a way of overcoming these gaps is the international issue.
International cartel prosecutions have increased dramatically during the past decade. Significant fines have been imposed on firms in a diverse range of industries and markets. A recent study concluded:
Cartel activity has occurred in a variety of industries--from commodities like cement and citric acid to specialized services like fine arts auctions and wastewater treatment facility construction. Chemical products top the list with thirteen different cartels. The next largest product category is transportation (seven cartels in our sample), followed by steel (four), carbon and graphite products (three), plastics and paper (two each), and several miscellaneous goods and services.
A noticeable feature of the recent upsurge in cartel prosecutions has been the number of states taking action. For example, in the vitamins cartel case significant fines were levied in the US, Australia, Canada and the EU amongst others. This enforcement activity, however, is mainly restricted to the industrialised states. Developing states have not yet played a significant enforcement role.
The evidence suggests that the majority of international cartels are made up of multinational producers based in the industrialised states. The products involved are largely sophisticated intermediate goods and services with a relatively high level of product homogeneity. The primary targets are the developed economies, although, because of their nature, international cartels also affect developing economies. In fact, the effect on developing economies is almost certainly more severe in relative terms than on the industrialised states.
Whether this increased prosecutorial activity indicates a growth in global cartel activity or merely improved detection coupled with an increased enthusiasm to prosecute is uncertain. What is clear, however, is that the cartels that have been uncovered have done significant harm. A report by the OECD Competition Committee concluded:
[R]ecent cases against large, international cartels suggest that the dimensions of the problem are even larger than previously thought. It remains difficult to place a monetary value on the harm, but it is surely significant, amounting to billions of dollars annually.
The same report estimated that the average overcharging in the international cartels selected for study was between 15 and 20 percent. The vitamins cartel, which involved companies from the US, Germany, Switzerland, France and Japan amongst others, is estimated to have increased prices of vitamins used in manufacturing livestock feed and products such as bread, rice and juice, by up to 35 percent to US customers. The graphite electrodes cartel raised prices in many parts of the world by between 45 and 90 percent. The price of lysine, a feed additive for livestock, rose by as much as 70 percent in the first six months after the formation of the lysine cartel.
Some care must be taken with these figures. Estimating the monetary harm done by cartels is a difficult task. This applies domestically as well as globally. Nevertheless, quantification difficulties do not alter the fact of harm nor the widespread belief that harm has been significant. Further, as commercial globalisation grows, the level of harm caused by international cartels is likely to increase, both in absolute terms and relative to harm caused by domestic cartels. The US Department of Justice has stated that international cartels currently account for a third of all the Antitrust Division’s grand jury investigations. According to the Deputy Assistant Attorney General for Criminal Enforcement at the Antitrust Division:
The subjects and targets of the Division's international investigations are located on 6 continents and in nearly 25 different countries. However, the geographic scope of the criminal activity is even broader than these numbers reflect. Our investigations have uncovered meetings of international cartels in well over 100 cities in more than 35 countries, including most of the Far East and nearly every country in Western Europe.
Cartels suffer from problems of coordination and policing. The reason for this is that individual members of the cartel have incentives to cheat on the cartel. By undercutting the cartel price a cartel member is able to increase market share. This is done without loss of profits because the cartel price is above the competitive market price. Therefore, cartels are often unstable. This inherent instability raises the question whether international cartels need to be actively disciplined.
Although there are some ‘principled’ objections to the use of competition rules against commercial organisational activities (such as cartels), the major objection rests essentially on a negative view of the effectiveness of government intervention.
According to this view, while a laissez-faire approach may allow some welfare-reducing cartels to exist longer than they otherwise would, the harm done by these cartels is less than the harm likely to be done by governments proscribing the wrong cartels (that is, welfare-enhancing cartels). Therefore, regulatory intervention into market activities normally cannot be justified, even in the case of hard core cartels.
Contrary to this assessment, however, the evidence suggests that international cartels are long-lived and cause substantial harm. They cause not only short-term economic welfare losses (in the form of higher prices and/or lower product choice) but arguably also long term structural damage (particularly to developing economies). Although it is extremely difficult to be confident in the absence of further empirical evidence, it does appear that international cartels have increased in recent years in the wake of growing commercial globalisation. Certainly most states are committed to disciplining international cartels.
Anti-cartel enforcement is a matter of domestic law. In recent years a number of anti-cartel tools have evolved. Investigatory tools include the use of surprise raids on suspected cartel members, seizure (or copying) of evidence, compulsory interviews, and phone taps and electronic eavesdropping. As important as the improved investigatory tools are, the undoubted key to modern anti-cartel detection and enforcement is the leniency or immunity program. This is particularly important in the case of private international cartels because it overcomes some of the difficulties associated with information sharing. There has also been considerable attention paid to improving sanctions in line with the prevailing object of finding an effective level of deterrence, particularly general deterrence. These initiatives appear to have achieved some success.
However, problems remain. First, the secret nature of international cartels means that ultimately there is no reliable evidence as to how successful the current initiatives have been. Second, states remain very conservative about sharing information. This reduces the effectiveness of investigatory tools such as the secret raid. Even where states are able to share information to the extent of co-ordinating raids, it does not mean that they will be able to share the results of the raid. Additionally, cartel participants have become adept at constructing international cartels to make domestic enforcement difficult. Third, very few of the developing states have been prepared to take action against international cartels. Developing states suffer from capacity constraints. The best way to involve the developing states in the enforcement process is for the industrialised states to share the information they collect. Fourth, while states agree that deterrence should underpin anti-cartel sanctions, states have not been able to agree on the appropriate character or level of sanctions. The whole notion that pecuniary penalties act as an effective deterrent has been questioned. The US has doubts that pecuniary penalties can solve the problem of international cartels as long as so many states remain inactive. Therefore, the US has pushed strongly for gaol sentences. Many states, however, are not prepared to take this step. This is not surprising given that until a decade ago cartels were not unlawful in many states. In fact, they were often encouraged.
Therefore, despite the successes of the past decade, there is room for improvement. It is doubtful, however, that any agreement on substantive rules is required. Most states already have legislative provisions dealing with hard core cartels. They also have the desire to enforce those provisions. What they often lack is capacity. The way to a more effective regulation of international cartels requires more states to build better prosecutorial capacity. This can be achieved through greater use (and co-ordination) of leniency programs, a greater willingness of states to share information and continuing progress in capacity building.
The international issues raised by mergers and monopolistic conduct are similar in nature. Monopolistic conduct includes refusals to deal or license, predatory pricing and all types of distribution restraints existing as a result of an exercise in market power. These include exclusivity deals, product tying, and customer and territorial restraints. A firm capable of engaging in monopolistic conduct will be referred to as a monopolist.
There are three principal reasons why international mergers (that is, mergers having substantial international effects) and international monopolistic conduct may need international rules (a GCA). First, the merger parties or monopolist may complain that being required to adjust their conduct to different national domestic rules is inefficient. This, however, is a normal cost of ‘doing business’ internationally. Different local rules generally reflect different social, economic and political conditions. They are not necessarily discriminatory or protectionist. There is no reason why the merger parties or the monopolist should not expect to be subject to the same rules applying to local firms. While there is certainly a case for better information sharing (especially in respect of pre-merger reporting), there is no real international problem to be overcome.
Secondly, the state may complain that it cannot enforce its domestic competition laws. For example, where the merger parties or the monopolist are located outside the jurisdiction, there may be problems in establishing jurisdiction, collecting evidence and implementing punitive and/or remedial orders. These problems, of course, are not exclusive to competition law. They are experienced to a greater or lesser extent in all areas of transborder law enforcement. In competition terms the issue is keenest in the case of the exporter as national champion. In such situations local law permits a merger (which it would otherwise prohibit) or monopolistic conduct (which it would otherwise prohibit) because many of the adverse effects are externalised. This has similarities to the issue of export cartels discussed earlier.
Thirdly, and particularly where global markets are involved, remedies are sometimes necessarily international in character. Where this occurs (and so long as competition law remains domestic), one state is able to impose its (national) remedy on other states. A domestic solution becomes an international solution. This is an international problem and provides a compelling reason for a GCA. It requires further explanation.
As a result of the reduction in transport and communication costs, the evolution of new economy markets and the creation of new forms of distribution (for example, the Internet), national boundaries have ceased in some industries to operate as market boundaries. In such cases the market has become international or even global in scope.
