Competing on the merits with pharmaceutical patents

Competing on the merits with pharmaceutical patents Name: Supervisor: Second Corrector: Master: Date: Emp: Anr: Fenna Douwenga Ms. T. Tseliou Prof. ...
Author: Crystal Walsh
3 downloads 1 Views 886KB Size
Competing on the merits with pharmaceutical patents

Name: Supervisor: Second Corrector: Master: Date: Emp: Anr:

Fenna Douwenga Ms. T. Tseliou Prof. P. Larouche International Business Law 11-06-2014 1255653 653414

Table of Contents Abbreviations .................................................................................................................................................. 3 1. Introduction................................................................................................................................................. 4 2. Tension between patent law and competition law..................................................................................... 7 2.1 Introduction........................................................................................................................................... 7 2.2 Objective ............................................................................................................................................... 7 2.2.1 Competition law objective ............................................................................................................. 7 2.2.2 Patent law objective ....................................................................................................................... 8 2.2.3 Comparison .................................................................................................................................... 9 2.3 Hierarchy and instruments .................................................................................................................. 10 2.3.1 Determination of hierarchy stemming from competition law ..................................................... 10 2.3.2 Determination of hierarchy stemming from IP law ..................................................................... 12 2.3.3 Overlapping instruments .............................................................................................................. 12 2.4 Conclusion ........................................................................................................................................... 13 3. Article 102 TFEU ........................................................................................................................................ 14 3.1 Introduction......................................................................................................................................... 14 3.2 Product market definition and dominance ......................................................................................... 14 3.3 Competition on the Merits .................................................................................................................. 17 3.4.1 Current Status Abuse.................................................................................................................... 17 3.4.2 Objective Justifications................................................................................................................. 21 3.4.4 Problems with current status of abuse and objective justifications ............................................ 24 3.4.5 Solutions ....................................................................................................................................... 26 3.4 Conclusion ........................................................................................................................................... 28 4. Analysis of cases ........................................................................................................................................ 30 4.1 Introduction......................................................................................................................................... 30 4.2 AstraZeneca ......................................................................................................................................... 30 4.2.1 Introduction.................................................................................................................................. 30 4.2.2 Relevant Product Market ............................................................................................................. 30 4.2.3 Abuse of a Dominant Position using SPCs .................................................................................... 32 4.2.4 Abuse of a Dominant Position by deregistering MAs................................................................... 33 4.2.5 Analysis ......................................................................................................................................... 33 4.3 Sanofi-Aventis ..................................................................................................................................... 35 4.3.1 Analysis ......................................................................................................................................... 38 4.4 Reckitt Benckiser ................................................................................................................................. 39 4.4.1 Analysis ......................................................................................................................................... 40 1

4.5 Pfizer.................................................................................................................................................... 41 4.5.1 Analysis ......................................................................................................................................... 42 4.6 Conclusion ........................................................................................................................................... 43 5. Synthesis and conclusion........................................................................................................................... 45 5.1 Introduction......................................................................................................................................... 45 5.2 Relationship between competition law and IP law ............................................................................. 45 5.2.1 Issues ............................................................................................................................................ 45 5.2.2 Solutions ....................................................................................................................................... 46 5.3 The ‘competition on the merits’ test .................................................................................................. 47 5.3.1 Issues ............................................................................................................................................ 47 5.3.2 Solutions ....................................................................................................................................... 48 Bibliography .................................................................................................................................................. 50 Cases.............................................................................................................................................................. 53

2

Abbreviations CFI ECJ EC Treaty EFPIA OECD TFEU TRIPS

Court of First Instance European Court of Justice Treaty establishing the European Economic Community European Federation of Pharmaceutical Industries and Associations Organization for Economic Co-operation and Development Treaty on the Functioning of the European Union The Agreement on Trade Related Aspects of Intellectual Property Rights

3

1. Introduction In 2008 the European Commission started multiple investigations into originator pharmaceutical companies who possibly delay the arrival of generic medicines by using pay-for-delay settlements (which can fall under the scope of article 101 TFEU) and methods to artificially elongate patent protection (which can fall under article 102 TFEU). The inducement for the investigations –the so-called broad competition inquiry- is the protection of consumers against harm which they can experience due to the delayed market entry of generic medicines.1 With the use of article 101 and 102 TFEU, the Commission and national competition authorities are using competition law as a tool to put a break on side-effects of patents that they find unwanted. The use of these competition law tools for restricting patent protection, is stirring up some commotion in the legal literature. The commotion is a direct result of the fact that both patent law and competition law strive for innovation. Competition law aims at stimulating competition so that companies want to invest in research and development which furthers innovation, while patent law stimulates innovation by giving inventors a right to be the sole user of that invention for a certain period of time. People wonder whether the Commission is stepping outside the territory of competition law and into other fields of law. Both patent law and European competition law applying to the same subject has the potential of causing legal uncertainty and violating the rule of law. Companies might not know what law applies when and why.2 Furthermore, the question rises whether the involvement of competition law in the field of pharmaceutical patents is even really necessary. Besides it causing a potential waste of time, money and effort, it might also harm competition. This, however, cannot be said with certainty. The Commission might also be stepping in where patent law cannot go, which means that there is no risk of an unnecessary involvement of competition law in the field of patents. Also, as long as it is clear what the Commission finds incompatible with European competition law, there is no need for companies to feel uncertain as to what is and what is not allowed and whether or not their behavior will be fined. In this light, the problem of legal uncertainty does not arise. The possible undue restraints on patent law using competition law as a result of the pharmaceutical sector inquiry is precisely what this research looks into. Therefore, some of the relevant case law is looked into. Since the inquiry by the Commission started in 2008, a few interesting cases have taken place. One of these interesting cases with regards to the pharmaceutical company Lundbeck. Lundbeck had agreed to delay the market entry of cheaper generic versions of Lundbeck’s antidepressant citalopram. The conduct constituted a pay-for-delay settlement and was therefore an infringement of article 101 TFEU. As a result, Lundbeck was fined with € 93,8 million by the Commission.3 Lundbeck appealed the decision by the Commission to the General Court and the result will most likely be published in two to three years from the date the fine was imposed.4 In another 101 TFEU case, the Commission issued a statement of objections to Les Laboratoires Servier and several of its generic competitors for their practices which potentially delayed the generic entry of perindopril, a cardio-vascular medicine. The delay appears a result of patent settlement agreements.5 Furthermore, the Commission imposed a €15 million fine on Johnson & Johnson and Novartis for infringing article 101 TFEU by delaying the market entry of fentanyl, a pain-killer.6 All of these cases have two things in common. First of all, all these cases are 101 TFEU cases. Second of all, they were not published.

1

European Commission, 2008; Neelie Kroes 2008 Geradin 2014, p. 17 3 European Commission 2013b 4 Lundbeck 2013 5 European Commission 2011 6 European Commission 2013a 2

4

Due to the lack of available motivation of the fines or statement of objections, the 101 TFEU cases are very difficult to analyze. Therefore, it is difficult to find out whether the Commission is excessively constraining patent law with this competition law based fines and these cases are not used. Once the Lundbeck decision is available, it is very interesting to look into the implications for patent law by applying 101 TFEU as well. More luck in the possibility to analyze pharmaceutical competition law cases can be found in the 102 TFEU cases. At a European level, the only ECJ case which took place since the sector inquiry is AstraZeneca. In short, the Court and the Commission found that AstraZeneca had supplied misleading information when applying for an SPC7. The Court fined AstraZeneca as this constituted an abuse under EU competition law because the conduct fell outside the scope of “competition on the merits”.8 In addition to the Commission, national authorities are also using article 102 TFEU in the field of artificially elongating pharmaceutical patent protection. Think of the Reckitt Benckiser case in the UK9, Sanofi-Aventis in France10 and Pfizer/Pharmacia in Italy.11 All of the national competition authorities and the Commission in these cases used the same test to decide whether there was an infringement of article 102 TFEU. Namely the test whether or not the behavior falls outside the scope of “competition on the merits”. One can argue that such a test is too broad to be in line with legal certainty. On the other hand, open norms in the law can help make the law more flexible and fair at times. The question is in which of the two area’s it falls. Also, the use of this test can cause competition authorities to fall too quickly into the area of patent law. At this point we come back to the question whether or not that is in line with legal certainty, the rule of law and the stimulation of innovation. Therefore, the question that I would like to answer in my thesis is: Does the competition on the merits test as it has been used in the last few years by the European Commission, the national competition authorities and courts put undue constraints on patent protection in the pharmaceutical industry? The structure to answer this main research question is as followed. In Chapter 2, the field of tension between patent law and competition law is discussed. In Chapter 3 the competition on the merits test is defined. In Chapter 4 the relevant pharmaceutical case law is discussed to see how the competition on the merits test has been applied and in Chapter 5 the main research question is answered using the information provided in the other chapters. Chapter 2 answers the question what the field of tension between patent law and competition law is to analyze how patent law and competition law work together. Also, the field of tension is used to evaluate whether competition law is going too far into the field of patent law by meddling with pharmaceutical patents to protect consumers. Therefore, the objectives of patent law and competition law are looked into and the instruments and hierarchy of patent law and competition law are discussed. The question is answered using academic literature. In Chapter 3 the question which is answered is what the literature says in regard to the “competition on the merits” test. Possible problems with the use of the competition on the merits test and solutions to those problems are also given. Besides the competition on the merits test, the chapter looks into the other issues for assessing the infringement of article 102 TFEU, namely the market definition and the dominant position 7

i.e., Supplementary Protection Certificate ECJ 6 December 2012 (AstraZeneca/Commission) 9 OFT 12 April 2011 (Reckitt Benckiser/OFT) 10 Autorité de La Concurrence 14 May 2013 (Sanofi-Aventis/Autorité de la concurrence) 11 Geradin 2014, para. III 8

5

test. Those two tests are only discussed briefly as they are not the main focus point of this research. The reason they are discussed is to better understand the analysis of the case law in Chapter 4 and to understand the whole picture of analyzing article 102 TFEU. In Chapter 4 the most important national and European case law is looked into to find how the competition on the merits test is used in relation to pharmaceutical patents. All cases that are analyzed here were decided after the publishing of the Commission’s Guidance Paper in 2009 and after the Commission announced their investigations into the originator pharmaceutical companies who supposedly delay the arrival of generic medicines. The reason not only European but also national cases are looked into is because the national Courts and national authorities tend to follow the example set at the European level. Furthermore, national cases in regard to pharmaceutical patents can influence the protection of pharmaceuticals all over Europe due to the internal market. If a product is put on the market in a European country where there is no patent protection, the product can be traded freely into other European countries where it might have such protection. Consequently, if a national authority finds that the patent protection was elongated unduly in one country, the decision can have indirect consequences for the protection the patent holder has all over Europe. First the European AstraZeneca12 case is discussed; it is discussed more elaborately than the national cases as it is the only relevant European case. After which the national cases of Reckitt Benckiser13, Sanofi-Aventis14 and Pfizer15 are analyzed in consecutive order. In Chapter 5 the main research question is answered using the information gathered in the other chapters.

12

General Court 1 July 2010 (AstraZeneca/Commission); ECJ 6 December 2012 (AstraZeneca/Commission) OFT 12 April 2011 (Reckitt Benckiser/OFT) 14 Autorité de La Concurrence 14 May 2013 (Sanofi-Aventis/Autorité de la concurrence) 15 Consiglio di Stato 12 February 2014 (Pfizer/Autorità garante della concorrenza e del mercato) 13

6

2. Tension between competition law

patent

law

and

2.1 Introduction The aim of this chapter is to determine the field of tension between patent law and competition law in the European Union. The field of tension is used to analyze how patent law and competition law work together. Also, the field of tension is used to evaluate whether competition law is going too far into the field of patent law by meddling with pharmaceutical patents to protect consumers. On that account, I look into the objectives of patent law and competition law. Furthermore, I discuss the instruments and hierarchy of patent law and competition law and offer some examples of the overlap between competition law and patent law.

2.2 Objective In order to determine the tension between the objectives of patent and competition law, first, the objective of competition law is discussed. Second, the objective of patent law is explored and, third, the two objectives are compared.

2.2.1 Competition law objective The main provisions of competition law are found in the Treaty on the Functioning of the European Union (“TFEU”), namely in articles 101 and 102 TFEU. As competition law is based on the TFEU, the objective of competition law is aligned with the objective of the TFEU: to achieve economic integration. Competition law aids in achieving this objective by keeping competition within the common market undistorted.16 Article 101 TFEU tries to achieve that goal by prohibiting “all agreements between undertakings, decisions by associations of undertakings and concerted practices which may affect trade between Member States and which have as their object or effect the prevention, restriction or distortion of competition within the internal market”.17 In the field of pharmaceutical patents, article 101 TFEU is used for prohibiting IP license agreements that are considered to prevent, restrict or distort competition. An example of such a prohibited agreement is a pay for-delay agreement.18 Moreover, article 102 TFEU aims to achieve a complete internal market by prohibiting “Any abuse by one or more undertakings of a dominant position within the internal market or in a substantial part of it (…) in so far as it may affect trade between Member States”.19 Article 102 TFEU is used for restricting the abuse of the dominant position of pharmaceutical producers and the patents they hold.20 As mentioned in chapter 1, the focus of this research centers on article 102 TFEU. Through creating a competitive market and maintaining effective competition, competition law also aims at stimulating innovation.21 In order to stimulate innovation, competition law seeks to achieve a number of efficiencies that would exist under perfectly competitive market conditions. First of all, competition law tries to increase competition to the point where the price that the consumer is willing to pay for a product

16

Article 101 and 102 TFEU; Whish & Bailey 2012, p.1 and 2; Anderman & Schmidt 2011, p.18 For a full understanding of the function and use of article 101 TFEU see for instance:Whish & Bailey 2012, chap.4 18 European Commission 2013b; European Commission 2013a; European Commission 2012 19 For a full understanding of the function and use of article 102 TFEU see for instance: Whish & Bailey 2012, chap.5 20 ECJ 6 December 2012 (AstraZeneca/Commission); General Court 1 July 2010; OFT 2011; Autorité de La Concurrence 14 May 2013 (Sanofi-Aventis/Autorité de la concurrence); Consiglio di Stato 12 February 2014 (Pfizer/Autorità garante della concorrenza e del mercato) 21 Anderman & Schmidt 2011, pp.12–14 17

7

is equivalent to the marginal utility that the consumer receives (“allocative efficiency” also known as Pareto efficiency).22 Second, competition law aims at productive efficiency in order to stimulate innovation. 23 Productive efficiency occurs when producers are pressured by the market to produce at the lowest possible cost due to the existence of competition in the market.24 Third, competition law tries to promote innovation by enabling competition in the market, so that producers invest in innovation and technological progress (“innovative efficiency”).25 Since perfect market conditions are hardly ever achieved in real life, some economists believe that innovation is not best achieved by competition law, as competition law would not always bring about the best results in markets on which competition is imperfect. An example of a factor that competition law does not handle well in imperfect market is the dynamic efficiency of patent law.26 Dynamic efficiency is productive efficiency over a longer period of time. Schumpeter, for instance, believes that innovation is best achieved in a market with a monopolist because potential competitors attempt to enter the market due to the high rewards from innovation.27 The points Schumpeter and other economists make have some validity, but to say competition law does not stimulate innovation may go a step too far. The reason is two-fold. On the one hand, competition law achieves innovation in other ways than by pushing dynamic efficiency, namely by aiming at innovative, productive and allocative efficiencies.28 On the other hand, competition law steps into other fields of law only where those fields fall short of preventing anti-competitive conduct; this includes fields of law which promote dynamic efficiency.29 As a basic rule, for instance, competition law tries not to interact with patent law - which provides for a limited monopoly on a certain invention – and thus respects the dynamic efficiency patent law tries to achieve. So even though competition law does not stimulate dynamic efficiency per se, it appears to respect fields of law which stimulate it.30 The reason that I use the term “appears” is because it is not clear at this moment whether this holds true for the relationship between competition law and patent law. The question whether competition law respects patent law is, after all, what this research aspires to answer. In the case that the answer is negative, competition law may be hampering innovation by focusing on short-term innovation and not respecting the instruments which promote long-term innovation. In this way, firms would not be able to strike a balance between investing in short-term and long-term innovation.

