A new CEPR research project on `Market Structure, Industrial Organization and Competition Policy in Europe' will apply recent theoretical advances in the academic literature on industrial organization to empirical research that will be of direct relevance to policy-makers. The project will also involve practitioners in the formulation of new areas of academic research. This project is led by Paul Seabright, Fellow and Director of Studies in Economics at Churchill College, Cambridge, and a Research Fellow in CEPR's Applied Microeconomics programme; Alasdair Smith, Professor of Economics at the University of Sussex and former Co-Director of CEPR's International Trade programme; and Xavier Vives, Research Professor at the Institut d'Anàlisi Econòmica (CSIC), Universitat Autònoma de Barcelona, and Co-Director of CEPR's Applied Microeconomics programme. The Centre gratefully acknowledges the financial support for the project provided by the Commission of the European Communities under its SPES programme.

This project was launched at two recent workshops which brought together academics and policy-makers to exchange views about the priorities for research on competition policy. The first workshop was organized by Research Fellows Carmen Matutes (Universitat Autònoma de Barcelona), Jordi Gual (Instituto de Estudios Superiores de la Empresa, Barcelona, and Universitat Autònoma de Barcelona) Patrick Rey (ENSAE, Paris), and Paul Seabright. Held at the Institut d'Anàlisi Econòmica on 24/25 April, it focused primarily on the implications of recent developments in the industrial organization literature for the design of competition policy. The second workshop, organized by Paul Seabright and held at the European Centre for Advanced Research in Economics (ECARE) at the Institut d'Etudes Européennes, Université Libre de Bruxelles, on 8/9 May, elicited policy-makers' views about the priorities for academic research on competition policy.

Suzanne Scotchmer (University of California at Berkeley) opened the first session of the Barcelona workshop with her paper, `Antitrust Policy, the Breadth of Patent Protection and the Incentive to Develop New Products', with Jerry Green. This considered markets characterized by `sequential innovation', in which the inventors of derivative improvements may undermine original innovators' profits through competition and hence undermine incentives to innovation. Scotchmer developed a simple, two-period model of two firms investing in R&D to consider how allowing cooperative agreements or broadening the first patent may mitigate such profit erosion. She found that ex ante agreements, in which firms share both the costs and proceeds of research, may increase the initial innovator's expected profits without inhibiting later research. Their incentives to innovate may nevertheless be too small to warrant such investment. Scotchmer favoured ex post agreements, which grant licences after research costs are sunk. Also, broadening the first patent does not always benefit the initial innovator, whose profits may be reduced if this raises the second innovator's bargaining power by giving it a credible threat not to invest unless the first innovator shares the costs. In practice, the optimal patent breadth will depend on details of the market and technology.

In his presentation, `Patenting Strategies in Innovation', Henning Bank (DG XIII, Commission of the European Communities) argued that the aim of a patent system is to protect the rights of inventors to the results of their intellectual effort rather than to foster innovation. Patents holders do not automatically gain monopoly powers; they must disclose their inventions in a manner that enables others to make use of their research and must initiate proceedings themselves against others who make commercial use of their patented innovation. Bank argued that the economic returns from patent protection often do not warrant the costs of their registration internationally and questioned innovators' ability in practice to establish whether or not their inventions are used by competitors. If the expected life-cycle of an invention or the product in which it is embodied is only a few years (as is often the case in electronics), obtaining patent protection may simply take too long to be worth while; and proving the `novelty' of any innovation is often fraught with difficulties.

The second session focused on the deregulation of key economic sectors: the electricity and airline industries. David Newbery (Department of Applied Economics, Cambridge, and CEPR) presented his joint work with Richard Green, `Competition and Regulation in the English Electricity Industry', which analysed the recent experience of vertical disaggregation of generation and distribution. He maintained that privatization of the generating sector as an asymmetric duopoly has produced an uncompetitive and inefficient market structure. If new entry can be deterred, inefficiency costs will arise from the deadweight losses of operating at high prices and low supply; on the other hand, with new entry, prices will approach competitive levels but there will be considerable unnecessary investment in new capacity. Recent tightening of the incumbent generators' sulphur emission targets, the development of highly efficient combined-cycle gas turbines and distributors' desire to develop a countervailing power against the generators all suggest that the second outcome is now the more likely. Newbery concluded that the rest of the world should learn from the UK's discovery through privatization of potentially cost-reducing changes in operating practice; well- regulated, state-owned enterprises elsewhere should be able to introduce the necessary administrative and regulatory reforms at lower cost.

