51316 v1 THE WORLD BANK POLICY NOTE 1 Competition Provisions in Regional Trade Agreements: Lessons for China 1 June 30, 2009 Main messages · Competition laws and policies are increasingly being established at the regional level, as they could be instrumental in supporting the benefits of trade and investment liberalization. · Competition laws in Regional Trade Agreements (RTAs) fall in the 4 models, ranging from fully centralized to fully decentralized. Their impacts have been mixed, because of institutional and behavioral reasons, relating to lack of competition culture or political will to promote implementation domestically. · So far RTAs in China have not included competition policy. Competition policy could be included in China's future RTAs, but through the decentralized model, with a focus on cooperation. · China may want to use the opportunity of these negotiations to (1) further discipline its state-owned enterprises; (2) carefully consider the possible role of antidumping policies; and (3) promote and lock-in domestic reforms aimed at improving its domestic competition policies. 1 This policy note was prepared at the request of the Ministry of Commerce, People's Republic of China. It draws on material prepared by Kamala Dawar, Henry Gao, and Gary Hufbauer for the Regional Trade Agreement Workshop co-sponsored by the MOFCOM and the World Bank that was held in Beijing on May 21-22, 2009. It benefited from the contributions of all workshop participants, with special thanks to Mr. Zhu Hong and Ms. Zhang Bin, respectively Deputy Director-General and Director, Ministry of Commerce. For more information, please contact Jean-Pierre Chauffour (jchauffour@worldbank.org) or Min Zhao (mzhao1@worldbank.org). 1 1. The costs imposed by anti-competitive practices were recognized by Adam Smith more than two hundred years ago. Today, countries widely appreciate Smith's insights and the potential for anti-competitive practices to undermine the benefits of trade liberalization. Even though the Doha round negotiators under WTO auspices have no mandate to negotiate a multilateral framework on competition policy, individual countries have signed rules on competition policy in at least one quarter of all bilateral and regional trade agreements (RTAs) in force or under negotiation. The inclusion of competition-related policies are part of the growing agenda on "behind the border" measures, ­ the effort to deepen trade agreements to cover liberalization in services, investment and labor markets, government procurement, monetary and financial integration, and environmental issues. 2. Competition policies may play a significant role in supporting the benefits of trade and investment liberalization. They prevent the economic and consumer welfare gains anticipated from liberalizing markets from being undermined by cross-border anticompetitive practices (e.g., cross-border mergers and acquisitions that could monopolize the domestic market). Competition provisions also tend to reduce collateral distortions from other policies, for example anti-dumping measures implemented with price or volume undertakings. They support region-wide competition policy institutions, and ensure that foreign firms can access open and competitive markets. 3. Competition provisions could be particularly useful for dynamic emerging countries such as China. Commitment to sound competition policies can help developing countries overcome powerful domestic constituencies that would otherwise block the reform process and carry on their cartel practices. Recent empirical evidence suggests that competition provisions are welfare enhancing in the long-term; the research takes issue with the view raised by some countries that competition provisions will compromise their development efforts. The main challenge facing developing countries such as China seems to be the cost to effectively enforce competition-related policies. 4. Among China's existing RTAs, none includes special chapters or side agreements on competition. To the extent that some of China's RTAs contain isolated provisions addressing competition policy, they are usually rather general and do not create specific legal obligations. Moreover, they often do not go beyond the existing WTO provisions. For example, Articles 69 and 70 of China's RTA with Singapore address monopolies and exclusive service suppliers and restrictive business practices, respectively. However, closer examination reveals that the two articles simply copied the language of Articles VIII and IX of the GATS without adding anything new. 2 2 Similarly, the language on the rejection of practices abusing intellectual property rights in Article 111.1(e) of the RTA with Chile simply reiterates the principles in Article 8 of the TRIPS. 2 5. The scarcity of competition provisions in China's existing RTAs does not necessarily mean that China is reluctant to engage in international cooperation on competition policy. Instead, during both the drafting and the enforcement of the Anti- Monopoly Law (AML), China has been cooperating with foreign competition authorities at both the formal and informal levels. This implies some interest by China to consider competition policy in its future RTAs. Several countries, such as Australia, New Zealand and Iceland, have made demands to China on competition provisions in their RTA negotiations. 