Antitrust Implications of Technology Consortia

Article 163 TEC

‘1. The Community shall have the objective of strengthening the scientific and technological bases of Community industry and encouraging it to become more competitive at international level, while promoting all the research activities deemed necessary by virtue of other Chapters of this Treaty.

2. For this purpose the Community shall, throughout the Community,

encourage undertakings, including small and medium-sized undertakings, research centres and universities in their research and technological development activities of high quality; it shall support their efforts to cooperate with one another, aiming, notably, at enabling undertakings to exploit the internal market potential to the full, in particular through the opening-up of national public contracts, the definition of common standards and the removal of legal and fiscal obstacles to that cooperation.*

3. All Community activities under this Treaty in the area of research and

technological development, including demonstration projects, shall be decided on and implemented in accordance with the provisions of this Title.’

* [emphasis added]

consistency through monitoring and assistance in the implementation of EC competition law and the establishment of appropriate antitrust enforcement. This is a challenge of itself, but at the same time the Commission must be in a position to prevent the wide substantive conditions in Article 81(3) from being misapplied. Illegitimate interventions can threaten consistency if these are triggered by national industrial policy considerations distorting competition within the common market or by the lack of a sound economic reasoning. This is a risk to be minimized throughout the EU whether old or new member states, but an important one within the scope of Article 163 to further the role of technology consortia in exploiting economies of scale and disseminating innovative technology more rapidly.

It is also the essence of technology consortia that gives rise to doubts as to the limits of competition law. The intellectual property rights (IPR) influencing the work of a consortium can initiate a dilemma. Whilst interference with IPR on the basis that these constitute a restriction of competition may be justified in very limited circumstances, it must be carefully analyzed and clearly stated in what circumstances an intervention is warranted. This is because the very nature of IPR is to restrict competition. The law at the intersection between competition and intellectual property issues is not entirely clear and remains an analytical challenge for the application of Article 81 to technology consortia. Similarly, Article 82 and its application to IPR puzzle business and antitrust authorities. In this area, the essential facilities doctrine may arguably have played a role in granting compulsory licenses raising the same dilemma as under Article 81. The appropriateness of the essential facilities doctrine and a refinement of circumstances in which Article 82 applies to IPR remain issues to be resolved in the quest to introduce more legal certainty.

In pursuit of addressing these issues it is attempted to show the

practical relevance of this study. To that end, case studies of real-life technology consortia have been carried out to better understand the incentives, competitive impact and contribution to technical progress of these inter-firm collaborations. The business press also features the practical relevance of technology consortia, whereas the empirical studies help to identify the crucial issues to be reconciled at the interface between competition and intellectual property law.

Competition law is to me the most exciting and fast-moving areas of the law. I have had the great benefit of excellent competition law courses at the University of Strathclyde, which have captured my interest since second year of university. In this context, I would like to thank Professor Barry Rodger for his topical and intellectually stimulating competition law sessions at undergraduate and Honours level as well as for his committed supervision of my Honours dissertation. It is a great pleasure to keep up the exchange of ideas through the Competition Law Scholars Forum.

Economic Reasoning, Competition Law and Intellectual Property Law Issues.

law issues with fellows of the university especially, Anna Roubier and Massimo Coluzzi. I could not express the enthusiasm any better than in your words Massimo, ‘it has always been just as much fun to talk about competition law as about football’. I would also like to express my gratitude to Stefanie for her endless love and support. A big thank you and appreciation also go to my parents Anne and Winfried, and my brother Stephan for their love and support making all this possible.

for being my thesis supervisors. Jens-Daniel has been very helpful in the supervision meetings in discussing the overall structure and some of the sources of the present thesis. Professor Dr. Koenig has made this Master year an outstanding experience through his representation of Doc Morris in the free movement of goods case before the ECJ. The discussion sessions with him on free movement as well as state aid have been a real experience.

Bundeskartellamt library and the librarians of the law school at the University of Strathclyde for letting me use their resources in the preparation of this thesis.

© Andreas Seip 2003.