State-based sovereignty – still the paradigm for competition regulation — can break down conceptually where markets are international. This breakdown occurs where regulation by two states is impossible in practical terms. For instance, where a structural remedy is imposed by one state in a global market, the effects will be experienced worldwide. One state, inevitably the state with the most restrictive competition law and the will and capacity to enforce it, becomes de facto the global regulator, even though it has nothing resembling a mandate to adopt such a role.
A similar situation may arise even where markets are national. For example, in some industries it may not be possible to isolate the consequences of compliance with an order to the geographic limits of the state making that order. Again, the firm will be forced to comply with the most restrictive order that it is unable to avoid or ignore. Those states which have adopted a less restrictive approach must tolerate the fact that their preferred policy has been rendered irrelevant.
The clearest example of the failure of state-based regulation is a merger between two firms involved in a single worldwide market. Because (in such a merger) there is only one market, there can be only one effective regulator, and inevitably that must be the state with the most restrictive merger regime. This occurred when the EC refused to permit the merger of General Electric and Honeywell, even though US authorities had previously approved it. As the relevant market was global, the EC’s decision to block the merger became the global default position. GE and Honeywell could not take advantage of US approval of their merger, even in the US, as long as the EC prohibited it.
Because the US would not have prevented the merger, the outcome is over-regulation. Over-regulation in these circumstances can lead to a certain degree of tension between states as some are denied the opportunity to pursue their preferred policy in their own jurisdiction. When Boeing and McDonnell-Douglas sought a merger in 1997, it was approved by the US Federal Trade Commission, but initially rejected by the EC. Both the US and the EU viewed the other’s decision as tainted by considerations other than competition concerns. The impasse between US and EU authorities during this case resulted in the US Vice-President threatening to wage commercial war. The EC’s rejection of the GE/Honeywell merger also drew strong political disapproval in the US, including from President Bush. The merger issue has been sufficiently serious for the US and the EC to engage in extensive talks culminating in an agreement on best practices in merger regulation. More recently the ICN has developed a Merger Guidelines Workbook.
Where the market is international, similar issues may arise for monopolistic conduct. Given that the market is international, more than one state is likely to want to impose its remedy on the monopolist. However, multiple solutions are sometimes not possible. Therefore, assuming that all interested states are able to exercise and enforce jurisdiction over the monopolist, the state with the most restrictive response will become de facto the international regulator. For example, if country A and country B both claim jurisdiction to regulate the activities of Firm F in an international market, and A permits the conduct but B prohibits it and if F cannot simultaneously comply with both orders, F must comply with B and cease the conduct (or otherwise conform to whatever remedy B considers appropriate). The result must be over-regulation, as A’s preferred outcome was to tolerate the conduct.
It is not necessary that A and B have different laws. They may have the same law but reach different conclusions. This results from the manner in which single firm conduct is assessed. For example, assume that both A and B assess liability for single firm conduct on the basis of aggregate welfare. Further assume that Firm F is located in A. If F’s conduct results in producer benefits that outweigh the consumer losses in A, A will permit the conduct. This follows because states invariably ignore foreign effects in determining liability and in fashioning an appropriate remedy. Therefore, A will ignore the consumer welfare losses in B. B, on the other hand, will prohibit the conduct because it is not in a position to capture any producer benefits. Its aggregate welfare outcome must be a loss.
The Microsoft cases demonstrate these possibilities. The legality of certain of Microsoft’s activities has recently been examined in the US, the EU and most recently Korea. Dealing with similar but not identical conduct the EC and the Korean Fair Trade Commission (‘KFTC’) have produced remedies that are much more restrictive of Microsoft’s freedom than those imposed in the US.
The EC has ordered Microsoft to make available to software developers information about its operating system software that will enable easier interoperability. It is difficult to see how this information could possibly be contained within the European environment. The EC also ordered Microsoft to make available in Europe a version of Windows without Microsoft’s Windows Media Player (‘WMP’). Again it will be difficult to contain this order because consumers often access software products and updates via the Internet. It is virtually impossible to impose geographic boundaries on the Internet. Consequently, the EC’s remedies become effectively global orders, much to the dismay of Microsoft, its US political supporters and even US antitrust authorities.
The KFTC adopted a remedial approach similar to the EC’s. Microsoft was ordered:
(1) To unbundle Windows Media Service from Windows Server Operating System; and
(2) To supply a version of Windows PC Operating System software stripped of Windows Media Player and Instant Messenger.
Indeed, the KFTC order goes further than the EC’s order. Microsoft is required to supply a version of Windows PC Operating System software ‘installed with “Media Player Centre” and “Messenger Centre” that will contain links to web-pages that allow consumers to download competing media players and instant messengers, so that competing softwares can be equally installed into Windows PC Operating System’. Microsoft must also supply consumers who have already acquired Windows PC Operating System a copy of “Media Player Centre” and “Messenger Centre” via CD or Internet. The content of “Media Player Centre” and “Messenger Centre” will be determined by the KFTC in consultation with a Supervisory Board established to oversee compliance with the order.
There is a clear case for international agreement in those circumstances where two or more states seek to regulate the same conduct and more than one remedy is impossible. In the absence of an agreement (and assuming that states wish to impose different remedies) the consequence is over-regulation and the possibility of international discord.
However, the chances of states agreeing to harmonise their rules on mergers and monopolistic conduct is quite remote. The EU and the US have formulated a set of best practices for merger regulation. The ICN has also done considerable work on merger standards. These may reduce, but cannot prevent, tension. No set of best practices exists for monopolistic conduct. Nor is it likely that one will be developed in the near future. The economic effects of most types of monopolistic conduct are hotly contested. States and commentators routinely disagree over what standards should be applied. The Chicago school, for example, argues for an efficiency standard. Under this standard, conduct is lawful unless it can be shown to be inefficient. Inefficiency is largely measured by examining whether the conduct will lead to a demonstrable reduction in output. Other competition experts prefer to apply a competition test: will the conduct lead to a substantial reduction in the level of competition in the relevant market? Still others prefer a standard that does not depend solely on economic considerations: thus, in determining whether monopolistic conduct should be proscribed, it is necessary to have regard to such factors as the protection of small business and the dispersal of economic power. The chances of harmonising rules for mergers and monopolistic conduct (even if desirable) remain extremely remote.
A more realistic possibility is that states might agree on a rule for allocating jurisdiction (a choice of law rule) in those cases where local remedies are impossible. Thus, where a matter involves such cases, states might agree that the law to be applied is the law of the state that has suffered or is likely to suffer the greatest consumer effects (measured in absolute terms). To improve welfare outcomes and hopefully a broader acceptance, states should agree to modify their domestic rule in these cases to take into account all effects (local and foreign). Unfortunately, there is no guarantee that the choice of law test will always isolate a single jurisdiction. However, it is difficult to envisage any more definitive test that does not involve rule harmonisation, which is a most unlikely prospect. Some oversight of the allocation process is probably necessary to prevent local bias. Again a possible choice would be the WTO.
There are a variety of competition-related matters which give rise to international problems. The nature and intensity of the international problem varies depending on the conduct in question. Private import barriers, export cartels and anti-dumping remedies exhibit strong protectionist elements. This is absent from the dialogue about international cartels. Although private import barriers, export cartels and anti-dumping remedies draw upon the common well of protectionism, they otherwise raise discrete issues which warrant separate treatment. Mergers and monopolistic conduct sometimes contain a protectionist element, but the essential regulatory issue for international purposes is the inappropriateness of state-based rules in those cases where only one remedy can prevail.
A GCA is needed in all cases except perhaps for international cartels. The rules required, however, are not the same. A rule of non-discriminatory application of domestic laws is appropriate for private import barriers. However, this will not do for export cartels because domestic laws vary in a critical manner. The minimum rule in the case of export cartels would appear to be one that requires exporting states to consider not only local but also foreign effects. A similar rule would also be useful for mergers and monopolistic conduct, but here something more is needed. As only one state can apply its law, a rule to allocate jurisdiction is indicated. The only other alternative is harmonisation. Harmonisation is not only unrealistic, at this stage, it is not necessarily desirable. Whatever in theory may be the preferred approach to anti-dumping, the reality is that the political will to replace anti-dumping laws with competition-based predatory pricing rules does not exist.