2.2.2 Patent law objective A patent is one of the intellectual property rights that a person can obtain in Europe. A patent is a legal title given by a government or another public organ to give the patent applicant, often the inventor of the invention, the sole right to make, use and sell that invention for a period of (usually) 20 years from the application date.31 A patent is granted for inventions from any field of technology if it is new, involves an inventive step and is susceptible of industrial application.32

22

Anderman & Schmidt 2011, p.14 Ibid., p.14 and 15 24 Feeling pressured to produce at a lower price can cause unsafe products and unsafe working circumstances which is why other fields of law regulate product manufacturing. 25 Anderman & Schmidt 2011, p.14 26 Ibid., p.13 27 Ibid, p.15 28 Ibid., p.13 and 14 29 Ibid., p.4 30 Ibid., p.16 31 Article 63 and 64 EPC; www.EPO.org. 32 Article 52 EPC; for a full understanding of how these requirements work and how a patent is granted see for instance Tritton 2008. 23

8

The objective of patent law has been the same since it first saw the light in Italy about 500 years ago; namely to stimulate innovation, stimulate follow-on innovation -using an existing piece of knowledge to further develop an invention- and reward the inventor (of the invention-not needed) by giving him an exclusive right.33 Patent law stimulates innovation and follow-on innovation by requiring the patent applicant to publish his invention in the patent database, so everyone can see the invention and use the information to work around the patent or further develop the invention once the patent has expired.34 As with competition law, some economists have a problem with patent law as well. Many economists, unlike Schumpeter, oppose to monopolies and view patents as a monopoly. These economists associate monopolies with inefficiency and unfair exploitation.35 It should be kept in mind, however, that the type of monopoly a patent provides for is not the same as a company that holds a monopoly in the market. A patent gives a monopoly on an invention for a maximum of 20 years. Other companies can work around that patent, so a diversity in products on the market and competition is very well possible, which causes innovative efficiency and dynamic efficiency. A company that has a monopoly position, on the other hand, has an extremely strong market position or is the only person operating in the market and can use that power to keep others out of the market. It can do so, for instance, by lowering its prices as soon as a competitor wants to enter the market. In such a market, competition is much less possible and this is what competition law tries to prevent. Due to the difference between a patent ‘monopoly’ and a monopoly position in a market, the problem that economists have with patents is not always justified.36 Often it is a result of a wrong definition of patents and the type of protection they give, because without patents everyone could use the knowledge without limitations.37 The result would be that innovation would be slowed down, either because inventors would no longer share their inventions or because they would no longer invest in inventions that other could easily copy without an investment being required.38

2.2.3 Comparison As we have seen above, both fields of law have a similar objective; they both try to stimulate innovation but in different ways. Patent law tries to stimulate innovation by rewarding the inventor for his invention with a patent, who then in turn publishes his invention so everyone can use it and further develop it. Without patents, every inventor would keep his invention a secret or would not invent at all which would cost innovation. While patent law tries to stimulate innovation in an indirect way with a reward system, competition law tries to stimulate innovation by protecting competition in the market so companies want to invest in innovation to stay ahead in the market. The interplay between the two fields and the effects both fields of law can have on innovation require a delicate balance. Patent law needs to function properly so patents are not granted to non-inventive, non-novel inventions, as this would slow down innovation. The justification for awarding a patent, after all, is that the inventor can get such a return for the investment he made on the invention that he is stimulated to invest the resources required to innovate. If the duration of the patent outlasts the time required for the patent holder to recoup the resources spent (and a margin for his efforts), the patent holder is put in an undeservedly strong position on the market vis-à-vis competitors, which causes innovation and competition problems. An undeserved patent causes innovation problems because other market players are not able to use the invention and further develop it. In addition, an

33

Spector 1989; Nard 2010, p.10 Guellec & van Pottelsberghe de la Potterie 2007, p.46 and 50; Spector 1989 35 Grosche 2006, p.285; Taylor et al. 1973, p.24; For a full analysis of the rationale of patent law see Douwenga 2014 (master thesis Law & Technology) 36 This paragraph was adapted from my other thesis: Douwenga 2014, para.42 and 43. 37 Spector 1989, p.49; Andersen 2004, p.430 38 Spector 1989, p.49; Andersen 2004, pp.423–428; Guellec & van Pottelsberghe de la Potterie 2007, p.49 and 50 34

9

undeserved patent causes competition problems because other market players are not able to use the invention and therefore cannot compete price-wise with the patent holder. Competition law, on the other hand, needs to function properly as well. If competition law attempts to prevail over patent law, then it significantly reduces the incentive to innovate. One of the reasons companies innovate, is to obtain a patent and receive a return on their investments. If such a patent cannot be acquired or cannot be acquired in such a way that it brings a return, then companies’ incentives to innovate disappear. Furthermore, too strict competition rules can cause long-term innovation problems because competition law focuses on short-term innovation due to not focusing on dynamic efficiency and not respecting the instruments which promote long-term innovation. The European Commission agrees with this balance between the demands of competition law and patent law, as the Commission says that both competition law and intellectual property law aim at promoting innovation and finds that intellectual property law can benefit competition. Intellectual property law, after all, encourages businesses to invest in new and better products and processes.39 In order to find this delicate balance between competition law and patent law and prevent competition law from excessively restricting patent law, there are some rules on the hierarchy between the two fields. Sometimes, there is also some overlap between the instruments both fields of law have. The hierarchy and instruments are discussed next.

2.3 Hierarchy and instruments As this research is aimed specifically at article 102 TFEU, the hierarchy and balance between patent law and competition law is mostly discussed in regard to that provision. Hierarchy problems between the two fields can specifically arise when applying article 102 TFEU to dominant market players who abuse their patents. Ways to determine the hierarchy between patent law and competition law stem from both competition law and patent law. In some cases, both fields of law offer instruments to fight patent abuse by patent holders; in these cases, it is not clear whether the drafters of the relevant instruments had a clear vision of the hierarchy between the two fields of law.

2.3.1 Determination of hierarchy stemming from competition law The first general rule in determining the hierarchy between patent law and competition law stems from competition law. It posits that the mere existence of an IP right cannot constitute an infringement under competition law, but a specific way of using that IP right can. This rule was posited with regards to the ownership of patent rights in the Parke Davis case by the European Court of Justice (“ECJ”).40 The ECJ concluded that the sole fact of owning a patent cannot be precluded by article 102 TFEU. The reason is that the prohibition of article 102 TFEU only applies when three elements are met: “the existence of a dominant position, the abuse of this position and the possibility that trade between Member States may be affected thereby”.41 The three elements are not met by giving someone a patent right, even though “a patent confers on its holder a special protection at national level”. 42 The explanation given in Parke Davis was confirmed in the Deutsche Grammophone case.43 In that case, the ECJ

39

European Commission 2011, para.269; European Commission 2004, para.7

ECJ 29 February 1968 (Parke Davis) Id., para. 72 42 Ibid. 40 41

43

ECJ 8 June 1971 (Deutsche Grammophon Gesellschaft/ Metro)

10

decided that merely exercising an exclusive right does not mean that a person is occupying a dominant position within the meaning of article 102 TFEU (then still article 86 EC Treaty).44 In the Magill and Parke Davis case the ECJ made clear that even though the use of a patent does not constitute an infringement of article 102 TFEU per se, this does not mean that the use of a patent in a particular way cannot form the abuse of a dominant position as prohibited by article 102 TFEU.45 Such an infringement by the use of a patent would require ‘exceptional circumstances’.46 What exactly these exceptional circumstances are is difficult to determine, but at the very least the use of the patent must mean that all three elements of article 102 TFEU are met. The same general rule applies to the practice of licensing IP rights to other parties. The ECJ has made clear that the refusal to grant a license for a patent is not enough to constitute an abuse of dominant position; additional circumstances are necessary to constitute such an infringement.47 In the 102 TFEU guidelines the European Commission made clear what these circumstances may be in the form of a non-exhaustive list. First, the refusal to give a license must relate “to a product or service that is objectively necessary to be able to compete effectively on a downstream market” (a market that makes use of the product for which the license is required). Second, the refusal of the license “is likely to lead to the elimination of effective competition on the downstream market.” Third, the refusal “is likely to lead to consumer harm.”48 In the Microsoft case, the ECJ applied these rules to assess whether refusing to grant a license was a violation of article 102 TFEU. Microsoft had abused its dominant position by not granting a license because the license was necessary for competing server operating system vendors to be able to interoperate with the Microsoft client and server operating systems. Without the license, making competing software would be undoable, due to the interoperability of programs in the software market.49 Software programs have to be able to communicate with each other, otherwise a new piece of software cannot be or will not be bought and used by consumers. It remains to be seen how the proof for the requirements formulated by the Commission can be formed in the field of pharmaceuticals. In this field it is quite possible to invent ‘around’ a patent (i.e. to invent something with similar effect that is not covered by the patent), as medicines are not dependent on each other. A producer of medicines does not need to use the composition of medicine A in the composition of medicine B. If, however, the ECJ would consider the refusal to license an infringement of 102 TFEU in the area of pharmaceutical patents, chances are that such a decision would greatly slow down innovation. Pharmaceutical producers, after all, invest large amounts of money in innovation with the goal of recouping these investments (and more) if they are able to find a cure for a certain disease.50 They will likely cease to make these investments if they cannot recoup their investments due to producers of generic medicines obtaining compulsory licenses from them. The second general rule in determining the hierarchy between patent law and competition law also stems from competition law, and holds that national legislation cannot safeguard an infringer from the consequences of competition law. In regard to trade-marks, the ECJ made clear in Consten/Grundig that a

44

Id., para.16 The ECJ made the Parke Davis and the Deutsche Grammophone decision under article 86 Treaty establishing the European Economic Community (‘EC Treaty’), which is the old version of article 102 TFEU. The text of the provision has not changed since the TFEU went into force in 2007, merely the lettering and numbering. For that reason, this research will refer to article 102 TFEU as often as possible, even when the court made the decision under an older version of 102 TFEU. 45 ECJ 29 February 1968 (Parke Davis), para. 72. 46 ECJ 6 April 1995 (Magill), para.50 47 48 49 50

ECJ 5 October 1988 (Volvo/Veng), paras.7–9; ECJ 6 April 1995 (Magill), para.48 European Commission 2009b, para.82 CFI 24 March 2004 (Microsoft v. European Commission) Vezzoso 2012, pp.522 - 523

11

trade-mark holder cannot hide completely behind a national right. Although, on the basis of 345 TFEU, the “treaty shall in no way prejudice the rules in Member States governing the system of property ownership”, limiting the exercise of a trade-mark may happen to the extent necessary to give effect to the prohibition under article 102 TFEU.51 Therefore, the fact that the acts of a trade-mark holder are allowed under national copyright law does not necessarily mean that the acts are also allowed under article 102 TFEU. We can wonder if the same rule applies to patent law; many authors argue that it does.52 This is not surprising, as applying the rule the ECJ formulated to patent law would not constitute an infringement of article 345 TFEU.

2.3.2 Determination of hierarchy stemming from IP law Not only competition law offers guidance on how to appreciate the hierarchy between competition law and IP law. Some intellectual property law provisions themselves grant room for competition law to apply. The Agreement on Trade Related Aspects of Intellectual Property Rights (‘TRIPs’) signal that additional and appropriate measures may be necessary to prevent an unreasonable restrain on trade or an adverse effect to the international transfer of technology by right holders who abuse their intellectual property rights. Any additional measure taken, however, does have to be consistent with the TRIPs.53 In my view, this has to do with legal certainty. A patent holder needs to be sure of the fact that the patent he is holding lawfully will not bring about an infringement in other fields of law.

2.3.3 Overlapping instruments In addition to the hierarchy issue between the two fields of law, there are also some instruments that overlap. One of the examples is that both competition law and patent law have ways to fight supplementary protection certificates (“SPC”) when an SPC was not supposed to be granted or is used to unduly prolong patent protection.54 An SPC provides protection for an additional period of time in addition to the protection time that is granted due to the original patent. The time period it protects is the same as the time between the filing of an application for a patent for a new medicinal product and authorization to place the medicinal product on the market, but no longer than 5 years. The reason for having these SPCs is to provide protection to prevent the penalizing of pharmaceutical research necessary for entering the market with the pharmaceutical product.55 In patent law, a supplementary protection certificate (SPC) which was unduly granted can be nullified on the basis of article 15(1) Regulation 1769/92.56 Competition law, however, can use article 102 TFEU to nullify the SPC and give a fine to the company with the dominant position who had unduly obtained an SPC. Recently, the ECJ decided in the AstraZeneca case on such a misuse of an SPC. The AstraZeneca case is further discussed in chapter 4. Another example of an overlapping instrument relates to licenses. Both fields of law can fight companies who do not want to grant a license for their patent. Under patent law, the government may step in and give out a license when a company refuses to (a so-called ‘compulsory license’). Article 31 of the TRIPsagreement and article 5A of the Paris Convention for the Protection of Industrial Property, after all, have provisions that make it possible for Member States to include such a provision for compulsory licenses in their legal systems. One remark must be made, though: even though many Member States have a provision

51

Id., para. 48; ECJ 13 July 1966, “Joined Cases 56 and 58/64,” para. 345. Speyart 2014, para.2.2 53 Article 8(2) TRIPs 54 ECJ 6 December 2012, C 457/10 (AstraZeneca/Commission); Maggiolino & Montagnani 2011 52

55 56

Council Regulation (EEC) No 1768/92 of 18 June 1992, Preamble Council Regulation (EEC) No 1768/92 of 18 June 1992

12

on compulsory licensing, they rarely use it.57 Under competition law, article 102 TFEU can be used when a dominant market player refuses to grant a license for a patent. In cases where both IP law and competition law offer instruments to combat undesirable use of IP rights, the instinctive behavior of competition authorities is to make use of the instruments offered by competition law. This may be because the mechanisms in patent law do not function properly (a guess that I make on the basis of the fact that compulsory licenses are rarely given under patent law). It may also be the case that competition law mechanisms feel more ‘natural’ to be used by competition authorities in part due to the fact that national competition authorities do not always have a background in intellectual property law and in part because the competition law mechanisms are the mechanisms they use every day. If that would be the reason, however, the use of these instruments may not be optimal. After all, the instruments provided by IP law were specifically crafted to deal with the exact problems they are addressing. The instruments provided by competition law, in comparison, are instruments of blunt force. A more substantive discussion on this topic will take place after the case law has been discussed in chapter 4, as we then have a clearer view of why the Commission chooses certain instruments and why they do or do not work properly.

2.4 Conclusion The aim of this chapter was to determine the field of tension between patent law and competition law in the European Union. As we have seen above, both fields of law have a similar objective, as they both try to stimulate innovation in one way or another. Competition law is directed more at short term innovation, while patent law aims at a long term innovation. It seems that we need both fields of law for the market to function properly, as they compensate for each other’s weaknesses. Therefore, competition law should be restrained from intervening in patent law too easily, as this could hurt innovation and competition in the long run. Thankfully, we can also ascertain that competition law cannot unrestrictedly intervene in the field of intellectual property law or patents, as a number of requirements will need to be met in order to do so. Sadly, we do have to conclude that the requirements the Commission and the ECJ have formulated in this regard are quite vague.

57

Bakels 2002, p.7

13

3. Article 102 TFEU 3.1 Introduction In this chapter the use of the competition on the merits test, in cases regarding pharmaceutical patents, is analyzed using academic literature. Possible problems with the use of the competition on the merits test and solutions to those problems are also given. Besides the competition on the merits test, the chapter looks into the other issues for assessing the infringement of article 102 TFEU, namely the market definition and the dominant position test. Those two tests are only discussed briefly as they are not the main focus point of this research. The reason they are discussed is to better understand the analysis of the case law in Chapter 4 and to understand the whole picture of analyzing article 102 TFEU. After all, in order for there to be an infringement of article 102 TFEU, the market player needs to have a dominant position in the relevant market and the market player needs to abuse that market power by acting in a way that falls outside the scope of competing on the merits.

3.2 Product market definition and dominance When assessing a possible infringement under article 102 TFEU, the Commission first defines the market to know whether a company is abusing its dominant position in that particular market.58 In defining the market, the competition authorities look at the geographical market and the product market to decide what the relevant market is. The geographical market concerns the geographical area in which a product is marketed, a country for instance.59 As this requirement is only remotely related to patents and the use of them in an anti-competitive way, the geographical market is not relevant for this research and not further discussed.60 The product market, on the other hand, is relevant and is discussed together with the requirement of dominance. The two are discussed together because they are very closely linked as dominance is measured within the relevant market. Overall, the product market is defined very narrowly by competition authorities.61 Due to a narrow market definition, competition authorities quickly decide that a certain market, is a single product market.62 A nail gun system by Hilti, for example, was seen as consisting of three separate products and therefore three separate product markets: the gun, the cartridge for the nails that was Hilti compatible and the Hilti compatible nails themselves. The Commission came to this conclusion because there were independent nail and cartridge strip makers who did not produce nail guns. Also, some nail gun manufacturers rely on other producers to produce some of the cartridges and nails. Furthermore, consumers purchase a nail gun so it can be used for a longer period of time, while the cartridges and nails are not used more than once.63 By using such a narrow product market, dominance is achieved more quickly as the amount of market power the undertaking has is measured over the specific relevant market. Dominance is tested by looking at the power the undertaking has to prevent competition on the relevant market and by looking at the power it has to move independent from its competitors.64 The broader the market, the bigger the chance

58

Anderman 2011, p.9 Whish & Bailey 2012, p.39 60 To read more about the geographical market see Anderman & Schmidt 2011, para.4.3; Whish & Bailey 2012, p.39 and 40 61 Anderman 2011, p.10; Anderman & Schmidt 2011, pp.40–46 62 Anderman 2011, p.10; Anderman & Schmidt 2011, para.4.2.2 59

63

ECJ 12 December 1991 (Hilti AG/Commission), para.18 and 66–78

64

ECJ 14 February 1978 (United Brands/Commission), para.65

14

there are competitors and that he is not (as) dominant. The smaller the market, the lesser the chance of competitors and the more likely it is that the company is found dominant.65 In addition, it is interesting to mention that both the gun and the cartridge had separate patents. One could say that having separate patents on these two products could have tipped the producer and the Commission off to the fact that the gun and the cartridge are considered part of separate product markets. A pharmaceutical producer can use the same analysis when having a patented medicine and wonder whether he has a dominant position in the market of that pharmaceutical product. Thankfully, having a patent does not necessarily mean that a company is dominant. The mere exercise of an exclusive right, like a patent, is not enough. The company needs to have the power to affect competition on the market over a considerable part of the market.66 Therefore, the companies who invented around the patent also matter. Obviously, such companies do not always exist and in those cases having a patent almost automatically means that you have dominance on the relevant market. Another consequence of a narrow market definition approach is that technology markets which also include patented technology, are built up of two separate markets.67 The first market is the licensed technology market and includes the licensed technology and the substitutes of that technology. The substitutes need to be able to constrain the exercise of the market power of the licensed technology in order to be part of the licensed technology market. The second market is the product market of the product that is built using the licensed technology. The licensed technologies are considered secondary markets, while the products are viewed as the primary market.68 Obviously, the product is not buildable without the licensed technology or a substitute of that licensed technology and the licensed technology cannot be marketed all by itself. Nonetheless, the licensed technology and the product are considered separate markets when the license is indispensable to conduct a specific business. The license is considered the indispensable upstream product for the supply of the downstream product.69 In such a market, the dominance of a patent holder is proven easily as well. The patent holder who has a patent has market power because it is essential for others to have a license of that patent to be able to compete.70 It can even go so far that the company with a patent is defined as a monopolist if the patent is a barrier to entry.71 In other words, if other companies cannot enter the market without a license, then the patent holder is a monopolist.72 In the area of pharmaceutical patents, this can mean that the patent holder of a pharmaceutical technology is the only player in the licensed technology market if there are no substitutes to the pharmaceutical technology. The pharmaceutical product provided by the patent holder and the competitors with the use of a license, are part of a separate market. Consequently, a dominant market power in pharmaceutical patent licensing cases is proved rather quickly due to the small market definition.

