Master thesis JFVQ Jan 29

Master Thesis European Business Law Recoupment of losses as a potential element of the predatory pricing test: its consistency with the Commission´s 2009 Guidance and the objectives of EU Competition Law Author: José Francisco Vega Quesada s4169352 Supervisor: Catalin S. Rusu Second Reader: Martijn Wools January, 2013 Contents Preface and acknowledgments 4 Introduction 5 1. The facts that provide relevance to the study 1.1.The rulings of the General Court and the European Court of Justice in France Télécom 1.2. Advocate General Mazák’s Opinion 1.3. The Co issio ’s Guida e 2. Relevance of the subject and purpose of the thesis 3. Methodology and disclaimers 4. Contents 5 5 6 8 9 Chapter I: The current treatment of Predatory Pricing under EU Competition Law 11 1. The concept of predatory pricing 2. Factors involved in the feasibility of predation 2.1. The structural features of the market(s) 2.1.1. The multi-market factor 2.1.2. The market power factor 2.2. The time factor and the amount of losses 2.3. Barriers to enter and re-enter the market 2.4. Effe ts of predatio o the predator’s reputatio 2.5. The recoupment of losses 2.5.1. Recoupment as part of the definition of predatory pricing 2.5.2. The importance of recoupment 3. Predatory pricing among the conducts forbidden by Article 102 TFEU 3.1. Predation as an exclusionary conduct 4. The cost assessment for predation 4.1. The purpose for the application of cost benchmarks 4.2. The cost measures 4.3. The cost assessment of the European Courts (AKZO) 4.4. The Co issio ’s ost assess e t i the Guida e 5. The treatment of recoupment of losses by the European Courts 5.1. Tetra Pak II 5.2. France Télécom 6. Recoupment of losses in the Guidance 7. Final Comments 12 14 15 15 16 16 16 17 17 19 20 22 22 24 24 27 28 31 33 33 35 37 38 Chapter II: The Guidance of the Commission and the recoupment factor 40 1. The process towards the Guidance 2. Some relevant characteristics of the Guidance 2.1. The Guida e’s li ited s ope 2.2. The Guidance and consumer welfare: the concept of anticompetitive foreclosure 41 44 44 2 5 47 2.3. The binding force of the Guidance: Is the Commission bound by the Guidance? 49 3. What is an effects-based approach? 4. Recoupment in the Guidance 4.1. Recoupment: The Commission´s approach in France Télécom 4.2. Recoupment a d the Guida e’s o je tive 4.3. The Commission may have conducted a dominance test in France Télécom 4.4. The Guidance opens a room for recoupment 5. The Post Danmark case 5.1. The facts 5.2. The questions referred to the ECJ 5.3. Main aspects of the ruling 5.3.1. Consumer welfare 5.3.2. Competition on the merits and use of the te as-efficient competitors 52 54 55 59 59 62 63 63 65 65 65 66 5.3.3. The cost-benchmark (AIC) and aspects directly related to price-based exclusionary conducts 5.3.4. Effects of the conduct 5.3.5. Objective justifications and efficiencies 5.3.6. Operative part of the ruling 5.4. The Guidance and Post Danmark 6. Final Comments 66 68 68 68 69 73 Chapter III: Consumer welfare as an objective of EU Competition Law a d Ad o ate Ge e al Mazák s Opi io a out recoupment 75 1. Consumer welfare as the main objective when enforcing Article 102 against predatory pricing 1.1. Definition of consumer welfare 1.2. Is consumer welfare the objective when dealing with predation? 2. AG Mazák’s Opi io i France Télécom 2.1. AG Mazák’s Opi io a out recoupment of losses 2.1.1. First ground of appeal 2.1.2. Seventh ground of appeal 2.2. Main aspects of the Opinion 2.2.1. About the first ground of appeal 2.2.2. About the seventh ground of appeal 2.2.3. Fe ell as Mazák s ai sou e 2.3. Secondary aspects of Mazák’s Opinion 3. Final comments 76 77 80 88 88 89 90 91 91 92 96 99 100 Chapter IV: Conclusions 102 List of cases, consulted bibliography, and others 115 3 Preface and Acknowledgments: I first considered writing my thesis on Competition Law during the first weeks of the Master Program. As part of the Introductory Course of the Program, Dr. Rusu taught some notions about Competition Law which motivated me to gather more information and read a bit more about this field. I considered very appealing to develop a subject related to rules having direct influence on the interests of some of the most important players in the world market. At the same time, I found very interesting to write about a legal system that considering the values of a modern market economy, has to find a balance between these values and the interests any democratic government must promote or protect. That is how I, after doing some research, came up with predatory pricing and the discussion about recoupment that took place in the France Télécom case. I would like to thank my Supervisor, Catalin Rusu, for his academic support during the