Zurich Open Repository and Archive University of Zurich Main Library Strickhofstrasse 39 CH-8057 Zurich www.zora.uzh.ch Year: 2011 Intellectual property rights and market integration Heinemann, Andreas Posted at the Zurich Open Repository and Archive, University of Zurich ZORA URL: https://doi.org/10.5167/uzh-53574 Book Section Originally published at: Heinemann, Andreas (2011). Intellectual property rights and market integration. In: Anderman, Steven; Ezrachi, Ariel. Intellectual property and competition law: new frontiers. Oxford, United Kingdom: Oxford University Press, 303-322. (p. 303) 13 Intellectual Property Rights and Market Integration Andreas Heinemann Introduction Since the Spaak Report of 1956,1 market integration has always been an important goal of European competition law. The abolition of frontiers for the free movement of goods and services should not be contradicted by private restraints. Intellectual Property Rights (IPRs) have played a major role in this respect—parallel imports of IP-protected goods must be possible in the internal market. Therefore, private practices, based on IPRs, tending to prevent cross-border trade have to undergo antitrust scrutiny. General principles have been laid down in European regulation. Others have been developed by the European Courts. Recently, a more cautious approach may be identified when it comes to the application of competition law to market partitioning practices. On the one hand, there is the more lenient perception of vertical restraints based on the effects-oriented analysis of the Chicago School which, to a certain extent, has found its way into European competition law via the more economic approach. On the other hand, there are sector-specific arguments militating for a larger margin in favour of innovative industries like for example the pharmaceutical sector. In addition, there have been issues of a more general or technical nature, for example regarding the concept of ‘agreement’ in the sense of Article 101 TFEU. The relevant case law is of great importance when it comes to the leeway undertakings enjoy in defining their distribution strategies. 1 Report of the Heads of Delegation to the Ministers of Foreign Affairs (‘The Brussels Report on the General Common Market’) (Brussels 1956) Introduction, Section II B. The French version of the Spaak Report is available at: http://www.unizar.es/euroconstitucion/library/historic%20documents/Rome/preparation/Spaak%20report%20fr.pdf (accessed 15 September 2010). The following analysis starts by sketching the traditional approach of European competition law to market partitioning practices (A). Next, the recent developments are (p. 304) described which put emphasis on the limits of competition law application in this field (B). This directly leads to the question what impact the specific features of innovative industries should have on the application of competition law (C). From this discussion, consequences will be drawn for the status which the aspect of market integration should enjoy in European competition law. A. The Traditional Approach to Market Partitioning Practices The triangle of integration, competition law and IP protection The establishment of an internal market is at the heart of the European Union, Article 3 (3) TFEU. Even if the Lisbon Treaty has considerably broadened the foundations and the goals of European unification, the setting up of an area without internal frontiers in which the free movement of goods, persons, services, and capital is ensured (Article 26 TFEU) stays a cornerstone of European integration. As regards competition policy, the reference to a system of competition which is not distorted (ex-Article 3(1)(g) EC) has been removed from the Treaties and shifted to a Protocol.2 However, as Protocols form an integral part of the Treaties (Article 51 TEU), it is expected that this modification will not have substantive effects on the status of European competition policy. 3 Therefore, competition policy will keep its important function in the context of the internal market. 4 2 Protocol No 27 on the Internal Market and Competition. The Protocol clarifies that also in future it will be possible to make use of the flexibility clause in Article 352 TFEU in the field of competition law (as it was done, e.g. in the case of the EC Merger Regulation). 3 However, this optimistic view cannot be taken for granted. Ex-Article 3(1)(g) EC has been used in a considerable number of cases in order to strengthen pro-competitive arguments. In Eco Swiss, e.g. the ECJ based the ordre public character of the competition rules on this article, Case C-126/97 Eco Swiss China Time Ltd v Benetton International NV [1999] ECR I-3055 para 36. 4 Article 3 (1)(b) TFEU attributes to the Union the exclusive competence to establish the competition rules necessary for the functioning of the internal market. One could also ask the question if the reference to a ‘social market economy’ in Article 3(3) TEU strengthens the weight of competition policy. This concept was coined in Germany by Alfred Müller-Armack and Ludwig Erhard and is based on the ordoliberalism of the Freiburg School which gave highest Finally, the role of IPRs has been strengthened by the Lisbon Treaty—whereas in the past, on the level of primary law, intellectual property was only mentioned as a justification for restrictions of the free movement of goods, Article 118 TFEU now provides for an explicit legal basis for the creation of uniform European IPRs, and Article 207 TFEU adds IP protection to the catalogue of EU competences in the field of the common commercial policy.5 (p. 305) Exhaustion as a means to extenuate tensions between IPRs and market integration IPRs normally confer exclusive rights including an exclusive import and export right. This may cause a conflict with the free movement rules. In Europe, this conflict has been resolved by the concept of EU-wide (respectively EEA-wide) exhaustion. Once a protected product has been put in circulation somewhere in the EU (or in the EEA) by the IP owner or with his consent, this product is allowed to move untroubled in the EU (and the EEA). The IP owner cannot prevent further transactions; her right of distribution is exhausted. The principle of exhaustion is derived from Articles 34, 36 TFEU.6 It has been extended to the EEA by EEA-Protocol 28 on Intellectual Property. With respect to some IPRs, the principle of exhaustion has been explicitly included into secondary law.7 priority to competition policy. See DJ Gerber, Law and Competition in Twentieth Century Europe: Protecting Prometheus (Oxford: Clarendon Press, 1998) 259–61. 5 Moreover, Article 262 TFEU jurisdiction may be conferred on the ECJ in order to judge on disputes relating to uniform European IPRs. 6 Groundbreaking judgments are Case 78/70 Deutsche Grammophon/Metro [1971] ECR 487; Case 15/74 Centrafarm/Sterling Drug [1974] ECR 1147. 7 See, eg, Article 4 (2) of the Directive 2009/24/EC of the European Parliament and of the Council on the Legal Protection of Computer Programs [2009] OJ L111/16: ‘The first sale in the Community of a copy of a program by the rightholder or with his consent shall exhaust the distribution right within the Community of that copy, with the exception of the right to control further rental of the program or a copy thereof.’ This rather simple starting point has been elaborated in subsequent case law.8 For our context, the following specifications are of particular importance—according to the ECJ, Member States are no longer free to opt for international exhaustion.9 There is no consent of the right holder and hence no exhaustion if the right holder was under a legal obligation to market the product in question (including mandatory licensing). On the other hand, mere ethical obligations are not sufficient to trigger exhaustion. Nor is the existence of price control, eg, in the field of pharmaceutical products, relevant to exclude the exhaustion effect.10 Exhaustion affects the distribution right (and rights directly related hereto, like, e.g. the right to use a trade mark for (p. 306) advertising purposes11), but not the reproduction right. Exhaustion applies to IP-protected goods, but (normally) not to services.12 The case law is not restricted to the fundamental principles of exhaustion, but also deals with specific details. A famous example is the repackaging cases in the pharmaceutical sector. The Court has set up check-lists which have to be respected if a firm wants to re-box, re-label or even re-brand pharmaceuticals in order to make them compatible with the legal requirements in the import state.13 8 For a detailed overview see S Enchelmaier, Europäisches Wirtschaftsrecht (2005) paras 72–82; C Stothers, Parallel Trade in Europe. Intellectual Property, Competition and Regulatory Law (Oxford: Hart Publishing, 2007) 40 et seq. 9 Case C-355/96 Silhouette International Schmied v Hartlauer Handelsgesellschaft mbH [1998] ECR I-4799 paras 22– 7. The Court said so with respect to the Trademark Directive (Council Directive 89/1004/EEC to Approximate the Laws of the Member States Relating to Trade Marks [1989] OJ L40/1). In Laserdisken (Case C-479/04 Laserdisken v Kulturministeriet [2006] I-8089 paras 19–27), the Court came to the same result in the field of copyright relying on the InfoSoc Directive (Directive 2001/29/EC of the European Parliament and the Council on the Harmonisation of Certain Aspects of Copyright and Related Rights in the Information Society [2001] OJ L167/10). It seems most probable that the Court would apply the same principle to all IPRs. For a broader perspective see T Cottier, ‘The Exhaustion of Intellectual Property Rights—A Fresh Look’ [2008] Intl Rev Intellectual Property and Competition L 755–7. 10 Joint Cases C-267/95 and C-268/95 Merck/Primecrown [1996] ECR I-6285 (hereinafter ‘Merck/Primecrown’). 11 Case C-337/95 Parfums Christian Dior/Evora [1997] ECR I-6013. An exception applies if the use of the trade mark in the reseller’s advertising seriously damages the reputation of the trade mark, ibid para 46. For a critique of this exception see H Ullrich, ‘Gewerblicher Rechtsschutz und Urheberrecht im Gemeinsamen Markt’ in U Immenga and E-J Mestmäcker (eds), Kommentar zum Europäischen Kartellrecht, Vol 1(2) (4th edn, 2007) para 91. 12 Case 62/79 Coditel/Ciné Vog Films [1980] ECR 881. 13 Case C-348/04 Boehringer Ingelheim [2007] ECR I-3391. Even though the doctrine of exhaustion is now in a very advanced stage, the development is not yet finished. The ECJ is continuously confronted with new questions as are the national courts. As firms have a natural tendency to test the limits of the exhaustion doctrine, national courts have to deal with these questions in numerous cases including the consequences in other fields of law, like unfair competition law. They do not always succeed in striking an appropriate balance. 14 The role of competition law with respect to the principle of exhaustion In the traditional view, one of the functions of European competition law is to protect free movement of goods and services against private restraints. Therefore, private agreements or unilateral conduct providing for territorial restrictions have to be examined as to their compatibility with the competition rules. Article 101 (1)(b)and (c) TFEU expressly mention agreements which limit, control or share markets. Firms are not allowed to circumvent the principle of exhaustion by means of agreements or abusive behaviour.15 An example is the Nintendo case—prices for consoles and game cartridges were very uneven in the European Union. In the UK, they were situated 65 per cent below the price level in Germany and the Netherlands. Nintendo agreed with seven European distributors upon preventing parallel imports. Distributors which were identified as the source of parallel imports were sanctioned by the reduction or even the refusal to supply furthermore. The European Commission fined all parties involved, not only the producer but also the distributors. The Commission did not accept the argument that the producer had exerted considerable pressure on the distributors. They should have come to the Commission in the first place. 16 (p. 307) The CFI 14 For Germany see A Heinemann, ‘Gebietsschutz im Binnenmarkt—Bewährungsprobe für ein prekäres Gleichgewicht’ in A Ohly et al. (eds), Festschrift für Gerhard Schricker (2005) 53, 56–9. 15 B Conde Gallego, Handelsbezogene Aspekte des Lizenzkartellrechts (2003) 225, considers the maintenance of parallel imports the main task of international IP antitrust law. 16 Nintendo Distribution [2003] OJ L255/33. reduced the fines of some parties for reasons of equal treatment, and upheld the decision as for the rest.17 The Nintendo case confirms the consistent practice of the European courts (beginning with Consten and Grundig18) according to which one of the tasks of competition law is to guarantee parallel imports within the Community.19 According to the case law, absolute territorial protection is anti-competitive whereas a certain territorial protection is compatible with Article 101 TFEU.20 The details were laid down in block exemption regulations. In vertical distribution agreements, a supplier may allocate exclusive territories to his buyers. The block exemption regulation on vertical agreements21 contains specific rules on the legality of such restrictions. According to the regulation, the supplier may agree with the buyer on a restriction of active sales into the exclusive territories of other enterprises (if the market shares of the parties do not exceed 30 per cent). However, passive sales, i.e. sales in response to unsolicited orders, must not be excluded.22 These rules apply to simple distribution agreements in which IPRs only play a minor role. To certain licensing contracts, the block exemption regulation on technology transfer agreements23 is applicable which—in the 17 Cases T-12/03 Itochu/Commission [2009] ECR II-883, T-13/03 Nintendo and Nintendo Europe/Commission [2009] ECR II-947, Case T-18/03 CD-Contact Data/Commission [2009] ECR II-1021. In the latter case, an appeal to the ECJ is pending, see case number C-260/09. 18 Joint Cases 56 and 58/64 Consten and Grundig v Commission [1966] ECR 321. 19 Other recent cases are Case T-67/01 JCB Service/Commission [2004] ECR II-49, confirmed in the aspects relevant here by Case C-167/04 P JCB Service/Commission [2006] ECR I-8935; European Commission, Souris/Topps, 26 May 2004, COMP/C-3/37.980, [2006] OJ L353/5. For an overview on the illegality of direct and indirect export bans see R Whish, Competition Law (6th edn, Oxford: OUP, 2009) 622–26. 20 SD Anderman, EC Competition Law and Intellectual Property Rights: the Regulation of Innovation (Oxford : Clarendon Press, 1998) 90 et seq. 21 Commission Regulation (EU) 330/2010 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of vertical agreements and concerted practices [2010] OJ L102/1. 22 For a definition of active and passive sales see Commission Guidelines on Vertical Restraints [2010] OJ C130/1 para 51. 