David Encaoua and Anne Perrot (Université de Paris I, Sorbonne) then presented their joint paper with Michel Moreaux, `Network Effects and Competition in Airline Markets', which argued that the current coexistence of Community-level regulation of intra- European flights, high protectionism for domestic flights and bilateral agreements between EC member and non-member countries for international flights inhibits the full exploitation of the network effects inherent in air transport and may place European airlines at a competitive disadvantage vis-à-vis their US counterparts. Using a simple, two-firm model to find price and departure time equilibria, he found from welfare comparisons that consumers may benefit from asymmetric traffic rights which increase compatibility of departure times, reduce firms' horizontal differentiation on certain routes and thereby increase competition. He proposed developing a `hub-and-spoke' system at the European level restricted to the traffic between primary and secondary cities.

Opening the second day with his paper, `Competition Policy in the United States', Lawrence White (New York University) described the institutional setting and main areas of application of US competition policy. In recent years privately initiated antitrust suits have outnumbered government law-suits by ten to one, and much antitrust law is therefore developed by judges in privately initiated cases. Government policies against horizontal price fixing and market allocations among competitors have remained stringent, while actions against `monopolization' have virtually ceased. The 1982 Merger Guidelines for concentrated markets significantly improved the definition of the relevant market within which to calculate market shares and established clear `safe harbour' and `presumptive challenge' levels of concentration. Policy on vertical restraints remains rather unclear: vertical price restraints (such as resale price maintenance) are illegal, but non-price restraints (such as territorial restrictions) are subject to a `rule of reason'.
In his presentation, `Competition Policy in Eastern Europe', Joe Phillips (OECD) described the legislative framework of competition policy in Russia and Poland in particular. He analysed flaws in their respective laws and their application, arguing that problems regarding their enforcement derive from the absence of a culture of competition and from the intrinsic indivisibilities of former state monopolies. Phillips advocated the introduction in Eastern Europe of guidelines of the type currently being established in Russia.

Colin Mayer (City University Business School, London, and CEPR) presented his joint paper with Julian Franks, `Corporate Control: A Synthesis of the International Evidence'. This compared the Anglo-Saxon or `outsider' system of corporate control, in which take-over bids act as disciplinary tools to control firms' managements, with the Continental European and Japanese `insider' systems, which involve the direct control of management by an advisory board. Such pronounced international differences in the structures of corporate ownership and control reflect differing views about the rights of different stakeholders and the merits of markets for corporate control. European/Japanese `committee' governance is essentially an extended form of credit analysis, which allows a high standard of screening and monitoring of firms engaged in well-established or basic activities, where known skills can be readily transferred from one firm to another. `Auction' governance as found in the UK and US appears better suited to new or speculative activities. The market for corporate control is really a market for corporate strategy in which managers compete for the rights to implement their preferred strategies. There is little information on which to base future assessments; venture capitalists therefore make subjective assessments of future prospects rather than retrospective evaluations of past performance, and divergent views should be encouraged.

Opening the first session of the Brussels workshop, on `Market Power and its Remedies', Patrick Rey drew attention to some unresolved issues. There have been no convincing measures of the deadweight losses associated with market power, while recent theories suggesting that such losses derive largely from productive inefficiencies deserve further study. Assessments of market power are difficult and should focus on potential competition and how the reduction of entry barriers, such as strategic barriers erected by incumbent firms, can enhance its effectiveness.

Martin Howe (Office of Fair Trading, London) noted that the US Merger Guidelines specify that a merger should be allowed if market entry is viable within two years; new insights on market entry might provide a means of verifying such a bench-mark in practice. Kurt Stockmann (Bundeskartellamt, Berlin) added that assessments of market power require an improved understanding of the exercise of corporate control, which increasingly involves instruments other than dominant shareholdings.