CONTENTS

1 INTRODUCTION............................................................... - 8 -

2 THE ECONOMIC REASONING OF TECH CONSORTIA................... - 10 - 2.1 The Incentives Mechanism ......................................... - 13 - 2.2 Contrasting Stability Issues......................................... - 16 -

2.2.1 Formalization of the Arrangement .......................... - 17 - 2.2.2 Penalizing any Innovation Shortfall ......................... - 17 - 2.2.3 Ejecting the Cheating Member ............................... - 18 - 2.2.4 Sustaining the Consortium despite Default ................ - 19 - 2.3 Benefits and Risks.................................................... - 19 -

2.3.1 Risk Diminishing Factors ...................................... - 21 - 2.3.2 Risk Increasing Factors ........................................ - 24 - 2.4 Issues Relevant for Assessment ................................... - 25 -

3 POLICY IMPLICATIONS FOR COMPETITION LAW ...................... - 27 - 3.1 General Standard Setting Cooperation........................... - 29 - 3.2 Specialised Cooperation - Technology Consortia .............. - 32 - 3.3 Competition Impact Assessment .................................. - 34 -

3.3.1 Market Power.................................................... - 37 - 3.3.2 Cooperative Momentum ....................................... - 39 - 3.3.3 IPR Reliance ..................................................... - 40 - 3.3.4 Ancillary Restraints ............................................ - 42 - 3.4 From Policy to Competition Law .................................. - 43 -

4 THE COMPETITION LAW PERSPECTIVE ................................. - 45 - 4.1 Article 81: General Scope .......................................... - 46 -

4.1.1 The Ins and Outs under Article 81 .......................... - 47 - 4.1.1.1 Induced Collusion ......................................... - 48 - 4.1.1.2 Limitation of Technical Development ................. - 49 -

Economic Reasoning, Competition Law and Intellectual Property Law Issues.

5 CONCLUSION: MAIN FINDINGS ........................................... - 86 -

Appendix………………………………………………………………………………………………….- 91 -

Bibliography…………………………………………………………………………………………..- 100 -

© Andreas Seip 2003.

This thesis is to provide guidance for the antitrust analysis of technology consortia which is challenged by virtue of the various forms the inter-firm collaboration may take, the pooling of intellectual property rights (IPR) and the ambivalent impact this may have on competition. The starting point to a meaningful antitrust analysis of technology consortia is an understanding of the underlying economics. The following chapter is to briefly discuss the incentives of firms to cooperate, the contrasting stability issues prevailing in an anti-competitive cartel as opposed to innovation driven consortia, and the resultant welfare implications in terms of the benefits and risks of cooperation.

This will allow an outline of the workable policy approach to be pursued in applying antitrust law. The third chapter focuses thereby on issues of antitrust analysis by distinguishing between two main types of technology consortia and their role in the innovation process. The assessment is to help the identification of the essential elements in antirust analysis ranging from relevant market definition to market power and intellectual property rights (IPR).

In the fourth chapter, EC competition law is specifically examined against the discussed policy approach. This includes a consideration of relevant anti-competitive conduct relating to technology consortia under Article 81, the relevance of block exemptions, and finally the self- assessment under Article 81(3). In addition to a discussion of the intersection between IPR and Article 81, this will continue to be relevant for the assessment of IPR under Article 82. This chapter will end with a recommendation as to how IPR policies of technology consortia should be formulated to alleviate some antitrust concerns.

Economic Reasoning, Competition Law and Intellectual Property Law Issues.

and competition law work towards the promotion of innovation provided that all stakeholders including firms, competition authorities and courts respect the innovation economics and legal sensitive issues. In order to promote such an awareness the identified uncertainties are addressed in tests, which are to evaluate the competitive implications of technology consortia, whereas the IPR policy is to support the prevention of an antitrust challenge. The refined analysis is then provisionally translated in the format of a guidance notice in the appendix to this thesis.