Finally, international hard core cartels are widely recognised as unacceptable. This shared understanding has already resulted in a degree of enforcement co-operation between domestic competition authorities. What is needed for the future is more widespread adoption of leniency programs and more information sharing between competition authorities. It is doubtful that any rule-based GCA is necessary.
[∗] LLB, B Com (Melb), PhD (Monash), Senior Lecturer, Department of Business Law & Taxation, Monash University.
 Although there is a trend towards market-oriented models, it cannot be said that the trend is towards one particular model. In 1999 the Antitrust Law and International Law and Practice Sections of the American Bar Association (‘ABA’) identified seven separate models of competition regimes. See ABA, Report on the Internationalization of Competition Law Rules: Coordination and Convergence: Executive Report (1999) American Bar Association at 1–2 <http://www.abanet.org/antitrust/at-comments/2000/reports/01-00/conv_exe.pdf> accessed 29 December 2006.
 In the early 1980s only 20 jurisdictions had a competition law. By 2000 the number of jurisdictions with some form of competition law had grown to 98. See Michal S Gal, Competition Policy for Small Market Economies (2003) at 8.
 Ideally a number of the solutions require some form of international oversight. Because of the connections between competition and trade the WTO is an obvious starting point. However, the choice of the WTO presents significant problems which cannot in this paper be analysed at anything like the depth required.
 General Agreement on Tariffs and Trade, opened for signature 30 October 1947, 55 UNTS 187 (entered into force provisionally 1 January 1948).
 See art 46 of the Havana Charter for an International Trade Organization (‘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 at 4 <http://www.abanet.org/antitrust/at-comments/2000/reports/01-00/conv_rpt.pdf> accessed 12 March 2008 (‘ABA Internationalization of Competition Law Rules’). See also John Braithwaite and Peter Drahos, Global Business Regulation (2000) at 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 at 558–9.
 Treaty Establishing the European Economic Community, opened for signature 25 March 1957, 298 UNTS 11, arts 85 and 86 (entered into force 1 January 1958).
 On the history of competition law in Japan see, for example, Alex Y Seita & Jiro Tamura, ‘The Historical Background of Japan’s Antimonopoly Law’ (1994) University of Illinois Law Review 115; Hiroshi Iyori & Akinori Uesugi, The Antimonopoly Laws and Policies of Japan (1994).
 Singapore Ministerial Declaration, WTO Doc WT/MIN(96)/DEC (1996) . The WTO was established by the Marrakesh Agreement Establishing the World Trade Organization, opened for signature 15 April 1994, 1867–9 UNTS 1, 33 ILM 1125 (entered into force 1 January 1995). The WTO Ministerial Conference in Marrakesh in April 1994 had identified competition policy as one issue amongst a number that required further investigation: See Agreement on Trade-Related Investment Measures, opened for signature 15 April 1994, 1869 UNTS 299, 33 ILM 1197, art 9 (entered into force 1 January 1995).
 The working group was known as the Working Group on the Interaction between Trade and Competition Policy (‘WGTCP’).
 See Eleanor Fox, ‘Competition Law’ in Andreas F Lowenfeld, International Economic Law (2002) at 379–383.
 See Doha Work Programme, WTO Doc WT/L/579 (2004) [1(g)] (Decision Adopted by the General Council on 1 August 2004).
 The ICN was created as a formal vehicle for domestic competition agencies to share ideas, strategies and some limited information with the aim of improving the enforcement of domestic-based rules. It derives its creative inspiration from the findings of the ICPAC Report: 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). The ICN is described by US authorities as a ‘virtual network’. It has no headquarters. 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).
 See Oliver Budzinski, ‘The International Competition Network: Prospects and Limits on the Road Towards International Competition Governance’ (2004) 8 Competition and Change 223.
 See, for example, David Gerber, ‘Is Reconciliation Possible? The US – European Conflict Over the Globalization of Antitrust Law: A Legal Experience Perspective’ (1999) 34 New England Law Review 123 at 125. See also Alan O Sykes, ‘Externalities in Open Economy Antitrust and their Implications for International Competition Policy’ (1999) 23 Harvard Journal of Law and Public Policy 89 at 93–4; Eleanor Fox, ‘Global Markets, National Law, and the Regulation of Business: A View from the Top’ (2001) 75 St John’s Law Review 383 at 392; Andrew Guzman, ‘Antitrust and International Regulatory Federalism’ (2001) 76 New York University Law Review 1142 at 1152–5. Under public choice analysis the welfare deficit is likely to be even greater.
 See, for example, the tensions created as a result of the application of US antitrust law to activities of the uranium cartel, a cartel formed in response to US policies designed to protect its domestic uranium producers by barring non-US producers from US enrichment services. See In re Uranium Antitrust Litigation  USCA7 143; 617 F 2d 1248 (7th Cir, 1980) (the Westinghouse Uranium case). For a discussion of the case see, for example, Warren Pengilley, ‘Extraterritorial Effects of the US Commercial and Antitrust Legislation: A View from Down Under’ (1983) 16 Vanderbilt Journal of Transnational Law 833.
 In the wake of Westinghouse Uranium  USCA7 143; 617 F 2d 1248 (7th Cir, 1980) evidence-blocking legislation was passed by Britain, Canada, Australia, New Zealand and South Africa.
 The literature reveals a preference for decentralised competition regulation, although the reasons for doing so vary. The literature is now fairly extensive. See, for example, Frederic Jenny, ‘Competition Law and Policy: Global Governance Issues’ (2003) 26 World Competition 609, 620 arguing that a comprehensive competition regime ‘is politically unrealistic at the multilateral level both because of the importance of the differences in levels of economic development of countries around the world and because of the importance of the abandonment of national sovereignty by each country that it implies’. See also Michal S Gal, Competition Policy for Small Market Economies (2003) (the appropriate competition regime varies with the size and level of development of the economy); Ulrich Immenga, ‘Internationalization of Competition Laws: Levels of Disparities’ in Tzong-Leh Hwang & Chiyuan Chen (eds) The Future Development of Competition Framework (2004) at 5–9 (states have different enforcement preferences); Karl Meessen, ‘Competition of Competition Laws’ (1989) 10 Northwestern Journal of International Law and Business 17; Karl M Meessen, Economic Law in Globalizing Markets (2004) at 94 (regulatory competition is more likely to produce optimal outcomes than a GCA); Wolfgang Kerber, ‘An International Multi-Level System of Competition Laws: Federalism in Antitrust’ in Josef Drexl (ed), The Future of Transnational Antitrust — From Comparative to Common Competition Law (2003). Finally, the International Competition Network (ICN), an organisation of national competition authorities, takes the same view: see ICN, A Statement of Mission and Achievements up until May-2005 at 2.
 Report of the Meeting of the 11–13 March 1998, WTO Doc WT/WGTCP/M/4 (1998)  (Note by the WGTCP Secretariat).
 The categorisation is somewhat ambiguous in that categories (ii) and (iii) could be interpreted as being based on geographic effects, that is, global markets (category (ii)) as opposed to national markets (category (iii)). This categorisation lacks rigour. A market may be international, but the effects on any two states quite different. Equally, just because markets are national does not mean that states must suffer different levels of harm. The interpretation in the text accords with the understanding of the Chair at the time of the WGTCP, Professor Jenny. See WGTCP Report, above n18 at .
 See General Agreement on Tariffs and Trade, art VI, opened for signature 30 October 1947, 55 UNTS 187 (entered into force provisionally 1 January 1948) and the Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade 1994, opened for signature 15 April 1994, 1868 UNTS 201 (entered into force 1 January 1995).
 Broadly speaking, hard core cartels are agreements between competitors designed to increase the welfare of the cartel members at the expense of both consumer welfare and aggregate welfare. See discussion in text accompanying n101.
 Since World War Two and the introduction of the GATT, tariffs have dropped from an average of 40 percent to 4 percent. See OECD Policy Brief:The Development Dimensions of Trade (2001) OECD at 2 <http://www.oecd.org/dataoecd/20/26/2431455.pdf> accessed 4 October 2006.
 But see Joseph Stiglitz, Making Globalization Work: The Next Steps to Global Justice (2006) at 15 arguing that the evidence that trade liberalisation has led to growth is mixed.