6. This policy note will review the international experience with competition provisions in RTAs with a view to drawing some possible lessons for China as it pursues its regional trade strategy. International experience 7. The scope of the competition provisions in existing RTAs around the world range from a comprehensive regional competition law to a minimal regime encouraging cooperation in preventing anti-competitive practices. A comprehensive regional law would usually address issues related to cartels and abuse of dominance, state aid and inter-agency cooperation, as well as procedural and transparency requirements. Some regional competition regimes incorporate consumer law to address cross-border consumer protection issues. This can complement competition policies' economic efficiency and social welfare objectives. However, the application of binding judicial dispute settlement mechanisms to support the implementation of competition law and policy is uncommon. 8. With or without a regional competition regime, provisions with competition objectives are often introduced in RTAs without negotiating a competition chapter. These provisions include the principles of non-discrimination and national treatment as well as competition related provisions in other areas, such as intellectual property and services. Nevertheless, these provisions do not present a comprehensive commitment from the parties to create an independent framework to tackle harmful cross-border restrictive business practices or to enable private parties to seek redress against such practices. 9. A useful taxonomy to understand the scope of competition provisions in RTAs would differentiate between substantive and procedural provisions, and between cooperative versus unification models. The cooperative model would promote separate, independent laws with information exchange, while the unification model would tend to harmonize the parties' substantive rules and institutions. 10. Using this taxonomy, one could distinguish between four basic models of competition law, ranging from a fully centralized regional competition law, to the 3 fully decentralized model (see Box 1). Both developed and developing countries have experienced all four models, with varying degrees of success. However, due to the relative immaturity of these regimes, there is still little evidence available on the impact of these provisions to make definitive conclusions about their effects. 11. While most RTA competition regimes focus on the supply side behavior of firms, it is also important to consider the possible demand-side market imperfections. Preventing barriers to market entry, cartels, monopolies, and other restrictive business practices works to protect both consumers and firms. Yet, dealing with these barriers does not directly address demand-side market imperfections caused, for instance, by lack of consumer information or the inability to switch suppliers. Enhancing notification, information sharing, and investigative assistance among the Member States can work to protect foreign consumers from domestic anti-competitive businesses practices and to protect domestic consumers from foreign anti-competitive business. It can also help to prevent inappropriate consumer policies becoming unnecessary barriers to trade. A number of RTAs have explicitly mainstreamed their consumer law provisions within or alongside the competition regimes to address the legislative and enforcement gaps in cross-border trade relating to consumer welfare. For instance, the Australian regional competition agreements with the United States, South Korea and Papua New Guinea incorporate consumer policy provisions. The COMESA treaty also addresses consumer welfare from within its competition regime. 12. While including competition provisions in a RTA makes sense in theory, implementation based on limited evidence from US and EU experience has proven difficult. One feature that clouds an evaluation is that the level of ambition of competition policies varies depending on the level of economic development among the signatories and the extent that its regulatory structure is shaped by the rule of law. The United States and the European Union have been most active in advocating ambitious models for competition provisions; Japan and New Zealand have taken a less comprehensive approach. 13. Implementation records have been particularly poor in developing countries. Lack of effective implementation and enforcement is more common for RTA members that did not possess a national competition law prior to accepting regional commitments; or in those RTAs with members at very different stages of competition regime development. This is partly due to lack of knowledge and experience of competition policies. But without an effective domestic law in all the member states of an agreement, there can be no legal basis for a member to take any action against practices organized in another member state in respect of the effects upon its own territory. And where regional regimes establish measures such as cooperation, notification, consultations to generate regional benefits, they are of little use nationally without the human resources and expertise to absorb or respond to the information. 4 Box 1. Models of Competition Regimes in RTAs Full Centralization. The most comprehensive regional competition regime consists of a supra- national law with a distinct regional jurisdiction. While the regional and domestic laws can overlap, the regional laws only address competitive practices taking place in the regional territory that is beyond the jurisdictional boundaries of just one member of the RTA. These competition laws are directly applicable within the territory of a Member and superior to any national law or judgment that is inconsistent with the regional law. Regional competition laws may also have `direct effect' in the Members' jurisdictions, providing firms or citizens the right to invoke the regional law in the domestic courts of the member countries. To fully centralize the regional competition law, a complementary regional institutional mechanism is established to conduct investigations, enforce actions and assess and levy penalties. Examples of this model include the competition laws established in the EC Treaty, the COMESA, UEMOA and the Andean Pact. Partial