WT/WGTCP/W/216 26 September 2002
Working Group on the Interaction between Trade and Competition Policy
The following communication, dated 24 September 2002, has been received from the Permanent Mission of India with the request that it be circulated to Members.
- Although National Treatment (NT) has been a core principle of GATT, it is important to appreciate its relevance before applying it to the area of competition policy. The importance of the national treatment principle is based on the original purpose of the GATT as an agreement to facilitate the reduction of barriers to international trade in goods. As such, national treatment is significant for three inter-related reasons, not all of which are automatically applicable to competition policy.
- First, national treatment ensures that concessions made by Members in respect of trade barriers at the border are not nullified by within-border discrimination between imported and domestically-produced goods, in the form of differential taxation or regulatory requirements by the government of the importing country. Competition policy, on the other hand, is concerned with regulating anti-competitive conduct by private parties. Mergers, or many other vertical restraints, will generally have anti-competitive effects at the cost of both domestic and foreign producers. These must be evaluated in relation to the possible efficiencies they generate, but it is difficult to see how they unfairly block market access to imports, which is the concern of the national treatment principle.
- Second, national treatment is designed to support trade liberalization, which is desirable because it is believed to be on the whole beneficial to the participating countries. This is because the bulk of world trade taken aggregatively is in products produced under competitive conditions. Competition policy, however, is by definition concerned with imperfect competition, where the presumption in favour of free trade is much less clear cut. It is now widely recognized that although there is a theoretical case for optimal trade interventions (tariffs and subsidies), the information requirements for such “strategic trade policy” are prohibitive. However, the same caveat does not apply to competition policy in respect of mergers and cartels, which obviously have very different welfare effects on different countries, depending on where the producers and consumers are located. This has led to conflict between the competition authorities of major advanced countries, for example in the Boeing/McDonnell Douglas merger, and has encouraged antitrust cooperation between them to minimize such conflicts in future. Such cooperation is likely to be mutually beneficial on average for industrialized countries, since for a large enough number of such competition cases, the affected producers and consumers are distributed across both jurisdictions. However, developing countries have very few products that are exported under conditions of imperfect competition, but they have to deal with oligopolistic market structures in respect of imports and inbound foreign direct investment. They are thus likely to be predominantly on the losing side of mergers and RBPs with cross-border effects, and have little to gain from the application of the NT principle in the competition law of developed countries. Until such time as developed countries are willing to consider the impact of mergers on consumers in foreign countries, to rescind the exemption of export cartels in their competition laws, to give serious consideration to enforcing the UNCTAD Set of measures to control RBPs, and to extend the benefits of “positive comity” in competition law enforcement to developing countries, the latter will have to retain the right to challenge foreign mergers and RBPs that have an effect on domestic consumers.
- At the same time, partly as a result of concessions made in multilateral trade negotiations, developing countries are undergoing a process of structural transformation in which many industries are being exposed for the first time to international competition. While there may be benefits from the downsizing and even elimination of uncompetitive industries, there is also a case for orderly restructuring, involving mergers and perhaps rationalization cartels, which have been practised in many developed countries. The need for such measures is, if anything, greater in developing countries. Given their imperfect land, labour and capital markets, these resources cannot be easily transferred to new activities. Although many developing countries, including India, are trying to reform these markets in order to impart greater flexibility to their economies, such institutional reforms are obviously far more difficult than simple trade liberalization. A relatively liberal treatment of mergers and cartels involving domestic firms would prevent the wastage and under-utilization of scarce resources, at least until arrangements are made for their redeployment in other activities.
- The third justification for the national treatment principle in trade agreements, again as a logical corollary of the liberalization of border measures, is that even in cases where a country can gain unilaterally from departing from free trade, the gains turn into losses if other countries act the same way. This is the case, for example, with optimal trade taxes where a country or its firms possess market power. Constraining countries’ ability to raise trade barriers unilaterally is thus in their own self-interest. However, it has not been established that departures from an international competition policy norm can be mutually destructive in the same way. The theoretical literature is quite ambivalent about whether harmonization of competition policy principles is desirable, and there are few robust results of the kind that dominate the theory of trade policy.