65

Anderman & Schmidt 2011, p.57

ECJ 8 June 1971 (Deutsche Grammophon Gesellschaft/ Metro), para.16 van Loon 2008, para.4.3 68 Anderman 2011, pp.10 - 11 69 CFI 24 March 2004 (Microsoft/Commission), para.319, 331–335, 691 and 1336; Anderman 2011, p.10 and 11; Anderman & Schmidt 2011, para.4.2.3 70 CFI 24 March 2004 (Microsoft/Commission), para.319, 331–335, 691 and 1336; Anderman 2011, p.63; Anderman & Schmidt 2011, para.4.2.3 71 Tritton 2008, para.11–044 66 67

72

Advocate General in: ECJ 5 October 1988 (Volvo/Veng)

15

The abovementioned way of finding the product market means that an intellectual property right or patent may actually cause a firm to have a dominant position, due to that patent as the market is quickly defined as broad as the patent is. As a consequence, a patent may give a company a good position in patent infringement proceedings, but in competition law a patent can get him in trouble.73 As a result, the way that competition law is defining the market may harm the innovative effect of patents due to the way competition law and its market definition detract from the rewarding effects of patents.74 Therefore, it might be better not to look at the product market, but at the innovation market. The difficulty is that this is currently only used under article 101 TFEU and still in development. The Commission, however, has made clear that it is prepared to use and make innovation market be a part of the relevant market.75 As there is no official way of defining the innovative market yet, it is not as clear how an innovation market should be defined as it is to define the product market. What is clear, is that for defining the innovation market the amount of innovation that can take place in the market needs to be taken into account and researched. If there is little room for innovation, then a small market definition should be used. But if there is a lot of room for innovation, then a broad market definition needs to be applied.76 The Horizontal Guidelines can be used to define the innovation market by looking at the competition in innovation requirements in those Guidelines.77 The competition in innovation requirements explain that competition authorities should take into account new products which can replace the current products or which can create a whole new intended use and create a completely new demand.78 In order to decide whether there is innovation, it is good to research whether competitors are spending money in research and developments as that can be a tip whether or not there is room for innovation. Also, looking at the barriers of entry into the innovation market can be of interest. If the patent holder has the power to keep competitors out of the innovation market then this can be a sign that he is taking a dominant position in the innovation market or that the innovation market is very small. Thus, the innovation market definition requires the Commission and other competition authorities to put their focus less on the consumers and more on the competitors. The advantage of using this approach is that a patent holder is not considered dominant too quickly on an innovation market as patents leave the possibility to innovate open. Only in interoperability cases, like in regard to software, a company stays dominant.79 Now that we have briefly seen how the product market is defined in all fields of competition law, it is time to zoom in on how the market for pharmaceuticals is defined. As we have seen the market is defined narrowly, as is the case in all fields of products, but how do competition authorities decide when a medicine has substitutes? Are older, less effective medicines part of the same product market? What about the application for the medicine? How about generics? Usually what the Commission does is look at the Anatomical Therapeutic Chemical (‘ATC’) Level 3. ATC Level 3 is an international classification standard for medicines which divides medicines into separate categories depending on their therapeutic and pharmacological indication. The problem with using ATC Level 3 is that the market definition can become too broad due to the fact that the different categories are set up broad or that it is too small as the categories merely look at patent-protected products. Therefore, it would be better to focus on the demand side instead of these categories as this would require looking into threats from generic producers and 73

van Loon 2008, para.4.2; Anderman & Schmidt 2011, p.8 van Loon 2008, p.231 75 Whish & Bailey 2012, p.39; van Loon 2008, p.231 76 van Loon 2008, p.248 77 Whish & Bailey 2012, p.39; van Loon 2008, p.248 78 European Commission 2011, para.119–122 79 van Loon 2008, para.4.5.7 74

16

patient usage. Such an approach already sounds more like the innovation market approach, than a product market approach and is already being used.80

3.3 Competition on the Merits After explaining how the relevant market and a dominant position is found, it is time to look into the question what competing on the merits is and when the behavior of a dominant pharmaceutical market player falls outside the scope of competing on the merits. Attention is merely paid to the academic literature and discussions and not yet to the case law on pharmaceutical patent abuse. The recent pharmaceutical case law is discussed in the next chapter after a full understanding is achieved on how competition on the merits is defined and should be defined to not excessively limit the positive effects of intellectual property on innovation. Furthermore, not all possible ways of acting in an abusive way are discussed, but merely the types of abuse relevant in regard to the pharmaceutical cases and their patents.81 Besides the current definition of competition on the merits, attention is also paid to the problems of that definition and possible solutions and alternatives. The order is as follows. First the current status of abuse is discussed. Secondly, possible objective justifications are discussed which can be used as counterarguments against arguments which build the proof for there being an abuse. Third, the possible problems which may arise out of the current status of abuse and its objectives justifications are looked into. Fourth, some of the solutions to those problems are mentioned. Fifth, the alternatives to article 102 TFEU which might also solve the problems of abusive conduct by dominant behavior are briefly explained.

3.4.1 Current Status Abuse The first question to look into is what the current status of abuse is. In other words, what does and what does not constitute abuse (and – thus – not competition on the merits). The current status of abuse is analyzed by looking at the theoretical underpinnings of the concept of abuse in article 102 TFEU and by giving examples of the way these theories are generally applied in practice. Also some of the recent developments in defining abuse are discussed. An opinion or a view on whether or not that approach to abuse is the best way to define abusive conduct is not discussed here but in para. 3.3.3, after the objective justifications have also been looked into.

3.4.1.1 Theoretical underpinnings of the test for abuse of a dominant market position: form-based and effects-based approaches In order to meet the critique on how abuse is established, the Commission made some changes to the test of abuse with the 2009 Guidance Paper.82 The Guidance Paper is not a piece of legislation as the Commission is not a legislative organ. What is more, their new approach is not even a guideline but a document on how to define their enforcement priorities.83 The question is whether the CFI, ECJ and the national Courts and competition authorities will follow the approach. Some authors even question whether the Commission will abide by their Guidance paper.84 Therefore, not only is the focus in this research put on how abuse was established after these changes, but also on how it was before. Using both views, a good understanding of the changes is realized for construing the possible advantages and disadvantages of the new approach. In addition, changes in law rarely happen overnight but are an evolving process. In order to recognize the newer way of determining abuse when it is applied, knowing the older way is necessary.

80

Priddis & Constantine 2011, pp.253 - 254 For a more elaborate explanation of the different types of abuse see for instance: Whish & Bailey 2012, pp.192–214 82 Gormsen 2013, p.224 83 Geradin 2010, p.46 84 Ibid., para.IV 81

17

The ECJ defines the concept of abuse of market power under article 102 TFEU as `an objective concept relating to the behavior of an undertaking in a dominant position which is such as to influence the structure of a market where, as a result of the very presence of the undertaking in question, the degree of competition is weakened and which, through recourse to methods different from those which condition normal competition in products or services on the basis of the transactions of commercial operators, has the effect of hindering the maintenance of the degree of competition still existing in the market or the growth of that competition.`85 In other words, abuse consists of a dominant company acting outside the scope of competing on the merits, while competing on the merits is permissible for strengthening a company’s economic market power whether that company is dominant or not.86 Holding an intellectual property right for instance, is not a problem, but using it to prevent others from putting a product on the market might be.87 Two theoretical approaches may be used to determine whether or not certain behavior is an abuse: the form-based approach and the effects-based approach.88 A form-based approach (also referred to as the orthodox approach) categorizes behavior into specific categories of what conduct is unlawful and what conduct is not, without looking at the actual effects the conduct had upon the market.89 The form-based approach bases abusive conduct on the presumption or capability of the conduct to create harmful effects.90 The difficulty with this approach is that companies do not get the opportunity to defend themselves because it is irrelevant whether or not their conduct was actually harmful for competition. In other words, certain conduct is regarded per se as abusive under the form-based approach.91 On the other hand, the effects-based approach, may permit behavior in one case and not in another, depending on the effect the behavior has on consumer welfare.92 Actual harmful effects of the abusive conduct need to be proven using empirical analysis under the effects-based approach.93 In the past –and depending on the effectiveness of the Guidance Paper maybe also in the future- the Commission and the Courts used a form-based approach.94 The approach was not only used for article 102, but in all areas of competition law. Due to the criticism on this approach, as it did not take actual harmful effects into account and was too legalistic, the law and policy of EU competition law moved more into an economic and therefore effects-based approach.95 Sadly, article 102 TFEU had been left behind in this modernization process. In 2005, Neelie Kroes made clear that a change in applying article 102 TFEU should be welcomed as well: “the exercise of market power must be assessed essentially on the basis of its effects in the market (…) We simply want to develop and explain theories of harm on the basis of a sound economic assessment for the most frequent types of abusive behavior to make it easier to understand our policy, not only as stated in policy papers but also in individual decisions based on Article 82. Article 82 enforcement should focus on real competition problems: In other words, behavior that has actual or likely restrictive effects on the market, which harm consumers. There are two main reasons for this: First, Enforcement Agencies should be cautious about intervening in the functioning of markets unless there is clear evidence 85

ECJ 13 February 1979, para.91; ECJ 9 November 1983 1983, para.70 Schillemans & van Bockel 2011, para.5 87 Waelbroeck 2010, p.5 88 Gormsen 2013, p.225 89 Drexl 2009, p.5; Ahlnorn & Evans 2007, p.803; Maggiolino 2011, p.73 and 74 90 Gormsen 2013, p.225 91 Ibid., p.233; Geradin 2010, p.48 92 Drexl 2009, p.5; Ahlnorn & Evans 2007, p.803 93 Gormsen 2013, p.25 94 Ahlnorn & Evans 2007, p.803; Larouche 2008, p.23 and 24; Drexl 2011, p.5 and 24 95 Geradin 2010, p.47; OECD Competition Committee 2005, p.18 86

18

that they are not functioning well. As you say in the States “If it ain’t broke, don’t fix it!” Secondly, Enforcement Agencies don’t have unlimited resources and need to focus their efforts on what makes a real difference.”.96 Finally, in 2009, due to the growing criticism on the out of date form-based approach still used in applying article 102 TFEU; the Commission wrote the 2009 Guidance paper.97 The difficulty in writing the Guidance Paper was that the Commission had to comply with the framework of article 102 TFEU, due to the fact that the Commission is not a legislative organ.98 In that sense, the amount of flexibility they had was limited.99 In finding the balance and defining how abuse is determined, the Guidance Paper is aimed at “anti-competitive foreclosure” (also referred to as exclusionary abuses).100 The Commission made clear that it will base its decisions on the degree of dominance, foreclosure and an objective justifications test.101 In order to achieve this goal, the Guidelines explain that abuse in the form of anti-competitive foreclosure is proven by complying with two steps.102 First, the Commission needs to prove that “effective access of actual or potential competitors to supplies or markets is hampered or eliminated as a result of the conduct of the dominant undertaking whereby the dominant undertaking is likely to be in a position to profitably increase prices”. Second, the Commission has to prove that the foreclosure proven with step to causes consumer harm or “detriment of consumers.”103 The second step of requiring consumer harm, shows a more effects-based approach because merely effecting the structure of competition is no longer enough, there needs to be an actual negative effect on consumers. Thus far the Court had not requested the proof of consumer harm. In using the Guidance Paper, it seems as if the Commission is moving away from the form-based approach into the effects-based approach.104 In using a more effects-based approach, the aim is to prevent the definition of “not competing on the merits” obtaining too broad a reach. If the definition of “not competing on the merits” is too broad, the prohibitions imposed by article 102 TFEU would apply to any firm that has a dominant position in the market, regardless of whether that firm’s behavior impedes competition in the market or not. By excessively limiting the space dominant firms have to manoeuver, smaller firms are given an unfair competitive advantage. In the end, this is not good for competition or innovation because even dominant companies must be able to compete on the market and have an incentive to invest resources in their innovative technologies.105 The use of an effects-based approach prevents dominant firms from being targeted indiscriminately by limiting the application of article 102 TFEU to only those dominant firms that negatively affect competition on their market. A shift from a form-based to a more effects-based approach implies that the Commission and the Court are to make a different determination of the facts as well. Rather than testing whether the behavior of a dominant firm falls under a pre-defined category of unlawful acts, it needs to be determined what the specific consequences of that behavior are for the degree of competition and consumer welfare on the relevant market. Such a determination is more exacting for the Commission and the Court, as more factors

96

Kroes 2005, p.2 Gormsen 2013, p.224; Geradin 2010, p.47 98 Geradin 2010, p.46 99 Ibid., p.47 100 Gormsen 2013, p.223 101 European Commission 2009b; Maggiolino 2011, p.75; Drexl 2009, pp.5–11 102 Geradin 2010, p.5; Monti 2010, p.3; European Commission 2009b, para.19 103 European Commission 2009b, para.19 104 Maggiolino 2011, p.74 and 75; Gormsen 2013, p.224 105 Waelbroeck 2010, pp.5 - 6 97

19

are to be taken into account and the analysis is not purely legal, but also of an economic nature. Still, if the behavior of the market relative to the real efficiency of conduct is used, one can see how such an approach is a well-thought out, well-motivated way of approaching competition on the merits. Besides the general advantages of a more effects-based approach in all areas of competition law, there is also a more specific -patent related- advantage. In analyzing potential abuse in the area of pharmaceutical patents, the difficulty is deciding when preventing others from putting a product on the market with a patent is competing on the merits and when it is not. Just as it is difficult to decide when a certain price used for a product is not competing on the merits or not normal behavior. The difficulty lies in the fact that most dominant company behavior is also behavior used by non-dominant companies. Both dominant and non-dominant undertakings may own and use property rights to prevent others from going on the market with a product, just like they may use competitive pricing arrangements. Therefore, whether or not the behavior of the dominant company is an abuse should not be decided by looking only at the behavior of the company but by looking at the “foreclosure effect of the behavior on the market relative to the real efficiency of conduct”.106 The advantage of this approach and using the two step test is that one can limit competition law reaching too far into the field of patent law and possibly hampering innovation.

3.4.4.2 Current practice with regards to the definition of abuse of a dominant position In Chapter 4, the recent cases in the pharmaceutical industry regarding article 102 TFEU are discussed. In the current paragraph, a general overview is provided of the way in which the Commission and the Court apply article 102 TFEU with regards to the abuse of a dominant position in other types of cases. Please keep in mind the recent Guidance Paper of the Commission and the fact that due to the Guidance Paper, the test of defining competition on the merits and abuse might change. However, for years a more form-based approach was common in EU competition law.107 The reason it is important to look at these “older” cases anyway is to fully understand the developments in the case law, signal certain difficulties and recognize the approaches used in the pharmaceutical cases discussed in Chapter 4. Furthermore, despite the fact that the Commission recognizes the importance of a more effects-based approach, their Guidance paper is not completely consistent in what approach to apply and sometimes leans back into the form-based approach.108 As mentioned earlier abuse can be defined using a more effects-based or form-based approach. At least until the 2009 Guidance Paper, the Court and the Commission had been using a more form-based approach. The ECJ merely requires proving the likelihood of harm to consumers for conduct to be seen as abuse instead of actual harm, the requirement for an infringement of 102 TFEU is that the conduct by the dominant undertaking “may” cause harm.109 Actual proof that the conduct causes harm, is not necessary. The goal of this approach is the prevention of damage to the market.110 Some conduct does not even require proving that conduct probably causes harm, but merely requires proving the risk of eliminating competition.111 Such conduct is always regarded as having exclusionary effects on the market.112

106

Waelbroeck 2010, p.5 For the development of the effects-based approach see for instance Gormsen 2013 108 Waelbroeck 2010, p.7 107

ECJ 15 March 2007 (British Airways/Commission), para.106; CFI 20 September 2003 (Michelin II); Anderman & Schmidt 2011, p.83 110 Anderman & Ezrachi 2011, p.20 111 CFI 24 March 2004 (Microsoft/Commission), para.620; Anderman & Schmidt 2011, p.84 112 Anderman & Ezrachi 2011, p.20 109

20

In order to make the analysis of what is and what is not competing on the merits a bit more concrete, the ECJ has given some guidelines. An abuse exists when there is a direct prejudice on consumers or the behavior indirectly prejudices consumers by diminishing competition.113 The abuse does not necessarily have to take place on the same market where the effect is felt, but can also take place in an associated market to the market where the effect is felt if the company is dominant in both.114 The abuse this research focuses on is the abuse of intellectual property rights. One of such abuses can, under certain circumstances, be the refusal to grant a license.115 In a case concerning cars, the additional circumstances required for a finding of abuse were described as “the arbitrary refusal to supply spare parts to independent repairers, the fixing of prices for spare parts at an unfair level or a decision no longer to produce spare parts for a particular model even though many cars of that model are still in circulation, provided that such conduct is liable to affect trade between Member States”.116 The word liable in this quote indicates that the approach used here is a more form-based one as liable is used by the Court to define a type of conduct as per se illegal. Another type of abuse which can be used by pharmaceutical patent holders is the abuse of government process. Under special circumstances, starting legal proceedings can be found to be a form of abuse. Under normal circumstances, access to a Court is a fundamental right and will therefore not constitute abuse. The only time such special circumstances exist is when Court proceedings are merely used to frustrate the other party and the proceedings are part of a plan to eliminate competition.117 Patent holders may use such proceedings to scare off generic producers. In another case, the General Court made clear that sometimes the acquisition of an exclusive license may constitute an abuse of a dominant position. The license in that case complied with the block exemption provision, but the acquiring of the license nonetheless constituted an abuse of a dominant position. The reason was that the license strengthened the dominance the acquirer of the license and had the effect of preventing or delaying the entry of other competitors into the market. Also, the market did not have a lot of competition.118 The above three examples from the case law of the Court illustrate the two-minded approach towards the requirement of abuse: the first two examples lean towards a form-based approach (“provided that such conduct is liable to affect trade between Member States”, penalizing the intention rather than the effect of a certain behavior), whereas the last example leans towards an effects-based approach (the effect of preventing or delaying market entry by competitors).