Centralization. The less centralized competition model establishes a regional competition law, but supports this with only a partially centralized agency. As with the fully centralized regional regime, the independent regional law has direct applicability and superiority over inconsistent national laws. However, although the central agency has the mandate to receive complaints and initiate independent investigations, it must work with the Member State competition agencies and national courts to process case actions leading to enforcement and remedies. Examples of this model include ANZCERTA and CARICOM. Partial Decentralization. Further down the scale of centralization are regional regimes which are not buttressed by a regional authority. While a regional law is established, different means of intergovernmental cooperation are responsible for promoting the regional law. There is no independent regional body with powers of investigation and enforcement. Thus, the application of the law is left entirely to the Members States. National competition authorities have the jurisdiction to bring cases and they are also the recipients of complaints addressed at any violation of the regional competition law. Mercosur is an example of this model. Fully Decentralized. The least centralized regional regime identified in a regional competition regime exists where the Members do not create a regional law but rather set out the necessary criteria for national law and cooperation principles to address anti-competitive practices that are detrimental to the functioning of the RTA. These provisions require the members to adopt or maintain measures to proscribe anticompetitive business conduct and to take appropriate action, but do not set out any more specific competition rules. The provisions formalize consultation and co-operation between the parties on the enforcement of those laws via mutual legal assistance, notification, consultation and the exchange of information. Examples of this model include NAFTA Chapter 15, Canada-Chile, Canada-Costa Rica and SACU. 14. The reasons for the poor implementation of regional competition provisions are both structural/institutional, and behavioral, relating to lack of competition culture or political will to promote implementation domestically. Relying on bottom-up implementation initiatives in south-south agreements is less appropriate 5 when both a competition culture and buttressing regional laws and institutions are absent. A well designed regional competition agreement needs to takes account of these local circumstances and act as a policy tool to create the national structural and behavioral environment necessary to benefit from regional competition provisions. 15. North-south agreements have better implementation records when the more developed party offers appropriate technical assistance, capacity building and ability to the less developed regional partner. For those members with nascent or non-existent competition regimes, technical assistance provisions should be designed to impart the necessary expertise and experience over the long term. This will assist in inculcating the necessary behavioral reforms. While RTA competition provisions can offer the legislative push and policy lock-in necessary for sustained reform, initially a more appropriate key objective may be to focus on establishing a culture that values competition at the national or sub regional level in the region. 16. In RTAs involving members with no or little experience of applying competition policy, the provisions should focus primarily on the exchange of information, technical assistance and capacity building. Subsequent negotiations can expand the agreement. More general commitments should only be implemented after the necessary expertise and cooperation mechanisms have been developed. Provisions could be included that oblige members, over a specified period of time, to adopt competition laws that can address the full range of private and government sanctioned anti- competitive practices and outcomes. Regional laws can also act as a temporary alternative to the expense of establishing and implementing domestic competition laws. 17. Soft law can be more appropriate if governments are uncertain of the underlying technical issues and judicial consequences of the provisions. If competition regimes are at an early stage of implementation, other non-judicial mechanisms may also be more appropriate, including voluntary peer review, consultations on implementation issues and informal diplomatic methods. These can subsequently be complemented by a non-binding competition policy review mechanism and non-binding consultations, promoting voluntary implementation of competition policy obligations and ultimately harmonization. 18. Mainstreaming consumer law within competition regimes can help to address the legislative and enforcement gaps in cross-border trade relating to consumer protection. This complements and advances the economic and social welfare objectives of competition policy. 19. Transition economies and other countries with a legacy of heavily state-led economies may find it more beneficial to include effective regulation on state aid and antidumping policy to address government created trade distorting practices. Also, in highly centralized regimes that establish a regional authority to assist implementation, the authority must be endowed with strong investigative powers, 6 adequate resources and expertise, cease and desist orders, as well as the power to collect fines if it is to avoid becoming a paper tiger. Possible lessons for China 20. As China becomes more integrated, cross-border market failures can emerge to undermine the benefits of its increased liberalization. These could include regional cartels, cross-border predatory pricing, restrictive vertical marketing agreements establishing price discrimination, and mergers designed to create regionally dominant positions. Unilateral competition measures undertaken by national governments cannot address such negative cross-border anti-competitive practices adequately unless there is some mechanism to promote inter-governmental cooperation. 