- These several distinctions between the national treatment principle as it applies to trade in goods and its potential applicability to competition policy caution us on the appropriateness of carrying it over to the latter. It is perhaps because some of these distinctions are relevant even in relation to cross-border supply of services that national treatment has not been uniformly applied in GATS; instead Members have been allowed to specify exemptions from national treatment in their commitments. Its applicability to competition policy in general is even less obvious.
- Much of the discussion so far has been based on the idea of promoting static allocative efficiency as the sole objective of multilateral agreements. In our earlier communication to the Working Group (WT/WGTCP/W/149), India drew attention to the dynamic objectives of promoting investment and total factor productivity growth, which are more important for developing countries. Here we would only supplement that observation by pointing out that as regards investments in technology, significant deviations from the idea of free markets are allowed in the WTO framework itself, for example, in the TRIPS agreement. In developing countries, where both private and public resources for R&D are limited, promotion of investment may require a stable degree of economic concentration. Increasing exposure to international trade consequent on trade liberalization can be relied upon to keep a check on market power and limit (static) resource misallocation. As argued in a Working Paper of the South Centre viz. ‘Competition policy, Development and Developing Countries’, “…it may be perfectly legitimate for a developing country competition authority to allow large domestic firms to merge so that they can go some way toward competing on more equal terms with multinationals from abroad. Even if the amalgamating national firms are on the horizontal part of the L-shaped static cost curve, bigger size may still promote dynamic efficiency for the reason that firms need to achieve a minimum threshold size to finance their own R&D activities. The competition authority may therefore quite reasonably deny national treatment to the multinationals and prohibit their merger activity (because they are already large enough to achieve either static or dynamic economies of scale in this sense). In these circumstances, a violation of the doctrine of national treatment is likely to be beneficial both to economic development and to competition” (p.17).
- The various WTO Agreements allow for departures from non-discrimination where foreign producers can be shown to have “unfair” advantages that cannot be related to their inherent competitive or comparative advantage. For example, foreign export subsidies can be countervailed; and furthermore it was stated in the Working Group that one justification for not tightening the disciplines on anti-dumping measures is that they are required to counter “unfair” foreign advantages arising from continuing protection and weak competition policies in exporters’ home markets. It is important in this regard to point out that a competition policy that ostensibly applies to all members equally is likely in practice to discriminate against firms in developing countries. Since prosecuting RBPs perpetrated by firms based abroad is going to be extremely difficult for countries with limited resources, domestic producers will in practice bear the brunt of a competition law that enshrines the NT principle, while allowing foreign producers to get away with similar infractions.
- Further, many developing country Members have repeatedly pointed out in this Working Group that they have little experience or expertise in regard to competition policy. This means that competition law principles drawn from countries with much more experience, apart from possibly being intrinsically inappropriate for developing countries, will impose much greater compliance costs. These costs will have to borne both by their governments, which will have to set up the necessary institutions, and by their firms, which will have to invest in legal resources. Such costs, which can in no way be related to comparative advantage or efficiency in production, unfairly discriminate against producers in developing countries and put them at a competitive disadvantage against their rivals in developed countries.
- Developing countries do not yet have the kind of well-developed safety nets that exist in industrial countries to provide for those displaced by import competition. There is thus a greater need to cushion its impact by suitable industrial restructuring measures of the kind mentioned above, which would also enable developing countries to embrace greater trade liberalization. In this sense, a discriminatory competition policy can be a concomitant to a non-discriminatory trade policy. Of course, care must be taken to ensure that these restructuring measures do not have trade-distorting features that nullify commitments made under the multilateral trade agreements. Such measures have been practiced in many developed countries until fairly recently, particularly in periods when they were adjusting to international integration of the kind that many developing countries are now undergoing.
- In the context of meeting the needs of developing countries, it is more appropriate to adopt the concept of non-discrimination in terms of the need to treat different countries with different capacities in a differential manner, and of the need and responsibility to provide assistance, positive measures and affirmative action to local firms and institutions in developing countries to ensure their viability and development so that they can become increasingly efficient and competitive.