 See Joseph Stiglitz ‘Trade and the Developing World: A New Agenda’ (1999) 98 Current History 387 at 389–390 arguing that tariff reductions, which have so far served the interests of the industrialised states much more than the developing states, must be more balanced.
 See 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 at 7–8 <http://www.abanet.org/antitrust/at-comments/2000/reports/01-00/ma_rpt.pdf> accessed 12 March 2008 (‘ABA Report on Market Access’); ICPAC Report, above n12 at 203-209. See also OECD, Trade and Competition Policies for Tomorrow (1999).
 Government involvement may be more direct than a failure to enforce the state’s competition law. Many competition regimes make provision for authorising or exempting particular conduct. In such cases the state, through its competition authority, directly authorises the private barriers.
 For a discussion on the issue of private import barriers see UNCTAD, Exclusionary Anti-competitive Practices, Their Effects on Competition and Development, and Analytical and Remedial Mechanisms UNCTAD/DITC/CLP/2005/4 (2005); Philip Marsden, A Competition Policy for the WTO (2003); Brendan Sweeney, ‘Globalisation of Competition Law and Policy: Some Aspects of the Interface between Trade and Competition’  MelbJlIntLaw 15; (2004) 5 Melbourne Journal of International Law 375.
 See OECD Joint Group on Trade and Competition Trade and Competition Policies: Options for Greater Coherence (2001) at 16; ABA Report on Market Access, above n25 at 24. ICPAC Report, above n12 at 224–5 concluded 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 ICPAC Report, above n12 at 211–219. See also the Annual Reports of the Working Group on the Interaction between Trade and Competition Policy (‘WGTCP’) at the WTO. The reports are available at the WTO website: see <http://www.wto.org/english/tratop_e/comp_e/comp_e.htm> (accessed 4 October 2006).
 See ABA Report on Market Access, above n25 at 8. Many of the complaints made by US traders 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. See also ABA, Report on Market Access, above n25 at 6–19.
 Contrast UNCTAD Report, Exclusionary Anti-competitive Practices, Their Effects on Competition and Development, and Analytical and Remedial Mechanisms, above n27 at 3.
 See Stiglitz, Making Globalization Work, above n23 at 16.
 Robert Hudec, ‘A WTO Perspective on Private Anti-Competitive Behavior in World Markets’ (1999) 34 New England Law Review 79 at 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 at 364.
 See Eleanor M Fox, ‘International Antitrust and the DOHA Dome’ (2003) 43 Virginia Journal of International Law 911 at 916; Jürgen Basedow & Stefan Pankoke, ‘General Report’ in Jürgen Basedow (ed), Limits and Control of Competition with a View to International Harmonisation (2002) at 27–8; William E Kovacic, ‘Extraterritoriality, Institutions, and Convergence in International Competition Policy (2003) 97 American Society of International Law 309 at 310. American Bar Association Sections of Antitrust and International Law, Comments and Recommendations on the Competition Elements of the Doha Declaration (2003) American Bar Association at 10 <http://www.abanet.org/antitrust/at-comments/2003/05-03/doha.pdf> accessed 12 March 2008 (Joint Comments and Recommendations to the United States Trade Representative).
 Blocking laws have been used both against the production of evidence and enforcement of judgments. Some states have also used ‘claw-back’ laws to allow defendants to claw-back damages payments. See, for example, Foreign Proceedings (Excess of Jurisdiction) Act 1984 (Australia); Protection of Trading Interests Act 1980 (UK) c 11. See generally Deborah Senz & Hilary Charlesworth, ‘Building Blocks: Australia’s Response to Foreign Extraterritorial Legislation’  MelbJlIntLaw 3; (2001) 2 Melbourne Journal of International Law 69.
 See Sweeney, above n27 at 401–411.
 See Japan — Measures Affecting Consumer Photographic Film and Paper, WTO Doc WT/DS44/R (1998) (Report of the Panel). The case is discussed in Patricia Isela Hansen, ‘Antitrust In The Global Market: Rethinking "Reasonable Expectations"’ (1999) 72 Southern California Law Review 1601. 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 at 334 arguing that Article XXIII of the GATT ought to be interpreted and applied in an expansive manner that furthers the object of trade liberalisation.
 Japan — Measures Affecting Consumer Photographic Film and Paper, WTO Doc WT/DS44/R (1998) (Report of the Panel).
 Agreement between the European Communities and the Government of the United States of America on the Application of Positive Comity Principles in the Enforcement of their Competition Laws  OJ L 173/28 (signed and entered into force 4 June 1998) art III. This Agreement supplemented the earlier Agreement between the Government of the United States of America and the Commission of the European Communities regarding the Application of their Competition Laws  OJ L 95/47 (signed and entered into force 23 September 1991); corrected at  OJ L 131/38.
 Contrast Marsden, above n27, who argues that a non-discrimination agreement is not sufficient. However, the problem that Marsden tackles is broader than the issue of private import barriers jeopardising trade agreements. Essentially Marsden argues for a harmonised competition rule.
 Fox, above n35 at 926; Andrew T Guzman, ‘Is International Antitrust Possible?’ (1998) 73 New York University Law Review 1501.
 Kevin C Kennedy, ‘Foreign Direct Investment and Competition Policy at the World Trade Organisation’ (2001) 33 George Washington International Law Review 585 at 623; ICPAC, above n12 at 23. See also Alan Wolff, ‘The (Notionally) Bridgeable Chasm between Antitrust and Trade Policy’ (2003) 47 New York Law School Law Review 167 at 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.
 Hudec, ‘A WTO Perspective’, above n33 at 88. See also Wolff, above n43 at 184–5, arguing that the WTO Secretariat demonstrated its inability to handle complex, factually intensive cases in the Japan-Film case. See also Mariana Bode & Oliver Budzinski, ‘Competing Ways Towards International Antitrust: the WTO Versus the ICN’ (2005) Marburg Papers on Economics at 16 <http://www.wiwi.uni-marburg.de/Lehrstuehle/VWL/WIPOL/downloads/free/Volkswirtschaftliche_Beitraege/2005-03-MPE-Gesamt-Bode-Budzinski.pdf> accessed 26 November 2005. See also Paul B Stephan, ‘Against International Cooperation’ in Michael S Greve & Richard A Epstein (eds) Antitrust Jurisdiction in the Global Economy (2004) at 86 arguing that ‘[a]nalogies to trade agreements do not work, because the assumptions, interests and legal tools at stake in competition policy are far less transparent, and therefore far less susceptible to oversight and satisfactory dispute settlement, than those presented by direct trade barriers’.
 See, for example, Hudec, ‘A WTO Perspective’ above n33 at 100.
 See John O McGinnis & Mark L Movsesian, ‘The World Trade Constitution’ (2000) 114 Harvard Law Review 511. The focus of this article is on the attempts to include labour and environmental issues in the WTO trade agreements. However, the arguments may also be applied to competition policy. McGinnis and Movsesian warn that any expansion of WTO authority from purely adjudicative trade responsibility into more broad-ranging regulatory powers will attract interest groups which will attempt to divert the WTO into covert protectionism.
 Daniel Tarullo, ‘Norms and Institutions in Global Competition Policy’ (2000) 94 American Journal of International Law 478 at 489–492. See discussion in Claus-Dieter Ehlermann & 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.
 See Bernard Hoekman, ‘Competition Policy and the Global Trading System: A Developing Country Perspective’ (Working Paper No 1735, World Bank, 1997) 19. See also David W Leebron, ‘The Boundaries of the WTO Linkages’ (2002) 96 American Journal of International Law 5 at 26, discussing the possibility of institutional bias where diverse areas of regulation are linked in a single institutional structure.
 Robert Litan & Carl Shapiro, ‘Antitrust Policy During the Clinton Administration’ in Jeffrey Frankel & Peter Orszag (eds) American Economic Policy in the 1990s (2002).
 See Edward Graham, ‘Internationalizing Competition Policy: An Assessment of the Two Main Alternatives (2003) 48 Antitrust Bulletin 947 at 952 arguing that the mistrust flows in large part from discontent over the results of the TRIPS agreement. The TRIPS Agreement sets out minimum standards for the protection of intellectual property: See Marrakesh Agreement Establishing the World Trade Organization, opened for signature 15 April 1994, 1867 UNTS 3 (entered into force 1 January 1995), annex 1C (Agreement on Trade-Related Aspects of Intellectual Property Rights) 1869 UNTS 299 (‘TRIPS Agreement’).