21. In principle, implementing effective regional competition law and policy can be used to address these negative practices and generate more efficient markets offering better quality goods for cheaper prices, and encourage investment. Further benefits may also be achieved from mainstreaming consumer policy into regional competition laws to develop the effective demand for competition. The option for cooperation to deal with cross-border consumer protection issues complements both economic and social development objectives. 22. China can draw on the international experience to maximize the benefits of negotiating competition provisions in its future RTAs. While more cost-benefit analyses of the different regimes and provisions discussed above are required to craft appropriate regional competition regimes addressing both the institutional and behavioral factors obstructing successful implementation, a number of preliminary conclusions can be drawn as below. 23. Negotiating appropriate competition provisions in an RTA necessarily need to take account of the local institutional and developmental context. Given China's development status, the rest of this policy note will review possible avenues for China in its future negotiations of competition provisions in RTAs, before providing some possible recommendations to further improve its domestic competition law regime. 24. First, with regard to the overall structure of the competition policy provisions in RTAs, as the competition regime of China is still very young, it is probably better to start with isolated competition provisions for selected sectors (such as services or IPR) and gradually move to broader chapter or side-agreement on general competition issues later. In this regard, China might want to start including such provisions in its RTAs with partners with similar background and experiences (e.g., Singapore) or those with higher integration levels (e.g., Hong Kong), then move on with other partners as it has built up sufficient experience. 7 25. Second, China could carefully balance its defensive and offensive interests on competition policy. On the one hand, as the "factory of the world", China wields considerable influence on the prices of industrial and consumer products. While such market power generally means more profits, it can also raise concerns from third countries' competition authorities. 3 On the other hand, due to the growing importance of its domestic market, many multinational corporations (MNCs) are interested in increasing their market shares in China, at times using their dominance in the international market to gain "unfair" advantage in the Chinese market either by dumping their products at extremely low prices or by raising their prices to generate large profits. Therefore, for the dual purposes of protecting the legitimate interests of Chinese firms in the foreign market and ensuring effective enforcement of China's own competition laws in the domestic market, China might consider including provisions on notification, negative comity, positive comity, coordination, and joint investigation in its RTAs. 26. Third, in its future RTA negotiations, China could usefully put the focus on the procedural provisions such as cooperation, consultation and information exchange, rather than substantive provisions that aim to harmonize or coordinate competition standards or rules. Due to its unique historical experience, the challenges facing the competition regime in China are rather different from those in other countries. At one end of the spectrum, because many sectors are still in early stages of development and the entry barriers are very low, these sectors are highly fragmented and even the market leaders only have very small market shares. Thus, the regulators have historically been focusing on the unfair competitive behaviors of small players and lack experience in dealing with monopolies. At the other end of the spectrum, administrative monopolies pose real problems in the Chinese market. Because these administrative monopolies are typically affiliated with government ministries and regulated by special laws and regulations, the reach of the AML is greatly limited. As the main reason for both problems is the legacy of planned economy in China, there is limited relevant international experience on how to handle these problems. 27. Fourth, China may want to use the opportunity of its future RTA negotiations to further discipline its state-owned enterprises (SOEs). Because of the dominance of SOEs in certain sectors, foreign negotiating partners might request China to include in future RTAs special provisions on SOEs, monopolies and enterprises with special and exclusive rights. While China might be reluctant to agree to such provisions in the short term, including these provisions is in China's long-term interest as competition from foreign firms is usually the only way the SOEs could be made to follow competition disciplines in practice. 3 As illustrated by the recent cases of antitrust actions against Chinese Vitamin C manufacturers and magnesite producers, compliance with foreign competition regulations have become a major challenge for some Chinese firms. 