firms operating in innovation markets. This is evidenced by the fact that often a single firm is a member to a great many different consortia. The welfare implications and therefore the justification for antitrust intervention may vary according to the different stages they operate in the innovation process. For the purposes of the present thesis, technology consortia have to be distinguished from joint ventures since consortia can already operate at the earliest stage of idea generation up to the development of a prototype. Although partly overlapping, a joint venture usually just starts its operation at the point of prototype development up to the full commercialization of the product.

basis of selective anti-competitive and abusive practices that may affect technology consortia either directly or indirectly. The focus will thereby be placed on the future practice once Regulation 1/2003 enters into force. In identifying crucial uncertainties that could limit the effectiveness of the antitrust enforcement regime, an assessment of the relevant decisional practice by the Commission and by the European Court of Justice (ECJ) cases is necessary.

© Andreas Seip 2003.

Technology consortia are a mechanism of technology trading that needs to be distinguished from the unilateral licensing of proprietary technology, which can precede the collaboration in a technology consortium. Technology trading within technology consortia involves firms trading a right to use one another’s technology, rather than the unilateral sale of such a right via licenses. 1 There is a multitude of arrangements possible that come within the scope of a technology consortium. The exchange of technology could thus be organized within a trade association that maintains a research facility and is financed by its members who also benefit from its findings.

Similarly, two or more firms may decide to set up a complementary facility engaging in R&D and use its findings for the participating undertakings’ purposes. 2 In addition, the exchange itself can take various forms including explicit communication of research findings on the companies’ initiatives, or answering questions on request of another participating firm, and plant visits by engineers and technical training of staff. 3 The table below illustrates the variety of possible collaborations that usually distinguish technology consortia from other types of inter-

1 Baumol, W.J. The Free-Market Innovation Machine – Analyzing the Growth Miracle of Capitalism (Princeton, New Jersey: Princeton University Press, 2002), at p. 93. For a concise treatment of economic choices open to firm, which is to determine whether to licence or not and the implications thereof see Beard, R.T. and Kaserman, D.L. ‘Patent Thickets, Cross-Licensing, and Antitrust’ 2002 Antitrust Bulletin, 345, at pp. 347 – 350. 2 Consortia are usually much larger than joint ventures in terms of membership see further Rigatuso, C., Tachi, T., Sylvester, D. and Soper, M. ‘Collaboration between Firms in Information Technology’ EE 290X Group G, at p.2, available at http://www-inst.eecs.

2003.

3 See also Immenga, U. and Mestmäcker, E.J. (Eds.) EG-Wettbewerbsrecht: Kommentar, Band I (Munich, Germany: C.H. Beck, 1997), at p. 1418.

Economic Reasoning, Competition Law and Intellectual Property Law Issues.

important know-how exchange in the consortium. Technology consortia may work on standard-setting 6 ensuring interoperability 7 , develop improved production and process technology 8 , or work on the improvement of product components. Among the most frequently cited by both businesspersons and scholars, is the sharing of the high cost burden in the innovation process from the initial R&D to reaching marketing maturity. 9 Although this is a valid reason, it lacks comprehensiveness and persuasion in explaining why firms frequently revert to technology consortia as a means of cooperation. There is an interdependent incentives mechanism discussed in economic literature that provides further understanding of technology consortia. 10 Obviously, profitability induces firms to participate in sharing their technology, which is enhanced by the fact that the firms engage in complementary research. If the undertakings were to carry out R&D in substitute innovations, which are mutually competitive they would not have an incentive to cooperate and would rather seek to avoid a spreading of their proprietary technology.

Complementary R&D is all the more compelling in high-tech industries with rapidly evolving technology, where the large scale of firms is often due to their investments in incremental improvements of products and process rather than through leapfrogging with revolutionary new products. 11 Thus, in outlining the incentives mechanism exhibited by technology consortia,

6 See further the MPEG example in section 4.1.1.2. Limitation of Technical Development. 7 See further Bluetooth SIG example in section 3.1. General Standard-Setting Cooperation. 8 See further SEMATECH example in section 3.2. Specialised Cooperation – Technology Consortia.

9 Rigatuso, C. et al. (2003), at p.3. See also Baumol, W.J. (2002), at p. 95.

10 Baumol draws on his experience in real life consultancy and economic theory. For the technology consortium model, see Baumol, W.J. (2002), at pp. 97 – 106.