 See ICPAC Report, above n12 at 267, citing responses of Kenya to the WTO.
 See UNCTAD Intergovernmental Group of Experts on Competition Law and Policy, Preliminary Assessment of the Set,  UN Doc TD/B/COM.2/CLP/45 (2004).
 See, for example, the concerns expressed by India in the Working Group on Trade and Competition at the WTO: WGTCP, Report of the Meeting of the 23–24 April 2002, WTO Doc WT/WGTCP/M17 (2002) . See also UNCTAD, Report, above n27 at .
 See Joseph E Stiglitz & Andrew Charlton, Fair Trade for All: How Trade Can Promote Development (2005) at 193 (estimating the cost of maintaining a functioning competition agency).
 See, for example, Stiglitz, above n23. See also Bernard Hoekman & Petros C Mavroidis, ‘Economic Development, Competition Policy and the World Trade Organization’ (2003) 37 Journal of World Trade 1 at 7.
 See, for example, Trade Policy Review — United States, WTO Doc WT/TPR/S/126 (2003) ix (Report by the Secretariat, Trade Policy Review Body), detailing various protectionist aspects of US anti-dumping and countervailing duty policy. See discussion below at text accompanying n 79.
 See Core Principles in a Trade and Competition Context WTO Doc WT/WGTCP/W/221 (2003) – (Communication from the OECD to the WGTCP).
 See Aditya Bhattacharjea, ‘Export Cartels: A Developing Country Perspective’ (2004) 38 Journal of World Trade 331 at 346–7, discussing the cost efficiencies that arise from export cartels.
 See Sykes, ‘Externalities in Open Economy Antitrust’, above n14 at 91.
 Wherever the members of an export cartel account for a substantial share of domestic production, their export decisions are likely to spill over and influence domestic supplies and prices. Additionally, the exchanges of information that are likely to be necessary to operate the cartel on such matters as prices, costs, capacities and sales policies could lead to conscious parallelism in the domestic market. See Ulrich Immenga, ‘Export Cartels and Voluntary Export Restraints between Trade and Competition Policy’ (1995) 4 Pacific Rim Law & Policy Journal 93 at 125–126.
 In Kaldor-Hicks terms, the outcome is efficient. A change in a given state of affairs is regarded as efficient in Kaldor-Hicks terms if the ‘winners’ from the change are able to compensate the ‘losers’ and still be better off than prior to the change, and the maximum amount the ‘losers’ would be prepared to pay to prevent the change is less than the amount the ‘winners’ are prepared to accept as a bribe to forgo the change. Although it must be possible to compensate the losers, actually doing so is not a requirement.
 See, for example, Simon Evenett, Margaret Levenstein, & Valerie Suslow, ‘International Cartel Enforcement: Lessons from the 1990s’ (2001) 4 The World Economy 1221 at 1232.
 See Report (1998) WTO Doc WT/WGTCP/2 (1998)  (WGTCP, report to the General Council); Report on the Meeting of 11–13 March 1998, WTO Doc WT/WGTCP/M/4 (1998)  (WGTCP, note by Secretariat); Report (1999) WTO Doc WT/WGTCP/3 (1999)  (WGTCP, report to the General Council); Report (2000) WTO Doc WT/WGTCP/4 (2000)  (WGTCP, report to the General Council); Report (2002) WTO Doc WT/WGTCP/6 (2002) , ,  (WGTCP, report to the General Council); Report (2003) WTO Doc WT/WGTCP/7 (2003) - (WGTCP, report to the General Council).
 See, for example, Submission from Japan, WTO Doc WT/WGTCP/W/156 (2000) 4 (WGTCP, Communication from Japan); Report on the Meeting of 1-2 July 2002, WTO Doc WT/WGTCP/M/18 (2002)  (WGTCP, note by Secretariat) for the EU view on export cartels responding to the US argument.
 Report on the Meeting of 1-2 July 2002, WTO Doc WT/WGTCP/M/18 (2002)  (WGTCP, note by Secretariat); Report on the Meeting of 20-21 February 2003, WTO Doc WT/WGTCP/M/21 (2003)  (WGTCP, report by Secretariat).
 Some support for this view is found in Evernett, Levenstein, & Suslow, ‘International Cartel Enforcement’, above n62 at 1230–2.
 Webb-Pomerene Act of 1918, c. 50, 40 Stat 516, codified at 15 USC §§ 61–66 (1994).
 Non-Discrimination in the Context of Competition Policy: National Treatment, WTO Doc WT/WGTCP/W/216 (2002)  (WGTCP, communication from India); Hardcore Cartels and Voluntary Co-operation, WTO Doc WT/WGTCP/W/241 (2003)  (WGTCP, communication from China). See also Report on the Meeting of 20–21 February 2003 WTO Doc WT/WGTCP/M/21 (2003)  (WGTCP, note by the Secretariat); Report on the Meeting of 26–27 May 2003 WTO Doc WT/WGTCP/M/22 (2003)  (WGTCP, note by the Secretariat) noting comments by China.
 See Immenga, ‘Export Cartels’, above n60 at 126. See also Bhattacharjea, ‘Export Cartels – A Developing Country Perspective’, above n 58, 351, who argues that industrial structures are much more concentrated in developing countries than developed countries.
 See Report (2002), above n63 at . See also Closer Multilateral Cooperation on Competition Policy: The Development Dimension, WTO Doc WT/WGTCP/W/197 (2002)  (WGTCP, communication from UNCTAD).
 See, for example, Daishowa International v North Coast Export Co. Ltd, 1982–2 Trade Cases (CCH) 64,774 (ND Cal, 1982), in which Japanese paper manufacturers formed a buyer’s cartel in response to US wood chip manufacturers organising themselves as a Webb-Pomerene association. A Webb-Pomerene association is exempt from US antitrust laws.
 See Bhattacharjea, ‘Export Cartels – A Developing Country Perspective’, above n58 at 349.
 Immenga, ‘Export Cartels’, above n60 at 129.
 See Immenga, ‘Export Cartels’, above n60 at 129. See also Spencer Weber Waller ‘The Ambivalence of United States Antitrust Policy Towards Single-Country Export Cartels’ (1989) 10 Northwestern Journal of International Law and Business 98 at 101, citing former Assistant Attorney General Baker who gave this as a reason why US policy in the 1970s was not actively to pursue foreign export cartels.
 See Waller, ‘The Internationalization of Antitrust Enforcement’ above n95 at 397–8.
 See Webb-Pomerene Act of 1918, c. 50, 40 Stat 516, 15 USC §§61–66 (2000); Export Trading Company Act, Pub L 97-290, 96 Stat 1233 (1982), 15 USC §§ 4001–21 (2000). In Australia see Trade Practices Act 1974 (Cth) s 51(2)(g). In India see Competition Act 2002 s 3(5)(ii).
 See discussion at text accompanying n43.
 Bernard Hoekman, ‘Competition Policy and the Global Trading System: A Developing Country Perspective’ (Working Paper No 1735, World Bank, 1997) at 4.
 See Final Act Embodying the Results of the Uruguay Round of Multilateral Trade Negotiations, Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade, opened for signature 15 April 1994, 1868 UNTS 201 (entered into force 1 January 1995) (‘Antidumping Agreement’).
 Id, art 2. Where this cannot be calculated the Antidumping Agreement provides two other methods for calculating the ‘normal value’.
 See, for example, Japan Ministry of Foreign Affairs, ‘Submission by the Government of Japan to the Government of the United States Regarding Regulatory Reform and Competition Policy’ (14 October 2001) and (23 October 2002).
 See Joseph E Stiglitz, Two Principles for the Next Round. Or, How to Bring Developing Countries in from the Cold (1999) World Bank at 5 <http://www.worldbank.org/knowledge/chiefecon/articles/geneva.pdf> accessed 4 October 2006; Anne O Krueger, ‘Challenges Facing the Multilateral Trading System’ in Lee-Jay Cho & Yoon Hyung Kim (eds) The Multilateral Trading System in a Globalizing World (2000) at 8; Hoekman, above n78 at 1. This is also the view of many international trade law specialists. See John H Jackson, ‘Challenges Facing the Multilateral System: Commentary’ in Cho & Kim at 49–50. See also Raj Bhala, ‘Rethinking Antidumping Law’ (1995) 29 George Washington Journal of International Law & Economics 1 at 8–23.