entire process and for providing me with valuable comments, constructive criticism, and very productive conversations about the subjects of the study. In particular I thank Dr. Rusu for making always the time for discussing my queries and reviewing my submissions. In my view my learning process was effectively complemented with my Supe iso s o sta t la if i g a d ea i gful i put. I also want to thank the people that made possible, and always gave me their support during the past year-experience as a student. Among them, I want to mention my parents, siblings, and friends: Martín, Tania, Carla, and Carolina. I could not forget to thank Daniel Chaverri for being my travel partner in Europe and for welcoming in his house during my university vacations. No less important are Joost and Harry who showed me some of the best ways of having fun, Dutch Style of course! Thanks to Rigo, Lucía, Laura, and Rosemary for always asking for me and continuously checking my progress via Skype. I also want to mention Pilar González and Guillermo Martín who read the thesis and provided me with their remarks and suggestions. Thank you, I really appreciate it! Finally, I would like to give special thanks to Lisa Schubert for being my library and lunch partner while both of us were working in our theses, and for making the writing process entertaining and fun. 4 Introduction 2. Facts that provide relevance to the study 2.1. The rulings of the General Court and the European Court of Justice in France Télécom On July 16, 2003 the European Commission (Commission) adopted a decision imposing a fine of €10.35 million on Wanadoo Interactive (WIN). The Commission found that WIN infringed Article 102 TFEU by charging predatory prices for its internet (ADSL) services in France, from March 2001 to October 2002. The decision was contested before the European Judiciary and confirmed by the General Court 1 (GC) and the European Court of Justice (ECJ). The ECJ gave its ruling on April 2, 2009, dis issi g the appeal agai st the GC s uli g a d o fi i g its previous assessment of predatory pricing in the AKZO and Tetra Pak II cases. Furthermore, the ECJ ruled out the appli a t s lai WIN a o di g to hi h the proof of the possibility of recoupment of losses2 incurred by the incumbent during the predation period is a necessary precondition to establishing the existence of predatory pricing. 3 In its submissions to the ECJ, the Commission argued against recoupment as part of the predation test and alternatively, considered that recoupment was implied in the dominance test.4 2.2. Advocate General Mazák’s Opinion For this same ruling (France Télécom) the Advocate General (AG) in charge of giving his Opinion to the ECJ as Ja Mazák, ho e o e ded setti g aside the GC s judg e t a d the e-examination of the case by said court. Among other reasons, AG Mazák considered that proof of recoupment is a requirement set up by the EU case law. 5 AG Mazák not only based his recommendations on EU case law a d o p e ious of his fello AG s Opi io s, ut also follo ed the ite ia gi en, inter alia, by the Economic Advisory Group on Competition Policy (EAGCP), the Organization for Economic Co-operation and Development, and the European Regulators Group.6 1.3. The Commission’s Guidance On December 3, 2008, the Commission adopted a Guidance Communication (Guidance) on its enforcement priorities when applying Article 102 TFEU7 with regard to exclusionary abuses by dominant 1 By then, the Court of First Instance (CFI) which, after the Treaty of Lisbon came into force, as a ed Ge e al Cou t . Regardless of the date of the rulings of such court, for the sake of clarity and order we will refer to it as the General Court (GC) in the content of the thesis. 2 We will continue using, indistinctively, the terms recoupment, recoupment of losses, proof of the possibility of recoupment, or proof of recoupment of losses, which for the purposes of the study have the same meaning. 3 Case C-202/07P, France Télécom v Commission, [2009] ECR I-2369, para. 110. 4 Case C-202/07P, France Telecom v. Commission [2009] ECR I-2369, para. 102. 5 Case C-202/07P, France Telecom v. Commission [2009] ECR I-2369, para. 69. 6 Case C-202/07P, France Telecom v. Commission [2009] ECR I-2369, para. 75. 7 By then, Article 82 EC. 5 undertakings. The document was published in the Official Journal of the EU on February 2, 2009. 8 The Guidance is the result of a process undergone with the aim of shifti g the Co issio s e fo e e t 9 actions from a form-based to an effects-based approach, with consumer welfare at its core.10 Predatory pricing is within the scope of the Guidance. 