11 Baumol, W.J. (2002), at p. 96. Although the process of leapfrogging and ‘creative destruction’ still holds true as a competitive pressure as discussed in Seip, A. ‘Merger Policy in the E-conomy’ (Glasgow, Scotland: University of Strathclyde, 2002), at pp. 27 – 29, 31 for background and p. 44 available at http://www.e-efficiency.de. See also Posner, R.

‘Antitrust in the New Economy’ 2001 Antitrust Law Journal, 925-943, Ordover, J.A. ‘Antitrust for the New Economy or New Economics for Antitrust’ Summary of Remarks at the FTC/DOJ Hearings “Competition and Intellectual Property Law and Policy in the Knowledge-Based Economy” Washington D.C., 20 February 2002, and Schmalensee, R. ‘Antitrust Issues in Schumpeterian Industries’ 2000 American Economic Review (90), 192, at p. 193. The relationship between competition for the market and

Economic Reasoning, Competition Law and Intellectual Property Law Issues.

evolving high-tech industries speed is of paramount importance, which makes a speedy transfer of technical information mandatory to prevent it from becoming obsolete. 15 The second incentive is the result of the penalizing effect of the cost-reducing incentive. The holdout undertaking is anxious not to lose money due to its non-membership. However, for as long as the undertaking is not a member of the technology consortia, it can expect higher manufacturing costs of its final products since it cannot revert to the R&D efforts of the technology consortium. This cost disadvantage will further grow cumulatively over time unless the holdout undertaking exhibits strong scale economies or other special circumstances that allow a higher expenditure on R&D, while still achieving better cost reductions than the members of the technology consortium.

There is also the incentive of taking advantage of risk insurance by virtue of being a member in the technology consortium. R&D is a costly and risky adventure, which will not always yield the desired results, or may even achieve only little or no value at all in the pursuit of commercial introductions. Yet, it is in these instances when technology consortia provide protection in a given year since the unsuccessful undertaking still benefits from the innovations of the remaining participating undertakings. 16 Beyond the incentives shown by the technology consortium model, there are also other incentives that may vary in strength depending on the individual economic circumstances at hand. Thus, as will be shown by the Bluetooth Special Interest Group (SIG) case study 17 , there may be strong incentives concerning compatibility and standardization of different technologies to allow for greater, interoperable variety of products to be

Costs and Patent Costs: An Empirical Study.’ 1981 Economic Journal 91 (December), 907 – 918.

15 The tremendous speed of technological change is also a factor which influences a firm’s choice towards trade secrets instead of patents. See also Beard, R.T. and Kaserman, D.L. (2002), at pp. 348 and 349.

16 For a real life-example see the Japanese FGCS project discussed in Rigatuso, C. et al. (2003), at p. 8.

17 See below Box 3.2. in Section 3.1. General Standard-Setting Cooperation.

Economic Reasoning, Competition Law and Intellectual Property Law Issues.

C o n t r a s t i n g S t a b i l i t y I s s u e s

2 . 2 .

Price setting cartels and technology consortia are very different. Not just because of the obvious reason that hardcore cartels are anti- competitive and technology consortia generally pro-competitive, but also because of the different incentives or disincentives for cheating that prevail in the respective equilibrium. In order to identify the incentives and disincentives it is sensible to distinguish between the short and long run. 22 In respect of price-fixing cartels, a member can profit in the short run by undercutting the prices of the other members achieving thereby sales above its quota. Similarly, the short run scenario for technology consortia would be the cheating member to retain its information, whilst benefiting from the other members’ disclosure of proprietary technology.

The crucial difference impacting on stability becomes apparent in the long term. In a price-fixing cartel, it is worth for the customer to act as a co-conspirator to the cheating cartel member. Lower prices make it also more difficult to detect such a cartel on the initiative of the demand side. 23 Yet, no customer’s acquiescence is needed for cheating in a technology consortium. Stability of technology consortia result from repetition of the exchange. In addition, there are monitoring criteria that indicate cheating of one technology consortium member such as long-lasting cost reductions, or changes in product characteristics or industry gossip that discourage undertakings from cheating. The following retaliatory mechanisms in technology consortia can be more effective in discouraging cheating protecting it thereby from collapsing.