8 28. Fifth, in a similar vein, China may want to consider the strategic use of antidumping measures for its own benefit. For a long time, Chinese firms have been the prime targets of antidumping actions. Indeed, some commentators even argue that, in some cases, the main reason that the Chinese firms form cartels to fix prices is to avoid accusations of dumping. While RTA provisions prohibiting the application of antidumping measures seem to be in the short-term interest of Chinese firms, it is unclear whether such provisions serve its long-term interest as China itself has been increasingly resorting to antidumping investigations and measures. In this regard, further study and better coordination between the antidumping and competition authorities in China would help formulate and establish a common position. As China has adopted a rather critical approach to the existing antidumping regime in the Doha negotiations, China should also coordinate its negotiating positions in both regional and multilateral settings if it decides to adopt a different approach on antidumping measures in future RTA negotiations. 29. Sixth, as the competition issues facing developed and developing countries are rather different, China might consider taking different approaches in RTA negotiations with different countries. Generally speaking, the competition laws and regulations in developing countries are less developed. Thus, in RTAs with developing countries, China might focus on information exchange or joint studies on issues such as the relationship between industrial policy and competition rules, regulations of SOEs, or unfair competition among smaller firms. Moreover, depending on the willingness of the other country, there might not be any competition provision at all in the RTA. In RTAs with developed countries, especially when the other country demands the inclusion of competition policy in the negotiations, China might consider requesting for transition periods, technical assistance, or even exemptions from particular obligations to soften the impact of over-ambitious competition provisions. 30. Finally, competition provisions in RTAs also provide the opportunity to promote and lock-in domestic reforms aimed at improving competition policies domestically. To reap the maximum benefits from the market access opportunities created by RTAs, China might wish to consider further improving its competition law regime as follows: · Better coordinating China's industrial, competition, and trade policies. Industrial policy typically involves fostering a few national champions in a sector to the detriment of the interests of other firms. For competition policy, the interests of consumers are the main concern. For trade policy, the interests of exporting firms and import companies usually take priority. Because of their different focus, these policies might well run into conflicts. For example, when the state decides to merge several firms together to promote the development of certain industry, such action might run afoul to the competition obligations in RTAs. Similarly, when several exporters decide to raise prices together to deal with anti-dumping investigations, they might violate the anti-trust laws of the importing countries. Coordination 9 among various ministries involved in competition is important but not always easy; similarly for coordination between central government and local government; and among local governments themselves. One way to mitigate the problem is to have joint meetings between the different agencies at regular intervals, or at least require an agency to notify and consult the other agencies when it makes changes to its policies. · Further improving the legislative framework. While the enactment of the AML is a great achievement, further improvement can be made to make it more effective. Clarification is needed on what industries include in the Article 7, which stipulates that the state protects the lawful operations of industries "which are crucial to the well-being of the national economy and dominated by SOEs". The ambiguity may lead to the abuse of this clause. For instance, as many commentators argue, while the telecommunication sector should be included, it is doubtful that value-added telecom services should be accorded the same importance as basic telecom services. Moreover, China may consider enhancing the effectiveness of the AML and furthering empowering the competition authorities to allow them to effectively deal with administrative monopolies. Article 51 states that, when dealing with administrative monopolies, the anti-monopoly enforcement agencies do not have power to sanction the administrative bodies directly. Instead, they may only make recommendations to the relevant bodies at a higher level. Doubt casts on the willingness of the supervising bodies to punish their subordinates. The same article also provides that, in the case that any other relevant laws and regulations have different provisions on administrative monopolies, the provisions in those laws should prevail over the AML. Again, this greatly reduces the effectiveness of the AML. · Streamline the competition enforcement process by further delineating the jurisdictions among different competition authorities. In the existing competition regime in China, the enforcement power is shared by several agencies with potentially over-lapping jurisdictions. While this could be justified by the need to develop expertise between agencies, the current practice of giving the jurisdiction to a case to the agency that first gets its hands on the case does not take the relative expertise of the agencies into consideration. In this regard, one possibility is to specify a clear job division and have a one agency in charge of certain subject matters. 10 References Gary Hufbauer and Yee Wong (2009) China's Trade Agreements: Lessons from Other Countries, background paper Henry Gao (2009) Supporting China's Regional Trade Agenda: China's Strategies on Negotiating Competition, Environment, and Mode 4 Provisions in its RTAs, background paper Kamala Dawar (2009) Competition Provisions in Regional Trading Agreements, background paper 11