|
Traditional analyses of international trade policy have tended to assume that markets are perfectly competitive. But this approach has not been able to provide an explanation for the observed behaviour of firms or government policy interventions. In recent years attention has been focused instead on the strategic aspects of international trade policy and the behaviour of firms. These analyses have relaxed the assumptions of free trade and allow for increasing returns to scale in production, differentiated products, and firms with significant market power. Many of the theoretical implications of this approach have been outlined by CEPR Research Fellow Tony Venables in Discussion Paper No. 74 and in his article in Bulletin No. 12.Further investigation of these theoretical insights has been an important focus of research in CEPR's Inter national Trade programme. Much of this work has taken place under the auspices of a major research programme conducted jointly with the National Bureau of Economic Research on `Empirical Studies of Strategic Trade Policy', with financial support from the Ford Foundation. On September 17/18, a conference was held by the NBER in association with CEPR to review work done to date under this programme. The conference, which followed an earlier workshop reported in Bulletin No. 19, was organized by Alvin Klevorick (Yale University and NBER), Paul Krugman (MIT and NBER), and Alasdair Smith, Co-Director of the Centre's International Trade programme. Highlights of the programme included: BULLET Alasdair Smith (University of Sussex and CEPR) and Anthony Venables (University of Southampton and CEPR) on `Trade and Industrial Policy under Imperfect Competition: Some Simulations for EEC Manufacturing' BULLET Robert Feenstra (University of California at Davis and NBER) on `Symmetric Pass-Through of Tariffs and Exchange Rates under Imperfect Competition: An Empirical Test' BULLET Marvin Lieberman (Stanford University) on `Learning-by-Doing and International Competition: Autos and Semiconductors in the United States and Japan' BULLET Richard Baldwin (Columbia University and NBER) and Paul Krugman (MIT and NBER) on `Modelling International Competition in High Technology Industries: Lessons from Aircraft and Semiconductors' BULLET Victor Norman (Norwegian School of Economics and Business Administration and CEPR) on `Incentive Problems in Discretionary Trade Policy: Two Examples' BULLET Val Eugene Lambson (University of Wisconsin) and David Richardson (University of Wisconsin and NBER) on `Tacit Collusion and Voluntary Restraint Arrangements in the US Auto Market' BULLET Andrew Caplin (Princeton University and NBER) and Kala Krishna (Harvard University and NBER) on `Tariffs and the Most-Favoured-Nation Clause: A Game-Theoretic Approach' VERs in the UK Car Market Similar work has also been presented to policymakers at a number of CEPR lunchtime meetings in a recent series on `Trade Policy and the New International Economics'. At the first of these talks, on 3 November, Alasdair Smith analysed the effects of voluntary export restraints (VERs) on the UK car market. His talk was based on research conducted with Caroline Digby and Anthony Venables, sponsored by the Ford Foundation and the ESRC, and reported in a forthcoming Discussion Paper. The opinions expressed by Smith were his own and not those of the Ford Foundation, ESRC or CEPR, which takes no institutional policy positions. Since 1977, by agreement between the Japanese and UK motor industries, the Japanese share of the UK car market has been held at 11%. Other European markets are similarly affected by VERs: the Japanese share of the French market has been restricted to 3% since 1977, and Japanese sales to Italy have been negligible since the 1960s. Strategic analyses of international trade are particularly appropriate to the motor car market, Smith argued. There is a relatively small number of producers, with significant economies of scale and producing widely differentiated products and as an instrument of trade policy; VERs themselves affect the competitive structure of the market. Smith noted that VERs had three important economic effects. By restricting the supply of particular cars the VER raises the prices of these cars like an ordinary import tax. Whereas an import tax raises revenue for the government, the VER increases the export revenues of the exporting firm. In addition, the VER reduces competition and may enable non-Japanese firms to sustain higher prices. In order to assess the welfare consequences of the VER, Smith, Digby and Venables had constructed a model of imperfect competition in the world car market. Smith outlined the nature of the model, based on 1985 data, which distinguished seven markets: France, Germany, Italy, UK, the rest of the EC, Japan, and the rest of the world. Eight groups of producers are identified: France; Germany, excluding US multinationals; Italy; the UK, excluding US multinationals; the European operations of US multinationals; Japan; North and South America; and other countries, including Sweden and Korea. The model was used to estimate the effects of removing the VER. The simulations suggested that the Japanese share of the UK car market would rise from 10.9 to 17.1%, and the share of UK producers would fall from 18.3 to 17.0%. The shares of other producers would fall, except for the US producers who, as market leaders, were assumed to reduce their prices in order to compete more effectively with the Japanese. The reduction in prices produced a welfare gain of around £127 million, or 2.1% of the value of consumption. Smith's calculations also indicated that the VER is a very costly means of preserving employment in the motor industry: £50,000 to £70,000 per job saved. This estimate was comparable to that of Hindley, he noted, even though Hindley had derived a higher estimate (£180 million) of the welfare effect of the VER. Smith also discussed the economic effects of replacing the VER with an `equivalent tariff' on Japanese imports, i.e. a tariff set at a level which would leave output and employment levels in the UK motor industry unchanged. Although Smith did not advocate such a tariff, which in any event is prevented by UK membership of the EC, this comparison provided a means of measuring the `excess' costs of the VER. The equivalent tariff was still costly, but it did lead to small reductions in prices, and around half the costs associated with the VER were eliminated by the shift to a tariff. This analysis of the motor industry demonstrated, according to Smith, that the new approach to trade theory could yield useful insights into policy issues. In particular it confirmed that VERs can be substantially more costly than tax instruments which have the same effect on output and employment. He cautioned, however, against using these new theories to justify widespread policy interventions. Although most industries have important features which are not well described by economists' traditional