23 Commission Regulation (EC) No 772/2004 on the application of Article 81 (3) of the Treaty [now Article 101(3) TFEU] to categories of technology transfer agreements [2004] OJ L123/11. For a comprehensive analysis see SD interest of IP protection—allows larger restrictions of competition. If the parties of a licensing contract are not competitors (and if their market shares do not exceed 30 per cent), even passive sales into territories allocated to other licensees may be excluded for a period of two years. This only applies to direct sales by the licensee. Once the licensed product has been put on the market, parallel imports must be admitted. (p. 308) B. Limits of the Traditional Approach For a long time, European competition law has been interpreted so as to complement the rules on exhaustion and to safeguard parallel imports. Recent developments have cast doubt on whether this tradition will be followed in the future. Relevant cases involve the question under which conditions of private behaviour may be qualified as an agreement or a concerted practice within the meaning of Article 101(1) TFEU. Another problem lies in the requirements concerning the right holder’s consent. Finally, the most fundamental question has been raised by economic theory—if vertical restrictions have to be dealt with in a more cautious way, do we have to say farewell to the close link of exhaustion and competition law which has been observed so far? The concept of ‘agreement’ and ‘concerted practices’ in the context of marketpartitioning strategies Article 101 TFEU only applies to agreements, decisions of associations of undertakings and concerted practices. In case of unilateral conduct, the control is effected by Article 102 TFEU which presupposes a dominant position. Therefore, the impact of competition law is heavily influenced by the concept of agreement and concerted practices. If these notions are defined narrowly, many cases will be subject to a mere control of abuse. The case law is ambivalent. On the one hand, the European courts give a broad definition of concerted practices. In T-Mobile Netherlands, the ECJ has come to the result that a single meeting between companies may Anderman and J Kallaugher, Technology Transfer and the New EU Competition Rules: Intellectual Property Licensing after Modernisation (Oxford: OUP, 2006). constitute a concerted practice.24 This finding is based on the consistent definition of a concerted practice as a, -‘. . . form of coordination between undertakings by which, without it having been taken to the stage where an agreement properly so-called has been concluded, practical cooperation between them is knowingly substituted for the risks of competition’.25 On the other hand, when applying the abstract definition to specific cases, the European courts are sometimes very restrictive. One example is the Adalat case concerning a pharmaceutical product used to treat cardiovascular diseases. The German company Bayer limited the supply of its product Adalat to the Spanish and French wholesalers in order to prevent parallel imports to the UK where the price for Adalat was much higher. The Commission came to the conclusion that the quantitative limitation was not unilateral conduct, but followed from an agreement between supplier and buyer.26 The CFI and the ECJ did not share this view and annulled the Commission’s decision. The mere coexistence of a sales contract and a unilaterally (p. 309) imposed limitation of supply does not, according to the courts, constitute a restrictive agreement in the sense of European competition law.27 This finding has to be situated into the context of other judgments trying to identify the thin line between, on one side, genuinely unilateral conduct and, on the other side, coordination even if it is disguised as unilateral behaviour.28 This distinction is essential, not only for competition law in general, but specifically for the topic of parallel imports since a narrow concept of coordination enlarges the leeway for the holders of IPRs to restrict parallel imports. In addition, according to the rules on the burden of proof, the authorities have to prove that there was a concurrence of wills. In our view, the delimitation has not always been convincing. In the Adalat case, eg, the European 24 Case C-08/08 T-Mobile Netherlands [2009] 5 CMLR 11 (judgment of the ECJ, 4 June 2009) para 26. 25 Ibid. 26 Adalat [1996] OJ L201/1. 27 Case T-41/96 Bayer/Commission [2000] ECR II-3383; Joined Cases C-2/01 P and C-3/01 P Bundesverband der Arzneimittel-Importeure [2004] ECR I-23. 28 Case C-74/04 P Commission/Volkswagen [2006] ECR I-6585; Case T-450/05 Automobiles Peugeot SA and Peugeot Nederland/Commission [2009] ECR II-2533. Commission found that Bayer limited supplies to a certain buyer when he had been identified as a source of parallel imports. Thus, the decrease in supply was a sanction aimed at eliminating parallel trading. The buyers reacted to this threat by stopping or at least hiding their parallel exports. This very evident connection of stimulus and response leads to a coordination of behaviour for which in European competition law the category of ‘concerted practices’ exists. Therefore, even if there is not an agreement between Bayer and its wholesalers, there might be a concerted practice which is equally caught by Article 101 TFEU. Unfortunately, the European courts did not examine the question if there is such a concerted practice. In our view, it should be sufficient for the application of the European cartel interdiction if the supplier, explicitly or implicitly, expresses certain behavioural expectations to the buyer, and if the buyer then complies or at least pretends to comply with the declared intentions of the supplier. Otherwise, it would be possible for the supplier to circumvent the cartel interdiction by taking measures which are unilateral only at first sight. It has to be added that a broader concept of the notion of ‘agreement’ and ‘concerted practice’ is not supposed to predetermine the outcome of substantive analysis. It simply allows behaviour to be brought into the scope of Article 101 TFEU which aims at the coordination between undertakings. This approach does not touch the question if the agreement or concerted practice has as its object or effect the restraint of competition, and if such a restraint could be justified under Article 101(3) TFEU. Consent of the right holder as a prerequisite for exhaustion Exhaustion only takes place if the product has been put on the market with the consent of the right holder. Thus, the notion of ‘consent’ is pivotal for exhaustion doctrine. Consent must have been expressed with respect to each individual product for which (p. 310) exhaustion is claimed. Therefore the fact, that a firm has put a certain product on the European market at all, does not justify parallel imports of identical or similar products which have been put on the market outside the EEA.29 Moreover, there is no consent simply because a product was marketed under the same trademark in another Member State if the identical trademarks belong to different and independent entities. If for example the holder of a trademark in EEA Member State A puts the trademarked 29 Case C-173/98 Sebago and Maison Dubois [1999] ECR I-4103 para 19. good on the market of state A, the holder of the identical trademark in state B can still prevent the import of that good into state B even if it was him who assigned the trademark for state A to the other firm, as long as there is no economic link between the two firms. Hence, the voluntary transfer of IPRs to another firm does not imply the consent with putting protected goods on the market. 30 Even if this result is convincing, it is obvious that the separate assignment of IPRs with respect to single Member States may be used to partition the internal market. From the very beginning, ie the Consten and Grundig case,31 the ECJ has clarified that assignment of IPRs must not be used in an anti-competitive manner. In Ideal Standard, the Court has explicitly referred to the cartel interdiction, - ‘It should be added that, where undertakings independent of each other make trade-mark assignments following a market-sharing agreement, the prohibition of anti-competitive agreements under Article 85 [now Article 101 TFEU] applies and assignments which give effect to that agreement are consequently void.’32 Hence, the entire legal and economic context has to be explored in order to assess the conformity of an assignment with Article 101 TFEU. The relevant question will be if the assignment of IPRs made for part only of the European Union aspires to fully exploit the economic potential of the rights in question (then it is perfectly legal), or if it is part of a plan to allocate markets in an anticompetitive manner.33 If the latter is the case, the assignment is void (Article 101(2) TFEU) and the 30 Case C-9/93 IHT Internationale Heiztechnik/Ideal Standard [1994] ECR I-2789 (hereinafter ‘Ideal Standard’), especially para 43. According to the court, ‘. . . the owner of the right in the importing State must, directly or indirectly, be able to determine the products to which the trade mark may be affixed in the exporting State and to control their quality. That power is lost if, by assignment, control over the trade mark is surrendered to a third party having no economic link with the assignor.’ For the ECJ to get to this result the long way, see WR Cornish, Intellectual Property: Patents, Copyright, Trade marks and Allied Rights (3rd edn, London: Sweet & Maxwell, 1996) [18–06] calling this development the ‘common origin adventure’. 31 Consten and Grundig (n 18 above). 32 Ideal Standard (n 30 above) para 59. 33 The Ideal Standard case concerns trademarks. The reasoning can be extended to other IPRs, eg, to patents, compare J Straus, ‘Gebietsbeschränkte Übertragung von Patenten und Patentportfolios und Erschöpfung des Patentrechts in der Europäischen Gemeinschaft’ in RM Hilty, J Drexl and W Nordemann (eds), Festschrift für Ulrich Loewenheim (2009) 309–21. assignee has no title to prevent imports into the state for which the right in question ineffectively was (p. 311) transferred to him.34 The same is true for strategies where independent firms file the same IPR in various Member States in order to divide up the internal market among themselves.35 The form of consent According to general legal principles, consent may be declared in an express or implied way. In Davidoff and Levi Strauss the ECJ held that ‘. . . in view of its serious effect in extinguishing the exclusive rights of the proprietors. . ., consent must be so expressed that an intention to renounce those rights is unequivocally demonstrated’.36 Normally such intention will be shown by an express statement. But it is possible—according to the Court—that consent is declared in an implied manner, ie it may deduced from facts and circumstances in the context of placing the protected goods on the market.37 However, consent cannot be inferred from the mere silence of the right holder. Therefore, there is no obligation of the IPR owner to attach an EEA import ban to her products when putting them on markets outside the EEA. Likewise, she is not obliged to impose contractual reservations preventing the import of the goods in question into the EEA.38 When is a protected good put on the market? Although the abovementioned principles seem clear, the goal of market integration depends to a great extent on their substantiation. When is an IP-protected good put on the market? And what is necessary to affirm the consent of the right holder? As regards the first question, a product is put on the market if there is a transaction allowing the right holder to realize the full economic value of his 34 Discussion of this and of similar problems by O-A Rognstad, ‘The exhaustion/competition interface in EC law—is there room for a holistic approach?’ in J Drexl (ed), Research Handbook on Intellectual Property and Competition Law (Cheltenham: Edward Elgar Publishing, 2008) 427, 441–3. 35 Case 144/81 Keurkoop/Nancy Kean Gifts [1982] ECR 2853 para 28. 36 Case C-414/99 Zino Davidoff and Levi Strauss [2001] ECR I-8691 para 45. 37 No matter if the products in question are put on the market outside or inside the EEA, Case C-324/08 Makro Zelfbedieningsgroothandel and others (ECJ judgment dated 15 October 2009) <http://eurlex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:62008J0324:EN:HTML> accessed 15 September 2010 paras 28–34. 38 Case C-414/99 Zino Davidoff and Levi Strauss [2001] ECR I-8691 paras 53–60. IPR.39 This will normally be a sale, but exhaustion should not be restricted to this kind of transaction from the outset. If a protected good is transferred for use only, exhaustion will take place if the duration of the contract covers the economic lifespan of the product in question.40 However, for exhaustion to take place, it is not sufficient to import the product, to stock it or to offer it with a view to selling it. The product must have actually been transferred, and exhaustion only takes place after the transfer has (p. 312) been completed.41 If the product is not transferred to a third party, but only moves within a group of companies which are under common control, IPRs are not consumed.42 Hence, in group of companies, the time of exhaustion is postponed—only when the protected good leaves the group, ie when it is sold by an affiliated company to an actor outside the group, exhaustion takes place.43 There are proposals in the literature to retard the time of exhaustion even further. Exhaustion shall not take place after a transaction between the members of a distribution system even if they are not part of the same group of companies. In this view, exclusive rights only exhaust when the product leaves the distribution system.44 The condition is added that the distribution system has to be compatible with competition law.45 In the same sense, Rognstad opts for a ‘holistic approach’—he criticizes the traditional ‘cumulation theory’ according to which compatibility with the exhaustion principle and the competition rules has to be examined in a separate way. Instead, he pleads for removing or at least reducing inconsistencies between the two 39 Compare the definition in Case C-16/03 Peak Holding [2004] ECR I-11313 para 40. 40 Otherwise it would be easy to circumvent the effect of exhaustion, compare H Ullrich (n 11 above) para 88 fn 615. 41 H Ullrich (n 11 above) paras 38–44 with respect to the Trademark Directive. 42 The concept of putting on the market is also relevant in transit cases. IPRs can only be infringed by goods in transit if they are put on the market of the state of transit which is not the case if the product is placed under the external transit procedure and if it is not subject to the act of a third party, Case C-281/05 Montex Holdings [2006] ECR I-10881. 43 For qualifications to this general statement see A Lessmann, ‘Erschöpfung von Patentrechten bei Konzernvertrieb’ [2000] Gewerblicher Rechtsschutz und Urheberrecht 741–50; E-J Mestmäcker and H Schweitzer, Europäisches Wettbewerbsrecht (2nd edn, 2004) § 27 paras 53–5. 44 K-H Fezer, ‘Wettbewerbsrechtlicher und markenrechtlicher Bestandsschutz funktionsfähiger Distributionssysteme selektiven Vertriebs vor Aussenseiterwettbewerb’ [1999] Gewerblicher Rechtsschutz und Urheberrecht 99, 105. 45 A Metzger, ‘Erschöpfung des urheberrechtlichen Verbreitungsrechts bei vertikalen Vertriebsbindungen’ [2001] Gewerblicher Rechtsschutz und Urheberrecht 210, 213–14. sets of rules.46 He adds the question if right holders should be entitled to assert their IPRs against imports which are in breach of territorial restrictions compatible with competition law.47 Qualified consent? This leads to the general question if there should be a refinement of the exhaustion principle. In this sense, simple consent would not be sufficient for exhaustion to take place. A qualified consent would be necessary requiring that the protected product is put on the market in conformity with all or at least the most important contractual obligations as far as they are compatible with competition law. This is of importance not only for exclusive and selective distribution but also for licensing agreements. A violation of the contract between producer and wholesaler or between licensor and licensee would have the consequence that there is no consent of the right holder to putting the product on the market. Such a violation could refer for example to territorial limitations, quantitative restrictions or quality specifications. Hence, if a (p. 313) licensee sells the protected product outside the appointed territory, if he produces more than agreed upon, or if he does not meet the quality standards stipulated in the contract, exhaustion would not take place (provided that the contractual commitments do not infringe competition law).48 The proprietor then would be able—in addition to his contractual and property claims against the contract partner—to assert his property rights against any party in the distribution chain. The ECJ did not accept this narrow concept of exhaustion. In the Copad case, the Court had to answer the question—with respect to trademark law and based on the Trademark Directive—if the disregard of a provision in a licensing agreement prohibiting the sale of the licensed products to discount stores would preclude consent thus preventing exhaustion.49 It held that a licensing 46 O-A Rognstad (n 34 above), 432–43. The proposals are presented in a cautious way; see p 437—it ‘. . . is at least worth considering abandoning the cumulation theory in favour of a more holistic or harmonized approach.’ 