3.4.2 Objective Justifications More information on how to define “competition on the merits” can be gleaned from the exceptions an alleged infringer can bring up to state that his behavior does not constitute an abuse of his dominant position. Such exceptions are called objective justifications. These objective justifications interconnect with

113

ECJ 13 February 1979 (Hoffman-La Roche/Commission), para.122

114

ECJ 14 November 1996 (Tetra Pak International/Commission), para.21–33

115

ECJ 5 October 1988 (Volvo/Veng), para.8 and 9 Id., para.9

116 117

Schillemans & van Bockel 2011, para.5(g); CFI 17 July 1998 (Promedia/Commission), para.60

118

Whish & Bailey 2012, p.804; CFI 10 July 1990 (Tetra Pak International (BTG License)/Commission), para.23

21

the concept of competition on the merits in different ways. In order to understand their implications for the concept of competition on the merits and the way in which this concept is applied, it is important to first define objective justifications. In the United Brands case, the ECJ made clear that only in the absence of an objective justification can there exist an abuse under article 102 TFEU.119 Under the 2009 Guidance Paper, the objective justifications got an even clearer place. The Paper says that parties may prove to the Commission that their conduct is justified because the conduct is indispensable and proportionate for achieving an objectively necessary goal or the conduct is indispensable and proportionate to produce substantial efficiencies which outweigh the anti-competitive effects on consumers.120 The wordings used by the Court on the one hand and the Commission on the other hand, prompt the question whether they are in complete agreement with each other; the Commission lists two objective justifications, whereas the Court leaves room for more. It has been argued in the literature that there is room for at least one more objective justification than the two already mentioned by the Commission, namely that the behavior of the alleged infringer amounts to “competing on the merits”.121 In total, therefore, there would be three objective justifications: the behavior constitutes competing on the merits, the behavior should be allowed on grounds of public interest considerations, and the behavior should be allowed on the ground of efficiency considerations which also benefit the consumer. In my opinion, the first – competition on the merits – is not an objective justification, as the term “justification” implies that otherwise reproachable behavior is condoned for reasons other than those normally applied to test whether the behavior is allowed. The reason that this ground is considered to be an objective justification by some is most likely the result of the fact that the defense of “competing on the merits” is raised in the same way as one would raise one of the two objective justifications (see also further below in this chapter). Regardless of whether the fact of competing on the merits can be invoked as an objective justification or not, it would not help in coming to a more elaborate definition of “competing on the merits”; defining competition on the merits as “competing in such a way that one can invoke the objective justification of competing on the merits” would lead to circular logic. Therefore, only the two objective justifications identified by the Commission are discussed. The first of these objective justifications is that certain behavior which has an adverse effect upon competition in the market is nevertheless indispensable and proportionate for achieving an objectively necessary goal. Examples given by the Commission are that the behavior is “objectively necessary for health or safety reasons related to the nature of the product in question”.122 This objective justification has never been awarded in the case law, despite alleged infringers having argued the justification’s applicability. In the Hilti case, it was argued that an exclusionary practice for the nail guns was necessary for public health and safety reasons, and in the Tetra Pak case it was argued that interoperability with other products and processes was objectively necessary. The ECJ did not go along with either line of argumentation.123 It appears as though this ground for objective justification is not easy to meet: the Commission has stated that it is not up to individual businesses to set and enforce health and safety standards, as this is normally the task of public authorities.124 Steps taken by the alleged infringer to exclude products from the market which are believed to be dangerous or inferior than its own products are therefore not easily justifiable. ECJ 14 February 1978 (United Brands/Commission), p.189 European Commission 2009b, para.28 121 van der Vijver 2012, p.122 122 European Commission 2009b, p.C45/12 119 120

123

van der Vijver 2014, p.65 and 66; ECJ 12 December 1991 (Hilti AG/Commission), para.118; ECJ 14 November 1996

(Tetra Pak International/Commission) 124 European Commission 2009b, p.C-45/12

22

The second objective justification, the efficiency consideration, pertains to situations in which conduct leading to foreclosure of competitors can be justified on the ground that the efficiencies created through that conduct are sufficient to guarantee that no net harm to consumers is likely to arise and that no lessanti competitive behavior would result in the same efficiencies. As an additional requirement, the behavior is not to completely eradicate all competition on the market; other market players must remain.125 These factors merit some more attention. The efficiency consideration requires that the exclusionary effect is less important than the efficiency advantages it brings for consumers, as described by the ECJ in the British Airways case: “It has to be determined whether the exclusionary effect arising from such a system, which is disadvantageous for competition, may be counterbalanced, or outweighed, by advantages in terms of efficiency which also benefit the consumer. If the exclusionary effect of that system bears no relation to advantages for the market and consumers, or if it goes beyond what is necessary in order to attain those advantages, that system must be regarded as an abuse.”126 In the area of intellectual property rights such an efficiency consideration requires the innovation incentives outweighing the exclusionary effects. This line of argumentation was brought up by Microsoft, but the argument was not substantiated enough for it to be accepted by the Court.127 The efficiency consideration, at its core, requires that, in the end, consumers are better off as a result of the behavior. In a sense, this objective justification overlaps with the economic principles behind the “effects-based approach” that is increasingly taken by the Commission and the Court. Some differences remain, however, notably that the effects-based approach is used to determine what behavior is allowed (i.e. competition on the merits), whereas the objective justification on grounds of efficiency considerations is used to determine whether behavior that would otherwise not be allowed (i.e. not competition on the merits) can be justified. This is also the reason why the requirements for finding an objective justification of efficiency considerations are much higher, requiring that the behavior was indispensable and proportionate and that effective competition remains possible in the market. The fact that these criteria are very hard to meet and may not always be in line with the rationale of stimulating innovation128 would suggest that – when determining whether behavior falls under the header of ‘competing on the merits” or under the header of “not competing on the merits, but perhaps justifiable because of efficiency considerations” – the onus should be on the Commission to prove that it is the latter category, rather than the first one. This, however, does not follow from the Guidelines, as the presumptions and evidentiary burdens of proof are stacked exactly the other way around.129 As can be expected from reading the above, the objective justification on grounds of efficiency reasons is not easy to meet (in fact, neither of the two types of objective justifications has ever been awarded by the Court in the ambit of article 102 TFEU).130 This difficulty is exacerbated by the restrictions imposed by the Commission upon application of the justification. The guidance paper states that “exclusionary conduct which maintains, creates or strengthens a market position approaching that of a monopoly can normally not be justified on the grounds that it also creates efficiency gains”.131 The more dominant the firm,

125

European Commission 2009b, p.C-45/12

ECJ 15 March 2007 (British Airways/Commission), para.86 CFI 24 March 2004 (Microsoft/Commission), paras.1144–1166 128 Waelbroeck 2010, p.passim. 129 van der Vijver 2012, p.132; Waelbroeck 2010, p.8 130 Bailey 2010, p.363; Monti 2010, p.9 131 Waelbroeck 2010, para.30 126 127

23

therefore, the more difficulty the undertaking has in proving that an objective justification should apply on the basis of efficiency considerations.132 The burden for proving any objective justification lies with the alleged infringer: “Although the burden of proof of the existence of the circumstances that constitute an infringement of Article 82 EC is borne by the Commission, it is for the dominant undertaking concerned, and not for the Commission (…) to raise any plea of objective justification and to support it with arguments and evidence. It then falls to the Commission, where it proposes to make a finding of an abuse of a dominant position, to show that the arguments and evidence relied on by the undertaking cannot prevail and, accordingly, that the justification put forward cannot be accepted.”133 The effective use of any of the objective justifications has as a consequence that the burden of proof is shifted back to the Commission. The Commission then needs to prove that there is no objective justification.134 The objective justifications can therefore help in invoking a more effects-based approach when the Commission has neglected to clearly motivate their decision to define a type of abuse as contradictory to competing on the merits.

3.4.4 Problems with current status of abuse and objective justifications In the last two paragraphs on abuse and objective justifications, an attempt was made to sketch the current status in a neutral way. In order to provide a neutral description of the current situation, opinions describing possible problems were left out: those opinions are given now. In this paragraph, the issues with the current approaches are highlighted. These issues can be kept in the back of the mind when analyzing the pharmaceutical cases in Chapter 4 to ascertain whether the issues arise there as well. As we have seen, the Commission’s 2009 Guidance Paper was written to show that the Commission would start to use more of an effect-based approach to abuse and how this could be proven. Due to the fact that the Commission is not entirely consistent in using the effects-based approach one can wonder how effective the Guidance Paper is in pushing the “modern” approach to abuse.135 Take for instance how the Commission explains that there are still circumstances under which a detailed assessment is not necessary for deciding there is an abuse: a clear form-based approach.136 Or the lack of actual mentioning of consumer harm in the test for deciding on anticompetitive foreclosure.137 Besides such an explanation -or lack thereofcasting doubt on the actual worth in promoting the effects-based approach by the Guidance paper, it also causes a lack of legal certainty.138 A possible reason for this two-minded approach might be that the Commission wants to keep its flexibility in determining which behavior falls under competition on the merits and which behavior does not. Furthermore, the reason many people are very skeptical of the more effects-based approach actually being applied is the track record of the Court, the Commission and highly influential people surrounding them. Take for instance new Advocate General Kokott, who is a firm believer in orthodox (form-based) approaches.139 Furthermore, what is worrying is that the Court has been even more hesitant in including

132

Geradin 2010, para.52 CFI 24 March 2004 (Microsoft/Commission), para.1144 134 van der Vijver 2012, pp.3–5; CFI 24 maart 2004 (Microsoft/Commission), para.1144 135 Anderman & Schmidt 2011, p.85; Geradin 2010, p.64; Monti 2010, p.3; Waelbroeck 2010, p.7 136 European Commission 2009b, para.22 137 European Commission 2009b, para.20 138 Geradin 2010, p.53; Monti 2010, p.3 and 7 139 Geradin 2010, p.49; Anderman & Ezrachi 2011, p.21 and 22 133

24

economic considerations than the Commission in the analysis of article 102 TFEU (whereas this is the other way around for article 101 TFEU, which has caused the Commission to adopt a more economics-based approach with regards to the latter article).140 On the other hand, not all judges appear to be in favor of the more form-based approach, so those judges might now have an opening to start applying a more effectbased approach.141 As long as the Court does not lean towards a more effects-based approach, it is unlikely that the Commission will put in the effort to prove an infringement on an effects-based approach. After all, it cannot be seen why the Commission would adopt a complicated approach when it can adopt a simple one to convince the Court as a simpler approach requires less time and energy to investigate. Their only possible incentive to investigate the more effects-based approach in addition to the form-based approach is to leave the dominant firms with the impression that the case was thoroughly analyzed which prevents companies from going in appeal against Commission decisions 142 Obviously, this is just speculation, so time will tell if these problems will actually arise. Besides the remarks on the Guidance paper, some problems with the definition of competition on the merits must be highlighted as well. In 2005 the Organization for Economic Co-operation and Development (“OECD”) had a roundtable discussion on the question what competition on the merits is and how it should be defined, as they wished to clarify the ambiguity that arose due to the use of the term.143 According to the OECD, the discussion arises due to the fact that the phrase “competition on the merits” is used without further clarification on what it means, with the presumption that the meaning of the wording is common knowledge even though this common knowledge does not exist. The only clarity that exists is that some conduct by a dominant undertaking is not permissible, but what conduct specifically should be defined as abuse is not easily agreed upon due to the fact that the perimeter and the underlying principles of competition on the merits remain unclear.144 As a result, the decisions on article 102 TFEU and the national provisions similar to article 102 TFEU are inconsistent in condoning or forbidding certain behavior in the sphere of, for example, refusal to deal and the awarding of fidelity rebates or discount.145 The inconsistency in what is and what is not competing on the merits not only exists between jurisdictions, but also within jurisdictions.146 Take for example the policy change in the form-based and effects based approach. As we have seen, the Commission is moving towards a more effects-based approach in applying article 101 TFEU, while such a shift has yet to occur in practice with regards to article 102 TFEU.147 Therefore, the competition on the merits test does not meet the general standards of EU law which require that any test should be clear, precise and predictable in its effect so parties know what their right and obligations are under EU law. In other words, the competition on the merits test does not meet the principles of legal certainty and the rule of law.148 However, the problem of the ambiguity of the competition on the merits test only arises when the Court or the Commission merely say that a type of abuse is not competing on the merits, without further explanation or clarification and without looking into the effects on the market of the alleged abuse.

140

Waelbroeck 2010, p.3; Monti 2010, p.7 Monti 2010, p.8 142 Anderman & Ezrachi 2011, p.21; Geradin 2010, p.58 and 59 “this approach would hardly give any incentives to the Commission to carry out a serious effects-based analysis as in any event the case can be won on the basis of the strict caselaw of the ECJ. At best, the effects-based analysis is added to the decision to give the defendant the impression that the matter was thoroughly assessed by the Commission even if that thorough assessment was not necessary in the end.” 143 OECD Competition Committee 2005 144 Id., p.17 145 Id., p.17 and 18; OECD Competition Committee 2008, p.18 146 OECD Competition Committee 2005, p.17 and 18 147 Id., p.18; Waelbroeck 2010, p.3 148 Geradin 2014b, p.2; ECJ 7 June 2007, para.79; Waelbroeck 2010, p.10 141

25

Using an effects-based approach or some form of explanation why a certain conduct is not competing on the merits, can therefore solve the unclear definition problem. The vagueness of the concept of competition on the merits and how the Guidance Paper is to be applied can create three negative scenario’s which hamper competition and innovation.149 First of all, due to the vagueness and legal uncertainty of what constitutes competing on the merits, companies may become cautious in their conduct so as to prevent an infringement of article 102 TFEU.150 The fact is that the consequences of an infringement of article 102 TFEU can greatly decrease the profits of a company by imposing a fine and by bringing the company in disrepute due to the negative press. If companies are not using the full potential of measures at their disposal (even legal ones) for fear of being prosecuted by the Commission, this might hamper innovation and/or economic production. Second of all, due to the vagueness of what constitutes competing on the merits, the Commission can limit the meaning to include any behavior they want. Consequently, the scope of competition law can be broadened too much151, even to the point where, effectively, most behavior by dominant companies is prohibited (regardless of the behavior’s effect on the market).152 In a sense, this not only increases the notion of “abuse” within the meaning of competition law, but also the application of competition law altogether. The scope of intellectual property law, for instance, can be limited by competition law intervening in the legal effects of patenting due to competition law’s broadening scope; the positive effects for competition may in such cases easily be offset by the negative effects for innovation. Third of all, not knowing what is and what is not abuse makes it difficult for companies to defend their behavior under the form-based approach as most conduct is seen as per se abusive.153 Also, even if an effects-based approach is applied, the objective justifications set-out in the Guidance Paper have an unclear meaning, scope and use and a very high standard so one can wonder if they can even ever be met.154

3.4.5 Solutions Strictly speaking, the purpose of this research is not to find a solution to the problems which may arise out of the current definitions of abuse and competing on the merits. The focus is primarily on the question whether using these definitions is putting constraints on patent protection in the pharmaceutical industry. Yet, signaling some (possible) solutions to the problems is of importance in order to know if there are possibilities to do it differently. The possible solutions are not extensively researched here, but are pointed out for possible future research of what can be investigated further. First, the solutions which can be applied within the scope of article 102 TFEU are discussed. Secondly, the solutions outside the scope of article 102 TFEU or even competition law are looked into.

3.4.5.1 Solutions within scope article 102 TFEU The main cause of the problems is the uncertainty both the concept of competition on the merits and the Guidance Paper have brought. It is not clear whether or not a more effects-based approach will be used in the future. The simple solution is that the Courts and the Commission start applying it in practice. Furthermore, the Court could acknowledge the more effects-based approach by specifically mentioning the 149

Monti 2010, p.7; Geradin 2014b, p.3 and 8; Waelbroeck 2010, p.2 Geradin 2014b, p.3; Waelbroeck 2010, p.7 151 Geradin 2014b, p.8 152 Waelbroeck 2010, p.2 153 Gormsen 2013, p.234 154 van der Vijver 2014, p.56; Monti 2010, p.9 150

26

Guidance Paper and its importance in a case. A more effects based approach is already used in applying 101 TFEU so there is no reason for article 102 TFEU to stay behind. In developing a more effects-based approach, the US consumer harm approach can be used as inspiration. Under Article 2 of the Sherman Act, the American version of article 102 TFEU, the showing of harm to consumers is necessary for concluding there is an abuse.155 In trying to find a more effects-based approach one should be tentative to the fact that a complete effects-based approach is not advisable either. The Commission or Court cannot always know and research the preferences consumers have.156 In that light, a wiser approach is to look at consumer choice, particularly in innovation related markets where the future is difficult to predict.157 Furthermore, the Sherman Act approach cannot be adopted completely, as the EU and US approaches to competition law are very different.158 In the US, the competition authorities do not enforce as actively as the competition authorities do in the EU.159 The tests to be used for a more effects-based approach can draw inspiration from the work done by the OECD. The OECD looked into four different economic tests for defining competition on the merits, brought forward by economists. Sadly, the OECD had to conclude that not one of the tests gave a one-size-fits-all solution, but each test does have advantages in different areas.160 The circumstances of the case should point towards a specific test. The first test is the profit sacrifice test, which is only helpful in the field of predatory pricing conduct.161 The test finds conduct unlawful when the conduct sacrifices profit in a way that would be irrational if the conduct did not have the tendency to eliminate or reduce competition. 162 The second test is the no economic sense test. The problem with this test is that not-licensing intellectual property right becomes an abuse very quickly, which limits the incentive to innovate.163 The question the test poses is if the conduct eliminates or reduces competition which causes economic benefit to the company merely because of that specific conduct.164 The third test is the equally efficient firm test or efficient competitor test.165 This test has also been adopted in the Guidance Paper for price-based exclusionary conduct.166 The test approaches the market as a fair place where you can only eliminate equally strong competitors with unfair or bad conduct.167 The difficulty with the test is that it is based on the probability that the conduct will cause problems instead of facts and is therefore more orthodox than it would seem at first sight.168 The fourth and final test is the consumer welfare balancing test. It analyses whether the conduct is unlawful in looking at both the positive and negative effects on consumer welfare of the conduct.169 The test is quite unclear in its application due to the fact that nobody knows over what period of time the advantages and disadvantages of the test must be measured. Also, when balancing both

155

Drexl 2009, p.2 and 3; Larouche & Schinkel 2013, p.7 Id, p.3 157 Id., pp.14, 31 158 See for a thorough analysis of the differences: Larouche & Schinkel 2013 159 Id., p.6 160 OECD Competition Committee 2005, p.10 161 Id., pp.19, 24–27 162 Id., p.25 163 Id., p.29 164 Id., p.28 165 Id., p.29 166 European Commission 2009b, para.23–27; Gormsen 2013, para.225 167 OECD Competition Committee 2005, p.29 168 Gormsen 2013, p.225 169 OECD Competition Committee 2005, pp.31 - 32 156

27

positive and negative consequences it becomes difficult to be objective and consistent, which causes legal uncertainty.170 The test therefore needs to be further clarified in order for it to be used. Besides the different tests, the OECD also gave some other advice which can help create more certainty. One of these is to also publish the motivation for not imposing a fine in cases where the Commission, after doing research into an undertaking, decides that there is not an infringement of article 102 TFEU. Publishing that motivation can help understand what competition on the merits is. The Irish competition authorities already use this approach and can be used as a model.171 Another solution can be found in broadening the market definition. After all, we saw in para. 3.2 that using a broader market definition which also includes the innovation market, causes the Commission to put their focus on the competitors and the possibilities those competitors have to enter the market. Such an approach leads to the Commission putting less of their focus on consumers. The advantage of using a broader market definition is that a patent holder is not considered dominant too quickly on an innovation market as patents leave the possibility to innovate open. If there is no dominance, then an abuse cannot be proven either, as dominance is a necessary requirement for liability under article 102 TFEU.