In the next presentation, Michel Catinat (DG XIII, Commission of the European Communities) focused on innovation and patents. The latter grant temporary monopoly power to their holders and also guarantee information disclosure, without which knowledge transfer may be delayed. Firms in international markets now make less use of the patent system, however, preferring to form clusters of firms to master new technologies and share know-how. Such cooperation often takes place without formal contractual links; the firm is now a less well-defined entity, and traditional antitrust analysis may be seriously misleading. Finally, the enforcement of a `level playing field' on firms from different countries which the European Commission has recently adopted for telecommunications and information technology requires a definition of `unfair' trade related to distributional criteria.

Patrick Rey suggested that recent theoretical work provides a suitable framework for the design of a patent system but has neglected the potential costs of keeping different national patent systems and their interaction with other incentives for R&D, such as research joint ventures and subsidies. Thomas Janicki (Bundesministerium für Wirtschaft, Bonn) criticized both the current emphasis on cooperation and policies that may induce firms to engage in R&D solely to qualify for subsidies. Richard Portes (CEPR and Birkbeck College, London) maintained that ex post research is now required to assess the effectiveness of the high levels of support currently available for research cooperation at national and EC levels. Catinat replied that a Community ban on state aids would have catastrophic consequences for Europe's high-tech industries, but evaluation of their effectiveness was difficult since they are mainly aimed at R&D at a pre-competitive stage.

Portes noted that British and American firms are much more vulnerable to take-over than their Continental European competitors and asked whether the European Commission had plans to open up the European market for corporate control. David Deacon (DG IV, Commission of the European Communities) replied that the US/UK take-over system represents a form of institutional market failure, since firms' owners are no longer involved in their risk. Moving towards such a take-over system is therefore undesirable; but there is increasing recognition of international aspects of competition policy, and antitrust agreements with the US and within the GATT are on the Commission's agenda. David Newbery (Department of Applied Economics, Cambridge, and CEPR) noted the further danger that a public firm faces a lower risk of default than a competing private firm in another country and can therefore secure lower- cost capital; identifying subsidies to public firms is also difficult.

Xavier Vives (Universitat Autònoma de Barcelona and CEPR) opened the second session, which focused on strategic interactions between horizontally or vertically related firms to reduce competition. The predictions of models derived from recent game- theoretic analyses of industrial organization depend critically on the fine details of markets, but some robust rules of thumb have emerged. Price fixing and similar overtly restrictive agreements should be banned; facilitating practices such as best- price clauses are potentially very effective in supporting high prices and deserve closer scrutiny; and all vertical restraints should be illegal when they increase rivals' costs and allow firms to exercise market power, unless they entail demonstrable efficiency gains.

David Deacon argued that, according to the Chicago School, vertical restraints are usually procompetitive, although those that segment markets according to national boundaries as in car distribution clearly fall foul of the single-market programme. The roles of minority interlocking shareholdings and long-run stable relationships among groups of firms as in Japanese keiretsus also deserve further research; they run contrary to current EC rules but often create significant productivity gains.

Opening the third session, on `Regulation and Institutional Design', Alasdair Smith (University of Sussex and CEPR) stressed that designers of competition policy form part of the process; they must consider the trade-off between sophisticated and simple policy designs and the relationship between the makers of policy and the agencies that implement it. The assignment of policy to different layers of government and of functions within the Commission are also significant in the European context. Smith also raised the issue of how much harmonization of policies on competition and employment practices is needed to prevent `unfair' trade.

Horst Reichenbach (DG II, Commission of the European Communities) called for further study of the intimate linkages between competition policy and industrial, trade and investment policies. He expressed concern that state aids may be substituting for former trade barriers; these remain too high and often unduly favour capital-intensive investments, even though discipline on state aids and competition policy were preconditions of the recent EEA agreements. Thomas Janicki proposed making state subsidies more transparent as a solution, citing the recent report by the German Economics Ministry as an example. Kotaro Suzumura (Hitotsubashi University and Japan Center for Economic Research, Tokyo) suggested basing both trade and competition policies on social welfare considerations, which clearly indicate that voluntary export restraints are a bad form of protection. Malcolm Bradbury (Department of Trade and Industry, London) stressed the importance of penalties in the design of competition policy rules: if penalties are low, firms have no incentives to obey rules, but they will face an unacceptably high level of uncertainty if high fines are combined with the exercise of significant discretion by the competition authorities.