11 However, even though the Guidance states that in cases of predation consumers are likely to be harmed if the dominant undertaking is likely to be in a position to benefit from sacrifice, 12 it also expresses that identifying consumer harm is not a mechanical calculation of profits and losses, and proof of overall profits is not required. 13 Hence, in the Guidance the Commission does not seem to embrace the recoupment factor as part of the predatory pricing test. Returning to the France Télécom case, AG Mazák is of the view that by avoiding the recoupment of losses as part of the assessment to ascertain predatory pricing, the protection of the position of competitors is preferred over the safeguard of the interests of consumers, which in his words is the primary purpose of Article 102 TFEU. 14 Although the AG and the Guidance follow the same aim (i.e., o su e elfa e as a ai o je ti e , the Co issio s ie a out recoupment is not the same as the AG s. Additio all , a gui g that recoupment is implied when ascertaining dominance, the Commission, in our view, does not take a clear approach as to the relevance and role of recoupment in ases of p edatio . O the othe ha d, if the AG s position is true as to the consistency of recoupment with a consumer welfare standard, the rulings of the EU Judiciary create serious doubts about the goals of EU Competition Law it intends to enhance. 3. Relevance of the subject and purpose of the thesis The 1) Rulings of the GC and the ECJ in the France Télécom Wa adoo ase, AG Mazák s Opi io , a d the 3) Guidance of the Commission are in our view clearly connected -at least- to one subject: The recoupment of losses as a potential factor within the assessment of predatory pricing under EU Law. We believe that in the light of these three elements the subject becomes current and relevant for EU Competition Law, due to the following reasons: 1. The contradictions over the recoupment factor between AG Mazák, the Commission, and the EU Courts. 2. The appa e t footi g of AG Mazák s Opi io o a effe ts-based approach with consumer welfare at its core, which is –in principle- the approach sought by the Commission in its Guidance. 3. Despite this apparent coincidence between AG Mazák and the Commission, they have divergent criteria in what refers to recoupment. 8 Communication from the Commission 2009/C 45/02 or Guidance (2009) as we will refer to it in the content of the thesis. 9 N. Petit (2009), p. 1. 10 Guidance (2009), inter alia, paras. 5, 6, and 19. 11 Guidance (2009), paras. 63 to 74. 12 Guidance (2009), para. 70. 13 Guidance (2009), para. 71. 14 Opinion of AG Mazák in Case C- 202/07 P, France Télécom v Commission, para. 74. 6 4. In spite of –according to some scholars- having the Commission applied an effects-based approach in the decision against Wanadoo Interactive, 15 the Commission still deems that the recoupment of losses is not a requirement for predation or that recoupment is implied in the dominance test. 5. As a result of the foregoing reasons, the existence of a possibility that EU Competition Law and its objectives would allow for a shift in the predation assessment (allowing the recoupment requirement to be part of the predation test), despite the rather contradictory treatment of recoupment by the EU case law. Considering the previous circumstances, this thesis aims to provide reasoned answers to the following problems: In light of the mentioned facts and reasons which we believe make the subject of this thesis current and relevant, our opinion is that the addition of recoupment to the predation test could be a possibility to consider that justifies an analysis from an academic perspective. Should this be a real scenario, there is a twofold interlinked situation posing a problem of legal relevance. The first one is determining whether or not the inclusion of recoupment to the predatory pricing test follows the effects-based approach that underpins the Guidance of The Commission, and having answered that, ascertaining whether or not such inclusion is consistent with the objectives of EU Competition Law. A second situation is that once –if possible- the requirement is added to the test, this will certainly raise the standard of proof for predatory pricing, making it more difficult for the Commission to prove a predation conduct related to prices, as an additional requirement will be included in the applicable assessment. This could facilitate that dominant undertakings drive competitors out of the market more easily, as their conducts may no longer be deemed as unlawful under EU Competition Law. The problem is therefore, in practice, whether or not this consequence will serve the objectives of EU Competition Law, and if the applied objective is consumer welfare, whether or not this would really benefit consumers. It follows from the previous lines that the twofold situation described is linked to a major issue: if EU Co petitio La a d the ele a t ele e ts ithi it allo fo the recoupment factor to be part of the predation test. B ele a t ele e ts 16 we mean: the Commission as the EU Competition Law Authority, the Guidance and the reform it intends to apply, the EU Courts and their case law, and the objectives of EU Competition Law. This means that once we have considered and analyzed both the apparent effects-based approach which would characterize the inclusion of the requirement, and the upraised standard of proof in the 15 N. Petit (2009), p. 2. The author states that the Commission first tried this approach not only in the Wanadoo case but also in the Microsoft I and Telefónica cases. 