models of perfect competition, the design of optimal trade policy interventions may be very sensitive to the exact form of competitive behaviour within an industry, Smith argued. Our ability to model imperfectly competitive behaviour is still rudimentary, so it is hard to make a strong case for activist policy intervention in imperfectly competitive industries. In addition, most of the theoretical gains from protection are at the direct expense of other countries, so from an international perspective, protectionism is more likely to be a game with a negative or zero sum than one with a positive outcome. After the talk, one member of the audience asked how the costs of the VER compared with those of an export subsidy that had the same effect on UK employment. Although Smith knew of little research on this, he conjectured that subsidies would be a less costly form of protection since they had a direct impact on production. Another participant argued that it would be unwise to abolish the VER, since it was the only bargaining lever with which the UK could encourage the Japanese to reduce the domestically produced share of their own market, which presently stood at over 98%. But Smith pointed out that the detrimental effects of Japanese protection on their own consumers' welfare should be sufficient to encourage them to increase access to foreign producers. SUBHEAD A Norwegian Example At a December lunchtime meeting, Victor Norman also expressed scepticism about whether interventionist trade policy could be successful. Norman is Professor of International Economics at the Norwegian School of Economics and Business Administration and a CEPR Research Fellow. He spoke at the second meeting in the CEPR series; the meeting received financial support from the Ford Foundation. The new approaches to international trade theory suggest that, at least in oligopolistic markets, interventionist trade policy can increase domestic welfare by shifting profits from foreign to domestic firms or by lowering prices for domestic consumers. But as Smith noted, these theories do not specify the precise nature of such policies, for this depends on the characteristics of the industry concerned. Norman's own research, reported in CEPR Discussion Paper No. 218 and conducted jointly with Sonja Daltung and Gunnar Eskeland, illustrated the difficulties of constructing such inter ventionist policies in two Norwegian industries with very different market structures, ski production and Caribbean cruise shipping. Theoretical analyses of strategic trade policy indicate that policy interventions may enable domestic firms to reap profits that would otherwise accrue to foreign firms, or make export competition between domestic firms less keen. Intervention may also benefit consumers, by inducing foreign firms to lower their prices in the domestic market or by encouraging new entry into an otherwise monopolistic industry. But Norman noted that the direction of the appropriate policies depends on the details of market structure. If firms perceive quantities as the strategic variable, an export subsidy could encourage domestic firms to produce more: foreign firms will reduce their production and profits will be shifted from foreign to domestic firms. If, on the other hand, firms perceive price as the strategic variable, domestic profits will increase if domestic firms can be encouraged to raise their prices: an export tax is therefore appropriate. Moreover, if the objective of policy is to increase domestic profits, domestic firms should be encouraged to produce less; if the objective is lower domestic prices, they should be encouraged to produce more. Since little can be said about the qualitative nature of optimal policies on the basis of theory alone, emp irical work is essential. In addition previous studies of trade policy intervention had indicated that the information needed to formulate optimum policy is of a kind which policy-makers are unlikely to possess, but which the firms in the industry may have. This asymmetry of information poses particular problems, because it can easily create perverse cost incentives. The study of the Norwegian ski industry provided a good illustration, Norman argued: the industry itself is not particularly significant, but it highlights the complexities of trade and industrial policy intervention. There is a small number of domestic producers facing international competition; to survive, two of the Norwegian firms (whose production costs are relatively high) have recently merged. It is not clear whether the merger will be sufficient to ensure the survival of domestic production, so a natural question is whether the industry should be protected. Norman's analysis sought to capture the effects on consumers of government policies designed to change either the domestic price of skis or the variety of both domestic and foreign skis available to consumers. The results suggested that in order to formulate an optimum policy, i.e. in order to maximize consumer surplus, the government must have precise information on conditions for entry into the industry and the cost reductions which can be achieved by the merged domestic firm. If there is no entry or exit of foreign firms into the Norwegian market, the optimum policy is to give domestic producers a 15% production subsidy. If, on the other hand, there is free entry and exit, the optimum subsidy is only 5%: a larger subsidy would force at least one foreign producer out of the domestic market, and consumers would suffer from a narrower range of available products. This creates perverse incentives for domestic firms: the merged firm is better off with a 15% subsidy and no cost reduction than with a 5% subsidy and substantial cost savings. As a result, Norman concluded, domestic welfare is likely to be lower with `optimal' intervention than with no intervention at all. This result is also likely to be valid for other countries and industries, Norman argued. Appropriate policies can only be formulated if detailed information on the industry is available (for example, on cost structures). This information is of a kind which governments are unlikely to possess, although it is available to firms, making policies formed in its absence liable to produce perverse incentives. Interventionist trade policies are difficult to devise and unlikely to succeed, Norman concluded: free trade may be best after all. In the discussion after the talk, one participant argued that the motivation for protection was not economic, but typically stemmed from a desire to preserve employment in existing industries or to sustain domestic production of strategic or defence goods. `Prestige' was also a motivation, and the European drive for high technology production was cited as an example. Another member of the audience was also worried about the consequences of encouraging `sunrise' industries: the edifice of European agricultural protection illustrated the consequences of protective measures. |