The first day concluded with a round table discussion on the most urgent research priorities. Xavier Vives suggested extending traditional analysis to take account of interactions among industrial, trade and regional policies, re-evaluating research on R&D in the light of new growth theory, applying organization theory to corporate governance, and emphasizing political economy issues more strongly in formal modelling. André Sapir (DG II, Commission of the European Communities, Université Libre de Bruxelles and CEPR) contrasted the shortage of empirical evidence with the abundance of policy advice based on theoretical arguments. Empirical research should evaluate the effectiveness of the Community's R&D and anti-dumping policies and could usefully assess governments' anticompetitive behaviour in the service sector, which is not facing deregulation. Miguel Angel Fernàndez Ordonez (Tribunal de Defensa de la Competencia, Madrid) noted that foreign competition will not constrain dominant firms in services, for which analysis of regulation is more important than evaluating the role of state aids. Refik Erzan (World Bank and Bogazici University, Istanbul) agreed that trade cannot increase competition in certain sectors but warned against underplaying the importance of free trade.

Martin Howe called for more empirical evaluations of the welfare effects of competition policy decisions and stressed the need for guidance on such basic questions as market definition, the assessment of entry conditions and the evaluation of likely responses to changes in market structure. Kotaro Suzumura proposed focusing on the nature of competition: Japan's successful research joint ventures in developing new integrated circuits had required both cooperation at the development stage and competition in the product market; R&D cooperation had been explicitly directed (and limited) to catching up with IBM. Colin Mayer (City University Business School, London, and CEPR) called for analyses of the interactions between national and international regulatory bodies and the criteria governing the level at which jurisdiction should lie.

The second day focused on competition policy for countries in economic transition. In his introductory remarks, David Newbery noted large firms' striking dominance of Eastern Europe's industrial structure and advocated breaking them up to reduce their ability to lobby governments for continued subsidies. This is more urgent for horizontally than for vertically integrated enterprises, since it will induce competition directly as well as indirectly by facilitating new entry. Foreign take-overs of the former state enterprises may further help to separate political from economic power.

Anna Fornalczyk (Anti-Monopoly Office, Warsaw) also advocated breaking up large firms and strengthening merger control, noting that Poland's Anti-Monopoly Office had already taken 60 decisions to divide such firms. Managers of state enterprises often organize `crisis-cartels' or join each others' supervisory boards, which is leading to price fixing. She dismissed the view that break-ups will weaken the position of Polish firms on world markets, since their structures are artificial and were only set up for plan implementation, but abandoning all regulation in reaction to the experience of the planned economy would be dangerous. Finally, Poland must attract investment from multinational companies but can not allow them to exert monopoly power; price setting by some foreign companies is already a cause for concern.

Thomas Janicki argued that restructuring should not precede privatization since a privatization agency might make investment decisions not in accordance with potential buyers' wishes. He also favoured protecting foreign investors to enhance foreign investment, but only under the condition of a credible and binding time-limit. Kurt Stockmann responded that East European governments were already offering too much protection to Western companies. Paul Seabright pointed out that if privatization's primary purpose was to delegate the restructuring process, the least efficient firms should be privatized first, which has not happened in practice. Malcolm Bradbury also advocated restructuring before privatization, although privatizing East European companies as single-product firms may be mistaken: if management skills are a scarce resource, product line diversification may be more important than in the West.

Ferenc Vissi (National Office of Economic Competition, Budapest) noted that the acquisition of firms by multinationals has increased the concentration of ownership, since they have bought neither production facilities nor capacity but rather the exclusive rights to new markets. Refik Erzan suggested that multinationals might be buying such licences for entry barriers or for protection; existing levels of protection in Eastern Europe are low and there is a historic opportunity to preserve this situation.