 See Krueger, above n82 at 8; Keith Maskus, ‘Commentary’ in Cho & Kim, above n82 at 114; Michael Trebilcock, ‘Competition Policy and Trade Policy: Mediating the Interface’ (1996) 30 Journal of World Trade 71 at 81. Traditional users of antidumping remedies were the US, the EU, Canada, Australia and New Zealand. These countries dominated antidumping cases until relatively recent times. Since the mid 1990s the number of countries with anti-dumping laws has grown rapidly. By 2001, over 90 countries had enacted such laws, although over 90% of cases still come from just 19 countries. See Chad P Bown, ‘Global Antidumping Database: Version 2.1’ (2006) <http://people.brandeis.edu/~cbown/global_ad/bown-global-ad-v2.1.pdf> accessed 4 October 2006. The number of anti-dumping cases reached a high in 2001 and has been dropping each year since then. See Maurizio Zanaradi, ‘Antidumping: What are the Numbers to Discuss at Doha?’ (2004) 27 The World Economy 403.
 See P D Ehrenhaft, ‘Is Interface of Antidumping and Antitrust Laws Possible?’ (2002) 34 George Washington International Law Review 363 at 369. For a comprehensive coverage of views on antidumping see Robert Z Lawrence (ed), Brookings Trade Forum: 1998 (1998). For a presentation of the arguments for and against anti-dumping see Alan Holmer, Gary Horlick & Terence Stewart, ‘Enacted and Rejected Amendments to the Antidumping Law: In Implementation or Contravention of the Antidumping Agreement?’ (1995) 29 International Lawyer 483.
 See Krueger, above n82 at 8.
 See Trebilcock, above n83 at 77, citing an OECD study which found that ‘dumping posed no threat to competition in more than 90 percent of cases in which the United States and the EU imposed anti-dumping duties in the 1980s’. There is no reason to believe that the situation has changed: see Joseph Stiglitz, Making Globalization Work, above n23 at 91–93, describing US practices.
 See Krueger, above n82 at 8–9.
 See Frederic M Scherer, ‘International Competition Policy and Economic Development’ in Cho & Kim, above n82 at 68.
 See Bruce A Blonigen & Thomas J Prusa, ‘Antidumping’ (Working Paper 8398, National Bureau of Economic Research, July 2001).
 See generally Observations on the Distinction between Competition Laws and Antidumping Rules, WTO Doc WT/WGTCP/88 (1998) (WGTCP, communication from US). See also Ehrenhaft, above n84 at 370; Hoekman, above n78 at 19–20.
 See, generally, Ehrenhaft, above n84. Trebilcock, above n83 at 77 has argued that non-economic rationales for anti-dumping measures (such as distributive justice or communitarian values) are only likely to be relevant in only a very small number of cases.
 See Ehrenhaft, above n84 at 372. It should be noted that temporary, limited relief against import competition is available for restructuring purposes under the Agreement on Safeguards, opened for signature 15 April 1994, 1869 UNTS 154, art 14 (entered into force 1 January 1995) applying art XIX of the GATT, above n4.
 See WGTCP, Observations on the Distinction between Competition Laws and Antidumping Rules, above n90 at 1 where the US argued that ‘there is no reasonable foundation for replacing the anti-dumping rules with competition laws or modifying them in a way that would make them reflect competition policy principles. Stated simply, the anti-dumping rules and the competition laws have different objectives and are founded on different principles, and they seek to remedy different problems.’
 Jurgen Basedow, ‘International Antitrust: From Extraterritorial Application to Harmonization’ (2000) 60 Louisiana Law Review 1037 at 1049. See also Frederic M Scherer, Competition Policies for an International Economy (1994). See also Krueger, above n82 at 9. Stiglitz & Charlton, above n54 at 129–30.
 Spencer Weber Waller, ‘The Internationalization of Antitrust Enforcement’ (1997) 77 Boston University Law Review 343 at 402. See also P J Pierce, Jr., ‘Antidumping Law as a Means of Facilitating Cartelization’ (2000) 67 Antitrust Law Journal 725 at 735 discussing the political attractions of anti-dumping measures. Thomas J Schoenbaum, ‘The International Trade Laws and the New Protectionism: The Need for a Synthesis with Antitrust’ (1994) 19 North Carolina Journal of International Law and Commercial Regulation 393 at 397–402.
 Waller, above n95 at 402. See also Pierce, above n95 at 735 discussing the political attractions of anti-dumping measures; Schoenbaum, above n95 at 397–402. See generally Raj Bhala, above n82 at 3–4 describing anti-dumping measures as the defensive weapon of choice from the 1980s onwards in states such as the US, Canada, Australia and the European Union.
 See, for example, Bernard Hoekman & Peter Holmes, ‘Competition Policy, Developing Countries and the WTO’ (Working Paper No 2211, World Bank, 1999) at 18.
 See generally Pierce, above n95. See also Maskus, above n83 at 116.
 See generally WGTCP, Observations on the Distinction between Competition Laws and Antidumping Rules, above n93. See also Ehrenhaft, above n84 at 363. There is considerable political pressure on the US not to change its stance. See House Members Oppose WTO Trade Rules Renegotiation (2001) US Department of State International Information Programs <http://usinfo.state.gov/topical/econ/wto/wwwh01110701.html> accessed 4 October 2006.
 Compare the US approach with the EU approach. In the US it is necessary to prove that prices were below cost and that recoupment of losses was a rational possibility: see Brooke Group Ltd v Brown & Williamson Tobacco Corp.  USSC 105; 509 US 209 (1993). The European Court of Justice takes a more expansive approach to predatory pricing: see AKZO Chemie BV v Commission  3 CMLR 273.
 See Ministerial Declaration, WTO Doc WT/MIN(01)/DEC/1 (2001)  (WTO Ministerial Declaration, Doha); WGTCP, Report (2002), above n63 at –; WGTCP, Report (2003) above n63 at 11. See also OECD, Recommendation of the Council concerning Effective Action against Hard Core Cartels (1998) C(98)35/FINAL, 3; OECD, Hard Core Cartels (2000); OECD, Second Hard Core Cartel Report (2003) CNM/COMP/TR(2003)7; OECD, Fighting Hard Core Cartels: Harm, Effective Sanctions and Leniency Programmes (2002) at 75. See also the cartel provisions in the UNCTAD Set of Principles for the Control of Restrictive Business Practices: 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). The UNCTAD Principles were recently reaffirmed.
 OECD, Hard Core Cartels: Recent Progress and Challenges Ahead (2003) at 8.
 In the 1990s after it had made the pursuit of international cartels an antitrust enforcement priority and over the opposition of the Justice Department, the Clinton administration bowed to industry pressure and permitted Alcoa to form a global cartel to stabilise aluminium prices. An attempt by the Bush administration to do the same for steel failed. See Stiglitz, Making Globalization Work, above n23 at 201.
 See generally WGTCP, Report (2002), above n63.
 See, for example, ICPAC Report, above n12 at 166–8. See also Scott D Hammond, Deputy Assistant Attorney General for Criminal Enforcement Antitrust Division, US Department of Justice, ‘An Update of the Antitrust Division's Criminal Enforcement Program’ (Speech delivered at the ABA Section of Antitrust Law Cartel Enforcement Roundtable: 2005 Fall Forum, Washington, DC, 16 November 2005). See also generally OECD, Hard Core Cartels (2000) and OECD, Joint Global Forum on Trade and Competition, Second Hard Core Cartels Report above n101.
 Margaret Levenstein & Valerie Y Suslow, ‘Contemporary International Cartels and Developing Countries: Economic Effects and Implications for Competition Policy’ (2004) 71 Antitrust Law Journal 801 at 806.
 See OECD, Competition Committee, Report on the Nature and Impact of Hard Core Cartels and Sanctions against Cartels under National Competition Laws (2002) DAFFE/COMP(2002)7, Annex A (‘OECD, Report on the Nature and Impact of Hard Core Cartels’). For a description of the vitamins case and its aftermath see John M Connor, Global Price Fixing: Our Customers are the Enemy (2001); Harry First, ‘The Vitamins Case: Cartel Prosecutions and the Coming of International Competition Law’ (2001) 68 Antitrust Law Journal 711.