22 The short term – long term distinction is crucial as the differences impacting on stability will become apparent in the long term. See further Baumol, W.J. (2002), at pp. 106 - 109. 23 The cheating firm will lower the prices by producing above the collusive output. However, lower prices do not necessarily indicate that someone cheated because a low price is consistent with other factors which shifted market demand, such as recession. The imperfection in monitoring collusion makes the sustainability of a cartel more difficult. See further Green, E.J. and Porter, R.H. ‘Noncooperative Collusion under Imperfect Price Information’ 1984 Econometrica 52, 87 – 100; for empirical illustration of the destabilising impact of recession see Suslow, V. ‘Stability in International Cartels: An Empirical Survey’ (Working Papers in Economics, E-88-7: Hoover Institution, 1988); Viscusi, W.K., Vernon, J.M., and Harrington, J.E.Jr. Economics of Regulation and Antitrust (Cambridge, Massachusetts: MIT Press, 3 rd Edition, 2000), at p. 120.

Economic Reasoning, Competition Law and Intellectual Property Law Issues.

other firm for any shortfall in the value of innovation exchanged. 28 Thus firms meet regularly to negotiate monetary compensation to be paid, if one firm fails to deliver the value in innovation expected before the next meeting. Consequently, the firms have a strong incentive to reveal their innovations since if a firm decides to conceal its results; it simply increases the amount it must pay to the other firm as compensation. Although Baumol acknowledges an adverse selection problem in that the innovating firm is more likely to know the true value of the invention, he suggests that this problem is less serious where the parties have gained trust in their respective integrity through repeated bargaining. 29

2.2.3. Ejecting the Cheating Member

Apart from the two options to formalize the nature of the technology consortium, there is the important disincentive to cheating with ejection from the technology consortium. This is an omnipresent and more automatic option inherent in the consortium of which the parties are all aware. The non-membership constitutes an ever growing opportunity loss, which is likely to be very costly in the long run. Non-members to a price cartel, in contrast, may even benefit from price increases by the cartel, whereas the non-member to technology consortium can only benefit from innovation spillover. 30 This is all the more pertinent in highly contestable markets, where external pressures through entrant competition strengthen the stability of the technology consortium because it gives further incentive to sustain the competitive advantage by virtue of being a member in the consortium. This contrasts sharply with price cartels where entry usually destabilizes collusion in the absence of a factor that prevents entry. 31 It follows that technology consortia and price cartels can also have very diverging internal

approach in the EU is the same. The mere exercise of IPR does not constitute an abuse, see Hoffmann-La Roche v. Centrafarm [1978] ECR 1139 28 Baumol, W.J. (2002), at p. 108.

29 Ibid.

30 ibid.

31 Viscusi, W.K. et al. (2000), at p. 120.

Economic Reasoning, Competition Law and Intellectual Property Law Issues.

of technology consortia as a form of cooperation with either vertical, horizontal or conglomerate elements. Technology consortia tend to generate welfare benefits in terms of wider and rapid use of innovations. 34 However, welfare implications are not clear-cut. One must first distinguish between compatibility and incompatibility. If faced with incompatibility, undertakings will compete for the market to create a de facto standard without cooperation. 35 If these undertakings, however, choose to cooperate they compete within the market through subsequent development of products that build upon the cooperative standard or technology. 36 An important result of the incompatibility – compatibility dichotomy may be that with incompatible standards each standard and the related product(s) constitute a narrow market, whereas a cooperative compatible standard with its related product(s) constitute a wider market or network. In the absence of cooperative standard-setting the undertakings see themselves confronted with a fragmented market, consisting of multiple and incompatible products, which may eventually contain only a single proprietary product. 37 It follows that cooperative standard-setting may prevent a firm from utilizing network effects that could lead to the adoption of a single proprietary standard monopolizing their products. 38 Although cooperative

first undertaking to submit evidence in an already undergoing investigation, it can achieve a reduction of up to 50 per cent in the level of the fine.