3.4.5.2 Solutions outside scope article 102 TFEU Sometimes competition law involves itself with abuse for which tailor-made solutions have been developed in other fields of law. In certain instances, such competition law conduct relates to intellectual property rights abuse where the intellectual property legislation itself could also be used to solve the problem, for example through the use of compulsory licenses.172 In other instances a dominant company might be badmouthing other producers via the press, which could be solved with fair trade legislation.173 In my opinion it would be better if the Commission or the Court would not intervene when a more specific piece of regulation outside the scope of competition law can be used to fix the problem, or at the very least see to it that that specific solution is tried before applying competition law. Getting involved undermines those other pieces of legislation and causes confusion for companies who do not realize that multiple public organizations can start investigations into them for the same conduct. Additionally, competition authorities may decrease some of the pressure on the competition law system by allowing private law solutions (such as the ones mentioned above) to the issues they face. It is true that it is easier and cheaper for companies to complain to the Commission than it is for them to start their own proceedings, but ease and cheapness on the side of companies are not what the competition law system is designed for.

3.4 Conclusion In regard to article 101 TFEU, the Court and the Commission have been moving towards a more effectsbased approach for a long time. It seems as though the approach to article 102 TFEU and its abuse is also moving in that direction. However, there is still some uncertainty as to the way the effects-based approach will be applied and if it actually will be applied at all. As a result, legal uncertainty arises and with that a possible constraint on innovation. After all, companies may become cautious in the conduct they apply to prevent an infringement of article 102 TFEU. Also, the Commission can limit the meaning of competition on the merits to include any behavior they want and with that, broaden the scope of competition law too much and possibly put constraints on patent protection. Lastly, it is difficult for companies to defend their behavior. 170

Id., p.33 Id., p.11 172 Anderman & Ezrachi 2011, pp.24-25 173 Geradin 2014b, p.8 171

28

The solutions to these problems are plenty, but they do need to be researched further. Inspiration can be taken from article 2 of the Sherman Act and from the OECD roundtable advice. However, arguably the best solution is if both the Court and the Commission start using a more effects-based approach in practice and motivate their finding of abuse thoroughly. In addition, the market definition can be adjusted a little in innovation sensitive markets by also including the innovation market in the market definition. Of course the more effects-based approach does have some limits as well which need to be taken into account. A more effects-based approach takes more time, because an economic analysis is necessary. Also, not all abuse and their effects on the markets is easily analyzed. The difficulty in analyzing the effects on the market may cause legal uncertainty because firms might not know what the effects of their conduct on the market are and if the effects are negative. Still, using a more effects-based approach trumps the use of a form based approach. At this moment, the form-based approach is very unclear because what conduct is per se abusive keeps broadening. A possible solution could be to make an extensive list of abuse, that way legal certainty can be given. The side-effects of such legal certainty however, is that flexibility is lost. Conduct which does not harm the market and might actually stimulate competition may be abusive under the formbased approach. Furthermore, new types of conduct which were not part of the “abusive conduct list”, but that do harm competition, can then not be qualified as abusive. Therefore, it would be better for the Commission to show what economic analysis they use to find abusive conduct under a more effects-based approach. That way companies can themselves analyze their conduct. Please take note of the fact that I am not proposing a “full-on” effects-based approach but the use of a “more” effects-based approach because, as was found in the OECD roundtable advice, a full-on effects based approach causes too much legal uncertainty and is too difficult to apply in markets. The economic tests to be used and yardsticks in applying a more effects-based approach are precisely what needs to be further researched. Besides clarifying the concept of abuse and competition on the merits, the Commission can also stop interfering in fields where there is field-specific legislation which lies outside the field of competition law. In the next chapter we look into the question of whether the above problems also arise in the field of pharmaceutical patents and what possible solutions might be.

29

4. Analysis of cases 4.1 Introduction The tension between competition law and patent law was described in Chapter 2. In Chapter 3 an overview was given of the way in which ‘abuse of a dominant position’ and ‘competition on the merits’ are defined, as well as the problems arising from the ambiguity of these definitions. In the current chapter the knowledge obtained from those earlier chapters is used to analyze the case law of article 102 TFEU at a European and national level in regard to pharmaceutical patents. This chapter focuses on the concept of ‘abuse of a dominant position’ and the ‘competition on the merits’ test used in the different cases. All cases that are analyzed here were decided after the publishing of the Commission’s Guidance Paper in 2009 and after the Commission announced their investigations into the originator pharmaceutical companies who supposedly delay the arrival of generic medicines. The reason not only European but also national cases are looked into is because the national courts and national authorities tend to follow the example set at the European level. First the European AstraZeneca174 case is discussed; it is discussed more elaborately than the national cases, since it is the only relevant European case. After this analysis the national cases of Reckitt Benckiser175, Sanofi-Aventis176 and Pfizer177 are analyzed in consecutive order. When analyzing the cases only the summary of the facts, grounds of appeal and findings of the Court that are relevant for this research are discussed. For each case, these items are followed by an analysis.

4.2 AstraZeneca 4.2.1 Introduction AstraZeneca is the producer of a medicine called “Losec”, which is the brand name for tablets and capsules which contain the active ingredient “Omeprazole”. Omeprazole has the effect of slowing down or preventing the production of stomach-acid. Although there are more medicines which treat stomach-acid related problems, Losec was the first medicine that acted directly on the enzyme in the stomach wall that pumps acid into the stomach (the so-called proton pump).178 The Commission accused AstraZeneca of infringing article 102 TFEU by preventing or delaying the entry of generic versions of Losec on the market by abusing the patent systems and the procedures for marketing pharmaceutical products.179 The case was first decided by the General Court, after which the ECJ made the final decision on appeal of AstraZeneca, the Commission and the European Federation of Pharmaceutical Industries and Associations (“EFPIA”) in 2012.

4.2.2 Relevant Product Market First of all, AstraZeneca did not agree with how the General Court had defined the relevant product market. The General Court described the relevant market as consisting merely of proton pump inhibitors (“PPIs”). Other medicines which could also be used for treating stomach-acid related problems, such as H2 blockers (which are less effective but also have less side-effects), were not found to be part of the relevant market.180

174

General Court 1 July 2010; ECJ 6 December 2012 (AstraZeneca/Commission) OFT 12 April 2011 (Reckitt Benckiser) 176 Autorité de La Concurrence, 14 May 2013 (Sanofi/Aventis) 177 Consiglio di Stato 12 February 2014 (Pfizer) 178 ECJ 6 December 2012 (AstraZeneca/Commission), para.15 179 Id., para.1 180 General Court 1 July 2010, para.28–222; ECJ 6 December 2012 (AstraZeneca/Commission), para.27 175

30

AstraZeneca was of the opinion that the General Court had made a mistake by finding AstraZeneca’s argument irrelevant that “the graduate [sic] increase in the sales of PPIs at the expense of those of H2 blockers meant that the latter exercised a significant competitive constraint over PPIs and, therefore, that H2 blockers should, for that reason, be included in the product market at issue”.181 AstraZeneca’s argument, which is not entirely clear from the above quote, was that PPI’s would have taken over the market much more rapidly if H2 blockers were strictly inferior; since the increased use of PPIs only came about gradually, H2 blockers were a rivalling product and should thus be included in the product market definition. The ECJ, however, decided that the General Court had rightfully rejected AstraZeneca’s argument, as the slow increase in sales could also be accounted to the fact that doctors were hesitant to use the PPIs due to possible carcinogenic (cancer-inducing) effects.182 Even if other methods were used to define the market, the Court would not come to a different conclusion. One such method would be to focus more on the demand side instead of the separate categories of stomach-acid problems. For example, if the PPIs were cheaper, would they then be prescribed in all cases? Or if the H2 blockers were not available due to delivery problems, would a doctor then prescribe PPIs? If a person no longer has PPIs in his household, would he then use H2-blockers similar to how we trade Paracetemol with Ibuprofen to treat a headache when the Paracetemol is out, even though Ibuprofen is meant for heavier headaches? Looking at the case of the General Court, these are precisely the type of questions that the Commission had answered. The Commission had looked at whether the price mattered. However, “since doctors and patients do not bear the bulk of the cost of prescription medicines, doctors are usually only slightly sensitive to the price of medicines when prescribing them. During the relevant period, prescribing doctors were primarily guided by the therapeutic appropriateness and effectiveness of medicines rather than by their price.”183 Also, the IMS Health data showed specifically what the PPIs and H2 blockers were used for. IMS health is the world’s leading information, services and technology company to make healthcare perform better.184 The IMS health data showed that the PPIs were used for the treatment of conditions that H2-blockers were not prescribed for, because the H2-blockers were not seen as appropriate or effective for treating those conditions.185 Furthermore, H2-blockers and PPIs were prescribed at separate stages of a treatment, depending on whether a more heavy medicine was necessary or not.186 Therefore, the narrow product market definition is the result of the fact that the market for PPIs is, in fact, very narrow. The burden of proof for using a different product market definition had therefore shifted to AstraZeneca, meaning that AstraZeneca had to prove that the slow increase in sales was actually the result of the H2 blockers competing for market share with the PPIs.187 In conclusion, AstraZeneca had to demonstrate that the market definition should include treatment for both intense forms of stomach-acid (PPIs) and less serious cases (H2 blockers).188

181

ECJ 6 December 2012 (AstraZeneca/Commission), para.39 Id., para.48 183 General Court 1 July 2010 (AstraZeneca/Commission), para.34 184 IMS Health 2014. The General Court is referring to the data the IMS Health had collected specifically for the AstraZeneca case and not the previous mentioned IMS Health case. 185 General Court 1 July 2010 (AstraZeneca/Commission), para.51 186 Id., para.52 187 ECJ 6 December 2012, C 457/10 (AstraZeneca/Commission), paras.46–49 188 Id., paras.41, 47 182

31

4.2.3 Abuse of a Dominant Position using SPCs Second, AstraZeneca did not agree with the decision that they were abusing their dominant position by giving misleading information to obtain a supplementary protection certificate (SPC).189 The first ground for this appeal was that the definition of “competition on the merits” used by the General Court was too narrow.190 Due to a lack of transparency in the legislation and how to interpret it AstraZeneca had interpreted the legislation on SPC applications –Article 19 of Regulation No 1768/92- in such a way that not the date of the first technical authorization of Losec had to be provided to the national patent offices, but the date of the publication of the prices of Losec.191 The latter is the date AstraZeneca provided to the national patent offices in many different European countries as the starting date of the SPC applications, while it should have provided the date of the first technical authorization.192 If AstraZeneca had used the first date, they would not have obtained an SPC in Germany and Denmark -a fact of which they were completely aware- so AstraZeneca used the wrong date in SPC filings all over Europe.193 The ECJ started its judgment by reaffirming that Article 102 TFEU “prohibits a dominant undertaking from eliminating a competitor and thereby strengthening its position by using methods other than those which come within the scope of competition on the merits”.194 The ECJ then went on to observe that AstraZeneca had not used the date of publication of the prices for six of its SPC applications regarding other medicines. Only for the medicines containing Omeprazole had they used the wrong date, which meant that AstraZeneca was fully aware of the fact that they were probably interpreting the legislation in the wrong way or that AstraZeneca was purposely exploiting the legal vagueness of the legislation to its own advantage.195 Nevertheless, AstraZeneca chose to use the wrong date without explaining how that date was established.196 Therefore, AstraZeneca knowingly accepted the fact that the SPCs would not have been granted if AstraZeneca had given the right date if their alternative interpretation would not be accepted.197 The ECJ therefore held that the General Court was right in deciding that “the submission to the public authorities of misleading information liable to lead them into error and therefore to make possible the grant of an exclusive right, such as the SPC, to which the undertaking is in actual fact not entitled, or to which it is only entitled for a shorter period, constitutes a practice falling outside the scope of competition on the merits and therefore an abuse of a dominant position.”198 The second ground of AstraZeneca’s appeal against the decision that their conduct constituted abuse was based on the argumentation that the mere fact of applying for an SPC was not sufficient to constitute an abuse. The Court should have also looked at the effect of the impugned conduct on competition. 199 If the General Court had looked at the negative effects, they would have come to the conclusion that there were none, as all but two of the SPC’s were either:  granted under the correct date;  not granted; 189

ECJ 6 December 2012, C 457/10 (AstraZeneca/Commission), para.62; General Court 1 July 2010, paras.352–382 For a description of SPCs see Chapter 2, para. 2.3.3. 190 ECJ 6 December 2012, C 457/10 (AstraZeneca/Commission), para.76 191Id., para.70 192Id., paras.78–85 193 Id., paras.65 and 78–81 194 Id., para.75 195 Id., para.80 196 Id., para.82 197 Id., para.96 198 General Court 1 July 2010, para.355–361; ECJ 6 December 2012, C 457/10 (AstraZeneca/Commission), paras.62 and 100 199 ECJ 6 December 2012, C 457/10 (AstraZeneca/Commission), para.101

32

 withdrawn before the patent had expired and the SPC could have had an effect on the market or;  revoked before the patent had expired and the SPC could have had an effect on the market.200 The two SPCs that gave additional protection to AstraZeneca for a five to six months period had no proven effect of restricting competition on the market. Also, AstraZeneca did not have a dominant market position during those months.201 The EFPIA supported this appeal of AstraZeneca.202 The ECJ, however, rejected the appeal as unfounded. It held that the General Court was right to decide that giving false information to obtain an SPC is an abuse if the false statements “are liable to lead” a patent authority to grant the SPC.203 The Court held that the unlawful application for SPCs by dominant players leads “to a significant exclusionary effect after the expiry of the basic patents, but they are also liable to alter the structure of the market by adversely affecting potential competition even before that expiry.”204 The fact that AstraZeneca was not actually capable of using the SPCs in most countries did not matter, because “the existence of an abuse is not affected by the fact that that strategy did not succeed in some countries”.205 Neither does it matter that AstraZeneca was no longer dominant at the time they wanted to use the SPCs still in place in the Netherlands and Belgium after the expiry of the patent as the conduct is measures to the time and place of the conduct.206

4.2.4 Abuse of a Dominant Position by deregistering MAs Third, AstraZeneca appealed against the second type of abuse that the General Court and the Commission had found them guilty of. AstraZeneca had deregistered the Marketing Authorizations (‘MAs’) for the Losec Capsules. Once a MA has been granted, other companies can use that same MA to bring similar medicinal products on the market without having to do their own clinical testing. The deregistration by AstraZeneca of its MA precluded or delayed other manufacturers from entering the market, as these manufacturers would have to fund and execute their own clinical tests.207 The Commission and the General Court had labelled this behavior on part of AstraZeneca as abuse of a dominant position. AstraZeneca argued that, in order to determine whether its behavior in fact constituted abuse, the test developed in the IMS Health case was supposed to be used. The test in the IMS Health case was developed to determine whether a compulsory license should be awarded to competitors of a patent holder and constitutes that “the mere exercise of a right lawfully afforded by Union law could at the most amount to an ‘abuse’ only in exceptional circumstances, namely where there is an elimination of all effective competition, a mere propensity to distort competition not being sufficient for that purpose.”208 The ECJ did not follow AstraZeneca’s line of argumentation because the IMS Health Case and the AstraZeneca case related to very different conduct and are in no way comparable.209

4.2.5 Analysis The approach used by the ECJ to define the market shows the narrow market definition which both the Commission and the Courts use.210 Both PPIs and H2 blockers treat stomach-acid problems; the first is just 200

Id., paras.101 and 102 Id., para.102 202 Id., para.101 203 General Court 1 July 2010, para.377; ECJ 6 December 2012, C 457/10 (AstraZeneca/Commission), para.106–113 204 ECJ 6 December 2012, C 457/10 (AstraZeneca/Commission), para.108 205 Id., paras.109 and 111 206 Id., para.110 207 Id., paras.117 and 118 208 Id., para.142 209 Id., paras.148–156 210 See para. 3.2 201

33

used for more severe forms than the other. The market definition used by the Court is based on treating the more severe form of stomach-acid problems separately. In this case, the ECJ could and should not have used a different approach to change the outcome of the relevant product market; looking at the demand side of the market, it leads to the same outcome. As we saw in chapter 3, the result of defining the product market as very narrow is that a finding of dominance is achieved more easily. Based on the facts and the ECJ’s considerations, AstraZeneca was fully aware of the fact that their interpretation of the law regarding the starting date of the SPC was false. If AstraZeneca had used the wrong date for all of its SPC filings and if it had not been so clear from their internal memoranda that they knew they would otherwise lose their SPCs, then AstraZeneca's behavior might have been excused. In my opinion, the ECJ made the right decision in rejecting the first ground of appeal against the finding of abuse of a dominant position. Lying to get an extension on your intellectual property right protection by knowingly giving false information to obtain the SPCs is in contravention with the law and under no circumstances competing on the merits. Furthermore, AstraZeneca had argued that the effect of the impugned conduct on competition should also have been taken into account. By rejecting that specific argument, the ECJ is saying that the fact that the bad behavior did not have the desired effect due to actions which lay outside AstraZeneca’s power, cannot be used to save them from being fined. After all, if their behavior would have succeeded then the competitive market structure would have, without a doubt, been affected negatively. This line of argumentation lies somewhere in between the effects-based and form-based approach. The negative effect did not get the chance to manifest itself, but it was necessary for the finding of abuse that the effect would have manifested itself if it weren’t for the intervention of third parties. Lying to obtain an SPC would only have constituted abuse if the duration of the protection of Omeprazole had been extended by obtaining the SPCs due to the lie. In using this approach, the ECJ is taking an approach similar to the approach used in criminal law: criminalization of attempted crimes. Attempts to commit crimes are criminalized for the deterrent effect the criminalization has, for the retribution of the attempted harm done to the legal order and to be able to intervene without destroying the possibility of prosecution.211 Under Dutch criminal law, as in many other national legal orders, the penalty given for attempted crimes is lower than the penalty for completed crimes because the shock to the legal order is smaller for an attempt to commit a crime than it is for the completion of the crime.212 Using that line of argumentation, AstraZeneca’s behavior should be penalized to deter other companies from lying as well in an attempt to get longer protection duration. After all, if there are no negative consequences for trying, why not make the attempt? Even though AstraZeneca should be fined, it would also be logical if the fine was lower because the market was not actually harmed. AstraZeneca attempted to raise this argument by stating that the absence of the anti-competitive effects should have been used to lower the fine.213 Regretfully, the ECJ decided that AstraZeneca “cannot take advantage, in the context of the calculation of the fine, of the fact that, thanks to the intervention of a third party, their highly anti-competitive conduct, which was likely to have a significant effect on competition, did not always produce the effects expected.”214 Personally, I do not agree with this line of argumentation of the ECJ. The fine should be based on the conduct, the possible effects of the conduct and the actual effects of the conduct. A differentiation should be made between a completed form of abuse and an incomplete form of abuse. AstraZeneca was not capable of profiting from their abusive conduct and the market was not actually harmed, which should be detracted from the fine as part of the fine is also to retrieve back the 211

de Hullu 2009, p.377 Id., p.378 213 ECJ 6 December 2012, C 457/10 (AstraZeneca/Commission), para.160 214 Id., para.165 212