 Levenstein & Suslow, ‘Contemporary International Cartels’, above n106 at 802.
 Ibid. See generally also Study on Issues Relating to a Possible Multilateral Framework on Competition Policy, WTO Doc WT/WGTCP/W/228 (2003) – (WGTCP, note from Secretariat); UNCTAD Intergovernmental Group of Experts on Competition Law and Policy, Preliminary Assessment of the Set,  UN Doc TD/B/COM.2/CLP/45 (2004). Julian L Clarke & Simon J Evenett, ‘The Deterrent Effects of National Anti-Cartel Laws: Evidence from the International Vitamins Cartel’ (2003) 48 Antitrust Bulletin 689.
 The evidence on whether global cartel activity has increased is equivocal. See ICPAC Report, above n12 at 164. But see OECD, Second Hard Core Cartel Report, above n101 at  unequivocally stating that international cartels are on the increase. No one denies that there has been an increase in enthusiasm driven partly by the use of more effective tools of detection: See OECD, Hard Core Cartels, above n101; ICPAC Report, above n12 at 163–200. See also Tarullo, ‘Norms and Institutions in Global Competition Policy’ above n47 at 479–80.
 OECD, Report on the Nature and Impact of Hard Core Cartels, above n107 at 9. See also OECD, Fighting Hard Core Cartels, above n101 at 72.
 OECD, Report on the Nature and Impact of Hard Core Cartels, above n107 at 9.
 Connor, above n107.
 See Levenstein & Suslow, ‘Contemporary International Cartels’, above n106 at 842.
 See Joel Klein, The Antitrust Division's International Anti-Cartel Enforcement Program (Speech delivered at the American Bar Association Antitrust Section Spring Meeting, Washington, DC, 6 April 2000).
 See WGTCP, Report (2002), above n63 at 18 in which it is argued that many of the calculations of cartel harm seem to be ‘rather crude estimates’. See also OECD, Report on the Nature and Impact of Hard Core Cartels, above n107 at 5–6.
 The difficulty in quantifying cartel harm is one reason many competition regimes do not require evidence of harm to establish a breach.
 See Scott D Hammond, An Overview of Recent Developments in the Antitrust Division’s Criminal Enforcement Program (Speech delivered at American Bar Association, Midwinter Leadership Meeting, Kona, Hawaii, 10 January 2005).
 See George J Stigler, ‘A Theory of Oligopoly’ (1964) 72 Journal of Political Economy 44, describing the factors at work in determining the success of a cartel. See also Frederic M Scherer & David Ross, Industrial Market Structure and Economic Performance (3rd ed, 1990) at 235–74. See also Peter Asch & Joseph Seneca, ‘Characteristics of Collusive Firms’ (1975) 23 Journal of Industrial Economics 223.
 There is a lively discussion about the overall harm done by anti-competitive collusion, particularly in oligopolistic markets. See Robert Bork, Antitrust Paradox: A Policy at War with Itself (with new introduction and epilogue, 1993) at 179–91 arguing that cartels are essentially fragile and in oligopolistic markets do little harm. Compare Jonathon B Baker, ‘Two Sherman Act Section One Dilemmas: Parallel Pricing, the Oligopoly Problem and Contemporary Economic Theory’ (1993) 38 Antitrust Bulletin 143.
 See, for example, Dominick T Armentano, Antitrust: The Case for Repeal (1999). Antitrust repealists are often driven by extreme notions of liberalism. Thus, antitrust interferes with the free exercise of private property and the freedom of markets. See Dominick T Armentano, The Myths of Antitrust (1972) at 53.
 Public choice theorists reject the notion that intervention is socially beneficial: See generally Charles Rowley & Anne Rathbone, ‘Political Economy of Antitrust’ in Manfred Neumann and Juergen Weigand (eds), The International Handbook of Competition (2004). Deep scepticism about government intervention has been a persistent theme in US thought: See, for example, William H Page, ‘Ideological Conflict and the Origins of Antitrust Policy’ (1991) 66 Tulane Law Review 1.
 These are referred to as negative errors. According to non-interventionists cartels will eventually collapse because barriers to entry are generally low and markets are quite robust in their ability to correct market imperfections. These assumptions, based on the organisational theories of economists such as George Stigler, are a common theme in the antitrust analysis of the Chicago school.
 These are referred to as positive errors. Unlike negative errors, there are no forces counteracting the regulatory mistake. See, for example, Fred S McChesney, ‘Talking ’Bout My Antitrust Generation: Competition for and in the Field of Competition Law’ (2003) 52 Emory Law Journal 1401 at 1412–3.
 Thus, Eleanor Fox has argued: The most libertarian wing of the Chicago School would have preferred no antitrust law at all, unless it prohibited only government interference with the freedom of business. Adherents believed that cartels would self-destruct faster than government intervention could catch them. And the most conservative wing of Chicago School believed that competitor collaboration was usually reasonably necessary to sustain a healthy and productive economy. Eleanor M Fox, ‘What is Harm to Competition? Exclusionary Practices and Anticompetitive Effect’ (2002) 70 Antitrust Law Journal 371 at 378, citing John S McGee, In Defense of Industrial Concentration (1971).
 See OECD, Second Hard Core Cartel Report, above n101 at – discussing the tools required to make an anti-cartel enforcement policy effective. See also OECD, Hard Core Cartels: Recent Progress and Challenges Ahead above n102 at 23–27.
 See US Department of Justice, Antitrust Division, Status Report: An Overview of Recent Developments in the Antitrust Division’s Criminal Enforcement Program (2004) at 8 <http://www.usdoj.gov/atr/public/guidelines/202531.htm> accessed 19 February 2006 (‘[T]he Leniency Program is the Division’s most effective generator of international cartel cases…’). See also Michael Sullivan & Josée Filion, ‘The Basics of International Cartel Enforcement in Canada’ (2004) Competition Bureau Canada <http://www.competitionbureau.gc.ca/internet/index.cfm?itemID=899 & lg=e> accessed 24 September 2005, arguing that ‘[b]y far the most powerful technique available to detect, investigate, prosecute and ultimately punish and deter cartels has been the use of immunity programs…’. Mr Sullivan is the major case director and strategic policy advisor at the Canadian Competition Bureau. But see Julian Joshua, ‘Competition Law Enforcement: Criminalisation, Cartels, Leniency and Class Actions: A Look into the Future’ (12 October 2004) Competition Law Insight at 3 arguing that as at October 2004 the EC’s leniency program had failed to have any major impact on competition enforcement in the EU. EC officials are much more positive about the role of leniency: See, for example, Neelie Kroes, The First Hundred Days (Speech delivered at the 40th Anniversary of the Studienvereinigung Kartellrecht 1965–2005, International Forum on European Competition Law, Brussels, 7 April 2005). Ms Kroes is Commissioner in charge of competition policy.
 See R Hewitt Pate, International Anti-Cartel Enforcement (Speech delivered at 2004 ICN Cartels Workshop, Sydney, Australia, 21 November, 2004).
 There is universal support amongst industrialised nations that an effective sanction should be based primarily (though certainly not exclusively) on deterrence rather than on notions of retributive punishment or compensation. See OECD, Report on the Nature and Impact of Hard Core Cartels, above n107 at . See also ICN Working Group on Cartels, Building Blocks for Effective Anti-Cartel Regimes (2005) at 53.
 In relation to capacity building a considerable amount of work is being done by the OECD, UNCTAD and by the ICN.
 Mostly international single firm conduct is about the activities of suppliers, although it also includes anti-competitive conduct by buyers, for example, by supermarket chains. See, for example, UNCTAD, Commission on Trade in Goods and Services, and Commodities, Market Entry Conditions Affecting Competitiveness and Exports of Goods and Services of Developing Countries: Large Distribution Networks, Taking into Account the Special Needs of LDCs, UN Doc TD/B/COM.1/EM.23/2 (2003). In practice, however, the main concern has been with supplier conduct.
 The problem of evidence gathering is likely to be quite different to that experienced in the case of international cartels. The monopolist is much less likely to be engaged in secretive activity. To the extent that its impugned activities are conducted secretly it is likely to be for genuine commercial reasons.