34 Much of the innovation of today requires cooperation to come about in the first place.

See Glader, M. (2001), at pp. 519 – 520 and Rigatuso, C. et al. (2003), at p. 8. A prerequisite to economic growth is the speedy transfer of technology. To that end, the innovation process and commercial exploitation can be efficiently facilitated by collaboration in a technology consortium. See further Sachwald, F. (2001), at pp. 3 and 7. 35 De facto standards are a result of market forces and must be distinguished from de iure

standards which have been promulgated by legislative bodies, official standard authorities or standard-setting organisations. See Koenig, C., Bartosch, A. and Braun, J.D. EC Competition and Telecommunications Law (The Hague, Netherlands: Kluwer Law International, 2002), at pp. 620 – 621.

36 See also Katz, M.L. and Shapiro, C. (2002), at pp. 29 -30, Lind, R.C. and Muysert, P.

‘Innovation and Competition Policy: Challenges for the New Millenium’ [2003] ECLR 87, at pp. 89 – 90.

37 See also Katz, M.L. and Shapiro, C. (2002), at pp. 27 - 29. 38 An example of an almost monopolisation constitutes Microsoft market share of 95% in PC

operating system, which is safeguarded by high entry barriers through the wider availability of compatible software and Intel compatibility. Similarly, the attempted monopolisation of the Internet browser. See also Gey, P. ‘Das Berufungsurteil in Sachen Microsoft –

Economic Reasoning, Competition Law and Intellectual Property Law Issues.

generally result in a positive feedback since it induces an increase in the profit maximizing oligopolist’s spending on innovation. 42 A prerequisite for this positive feedback to arise is an implication of the Cournot Model. 43 According to Cournot, if two firms were to produce identical products in terms of being interchangeable, then the profits of each firm will depend not only on its own output quantity, but also on the quantity of the other firm. The profit/quantity relationship clearly characterizes the interdependence of an oligopoly in that the more one competitor produces, the lower is the market price for the good, causing the revenue and therefore the profit of the other firm to fall. 44 Therefore, the undertakings will follow the most profit-maximizing output given the respective output of their competitors.

In the case of complementary products developed by technology consortia the reverse reasoning on the basis of Cournot can be established. The oligopolists’ interdependent relationship is no longer affected by each other’s quantity decisions, but rather on their individual contribution to the innovative success. The more firms in a technology consortium can innovate, the more they can raise profits and the more they are willing to invest in further R&D. It follows that the quantity/price relationship in case of identical products and the innovation quota/profit relationship in case of complementary products are the crucial determinants that can be clearly discerned from Cournot.

Price-Makers. Given that the firms are large scale they know that their output decisions will affect market prices and are then not price

41 Shapiro, C. ‘Navigating the Patent Thicket: Cross Licenses, Patent Pools, and Standard- Setting’ in Jaffe, A., Lerner, J. and Stern, S. (eds.) Innovation Policy and the Economy (Cambridge, Massachusetts: MIT Press, 2001), at p. 23.

42 Baumol, W.J. (2002), at pp. 101 – 02.

43 Cournot has influenced many oligopoly theories and is most widely used still today. Developed by Augustin Cournot in 1838. Cournot, A. Research into the Mathematical Principles of the Theory of Wealth English Edition, translated by Bacon, N.T. (New York, New York: Kelley, 1960). See also Morasch, K. Industrie- und Wettbewerbspolitik – Zentralisierung oder Dezentralisierung (München, Germany: Oldenbourg, 2003), at pp. 18 - 27. Schmidt, I. Wettbewerbspolitik und Kartellrecht (Stuttgart, Germany: Lucius & Lucius, 2001), at p. 5 and Folz, C.H. Technologiegemeinschaften und Gruppenfreistellung (Köln, Germany: Carl Heymanns, 2002), at pp. 61 – 62.

44 Viscusi, W.K. et al. (2000), at p. 103.