34

profits they made with the bad conduct. The ECJ is putting abuse which has actual negative effects on the market on the same level as abuse which does not have those negative effects. Furthermore, it is not certain that AstraZeneca would have actually used the SPCs. Currently; AstraZeneca is penalized for a crime that did not happen or may not even have been committed. In order to illustrate the difference between the actual commission of a crime and the potential commission of a crime, consider the difference between a person committing a murder and a person who is arrested by the police after buying a weapon. In the first case, the crime has already been committed, whereas in the second case it is all but sure that the same crime (killing a person) would eventually be committed; the potential perpetrator might change his mind before executing the deed. Therefore, the punishments inflicted for the first crime is larger than for the second crime. The ECJ in the AstraZeneca case finds that both are on the same level because to the ECJ it does not matter that the conduct failed to achieve the desired effect or that AstraZeneca might have decided to not use the SPCs as there was still time to withdraw them. Although the comparison is not 100% accurate, it does show to illustrate that there may be circumstances in which the level of punishment may be dependent upon whether a perpetrator actually carries out the deed all the way to the final result. So even though I agree with defining AstraZeneca’s conduct as abusive, I also find that the fine should have been lowered. Lastly, the ECJ made clear that the dominance of the undertaking should be measured at the moment of the conduct and not at the moment the conduct would have its effect. Therefore, the ECJ is using an ex tunc approach in defining dominance instead of an ex nunc approach. I find it logical to take the moment of the abusive conduct to measure conduct, as the abuse should be done by a dominant undertaking. Switching to an ex nunc approach could mean that conduct done while a company was not dominant becomes abusive if the company becomes dominant in the market between the moment it conducts the behavior and the moment that behavior has its effect. As a result, conduct which was within the meaning of “competition of the merits” when it was conducted could unintentionally become abusive conduct.

4.3 Sanofi-Aventis In 2013, two very similar cases were decided by the French Competition Authority (“FCA”): the SanofiAventis case and the Schering-Plough case.215 As the facts and motivation were very similar in both cases, only the Sanofi-Aventis case is discussed. After all, the purpose of this research is to investigate how the competition of the merits test is used; the Schering-Plough case did not change the approach used by the FCA in Sanofi-Aventis and is therefore left undiscussed. Sanofi-Aventis (“SA”) is the sixth largest pharmaceutical company in the world, the largest in France, and the producer of Plavix.216 Plavix is SA’s registered trademark for a tablet-form medicine containing Clopidogrel.217 Clopidogrel is used to prevent blood clots and is the fourth most prescribed drug in the world. Therefore, Plavix is also known as a blockbuster.218 After the expiry of SA’s patent on Clopidogrel, the first generic Clopidogrel versions were put on the market by competitors of SA in 2009.219 For the purposes of this research, generic versions produced by SA containing Clopidogrel are called generic Plavix, generic versions produced by competitors of SA containing Clopidogrel are called generic Clopidogrel.

215

FCA 18 December 2013; Autorité de La Concurrence 14 May 2013 (Sanofi-Aventis/Autorité de la concurrence) Autorité de La Concurrence 14 May 2013 (Sanofi-Aventis/Autorité de la concurrence), paras.17–21. 217 Id., para.26 218 Id., paras.2, 27 and 43 219 Id., paras.38 –42 216

35

In 2013, the French Competition Authority (“FCA”) fined SA with 40.6 million Euros. The FCA found SA had infringed Article L.420-2 of the French Commercial code (the French version of article 102 TFEU) and Article 102 TFEU. The abuse consisted of implementing a practice where SA purposely gave generic Clopidogrel producers a bad name on the French Clopidogrel market by exaggerating differences between generic Clopidogrel and Plavix and by making up possible shortcomings of generic Clopidogrel.220 In order to define the relevant product market, the FCA held that the perception of prescribers should be used to define the relevant market.221 In accordance with ECJ case law and national French case law, the perception of prescribers is found using ATC Levels. ATC is the anatomical therapeutic chemical classification system to classify drugs using their therapeutic and chemical characteristics. Each ATC level looks at a different aspect of the drug, ATC level 3 for instance looks at the therapeutic sublevel the pharmaceutical product is a part of.222 Here, ATC level 5 was best suitable, which described the active ingredient of Plavix being Clopidogrel, as in the perception of prescriber’s only medicines containing Clopidogrel could replace Plavix. All medicines containing Clopidogrel were interchangeable, as they all contain the same active ingredient and are each other’s efficacy and safety equivalent. There are no other products with that effect: despite the high price of Plavix, prescribers only wanted to prescribe medicines containing Clopidogrel.223 Therefore, the relevant product market was all medicines containing Clopidogrel.224 In regard to the dominance of SA on the market, SA had a dominant position during the time they were holding the patent as they were then in a monopoly position.225 After the expiration of the patent SA also has a dominant position because their market share remained over 50%.226 Lastly, in regard to the question of abuse, the FCA looked into the conduct of SA. After the SA’s patents on Clopidogrel expired, SA wanted to bring her own generic (cheaper) version of Plavix on the market. In order to pave the way for generic Plavix, SA launched a strategy to communicate with health professionals to get them introduced to generic Plavix.227 In SA’s first public statement on the introduction of generic Plavix -in May 2009- SA highlighted the differences between generic Plavix and generic Clopidogrel. Generic Plavix used a different type of salt than generic Clopidogrel and according to SA the latter type of salt did not have the same established profile of efficacy and safety.228 Later that year, the CEO of SA gave an interview for Quotidien du Médicin, the premier French news source for medical professionals, in which he told that even though the salts used in SAs generic Plavix version were different from the generic Clopidogrel versions of other companies, the European and French agencies gave authorization to market those generic

220

Id., para.120 Id., para.282; World Health Organization 2003, para.5.2 222 See para. 3.2 on the explanation of ATC levels and its use in market definitions; “La pratique décisionnelle et la jurisprudence, tant interne qu’européenne, s’appuient à cet égard sur le système de classification « Anatomical Therapeutical Chemical » (ci- après la « classification ATC ») reconnue et utilisée par l’Organisation mondiale de la santé pour définir les marchés pertinents [voir, par exemple, l’arrêt de la Cour de cassation du 15 juin 1999, précité, et la décision C (2005) 1757 final de la Commission européenne, du 15 juin 2005, relative à une procédure d’application de l’article 82 CE et de l’article 54 de l’accord EEE (affaire COMP/A.37.507/F3 – AstraZeneca)]. Ce système classe les produits pharmaceutiques en fonction de leurs indications thérapeutiques. La classification se décline en « niveaux ».” Autorité de La Concurrence 14 May 2013, 13-D-11 (Sanofi-Aventis/Autorité de la concurrence), para.289 223 Id., paras.296 and 297 224 Id., paras.299–301 225 Id., paras.317–320 226 Id., paras.321–324 227 European Commission 2009a, para.98 228 Autorité de La Concurrence 14 May 2013 (Sanofi-Aventis/Autorité de la concurrence), para.107 221

36

Clopidogrel versions.229 He stated that even though authorizations were given out, one should realize that the differences between generic Clopidogrel and generic Plavix are an objective fact. Generic Clopidogrel uses different salts than Plavix and does not have the same history as Plavix in dealing with acute coronary syndrome. Furthermore, SA’s CEO emphasized that despite the Medicines Agency (L’Agence du Médicament) not finding a difference so far, the fact remained that generic Plavix is produced in the same way on the same sites as original Plavix, which has already proven to be effective, while generic Clopidogrel does not have that same history.230 Furthermore, SA gave out information to “les visiteurs médicaux” (pharmacies and other health professionals, such as doctors) on the market entry of generic Plavix and generic Clopidogrel.231 Les visiteurs médicaux provide information on medicines to healthcare professionals.232 Once again, the difference between the salts used in the medicines was highlighted. In addition, SA advised against prescribing generic Clopidogrel due to the different salts used. Furthermore, SA advised to add the letters “NS” (non-substitutable) to the order of generic Plavix if the medicine was to be used in combination with aspirin to prevent the risk of bleeding. Generic Clopidogrel did not have the same established practice as Plavix in preventing bleeding when used in combination with aspirin.233 The message sent by SA was false or a strongly exaggerated highlighting of the differences. The different use of salts does not matter, as the salts do not have significant different properties in regard to safety or effectiveness. Therefore, according to the FCA, there was no need or use to stress the difference between the two salts or their possible different effects.234 Furthermore, the FCA looked into the effects of the conduct by SA. After the market entry of the Plavix and Clopidogrel generics, it took a lot longer for the orders of generic Clopidogrel to go up than is normally the case after the introduction of generics: the market share of the brand Plavix stayed more or less the same due to the sale of generic Plavix.235 The stagnation in Clopidogrel generic growth was the result of the NS label used, as no practitioner or pharmacy wanted to take the risk being held liable for ignoring such a label.236 Furthermore, Winthrop -the generic brand of SA- had established a very strong market segment in the area of Clopidogrel medicines, which is also atypical.237 Winthrop generic Clopidogrel had a four times larger market share in the area of Clopidogrel than it had in any other generic market.238 Based on the above, the FCA came to the conclusion that SA had infringed Article 102 TFEU (and Article L.420.2 French Commercial Code) by using conduct that is not considered competing on the merits. 239 Health professionals are very careful in prescribing (generic) medicines as they want to prevent liability for

229

Id., para.108 “Il ne s’agissait pas de polémique, mais d’un fait objectif: les génériques du clopidogrel, qui sont des sels différents de ceux de Plavix, n’ont pas les indications de Plavix dans le syndrome coronaire aigu. L’Agence du médicament a considéré que pour autant, on ne pouvait retenir de différenciation. Ce que nous disons simplement, c’est que le générique que nous lançons est fabriqué exactement sur les mêmes sites que le Plavix et qu’il a démontré son efficacité dans toutes les indications.” Id., para.109 231 Id., paras.117 and 118 232 Institut Superieur des visiteurs medicaux 2014 233 Autorité de La Concurrence 14 May 2013 (Sanofi-Aventis/Autorité de la concurrence), paras.118–120 234 Id., paras.113–115 and 130–174 235 Id., paras.223–238 236 Id., paras.239–249 237 Id., paras.250–254 238 Id., para.252 239 Id., paras.360 and 584 230

37

prescribing patent infringing generics or causing health issues by prescribing a certain medicine.240 SA abused their cautiousness by delivering partly false information.241 The FCA decided that it is perfectly fine to illustrate the differences between two pharmaceutical products and to highlight the objective qualities of a medicine. However, the conditions mentioned in this case are not to be considered as such substantial differences that raising doubt over the objective qualities of the competing generic products is permitted under the pretention that SA is protecting medical practitioners against making a mistake.242

4.3.1 Analysis I agree for the largest part with the approach the FCA used. When defining the relevant market, the FCA looked further than ATC Level 3 to make a more fitting product market definition. As discussed in para. 3.2, also taking into account the demand side is the fairest and clearest way to define the market, as threats from generic producers are then also taken into account. In regard to the abuse, the FCA clearly took into account the effects of the conduct on the market. The effects were not mere liable or possible effects, but actual consequences which happened due to the conduct. Although I agree with the way in which the FCA applied the tests for finding abuse of a dominant position, I disagree with the fact that they tested for abuse of a dominant position in the first place. In both the SA and Schering-Plough case, the test should not have been whether article 102 TFEU was infringed upon, as the problems in these cases were not primarily of a competition law nature. This opinion was also advanced by Geradin, who finds that the scope of application of competition law is going too far by claiming that the conduct falls outside of competing on the merits. There are more fitting, specific legal measures for sanctioning the type of conduct SA showed here. An example referred to by Geradin are the general tort provisions (article 1382 and article 1383) in the French Civil Code.243 According to Geradin, the generic suppliers should have used tort claims to seek injunctions and damages via the French civil courts instead of taking the easy route via the FCA so that they would not have to put in the time and effort to win the case. The approach used by generic suppliers and affirmed by the ECJ “poses a serious threat to legal certainty, as the generic suppliers and national authorities are using competition laws to circumvent the clearer legal standards set out in the existing laws.”244 Although I understand the point Geradin is making, I only agree with it in part. I agree with him that there is better suited legislation for this case, but the example he used is poorly chosen. The tort actions he refers to are not necessarily more fitting to the issue at hand, while it is also not unthinkable that tort claims lodged by civil parties co-exist with fines imposed by public authorities. A better example of more specific legislation would be the regulatory framework taken from Directive 97/55/EC concerning misleading advertising so as to include comparative advertising, which has been implemented in the French Code de la Consommation (articles L121-8 and further). Under the Directive,

240

Id., para.359 Id., para.360 242 “Il ressort de ce qui précède que, s’il est parfaitement loisible à un laboratoire pharmaceutique de mettre en évidence les qualités objectives d’un produit, le fait de mettre en évidence non pas seulement des qualités, mais des différences qui, dans le contexte du discours tenu et des conditions dans lesquelles il est entendu, ne peuvent se comprendre que comme des différences substantielles, de nature à soulever un doute objectif sur les qualités des spécialités génériques concurrentes, peut témoigner d’une volonté d’induire le praticien en erreur et être constitutif d’un abus de position dominante.” Id., para.373 243 Geradin 2014a, p.8 244 Ibid. 241

38

each country needs to implement appropriate rules to “ensure that adequate and effective means exist to combat misleading advertising and for the compliance with the provisions on comparative advertising in the interests of consumers as well as competitors and the general public”.245 By applying article 102 TFEU to his case as well, the scope of what is not competing on the merits is pushed too far. The same goes for the Schering case, where a similar judgment was written. As is also mentioned by Tayar, it is not unthinkable that other NCAs or the Commission will follow this broad interpretation of article 102 TFEU.246 This could, in the worst case scenario, mean that all conduct which is regulated by specific legislation will end up falling under not competing on the merits. This, again, would lead to the circumvention of the clearer and more apt legal standards set out in the more specific legislation. A complication with pointing towards a more specified piece of legislation might be that the company who filed a complaint feels unheard or gets the idea that the Commission or national competition authority is not thoroughly doing its job. Also, the competition authority might not always be familiar with other pieces of legislation and their possibilities. Therefore, the company which is accused of the conduct should be the one pointing at the more specialized piece of legislation in the form of a well-motivated counter-argument, preferably with the use of a written statement of an expert who works in the field of law better suited for the case. The burden of proof should be on the company. After the company has brought such a point forward and motivated it thoroughly, the burden of proof is moved to the Commission or national competition authority who is then to show that the suggested field of law is not better suited than competition law. Depending on the stage in which the case is, either the Court or the Commission decides whether or not there is a better suited piece of legislation and can then dismiss the case on those grounds.