 General Electric/Honeywell (COMP/M-2220)  OJ C 074/06. See Alec Burnside, ‘GE, Honey, I Sunk the Merger’ (2002) European Competition Law Review 107. An application for annulment was dismissed by the Court of First Instance: See Honeywell International Inc / General Electric Company v Commission (T-209/01, 210/01)  OJ C 48/26.
 See Department of Justice, ‘Statement by Assistant Attorney General Charles A James on the EU's Decision Regarding the GE/Honeywell Acquisition’ (Press Release, 3 July 2001). See Charles A James, ‘International Antitrust in the 21st Century: Cooperation and Convergence’ (Address delivered to the OECD Global Forum on Competition, Paris, France, 17 October 2001).
 See Daniel J Gifford & E T 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 EC was surprised that the US Federal Trade Commission did not challenge the merger given its aggressive attitude to other mergers, while the US viewed the EC’s stance as designed primarily to protect the European manufacturer, Airbus Industrie (which was in receipt of subsidies from three European governments): See ICPAC Report, above n12 at 55. This has been a constant theme of US critiques of EU merger policy, namely that EU policy sometimes protects competitors rather than the process of competition.
 Alina Kaczorowska, ‘International Competition Law in the Context of Global Capitalism’ (2000) 21 European Competition Law Review 117 at 117. Concern was also expressed by President Clinton and some members of Congress: See James P Griffin, ‘Jurisdiction and Enforcement: Foreign Governmental Reactions to US Assertions of Extraterritorial Jurisdiction’ (1998) 6 George Mason Law Review 505 at 518. See also generally Barbara Crutchfield George, Lynn Vivian Dymally & Kathleen A Lacey, ‘Increasing Extraterritorial Intrusion of European Union Authority into US Business Mergers and Competition Practices: US Multinational Businesses Underestimate the Strength of the European Commission from GE-Honeywell to Microsoft’ (2004) 19 Connecticut Journal of International Law 571, discussing the tensions between US and EU authorities in a variety of cases.
 See Burnside, above n134; George, Dymally & Lacey, above n138 at 596–7.
 See US-EU Merger Working Group, Best Practices on Cooperation in Merger Investigations (2002) <http://europa.eu.int/comm/competition/mergers/others/eu_us.pdf> accessed 4 October 2006.
 See ICN Merger Working Group: Investigation and Analysis Subgroup, ICN Merger Guidelines Workbook (2006). See also ICN Merger Working Group, Investigative Techniques Subgroup, Investigative Techniques Handbook for Merger Review (2005).
 Of course, if B is unable to enforce its judgment against F and F chooses to ignore that judgment, then the result is under-regulation.
 See generally Amanda Cohen, ‘Surveying the Microsoft Antitrust Universe’ (2004) 19 Berkeley Technology Law Journal 333.
 For the orders made in the US Microsoft case see US v Microsoft Corporation, 231 F Supp 2d 144 (DDC, 2002). The consent order was made pursuant to Antitrust Proceedings and Penalties (Tunney) Act, codified at 15 USC §§ 16(b)-(h). A similar order was made in respect of those actions that had not been settled by compromise. An appeal against this order was dismissed in its entirety: See Commonwealth of Massachusetts v Microsoft Corp,  USCADC 140; 373 F 3d 1199 (DC Cir, 2004).
 Commission Decision of 24 March 2004 relating to a proceeding under Article 82 of the EC Treaty (Case COMP/C-3/37.792 Microsoft)  EC Comm 12 (‘EC-Microsoft decision’) upheld by the Court of First Instance in Microsoft Corp v Commission of the European Communities (T-201/04)  ECR I-xxx. See EC, ‘Commission concludes on Microsoft investigation, imposes conduct remedies and a fine’ (Press Release IP/04/382, 24 March 2004). The Commission’s remedies were altered slightly by the Court of First Instance.
 See, for example, the response of certain members of the US House of Representatives International Relations Committee to the EC’s Microsoft decision in 2004: ‘Open letter to European Union Competition Commissioner Mario Monti’ CNET (2004) <http://news.com.com/House+letter+Windows+issue+not+a+concern+for+EU/2100-1012_3-5178964.html?tag=nl> accessed 28 February 2006. The complaining representatives were evenly split between Republicans and Democrats, suggesting a national rather than a partisan reaction.
 See Department of Justice, Antitrust Division, ‘Assistant Attorney General for Antitrust, R Hewitt Pate, Issues Statement on the EC's Decision in its Microsoft Investigation’ (Press Release, 24 March 2004). See also R Hewitt Pate, ‘Roundtable Conference with Enforcement Officials’ (Paper delivered at American Bar Association, Section of Antitrust Law Spring Meeting Washington, DC, 2 April, 2004).
 See Korea Fair Trade Commission, The Findings of the Microsoft Case (7 December 2005) 3–4 <http://ftc.go.kr/data/hwp/micorsoft_case.pdf> accessed 24 March 2006. See also Department of Justice, ‘Statement of Deputy Assistant Attorney General J Bruce McDonald Regarding Korean Fair Trade Commission's Decision in Its Microsoft Case’ (Press Release, 7 December 2005). Microsoft has appealed.
 KFTC Findings, above n149 at 3–4.
 Id at 4.
 See above n140.
 See above n141.
 The ICN has proposed a working group to look into unilateral conduct. See ICN, ‘Unilateral Conduct Working Group Draft Mandate’ (2006) <http://www.internationalcompetitionnetwork.org/media/library/conference_5th_capetown_2006/UnilateralConductWorkingGroupDraftMandate.pdf> > accessed 13 March 2008.
 See, for example, Robert Bork, Antitrust Paradox: A Policy at War with Itself (with new introduction and epilogue, 1993); Richard A Posner, ‘The Chicago School of Antitrust Analysis’ (1979) 127 University of Pennsylvania Law Review 925; Frank H Easterbrook, ‘The Limits of Antitrust’ (1984) 63 Texas Law Review 1.
 See, for example, Joseph F Brodley, ‘The Economic Goals of Antitrust: Efficiency, Consumer Welfare, and Technological Progress’ (1987) 62 New York University Law Review 1020. More recently post-Chicago scholars have accepted the Chicago school’s concentration on efficient outcomes but have insisted that strategic market conduct is an important factor in determining those outcomes. See, for example, Thomas G Krattenmaker & Steven Salop, ‘Anticompetitive Exclusion: Raising Rivals’ Costs to Achieve Power over Price’ (1986) 96 Yale Law Journal 209. For an analysis of post-Chicago antitrust see Herbert Hovenkamp, ‘Antitrust Policy after Chicago’ (1985) 84 Michigan Law Review 213. For a useful overview of the debate see Michael S Jacobs, ‘An Essay on the Normative Foundations of Antitrust Economics’ (1995) 74 North Carolina Law Review 219; Stephen G Corones, Competition Law in Australia (3rd ed, 2004) at 20–25.
 Diversity may be preferred to harmonisation for a variety of reasons: to reflect different competition objectives; to cater for different economies and different levels of development (see, for example, Michal S Gal, Competition Policy for Small Market Economies (2003); to foster a competition of competition regimes (see Karl Meessen, ‘Competition of Competition Laws’ (1989) 10 Northwestern Journal of International Law and Business 17; Karl M Meessen, Economic Law in Globalizing Markets (2004) at 94); to avoid rule lock-in that can occur with an internationally-agreed set of rules; to reflect federalist theories of regulation (see Paul Stephan, ‘The Political Economy of Choice of Law’ (2002) 90 Georgetown Law Journal 957).
 A consumer effects test is a better rule for allocating jurisdiction than an aggregate effects test. The proposed rule is derived from choice of law methodologies that rely on jurisdiction-selection rather than those, popularised in the US, that rely on rule-selection. On choice of law methodologies see generally Peter Machin North & James J Fawcett (eds), Cheshire and North’s Private International Law (13th ed, 1999); Michael Tilbury, Gary Davis & Brian Opeskin, Conflict of Laws in Australia (2002); William Tetley, ‘A Canadian Looks at American Conflict of Law Theory and Practice, Especially in the Light of the American Legal and Social Systems (Corrective vs. Distributive Justice)’ (1999) 38 Columbia Journal of Transnational Law 299; Michael J Whincop & Mary Keyes, Policy and Pragmatism in the Conflict of Laws (2001).
 See discussion at text accompanying n155.