47 O-A Rognstad (n 34 above) 439. 48 In this sense Dior and the French Government in Case C-59/08 Copad SA v Christian Dior couture SA [2009] ECR I- 3421 para 39 (hereinafter ‘Copad’) ‘. . . that any contravention of the licence agreement by the licensee precludes exhaustion’. 49 The licensing agreement concerns luxury goods. The interdiction of selling to discount stores was supposed to maintain the reputation and the prestige of the trade mark. agreement does not constitute the absolute and unconditional consent of the proprietor. But only the licensee’s contravention of clauses triggering trademark-based claims (as opposed to mere contractual claims) would hinder exhaustion.50 The starting point of the Court deserves approval. There can be no exhaustion if the licensee exceeds the limits of his competences as regards for example time or scope of his license. In these cases, the proprietor did not have the possibility to reap the benefits of her IPRs. In trademark law, the restrictions directly based on the exclusive right have been listed by the legislature. 51 Hence, as regards the examples mentioned above, trade mark rights erga omnes would exist if restrictions of the scope of the licence or territorial restrictions are violated, but not in the case of disregard of quantitative restrictions (which are not mentioned in the list of Article 8(2) Trademark Directive).52 For other IPRs, the competences accruing directly from the exclusive right and having erga omnes effect have to be determined by way of interpretation based on the protective goal of the IPR in question.53 Likewise, it seems convincing not to reject exhaustion in case of disregard of mere contractual obligations. One of the goals of the exhaustion principle is legal certainty. In commerce, purchasers should be allowed to suppose that IP-protected goods which have been sold by the proprietor or an authorized person now can move on without interference of the right holder. Of course, this certainty (p. 314) is not absolute.54 It may always be the case that for example a licensee did not respect obligations covered by the exclusive right so that exhaustion did not take place. But this is not an argument to completely undermine legal certainty by attributing claims ad rem where mere contractual obligations were violated. For the sake of legal certainty, it is important to reduce 50 Copad (n 48 above) paras 47–51. 51 Article 8 (2) of the Trademark Directive. This list is exhaustive, Copad (n 48 above) para 20. See also Article 22 (2) of Council Regulation (EC) 40/94 on the Community trade mark [1994] OJ L11/1. 52 In this sense H Ullrich (n 11 above) paras 78–9. Compare opinion of AG J Kokott (Case C-59/08 Copad [2009] ECR I-3421 para 20) who leaves the question of quantitative restrictions open. 53 An analysis differentiating between different types of IP is made by H Ullrich (n 11 above) paras 78–9. 54 O-A Rognstad (n 34 above) 436. uncertainty as much as possible by defining a simple concept of consent. 55 The exhaustion principle tends to reconcile IP protection and the free movement of goods. The proprietor has the right to decide on the first sale of an IP-protected good. This balance should not be upset by a narrow concept of exhaustion giving the right holder also the power to decide on the second and third sale. It is the task of competition law to determine to what extent resale restrictions shall be legal. This reasoning may be underpinned by more fundamental reflections. It is true that the free movement rules and the competition rules have certain goals in common and complement each other. But the two sets of rules are situated on different levels: The free movement rules—in the context of IP protection—confine the legal power which may be attributed to right holders whereas competition law limits the use which may be made of these rights in the market.56 One of the major achievements of modern IP competition law is the emancipation of competition law from the scope of IPRs. IPRs are not immune against the application of competition law. The fact, that a certain behaviour is covered by an exclusive right (for example a refusal to license), does not mean that competition law is inapplicable.57 Neither is it possible to justify the extension of IPRs with the argument that underlying contracts do not violate competition law. It is not the task of IP protection, even less of competition law, to vest distribution systems with an effect going beyond the privity of contract. IPR owners have sufficient possibilities to obtain this result by limiting licences carefully. In our view, the ECJ has gone too far as regards the erga omnes effect of a breach of contract. In the Copad case, the Court held that the disregard by a licensee of a clause prohibiting sales of luxury products to discount stores constitutes a violation of the rights conferred by the trade mark if this damages the ‘aura of luxury’ of the products in question.58 Starting point is Article 8(2) of the Trademark Directive which allows the proprietor of a trade mark to invoke the rights conferred by 55 The importance of legal certainty in the context of property rights is underlined by S Enchelmaier, ‘Intellectual property, the internal market and competition law’ in J Drexl (ed), Research Handbook on Intellectual Property and Competition Law (Cheltenham: Edward Elgar Publishing, 2008) 405, 420–23. 56 H Ullrich (n 11 above) paras 105–9. 57 Therefore, in our view, the concept of ‘specific subject matter’ which limits the application of the free movement rules to IPRs, cannot be extended to the application of the competition rules to IPRs, see A Heinemann, Immaterialgüterschutz in der Wettbewerbsordnung (2002) 327–8. 58 Copad (n 48 above) paras 30–2. that trade mark (p. 315) against a licensee who contravenes any provision in his licensing contract with regard inter alia to the quality of the goods manufactured by the licensee. Thus, the Community legislature has, to a certain extent, put a value on a quality function of trademarks complementing the traditional origin function. It has to be stressed that therefore violations of quality specifications in licensing agreements not only trigger contractual and property claims against the licensee but also property claims against third parties since exhaustion is denied. The Copad decision goes one step further—even if the properties of a product perfectly comply with the specifications in the licensing agreement, exhaustion does not take place in case of disregard of selective distribution rules since, -‘. . . the quality of luxury goods. . . is not just the result of their material characteristics, but also of the allure and prestigious image which bestows on them an aura of luxury’.59 -As a result, the exclusive right is no longer limited to allow its proprietor to control the first marketing in the EEA, but serves to control further transactions and to stabilize a selective distribution system. This happens in an absolute way since outsiders not standing in a contractual relationship with the right holder can be sued for infringement of the trade mark.60 This does not seem compatible with the functions of trade mark protection even if the quality function is accepted. Distribution systems should be shaped by contracts inter partes within the limits of competition law. It is not the goal of IP protection to waterproof distribution systems. The term ‘quality’ in the meaning of Article 8(2) Trademark Directive should therefore be restricted to material properties. The challenge of modern economic theory to the assessment of territorial restrictions 59 Copad (n 48 above) para 24. 