4.4 Reckitt Benckiser Reckitt Benckiser (“RB”) is the producer of Gaviscon, which medicine is used to treat acid reflux by creating a barrier on top of the contents in the stomach so acid cannot get into the esophagus.247 RB has two Gaviscon medicines which are important for this case. The first is Gaviscon Original Liquid (“GL”), the patent of which expired in 1997 after having been awarded in 1977.248 The second is Gaviscon Advance Liquid (GA), the patent protection for which lasts until 2016 and which was launched in the same year as the patent on GL had expired (1997).249 From 1997 until 2005 GL and GA were both available in the UK through the National Health Service (“NHS”). NHS is used by pharmacies for prescribing drugs. Pharmacies enter the name of a drug into the computer and then click on a button which shows what generics are available for that medicine.250 The advantage of this system is that it pushes price competition between the several pharmaceutical companies which deliver the same type of medicine and limits the necessity and incentive for promoting medicines directly to physicians by sales representatives (so-called detailing activities).251 After word came out that a generic version of GL would come on the market in 2005, RB quickly removed GL from the NHS-list.252 In doing this, RB had found a way to work around the button that showed the generic versions of a medicine. If pharmacies entered Gaviscon in the NHS system, only GA would come up. After clicking on the button to look for generics of GA, none would come up, as GA still had patent protection. As a result, mostly RB’s GA was sold. This conduct resulted in an investigation by the UK’s Office 245

Article 4 Directive 97/55/EC Tayar 2014, p.4 247 OFT 12 April 2011 (Reckitt Benckiser), para.1.3 248 Id., para.2.35 249 Id., paras.1.4 and 2.34 250 Id., paras.1.7 and 2.100–2.113 251 Id., para.1.8 252 Id., para.1.5 246

39

of Fair Trading (“OFT”) for an alleged infringement of article 102 TFEU and Section 18(1) Competition Act 1998. Section 18(1) is the UK’s national version of article 102 TFEU.253 The OFT had defined the relevant product market as consisting of the supply of alginates and antacids in the prescription channel, a definition that RB did not contest.254 Both GL and GA fall under this market definition.255 Because the ATC classification system did not give a clear view of the market, the OFT looked into other aspects, such as how the medicines were defined in the literature and the influence of the different medicines on each other.256 It was also found that RB had a dominant position due to the fact that it had retained a market share of over 80 per cent between 2004 and 2008 and there were high barriers of entry.257 RB did not contest this either.258 Last, the OFT looked into the question whether RB had abused their dominant market position. From the internal documents of RB it became clear that removing GL from the list would be loss-making if the result would not be the hindering of generics to enter the market.259 If the aim would not have been to hinder the development of full generic competition, the behavior would be unwise. The conduct “cannot therefore be regarded as 'normal competition' or 'competition on the merits'.”260 Also, due to the removal of GL, it was reasonable to expect that the conduct would result in a restriction of competition by hindering generic competition. The actual effects on the market also showed this.261 Therefore, the OFT was of the opinion that by removing GL from the NHS list, RB had abused its dominant position in the market for the supply of alginates and antacids.262 The UK’s Office of Fair Trading (“OFT”) fined RB in a settlement with 10.2 million pounds for infringing Section 18(1) Competition Act 1998 and Article 102 TFEU.263

4.4.1 Analysis One might say that all RB did was find a smart way around a computer program not functioning completely as it should be by using the right RB had to withdraw their product. Apparently having the right to withdraw the product does not necessarily mean that using the right is not considered abuse. In the specific circumstances of the case, RBs circumvention of the generic search option, constituted an abuse of their dominant position: a finding I agree with. As we saw in para. 3.3.1, abuse or acting in a way that is not competing on the merits consists of directly prejudicing consumers or of indirectly prejudicing consumers by diminishing competition through methods different from normal competition in products or services on the basis of the transactions of commercial operators. RB’s GA was a better product as it was patented and the patent was a part of normal competition. Under normal circumstances, de-listing a medicine would also be okay, as it is a right a company is entitled too. However, de-listing GL was not a smart commercial move under normal circumstances because under normal circumstances it would cause a loss to the company. The strategy was not part of normal competition, but part of a strategy to keep generic off the market by filtering the generic 253

Id., para.1.19 Id., paras.4.143 and 4.144 255 Id., paras.4.1 and 4.34 256 Id., paras.4.47–4.142 257 Id., paras.5.1–5.53 258 Id., para.5.2 259 Id., paras.1.11 and 2.168 260 Id., para.6.1 261 Id., paras.1.15 and 1.17 262 Id., paras.1.2 and 6.1 263 OFT 12 April 2011 (Reckitt Benckiser) 254

40

options out of the system. As a result of RB’s conduct, competition was greatly diminished. Most likely, if de-listing GL had been smart under normal circumstances for deducting costs or because it was a product no longer of interest to patient, then it would not have been an abuse. But under the abovementioned circumstances it was. In my opinion, what plays a role as well, is the fact that health care prescribers had been prescribing Gaviscon for almost 20 years. As a result, generic producers had little power in marketing their generic brands to doctors as Gaviscon was hardwired into their system for the situations when a doctor prescribed a drug to treat acid reflux. Also, doctors have little incentive to learn the generic terms because to them it does not matter whether the original or generic version is prescribed. The effect of RBs conduct was a great hampering of competition on the market, without generics being able to find an effective way around the computer program problem. Furthermore, it is of interest to mention that the OFT did not just look at the possible effects of the conduct on the market but looked at the actual effects it had. Also, in defining the market many different factors were taken into account besides the ATC levels.

4.5 Pfizer Pfizer is the owner of the patent for the pharmaceutical product latanoprost (EP 0364 417 “EP 417”), which medicine was traded by Pfizer under the trade mark Xalatan. The medicine Xalatan and its main ingredient latanoprost treat high eye pressure, which can cause glaucoma.264 Pfizer’s patent EP 417 was to expire in September 2009, but to elongate the protection period Pfizer had requested SPC’s for most European countries. The SPC’s that were granted caused Pfizer to have protection on Xalatan until 2011. However, for unknown reasons, Pfizer did not request an SPC for Xalatan in Italy and the date for requesting an SPC on EP 417 had passed. Without the SPC, the protection of Xalatan in Italy would end in September 2009. Pfizer’s competitors were aware of the fact that the protection in Italy would be expiring early, due to the lack of an SPC, so competitors made plans to enter the Italian market with generic versions of Xalatan using latanoprost.265 In order to prevent the generics from entering the market, Pfizer filed for a divisional patent (EP 1225 168 “EP 168”) in 2002 so Pfizer could request an SPC on the divisional patent266 A divisional patent uses subjectmatter from a prior patent application and gets the priority date of the original “main” patent (EP 417). The general rule is that a divisional patent can only be requested if the “main” patent request is still pending, which would imply that Pfizer was not capable of applying for the divisional patent, as the ”main” patent (EP 417) was granted long before the filing of EP 168.267 Yet, there is an exception to the main rule called a second generation divisional application: not only can a divisional patent be filed during the period that the main patent is still pending, it can also be filed during the period that another divisional patent on the main patent is pending. Filing a second divisional patent using an earlier divisional patent is completely legitimate, as long as the earlier divisional patent was filed legitimately and still pending at the time the second divisional patent was filed.268 Using the second generation divisional application, Pfizer was able to file for the divisional patent EP 168 by basing the filing on an earlier divisional patent ( EP 0569046 “EP 046”), as EP 168 described the use and dosage of latanoprost in eye-drops. EP 046 was filed in September 1989, during the pending period of main patent EP 417, but was granted only much later: in November

264

Stjernschantz & Resul 1989; Di Santo 2014, p.1 Di Santo 2014, p.1; Geradin 2014a, p.5; Consiglio di Stato 12 February 2014 (Pfizer); Stjernschantz & Resul 2002 266 Ibid. 267 Article 36 and 76 EPC; Administrative Council 2013 268 European Patent Office 2014b, para.1.1.1.4 265

41

2002.269 Therefore, until November 2002 EP 046 had been pending, so Pfizer was capable of filing EP 168 in 2002. Pfizer obtained EP 168 in January 2009 and used that patent to request a SPC and pediatric extension for Xalatan in Italy. Using EP 168, Pfizer extended the duration of the protection until July 2011 and the pediatric extension extended the protection until January 2012.270 The patent protection itself would end on the same date as EP 417, September 2009, as that patents’ priority date was decisive. After the granting of EP 168, Pfizer sent letters to its pharmaceutical competitors to make them aware of the SPC now in place on Xalatan. Also, Pfizer warned them that a market entry by the generics would be followed by patent infringement proceedings. Furthermore, the Italian Drug Agency (‘AIFA’) denied the market authorizations for the generics of Pfizer’s competitors. Consequently, various pharmaceutical companies filed a complaint with the Italian Competition Authority (“ICA”) and an investigation commenced. The ICA was of the opinion that Pfizer had infringed article 102 TFEU with their strategy to artificially extend Xalatan’s patent protection in the period from September 2009 until July 2011 and fined Pfizer for 10.7 million euros.271 In using the conduct to apply for a divisional patent and applying for a SPC and pediatric extension, Pfizer had abused the administrative and litigation procedures as a dominant company to prevent generic suppliers from entering the market. An important fact for the ICA to come to the conclusion that the administrative procedures had been abused, was based on the EPO revoking EP 168 and that Pfizer had not placed any new pharmaceutical products on the market based on the divisional patent. Therefore, Pfizer had abused the administrative procedures in a similar way as AstraZeneca had done. However, Pfizer did appeal the EPO’s ruling for revoking EP 168, which ruling had stayed pending until the end result of the decision. Furthermore, using extended patent protection, Pfizer had threatened the generic suppliers with proceedings and had prevented market authorizations to be granted. The conduct of Pfizer had the effect of keeping generic Xalatan from entering the market by creating legal uncertainty on whether or not Xalatan was still protected. Such conduct is a practice which does not comply with competing on the merits and is therefore an infringement of article 102 TFEU.272 On the 12th of February, 2014, the Consiglio di Stato (Italian Council of State) confirmed the decision of the ICA.273

4.5.1 Analysis The definition the FTO gave to conduct not considered being competition on the merits is conduct which is allowed under patent law. The FTO found that the conduct used by Pfizer was similar to the conduct of AstraZeneca and therefore was an abuse. However, there is one fundamental difference between the two cases. In the AstraZeneca case, AstraZeneca had lied to get the SPC protection while Pfizer had used a completely legitimate way to obtain the SPC. As the conduct was completely legitimate, the behavior of Pfizer should not be considered contrary to competition on the merits. The argument the FTO used, namely that the patent on which the SPC was based had been revoked by the EPO, cannot change the fact that the competition should not have been regarded as abuse. The reasons are as follows. First of all, the EPO does not grant patents lightly but has a specific procedure it goes through before one is granted. Among other things, the EPO examines the filing and gives a preliminary opinion on patentability before a patent is granted. Therefore, once a patent is granted like was the case here, the 269

Stjernschantz & Resul 2002 Di Santo 2014, p.1; Geradin 2014a, p.5; Consiglio di Stato 12 February 2014 (Pfizer) 271 Geradin 2014a, p.6 272 Id., p.5; Di Santo 2014, p.2 273 Consiglio di Stato 12 February 2014 (Pfizer), p.27 270

42

patent holder is right to believe he has a solid right. Any competitors or competition authorities must respect that patent right as well.274 Secondly, once the patent is granted the patent holder may exercise his rights as the patent holder then has a patent. The possibility of revocation or forfeiture of the patent in the future can always exist as the opposition proceedings take place after the grant of the patent.275 The possibility of the patent being revoked in the future does not affect the exercise of someone’s patent rights. However, once opposition proceedings have started one might argue that the patent holder may have to be careful in exercising its patent right. Therefore, as long as a patent is still in place and there appears no immediate risk that a patent may be revoked, a patent holder may start infringement proceedings against infringers or send warning to competitors who might be about to infringe a patent. As the patent was still in place when Pfizer sent out the warnings and there were no opposition proceedings, there was no reason why Pfizer would not be allowed to exercise its patent rights. Pfizer even waited until the patent was granted to send out the letters, instead of already doing it during the period that the patent was still pending. Thirdly, not only was what Pfizer did perfectly legal, had Pfizer not forgotten to request an SPC then the protection period would have been just as long. Pfizer found a creative way to solve a mistake she had made, to obtain a right Pfizer should have had to begin with had Pfizer not been so forgetful. The consequences of the decision may be larger than at first may appear. According to the judgment, any patent which is later revoked but the rights of which were used by the patent holder can constitute an abuse contrary to competition on the merits. Second of all, not allowing Pfizer’s strategy in Italy not only causes a shorter protection period in Italy, but indirectly for all of Europe, due to the internal market and the free movement of goods. When a product is rightfully put on the European internal market in one European country, the product can be exported to all other European countries. The fact that a product does not have a patent in the European country where the product was put on the market and the product is then moved to another country where it does have patent protection, does not matter. The patent right is then exhausted.276 As a result of the judgment, smart generic producers could produce a generic version of Xalatan in Italy, put it on the market there and then transport it to the rest of Europe where the SPC´s are still in place. The judgment is therefore detrimental to patent protection and the rights it gives. Lastly, Geradin made an interesting point by explaining that the government in the Italian Pfizer case is both the customer that pays for the products and the regulator. The biasedness requires that clear legal principles are established so the government cannot twist and turn the law to end up with a judgment which is most to their advantage.277

4.6 Conclusion The aim of this chapter was to analyze the various cases in which the national competition authorities and the Commission have defined competition on the merits and when conduct does not fall under that scope. As a thorough analysis was given for each case, only the highlights are taken out for the purposes of this conclusion. The market definition is rather narrow in most cases because pharmaceutical product markets are often defined as broad as the active ingredient of the medicine. It is difficult to change the outcome as the competition authorities are already taking into account multiple factors in defining the market amongst 274

European Patent Office 2014a, para.130 Id. 276 For a thorough explanation on the exhaustion of right, intellectual property and the movement of goods see for instance Tritton 2008, chap.7 277 Geradin 2014b, p.16 275

43

which the demand-side of the products. I do not see a better approach that could have been used. The consequence of the market definition is that dominance is found quickly. As soon as a pharmaceutical producer has a patent on the active ingredient which had defined the relevant product market, dominance is found. Overall, in defining competition on the merits a more effects-based approach was used. All conduct had negative effect on competition on the market or would certainly have had a negative effect on competition on the market. Nevertheless, not all cases were completely clear on why a type of conduct was considered abuse or conduct contrary to competition on the merits. Neither was it always clear why competition law was used, instead of a more specific piece of legislation for the circumstances of the case. Sadly, the problem of legal uncertainty cannot be completely solved with using an effects-based approach but still requires a clearer meaning of competition on the merits. After all, all types of competition, including competition on the merits, can have exclusionary effects when a dominant company has the best product on the market, the cheapest production method or the best marketing team behind him. In the next chapter, these cases and the previous chapters are used to finally conclude whether the competition in the merits test as it has been used in the abovementioned cases puts constraints on patent protection in the pharmaceutical industry.

44

5. Synthesis and conclusion 5.1 Introduction In this final chapter, the conclusion is drawn as to whether the ‘competition on the merits’ test as it has been used in the last few years by the European Commission, the national competition authorities and courts puts undue constraints on patent protection in the pharmaceutical industry. In doing so, the question is divided in two parts: 1) Do the findings of abuse of dominant market positions made by the European Commission, the national competition authorities and courts in the cases discussed, impinge on the terrain of IP law?, and 2) Is this the result of the ‘competition on the merits’ test’s ambiguity?

5.2 Relationship between competition law and IP law During the analyzing of the four cases discussed in chapter 4, it became clear that the concept of abuse has not been clearly delineated when it comes to competition law cases concerning the use of IP rights in the pharmaceutical industry. In all four cases discussed, the relevant authorities and courts ultimately came to a finding of abuse. From the analysis above it becomes clear that this is not the result of defining the relevant market too narrowly or assuming a dominant position too easily; these aspects of the cases have not been criticized by other authors and I believe that this is rightfully so. Therefore, any solutions to the problems discussed in this research cannot be found in adapting a different market solution (see chapter 3.3.4.1, last paragraph).

5.2.1 Issues However, legal writers have taken issue with the finding of abuse in many of these cases. The impression is easily raised that these critiques largely stem from a ‘gut feeling’ that IP law is ‘taken over’ by competition law. While I do agree with the criticism raised for these judgments, I believe that such instinctive reactions mix up two – equally important, but different - aspects of the interaction between competition law and IP law. The first of these interactions is a material one, as the message conveyed by these judgments is that provisions of competition law are more important than provisions of IP law. After all, the behavior of the companies involved in these cases was completely legal under IP law (Reckitt Benckiser and Pfizer cases), or at least might arguably have been legal under IP law (AstraZeneca case), and was nonetheless sanctioned by means of competition law. Illustrative in this regard is the opinion of Geradin with regards to the Pfizer case, which could, to a lesser extent, also be applied in regards to the other three cases: “The validity of Pfizer’s action under the patent system is relevant, and indeed critical, to the analysis of whether Pfizer’s action constituted an abuse under the EU competition laws. Patents are explicitly designed to allow innovators to prevent competitors from entering the market. These rights are granted as a reward for past innovators and an incentive for future innovators to invest the hundreds of millions of euros necessary to invent, develop, and bring new medicines to the market. If an originator obtains (not misleadingly) a valid patent right, any strategy implemented by that company to reap its reward by excluding competitors with its patent rights should be allowable under the EU competition law. After all, it is not because a patent has the effect to exclude that enforcing it can be assimilated to an exclusionary behavior.”278 The second interaction is of a more procedural nature, as the message conveyed by these judgments is that the instrument of competition law trumps the more apt legal measures available in other fields of law than

278

Ibid.

45

competition law to resolve the issues presented in these cases like in the Sanofi Aventis and the ScheringPlough cases. As noted by Geradin, this may pose “a serious threat to legal certainty, as the generic suppliers and national authorities are using the competition laws to circumvent the clearer legal standards set out in the existing laws”.279 The above two interactions may be seen as two sides of the same coin. The willingness of competition authorities to prevent or sanction unwanted behavior may lead them to limit the definition of ‘competition on the merits’ to such an extent that it excludes behavior that is 1) legal according to IP law, or 2) behavior that is unwanted according to IP law but for which IP law has devised its own solutions. As a result,, ‘competition on the merits’ is defined in such a way that it directly limits the legal effects of IP law provisions or the usability of IP law measures, respectively. I feel that both interactions are currently lopsided, in that competition law supersedes IP law at the cost of both legal certainty and the objectives of IP law (long-term innovation and potentially long-term consumer welfare).

5.2.2 Solutions In order to make the two interactions between competition law and IP law more balanced, two approaches can be taken. A first approach is to address the problems associated with the competition on the merits test as a more lenient test which would allow more behavior to constitute competition on the merits. A more lenient test brings less behavior that is either legal under IP law or for which IP law has devised its own solutions under the ambit of article 102 TFEU. This solution in the form of a more lenient test is discussed in the next paragraph of this chapter when discussing the ambiguity of the competitions on the merits test. The second approach is to propose solutions stemming from the hierarchy between competition law and IP law. As explained in chapter 2, three rules determine the interplay between competition law and IP law. From competition law legislation it follows that (only) specific circumstances can make the exercise of an IP right constitute an infringement of article 102 TFEU and that national legislation cannot safeguard an infringer from the consequences of infringing article 102 TFEU. From IP law it follows that, at least for those IP rights regulated by the TRIPS, restrictions on the exercise of IP rights need to be consistent with the TRIPS (the general system of IP law). While on a material level the norms imposed by competition law take precedence when applying competition law instruments, for example by outlining which specific circumstances can make the use of an IP right constitute an infringement of article 102 TFEU, no such hierarchy necessarily exists when determining which instrument – competition law or otherwise – to apply. The use of civil claims, criminal sanctions or other non-competition law public sanctions is still possible, despite the fact that the behavior for which these claims or sanctions are brought is (also) regulated by competition law. Once a choice has been made to let a case be handled by competition authorities this may preclude other legal measures from being taken due to the prohibition on double jeopardy or due to the fact that having a competition authority take care of the matter is simply cheaper for the aggrieved party, but this is not a result of a higher place that competition law would have in a hierarchy between different instruments. I suggest therefore to give parties the opportunity to bring forward the argument that there is better suited legislation. Such an argument should be thoroughly motivated, possible with the help and referral to an

279

Geradin, 2014a, p. 5

46

expert in a specific field of law. The court or competition authorities should then see whether or not there is more specific legislation suited for the case. If the finding is that there is a more specialized field of law, the case should be dismissed and referred to another more suitable court or legal institution. Examples of such potential referrals might be to patent authorities for the granting of compulsory licenses and to specialized publicity authorities for the addressing of issues of false (comparative) publicity, such as in the Sanofi case. The advantages of such an approach are threefold. First, such an approach is more suited to deal with the exact issues that arise in each case, as the measures that such authorities may take are specifically suited to deal with the issue within the context of the relevant field of law. Such would also be in line with the requirement of the TRIPS that any restriction on the use of IP rights is in conformity with the (system of the) TRIPS itself. Second, this approach would relieve the need for competition authorities to qualify any unwanted behavior as not behaving on the merits, which allows the concept of ‘competing on the merits’ to be defined much clearer. Ultimately, this would bring much more legal certainty in a field of law that direly needs it. Third, a more balanced work division between competition authorities and specialized authorities would relieve some of the pressure felt by many competition authorities in having to deal with a multitude of cases for which they are not always well-equipped (such as complicated IP law related cases). In order for such an approach to work, the competition authorities making the referral would need to convince themselves of the fact that the authorities to which the case is referred have the legal instruments required to achieve an outcome that could – in case of a finding of abuse of the alleged infringer – achieve the same end results as would be desired by competition law. If this were not the case, the issue would find its way back to the competition authorities eventually anyways, rendering a referral a moot option and giving the party who complain the feeling he is being sent from pillar to post. Therefore, at the very least, the authorities specialized and responsible in executing the more suitable piece of legislation must be able to come to a solution that will increase competition on the market, deter future abusive behavior and, eventually, increase consumer welfare. Without going into too much detail, it appears that the granting of compulsory licenses (perhaps at a lower than fair market price in case of abuse by the would-be licensor) would be able to obtain these results. A mere possibility for competitors to claim damages in civil court would not, as it would not necessarily allow these competitors to compete on the market and the rewards obtained from these proceedings may not necessarily benefit consumers. After a case has been referred to another field of law which is more suited for the case, the competition authority should no longer be able to impose a fine for the same behavior. The exception thus being that the other (more specific) field of law has not been able to render a decision on the matter. In this way, a clear delineation is made between the cases decided on by the competition authorities and the cases decided on by authorities of other legislation like patent courts.