60 The case concerns a licensing agreement allowing the licensee to produce and to sell certain goods under the licensed trade mark. The situation would be different in a simple distribution agreement: The Court held that the sale by the proprietor of the trade mark automatically leads to exhaustion no matter if restrictions on the further marketing of the good have been stipulated between the seller and the purchaser, Case C-16/03 Peak Holding AB v Axolin-Elinor AB [2004] ECR I-11313 paras 50–6. The discussion on the prerequisites and the limits of exhaustion is embedded into the general debate on the appropriate treatment of vertical restraints in the light of modern economics. 61 The starting-point is well-known—under the influence of the Chicago School, the assessment of vertical restraints has undergone an important development towards a more lenient competition law treatment. Whereas initially scholars of the Chicago School postulated the per se legality of vertical restraints, today a more qualified position prevails. Vertical restraints may—according to the (p. 316) circumstances of a particular case—be beneficial or detrimental.62 In the context of market integration, it has been stated that strict rules against market segmentation disregard the free-rider problem and encourage vertical integration.63 Already the former block exemption regulation on vertical agreements recognized the efficiency-enhancing effects of vertical restraints and underlined the importance of market power, ie the degree of interbrand competition in this context.64 The new regulation has integrated even more recent economic thinking, but, at the same time, it has maintained the negative assessment of absolute territorial restrictions and resale price maintenance.65 Competition law has of course to be adapted to economic insights corroborated in theory and in practice. However, reforms should be proportional. The fact that in single cases a free-rider problem may occur, is not sufficient to introduce the per se legality of resale price maintenance or of absolute territorial protection. The current system is flexible enough to take into account efficiency gains via an individual assessment based on Article 101(3) TFEU. The new guidelines on vertical restraints explain in detail under which conditions even hardcore restrictions may be justified, for example if they are necessary for the introduction of a new product. 66 In our context, it is important to note that the goal of efficiency pursued by the more economic approach is not contradictory to 61 Compare the remarks of Metzger (n 45 above) 213–14, and Rognstad (n 34 above) 434, 438. 62 M Motta, Competition Policy—Theory and Practice (Cambridge: Cambridge University Press, 2004) Ch. 6. 63 G Monti, EC Competition Law (Cambridge: Cambridge University Press, 2007) 363–6. 64 Commission Regulation (EC) 2790/1999 on the application of Article 81 (3) of the Treaty to categories of vertical agreements and concerted practices [1999] OJ L336/21 recital 6–7. 65 Commission Regulation (EU) 330/2010 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of vertical agreements and concerted practices [2010] OJ L102/1. 66 Commission Guidelines on Vertical Restraints [2010] OJ C130/1 para 47. the goal of market integration. The rules on exhaustion which—in the field of IP protection—are a key component of market integration, make it possible that an internal market of considerable size is constituted which leads to huge efficiency gains. It is important to keep this context in mind when shaping the details of IP competition law. C. Special Rules in Favour of Innovative Industries? To what extent do the general rules apply to innovative industries, like for example the IT sector or the pharmaceutical industry? As regards the IP aspect, it has to be repeated that property rights need a precise definition which must not vary subject to market conditions. Moreover, the exhaustion principle has to be kept simple. Therefore, and in conformance with the case law of the ECJ, exhaustion takes place even if there is state influence into pricing as it is the case in many Member States (p. 317) with respect to pharmaceutical products.67 As regards competition law, aspects of innovation are relevant in all sub-fields. The following reflections focus on Article 101 and Article 102 TFEU.68 Restrictive agreements As regards restrictive agreements, the impact on innovation has become relevant in the context of quantitative restrictions and of ‘dual pricing’. As regards quantitative restrictions, the control of supply by fixing allotments for purchasers or by imposing output restrictions in licensing contracts may pursue the aim of eliminating parallel trade. By limiting supply, the other party may get just the number of products necessary to satisfy domestic demand. Thus, no quantities are left which could be exported. The same goal may be pursued by ‘dual pricing’—in this case, purchasers have 67 Merck/Primecrown (n 10 above). It is of course different if the legislature has opted for another solution. For the ‘Specific Mechanism’ with respect to the new Member States of Middle and Eastern Europe see C Heath, ‘Parallel Imports of Patented Pharmaceuticals from the New EU Accession States’ [2004] Intl Rev Intellectual Property and Competition L 776–87. See also Swiss patent law; since 2009, Switzerland has (unilaterally) introduced EEA-wide exhaustion into patent law, but provides for an exception in Article 9a (5) of the Federal Patent Law with respect to products whose price is imposed by the state. 68 For merger control see A Heinemann, ‘Intellectual property rights and merger control: How to secure incentives to innovate in the long run’ in J Drexl et al. (eds), Technology and Competition—–Contributions in Honour of Hanns Ullrich (Brussels: Larcier, 2009) 601–26 (hereinafter ‘Technology and Competition’). to pay an ‘export tax’, ie a higher price for goods going to be exported than for goods that are sold on the domestic market. Limitation of parallel trade as the object or effect of an agreement? Both strategies restrict intrabrand competition and hence fall under Article 101 (1) TFEU if the other conditions of this Article are fulfilled.69 Although this starting point is widely accepted, the CFI has created some confusion in the Spanish Glaxo case. The European Commission prohibited a system of dual pricing stating that the agreement entailed restrictions of competition ‘by object’, and that, according to settled case law, it was not necessary to take account of the actual effect of this agreement.70 The CFI disagreed. According to the Court, parallel trade is not to be protected ‘as such’, but only ‘. . . in so far as it gives final consumers the advantages of effective competition in terms of supply or price’.71 The Court denied a restriction ‘by object’ in this sense, but affirmed a restriction ‘by effect’ because of the cost savings of patients in the countries of destination.72 (p. 318) This approach is not convincing.73 If it were necessary to show the direct consequences for final consumers in order to prove the restrictive object of an agreement, even hardcore cartels would not have an anti-competitive object if the disadvantages are not passed on to consumers. The participation of consumers in the resulting benefit is expressly mentioned in Article 101(3) TFEU and does therefore not belong into the analysis of the anti-competitive object of the agreement. On appeal, the ECJ has held in this sense—according to the Court, 69 In the case of quantitative restrictions, the existence of an agreement or a concerted practice has to be shown, see part B above. If there is mere unilateral conduct, only Article 102 TFEU is applicable. 70 Glaxo Wellcome [2001] OJ L302/1 (hereinafter ‘Glaxo Wellcome’). 71 Case T-168/01 GlaxoSmithKline Services/Commission [2006] ECR II-2969 para 121 (hereinafter ‘Case T-168/01 GlaxoSmithKline Services’). 72 Ibid para 148 et seq, especially para 181 et seq. 73 See the critique of P Behrens, ‘Parallelhandelsbeschränkungen und Konsumentenwohlfahrt’ [2008] Zeitschrift für Wettbewerbsrecht 20, 37–9; A Heinemann, ‘Geistiges Eigentum im internationalen und regionalen Kartellrecht’ in A Ohly and D Klippel (eds), Geistiges Eigentum und Gemeinfreiheit (2007) 199, 212–4. As regards the general difficulty of determining the effects on consumer welfare see H Schweitzer, ‘The role of consumer welfare in EU competition law’ in Technology and Competition (n 68 above) 511 et seq. -‘. . . for a finding that an agreement has an anti-competitive object, it is not necessary that final consumers be deprived of the advantages of effective competition in terms of supply or price’.74 Thus, the case law holds true according to which agreements aimed at restricting parallel trade have as their object the restraint of competition. Dual pricing strategies belong to this category and therefore have a restriction of competition as their object. Therefore, a justification is necessary. Block exemption The justification may be either general in the form of a block exemption regulation or individual. As regards output restrictions in licensing agreements, the legal situation has considerably changed. Under the 1996 block exemption regulation on technology transfer agreements, quantitative restrictions in licensing contracts were blacklisted in order to avoid the same result as export bans.75 In the 2004 Regulation76 the treatment of quantitative restrictions in licensing contracts has been relaxed.77 Output restrictions are only blacklisted in certain situations.78 According to the European Commission, the pro-competitive aspects (dissemination of the technology in question) outweigh the anti-competitive risks (reduced intra-technology competition).79 However, in order to avoid the danger of partitioning national markets, the Commission has added some caveats in the Technology Transfer Guidelines—quantity (p. 319) limitations must not be used in order to restrict passive sales. An indication thereof would be the adjustment of quantities over time to cover only local demand.80 Individual exemption 74 Joined Cases C-501/06 P, C-513/06 P, C-515/06 P, C-519/06 P GlaxoSmithKline Services Unlimited et al ./Commission et al. (judgment of the ECJ 6 October 2009) [2010] 4 CMLR 2 para 63 (hereinafter ‘Case C-501/06 P GlaxoSmithKline Services Unlimited’). 75 Article 3 (5) of Commission Regulation (EC) No 240/96 on the application of Article 85 (3) of the Treaty to certain categories of technology transfer agreements [1996] OJ L31/2. 76 Regulation No 772/2004 (n 23 above). 77 However, it has to be repeated that the block exemption only applies to individual market shares up to 30 per cent (in case of competitors up to a combined market share of 20 per cent), see Article 3 Regulation 772/2004. 78 That is reciprocal output restrictions between competitors, see Article 4 (1) (b) of Regulation 772/2004. 79 European Commission, Technology Transfer Guidelines [2004] OJ C101/2 para 175 et seq. 80 Ibid, para 98. Can market-partitioning practices be justified under Article 101(3) TFEU, and if so, under which conditions? The Commission and the European Courts had to deal with these questions in the Spanish Glaxo case. The Commission came to the result that the dual pricing scheme did not meet the conditions for an exemption. According to the Commission, Glaxo did not prove a causal link between parallel trade and R&D investments, and could therefore not show that consumers will receive a fair share of the resulting benefit.81 According to the CFI, the Commission did not sufficiently examine the question if dual pricing improves innovation.82 The ECJ shared this view.83 As a result, the Commission has to re-examine the application of Article 101(3) TFEU. The Court decisions bring up at least two questions. As regards procedure, it has to be asked, how detailed the Commission’s analysis has to be when assessing the conditions of Article 101(3) TFEU. And as regards substance, can private restraints aiming at hampering parallel trade be justified by effects on innovation? As regards the first question, the Commission has analysed the determinants of R&D expenditure. It came to the conclusion that there is no evidence that parallel trade has led to cuts or has prevented the growth of the R&D budget.84 The Courts criticized that the Commission did not analyse sufficiently the economic studies submitted, and that the Commission’s conclusions are lapidary.85 This stands in contrast to the generally accepted starting point that the European Courts only have a limited power of review of complex economic assessments made by the Commission.86 In our view, the requirements of the European courts in this case go beyond a confined control. An obligation to respond to all details of the studies submitted oversteps the mark and opens the way to defence strategies trying to stall proceedings by the quantity of evidence submitted. This point of view may get support from the answer to the second question. The starting point is quite simple: Of course, restrictive agreements may be justified under Article 101(3) TFEU by 81 Glaxo Wellcome (n 70 above) para 153 et seq. 82 Case T-168/01 GlaxoSmithKline Services (n 71 above). 83 Case C-501/06 P GlaxoSmithKline Services Unlimited (n 74 above). 84 Glaxo Wellcome (n 70 above) paras 155–69. 85 See for example paras275–9 in Case T-168/01 GlaxoSmithKline Services (n 71 above). 86 Case C-501/06 P GlaxoSmithKline Services Unlimited (n 74 above) para 146. effects on innovation since the promotion of technical or economic progress is expressly mentioned in this article. However, all conditions of this paragraph have to be met. The contribution of the restrictive agreement to the promotion of innovation has to be shown, and there has to be a direct link (p. 320) between both. It is not sufficient to claim a general context between increased profits and stronger R&D activities. Such a general reflection would be applicable to hardcore price cartels as well.87 Therefore, it is not sufficient to say that without parallel imports the undertaking in question would earn higher profits which would be invested into R&D. This leads to the fundamental question on which legal level the innovation aspect of parallel trade should be taken into account. In fact, the rules on exhaustion have multiple effects, not only on the pharmaceutical companies but also on distributors, the national or private health insurances and on consumers/patients. The long term effects have to be taken into account as well as the opposing interests of consumers in countries with high or low prices. 88 In our view, these questions have to be solved on the level of the exhaustion rules themselves. If a comprehensive economic analysis had the result that—because of the beneficial effects of regional price discrimination and in the interest of innovation—parallel trade should be prohibited, the principle of national exhaustion should be introduced.89 If, on the other hand, the answer is, with good reason, in favour of parallel trade (including the pharmaceutical sector), the rules on regional exhaustion should be upheld. Then it is consequent for competition law to support this fundamental decision by denying an exemption to agreements which try to suppress the viability of parallel imports. The IP holder still has the possibility to minimize the effects of parallel trade by unilateral measures. In this field, limitations are only set by Article 102 TFEU which presupposes market dominance and which does not prohibit any restraint of competition but only abuse. The abuse of a dominant position 87 E-J Mestmäcker and H Schweitzer (n 43 above) § 13 para 48. For further development of this argument see J Drexl, ‘Healing with bananas—How should Community competition law deal with restraints on parallel trade in pharmaceuticals?’ in Technology and Competition (n 69 above) 571, 586–91. 88 For an economic analysis of parallel trade see P Behrens (n 73 above) 20 et seq; J Drexl (n 87 above) 575–86. 89 Compare J Drexl (n 87 above) 575 who raises the question if the outcome of economic competition law analysis should not influence the design of the exhaustion doctrine as well. Every undertaking is free in formulating its distribution system. It can sell its products via branches, subsidiaries or agents, or over independent wholesalers and retailers. Parallel trade occurs in the latter case. The producer may prevent or at least diminish parallel trade by imposing quota based on the demand in the respective country. The question arises if such allocations, practised by an undertaking having a dominant position, constitute an abuse if their aim is to hamper parallel trade. The ECJ has dealt with this question in the Greek Glaxo case. The Court weighs the importance of parallel trade against the existence of price and supply regulation in the pharmaceutical sector. The result is a compromise—price regulation does not justify the complete refusal to supply wholesalers. But it is not abusive to refuse (p. 321) orders which go beyond the ‘ordinary’. It is for the national court to determine whether orders are ordinary in the light of the size of the market and of the previous business relations with the wholesalers concerned. 90 As regards the effect on innovation, the Court does not see the necessity to examine the argument that the limitation of parallel exports might be necessary in order to avoid the risk of a reduction of R&D expenditure.91 The judgment is ambiguous—on the one hand, the importance of parallel trade is underlined.92 On the other hand, the IP holder has the right to defend his own commercial interests, as long as the measures taken are reasonable and proportionate.93 The result is a limited obligation to contract— the orders which have to be met are capped at the level of domestic demand including a certain safety margin. In the first place, this favours dealers which have been active on the market and who see their supplies cut. Orders, which have been met in the past, will probably qualify as ‘ordinary’ and have to be met in the future, too. However, this should not exclude newcomers. If new firms are interested in being supplied, the dominant firm is obliged to reallocate the ‘ordinary’ quantity for 90 Joined Cases C-468/06 to C-478/06 Sot. Lélos kai Sia et al v GlaxoSmithKline AEVE Farmakeftikon Proïonton [2008] ECR I-7139 paras 58–77 (hereinafter ‘Joined Cases C-468/06 et al Sot. Lélos kai Sia’). See the in-depth analysis of the GlaxoSmithKline saga by J Drexl (n 87 above) who criticizes the Court for the lack of economic reasoning. 91 Joined Cases C-468/06 et al Sot. Lélos kai Sia (n 90 above) para 70. 92 Albeit in a half-hearted manner: ‘Indeed, parallel imports enjoy a certain amount of protection in Community law because they encourage trade and help reinforce competition’ (Joined Cases C-468/06 et al Sot. Lélos kai Sia (n 90 above) para 37). 93 Joined Cases C-468/06 et al Sot. Lélos kai Sia (n 90 above) para 69. the respective country on all interested parties.94 Against the backdrop of Article 102(b) TFEU (limitation of production, supplies or technical development), the result of the Glaxo case seems convincing—a dominant undertaking has to satisfy the national demand in a given Member State. It would have been enlightening, though, if the Court had stated more precisely its view on the relationship between parallel trade and the incentives to innovate. Conclusion: The Status of Market Integration in European Competition Law There are some indications that the importance of market integration is decreasing. Marketpartitioning strategies have been qualified as unilateral so that the cartel interdiction is not applicable.95 The CFI has held, that the effect on final consumers has to be shown in order to establish the anti-competitive object of an agreement.96 (p. 322) Exhaustion has been denied in the case of luxury products sold to discounters in contradiction to the terms of a licensing agreement.97 Generally, modern economic thinking has underlined the beneficial effects of territorial segmentation.98 These developments have provoked contradiction, though—regardless of whether a more economic approach is accepted or not, a microeconomic ‘absolutization’ of the (consumer) welfare criterion loses sight of the institutional framework which alone guarantees the maintenance of a competitive system in the long run.99 In recent case law, the ECJ has clearly rejected positions 94 W Wurmnest, ‘Einseitige Beschränkungen des Parallelhandels durch marktbeherrschende Arzneimittelhersteller im Binnenmarkt’ [2009] Rabel J Comparative and Intl Private L 962, 983. The ECJ did not have to rule on this question. 95 See text to n 27 above. 96 Case T-168/01 GlaxoSmithKline Services (n 71 above). 97 See text to n 58 above. 98 Compare G Monti (n 63 above) 365: ‘In sum, the importance of market integration wanes as economic analysis shapes competition policy.’ 99 Compare P Behrens (n 73 above) 37–9; J Drexl (n 87 above) 596; D Zimmer, ‘Competition law de lege ferenda’ in R Zäch et al. (eds), The Development of Competition Law: Global Perspectives (Cheltenham: Edward Elgar Publishing, 2010) 322-324; R Zäch and A Kuenzler, ‘Reining in the “more economic approach”: Some overriding constraints from constitutional law and economics’ in Technology and Competition (n 69 above) 541, 550–4. according to which the application of competition law systematically requires the effects on consumers. According to the Court, European competition law, -‘. . . is designed to protect not only the immediate interests of individual competitors or consumers but also to protect the structure of the market and thus competition as such.’100 The same holds true for the goal of market integration. The construction of the internal market has offered far-reaching opportunities to undertakings and to consumers. It would not be convincing to grant free market access within the internal market while reserving the right to create private barriers where it is deemed advantageous.101 In the IP context, the back-up function of competition law with respect to market integration is of particular importance since IPRs provide additional opportunities for segmenting markets according to national borders. This article has tried to illustrate how diverse the legal questions are which have to be assessed in this context. Territorial protection may be necessary for the development or the dissemination of innovations. However, the high value of market integration has to be taken into account whenever restrictions of cross-border trade are to be assessed. 100 Case C-08/08 T-Mobile Netherlands (n 24 above) para 38. Confirmed by Case C-501/06 P GlaxoSmithKline Services Unlimited (n 74 above) para 63. 101 The opinion is widespread that the integration function of EU competition law is or should be of unchanged importance, compare J Drexl (n 87 above) 597–9; A Heinemann (n 14 above) 67; D Hildebrand, The Role of Economic Analysis in the EC Competition Rules (3rd edn, Kluwer Law International, 2009) 16–18. According to others, its significance is diminishing, compare B Rodger and A MacCulloch, Competition Law and Policy in the EC and UK (4th edn, Routledge-Cavendish, 2009) 19. The market integration function is underlined by the ECJ (Case C-501/06 P GlaxoSmithKline Services Unlimited (n 74 above) para 61) and upheld by the European Commission in the new Commission Guidelines on Vertical Restraints (n 66 above) para 7, 100 (d).