5.3 The ‘competition on the merits’ test As indicated above, the four cases analyzed in this research present the view that the competition authorities have wanted to penalize behavior that may have been legal under IP law and have used the ambiguity of the ‘competition on the merits’ test to achieve their objective.

5.3.1 Issues It is the umbrella concept of ‘competition on the merits’ that legal writers have spoken out against, claiming that it is a “vague standard”280 for which it is required to refer back to the “practices of the competition

280

Geradin, 2014b, p. 16

47

authorities and the courts”281 and which has led to “inconsistent interpretations and therefore to unpredictable results”282. The ambiguity of the term has been acknowledged by some on the side of the competition authorities as well.283 Not only may legal uncertainty result from the use of the phrase because it is not entirely clear what the term encompasses, but the ambiguity of the term may also lead to increasing uncertainty in the future. This can be seen from the way in which behavior that was deemed by IP lawyers to fall inside the scope of ‘competition on the merits’ is placed outside that scope by competition authorities if the case ‘requires it’.284 The vagueness of the phrase ‘competition on the merits’ is thus used to flexibly address any issue that competition authorities have, regardless of whether it the behavior of the alleged infringer is in fact detrimental to competition on the market or the wellbeing of consumers. This effect is felt especially in the field of IP law, which, due to its characteristics (see also chapter 2) is inherently at odds with the basic tenets of competition law. In order for IP law and competition law to coexist, therefore, it is essential that the concept of ‘competition on the merits’ is defined especially well with regards to the use of IP rights. Until that time, the issues indicated above – legal uncertainty, jeopardizing the objectives of IP law – remain. Although this concludes the discussion of the main question of this research – whether the ‘competition on the merits’ test as it has been used in the last few years by the European Commission, the national competition authorities and courts puts undue constraints on patent protection in the pharmaceutical industry – it is a rather unsatisfying conclusion, as no solution has yet been found to the issue at hand. I will therefore leave the reader with some ideas of where a possible solution could be found.

5.3.2 Solutions Three proposals for a better working of the competitions on the merits test are to devise a new test for alleged infringements that stem from the use of IP rights, an ‘assumption of competition on the merits’ for IP rights holders, and the recognition of norms imposed by IP law. A first option would be to install an ‘assumption of competition on the merits’ in favor of those who allegedly infringe upon article 102 TFEU by making use of their IP rights. As held in earlier cases, “the mere exercise of a right lawfully afforded by Union law could at the most amount to an ‘abuse’ only in exceptional circumstances, namely where there is an elimination of all effective competition, a mere propensity to distort competition not being sufficient for that purpose”.285 This would imply that, when it is uncertain whether behavior is to be allowed, the benefit of the doubt should be given to the IP rights holder. This way, innovation is most encouraged.286 A second option would be to devise a new test in order to determine whether certain behavior constituted ‘competition on the merits’. Existing economic tests, such as those discussed during the OECD roundtable discussion, are mostly concerned with predatory pricing practices and are not well-adapted to dealing with

281

Schrijvershof, 2011, p. 102 OECD Competition Committee, 2005, p. 9 283 Kroes, 2005, p. 4 284 Geradin, 2014b, p. 16 285 ECJ 6 December 2012, C 457/10 (AstraZeneca/Commission), para. 142 286 Elhauge, 2003, p. 255 282

48

the use of IP rights.287 A possible test would be somewhat akin to the reasoning used in Reckitt Benckiser: the (exclusive) use of IP rights is permitted, as long as the behavior of the IP rights holder does not have the sole purpose of excluding competition in a way that would not have been possible through the use of IP rights that were valid and (could legally have been) in the possession of the IP rights holder at the time the behavior occurred. This test would lead to the result that a finding of abuse would occur in the cases of AstraZeneca, Sanofi Aventis and Reckitt Benckiser, but not in the case of Pfizer. A third option would be to ‘whitelist’ certain behavior on the basis of IP legislation. If, for example, under the predominant part of IP law systems in the Member States it would be allowed to deny third parties a license, this should be available as a defense in competition law cases. Based on research into the various IP law systems, a list could be drawn up that would indicate the types of behavior that are and are not allowed according to the Member States, which could then be published by the Commission as part of another Guidance Paper to indicate behavior that will not be penalized.

287 See e.g. (OECD Competition Committee, 2005, p. 29), where it is held that the test discussed limits the incentive to innovate

due to its tendency to impose the compulsory granting of licenses.

49

Bibliography Administrative Council, 2013. Revision of the requirements for the filing of divisional applications. European Patent Organisation. Available at: http://www.epo.org/news-issues/news/2013/20131018.html. Ahlnorn, C. & Evans, D.S., 2007. The Microsoft Judgment and its implications for competition policy towards dominant firms in Europe. Antitrust Law Journal, 3601, pp.801–846. Anderman, S., 2011. The IP and Competition Interface. In Intellectual Property and Competition Law New Frontiers. New York: Oxford University Press, pages 3–25. Anderman, S. & Ezrachi, A., 2011. Intellectual Property and Competition Law, New York: Oxford University Press. Anderman, S. & Schmidt, H., 2011. EU Competition Law and Intellectual Property Rights- The Regulation of Innovation, New York: Oxford University Press. Andersen, B., 2004. If “intellectual property rights” is the answer, what is the question? Revisiting the patent controversies. Economics of Innovation and New Technology, 13(5), pp.417–442. Bailey, D., 2010. Presumptions in EU competition law. 2European Competition Law Review, 31(9), pp.362–369. Bakels, R.B., 2002. Softwareoctrooien: een vanzelfsprekendheid of een gevaarlijke ontaarding? Computerrecht, (6), pp.347–352. Council Regulation (EEC) No 1768/92 of 18 June 1992 concerning the creation of a supplementary protection certificate for medicinal products. Available at: http://eur-lex.europa.eu/legalcontent/EN/ALL/?uri=CELEX:31992R1768. Douwenga, F., 2014. Software Patents: Conformity with the rationale of patenting programs for computers. Master Thesis Law and Technology. Drexl, J., 2011. Astrazeneca and the EU sector inquiry: when do patent filings violate competition law? Max Planck Institute for Intellectual Property and Competition Law Research Paper Series, (12). Drexl, J., 2009. Real Knowledge is to Know the Extent of One’s Own Ignorance: On the Consumer Harm Approach in Innovation-Related Competition Cases. Max Planck Institute for Innovation and Competition Law, 9(15). Elhauge, E. (2003). Defining better monopolization standards. Stanford Law Reviewford Law Review, 56(2), 253–344. European Commission, 2013a. Antitrust : Commission fines Johnson & Johnson and Novartis € 16 million for delaying market entry of generic pain-killer fentanyl. Press Release, (December). Available at: http://europa.eu/rapid/press-release_IP-13-1233_en.htm. European Commission, 2012. Antitrust : Commission sends Statement of Objections on perindopril to Servier and others. Press Release. Available at: http://europa.eu/rapid/press-release_IP-12835_en.htm?locale=en. European Commission, 2009a. Antitrust: Commission imposes fine of € 1.06bn on Intel for abuse of dominant position; orders Intel to cease illegal practices. IP/09/745, (May), pp.1–4.

50

European Commission, 2013b. Commission fines Lundbeck and other pharma companies for delaying market entry of generic medicines. Available at: europa.eu/rapid/press-release_IP-13-563_en.htm. European Commission, 2009b. Guidance on the Commission’s enforcement priorities in applying Article 82 of the EC Treaty to abusive exclusionary conduct by dominant undertakings. C 45/7. European Commission, 2011. Guidelines on the applicability of Article 101 of the Treaty on the Functioning of the European Union to horizontal co-operation agreements. C 11/1. European Commission, 2004. Guidelines on the application of Article 81 of the EC Treaty to technology transfer agreements. C 101/2. European Patent Office, 2014a. Guide for applicants, Part 1: How to get a European patent. Available at: http://www.epo.org/applying/european/Guide-for-applicants.html European Patent Office, 2014b. Guidelines for examination. Available at: http://www.epo.org/lawpractice/legal-texts/html/guidelines/e/a_iv_1_1_1_4.htm. Geradin, D., 2010. Is the Guidance Paper on the Commission’s Enforcement Priorities in Applying Article 102 TFEU to Abusive Exclusionary Conduct Useful? In F. Etro & I. Kokkoris, eds. Competition Law and the Enforcement of Article 102. New York: Oxford University Press, pages 46–63. Geradin, D., 2014a. The Uncertainties Created by Relying on the Vague “ Competition on the Merits ” Standard in the Pharmaceutical Sector : The Italian Pfizer / Pharmacia Case. Journal of European Competition Law & Practice, pp.1–9. Geradin, D., 2014b. When Competition Law Analysis Goes Wrong – The Italian Pfizer/Pharmacia Case. SSRN. Available at: ssrn.com/abstract=2393383 or http://dx.doi.org/10.2139/ssrn.2393383. Gormsen, L.L., 2013. Are Anti-competitive Effects Necessary for an Analysis under Article 102 TFEU? World Competition, 36(2), pp.223–245. Grosche, A., 2006. Software Patents-Boon or Bane for Europe. International Journal of Law and Information Technology, 14(3). Guellec, D. & van Pottelsberghe de la Potterie, B., 2007. The Economics of the European Patent System: IP Policy for Innovation and Competition, New York: Oxford University Press. H. Lundbeck A/S, 2013. Lundbeck appeals European Commission decision. Corporate Release, (509). De Hullu, J., 2009. Materieel Strafrecht, Deventer, Netherlands: Kluwer. IMS Health, 2014. About IMS Health. Available at: http://www.imshealth.com/portal/site/imshealth/menuitem.051a1939316f851e170417041ad8c22a/?vg nextoid=7311e590cb4dc310VgnVCM100000a48d2ca2RCRD&vgnextfmt=default Institut Superieur des visiteurs medicaux, 2014. Visiteur medical ou delegue medical. Available at: http://www.isvm.com/visiteur_medical.asp. Kroes, N., 2005. Preliminary Thoughts on Policy Review of Article 82. European Commission Speech/05/537. Larouche, P., 2008. The European Microsoft case at the crossroads of competition policy and innovation. Tilburg Law and Economics Center, (May).

51

Larouche, P. & Schinkel, M.P., 2013. Continental drift in the treatment of dominant firms: article 102 TFEU in contract to para. 2 Sherman Act. Tilburg Law and Economics Center. Van Loon, S., 2008. Licentieweigering als misbruik van machtspositie -Intellectuele eigendom, artikel 82 EG en de belemmering van innovatie, Nijmegen: deLex. Maggiolino, M., 2011. Intellectual property and Antitrust- A comparitive Economic Analysis of US and EU Law, Cheltenham: Edward Elgar Publishing Limited. Maggiolino, M. & Montagnani, M.L., 2011. AstraZeneca’s abuse of IPR-related procedures— A hypothesis of antitrust offence, abuse of rights, and IPR misuse. World Competition, pp.245–259. Marsden, P., 2010, Exclusionary Abuses and the Justice of ‘Competition on the Merits’ in Ioannis Lianos and

Ioannis Kokkoris, The Reform of EC Competition Law: New Challenges, Alphen aan de Rijn: Kluwer, pp. 411–418 Monti, G., 2010. Article 82 EC: What Future for the Effects-Based Approach? Journal of European Competition Law & Practice, 1(1), pp.2–11. Nard, C.A., 2010. Patent Law, Aspen Pub. OECD Competition Committee, 2005. Competition on the Merits. DAF/COMP(2005)27. OECD Competition Committee, 2008. Fidelity and Bundled Rebates and Discounts. DAF/COMP(2008)29. Priddis, S. & Constantine, S., 2011. The Pharmaceutical Sector, IPRs, and Competition Law. In Intellectual Property and Competition Law New Frontiers. New York: Oxford University Press, pages 241–275. Di Santo, G.P., 2014. Pfizer’s misuse of patent/SPC rights results in an abuse of dominant position. Journal of Intellectual Property Law & Practice, pp.1–2. Schillemans & van Bockel, 2011. Verbod van misbruik van economische machtsposities bij: Mededingingswet, artikel 24. Tekst en Commentaar Mededingingswet. Schrijvershof, D. (2011). De geneesmiddelensector: de “tool-box” en (“evergreening” van) het mededingingsrecht. Een tussenstand, (4), 93–103. Spector, H.M., 1989. An outline of a theory justifying intellectual and industrial property rights. European intellectual property review, 11(8), pp.270–273. Speyart, M.H.M., 2014. Intellectuele Eigendom en Reclamerecht, Mededingingsrechtelijke verkenningen in het IE-recht: een inherente spanning. IER, 23. Stjernschantz, J. & Resul, B., 1989. Prostaglandin derivatives for the treatment of glaucoma or ocular hypertension. Prostaglandin derivatives for the treatment of glaucoma or ocular hypertension Prostaglandinderivate, (19). Available at: https://data.epo.org/publication-server/pdfdocument?pn=0364417&ki=B9&cc=EP. Stjernschantz, J.W. & Resul, B., 2002. Prostaglandin derivatives for the treatment of glaucoma or ocular hypertension. , (19). Available at: https://data.epo.org/publication-server/pdfdocument?pn=1225168&ki=A2&cc=EP. Tayar, D., 2014. Spreading Misleading Information on a Competitor ’ s Product as an Abuse of a Dominant Position : a French Pharmaceutical. Journal of European Competition Law & Practice, pp.1–5.

52

Taylor, C.T., Silberston, A. & Silberston, Z.A., 1973. The Economic Impact of the Patent System: A Study of the British Experiment, Cambridge: Cambridge University Press. Tritton, G., 2008. Intellectual Property in Europe, Sweet & Maxwell. Vezzoso, S., 2012. Towards an EU Doctrine of Anticompetitive IP-Related Litigation. Journal of European Competition Law & Practice, 3(6), pp.521–535. Van der Vijver, T., 2012. Article 102 TFEU : How to Claim the Application of Objective Justifications in the Case of prima facie Dominance Abuses? Journal of European Competition Law & Practice, 4(2), pp.121–133. Van der Vijver, T., 2014. Justifications and Anti-competitive Unilateral Conduct: An International Analysis. World Competition, 37(1), pp.27–51. Waelbroeck, D., 2010. The Assessment of Efficiencies under Article 102 and the Commission’s Guidance Paper. In F. Etro & I. Kokkoris, eds. Competition Law and the Enforcement of Article 102Article 102. New York: Oxford University Press, pages 138–152. Whish, R. & Bailey, D., 2012. Competition Law, New York: Oxford University Press. World Health Organization, 2003. Introduction to Drug Utilization research. Available at: http://apps.who.int/medicinedocs/en/d/Js4876e/

Cases Autorité de la concurrence 14 May 2013, 13-D-11. Sanofi-Aventis/AUtorité de la concurrence. Consiglio di Stato 12 February 2014, N. 09181/2012. Pfizer/Autorità garante della concorrenza e del mercato. Court of First Instance 10 July 1990, Case T-51/89. Tetra Pak International (BTG License)/Commission. Court of First Instance 17 July 1998, Case T-111/96. Promedia/Commission. Court of First Instance 20 September 2003, T 203/01. Michelin II. Court of First Instance 24 March 2004, T 201/04. Microsoft v. European Commission. European Court of Justice 12 December 1991, Case T-30/89. Hilti AG/Commission. European Court of Justice 13 February 1979, Case 84/76. Hoffman-La Roche/Commission. European Court of Justice 13 July 1966, Joined Cases 56 and 58/64. Consten and Grundig. European Court of Justice 14 February 1978, Case 27/76. United Brands/Commission. European Court of Justice 14 November 1996, Case C-333/94 P. Tetra Pak International/Commission. European Court of Justice 15 March 2007, Case C-95/04. British Airways/Commission. European Court of Justice 29 February 1968, Case 24/67. Parke Davis. European Court of Justice 5 October 1988, C-238/87. Volvo/Veng.

53

European Court of Justice 6 April 1995, Joined cases C-241/91 and C-242/91. Magill. European Court of Justice 6 December 2012, C 457/10. AstraZeneca/Commission. European Court of Justice 7 June 2007, Case C-76/06 P. Britannia Alloys and Chemicals/Commission. European Court of Justice 8 June 1971, Case 78/70. Deutsche Grammophon Gesellschaft/ Metro. European Court of Justice 9 November 1983, 1983. Case 322/8. Michelin/Commission. FCA 18 December 2013, 2013. Décision n ° 13-D-21. Schering-Plough. General Court 1 July 2010, T 321/05. AstraZeneca/Commission. Office of Fair Trading 12 April 2011, Case CE/8931/08. Reckitt Benckiser/Office of Fair Trading.

54