16 We are not stating that these are the relevant elements of EU Competition Law as a matter of concept or theory. We use the two-word term to appoint the factors upon which we believe the inclusion of recoupment majorly depends on. 7 potential new test for predation , ultimately this thesis aims to provide an answer to the following questions:    After considering the current situation of EU Competition Law, is the inclusion of recoupment in the predation test feasible? If recoupment of losses is added to the test of predation, what would be a likely recoupment screen under EU Competition Law? Under the consumer welfare standard of EU Competition Law, would recoupment of losses play an important role in enhancing consumer welfare? 4. Methodology and disclaimers In developing the contents of the thesis we will conduct and analysis based on three types of sources: case law of the General Court and the European Court of Justice, decisional practice of the Commission, and contributions from EU Competition Law scholars. Additionally, but with ancillary importance, the analysis will extend to literature and legal sources from other jurisdictions, including US Competition Law. We explain the reasons for this approach hereafter. The scope of this thesis is related to a potential shift in the assessment of the Commission and the EU Courts of a particular conduct: predatory pricing. The main goal is to discover if such a shift is possible under EU Competition Law, considering the current situation of this system and if by doing so, the consumer welfare objective will be better promoted. Such study of EU Competition Law entails also the analysis of a particular state of affairs: the reform on the application of Article 102 TFEU. The reform has produced an appropriate environment for changes, which could eventually include also a change in the assessment of certain abusive conducts, including predation. However, any change in this regard depends on various factors, and not only on the sole impulse, for example, of the Commission. The EU Courts also have an important role to play; therefore, an analysis of EU case law is a must. The thesis is a legal study and the analysis will be based almost exclusively on legal considerations, focused on EU Law. Although a description of certain economic-related issues is required, their use is limited and aims at clarifying the legal issues we will address. For example, the analysis of the concept of predatory pricing and the cost benchmarks used for its assessment, necessarily entail the recourse to contributions based on economic theory, such as the Areeda and Turne s stud .17 Other academic sources that we used are both related to economic and legal aspects; this is the case of Scott Hemphill,18 whose analysis is intended to construct a theory about a new recoupment screen. However, this type of sources is mostly used when preparing the first chapter of the study, which deals more with theoretical rather than practical matters. The rest of the contribution is chiefly based on legal-related sources, i ludi g ase la a d ele a t AG s opi io s. We must say that the thesis is not a work of Comparative Law and thus, that the probably traditional approach of comparing EU and US Competition Law is not the substantive component of it. As we said, 17 18 Areeda and Turner (1975). Hemphill (2001). 8 even though at times it is necessary to address elements of US Law, such as the landmark Brooke Group ruling,19 this is done for illustrative purposes. Indeed, the recourse to such a ruling is used to point out the differences between both jurisdictions, but not to assess in depth the characteristics of both systems and draw conclusions out of the comparison. Rather than that, the effort is majorly focused on the analysis of a specific juncture of EU Competition Law (the reform in the application of Article 102 TFEU) and the different elements that may bring about (or not) an appropriate environment for a change. To this end, we believe that the proper exercise deals primarily with analyzing the different factors on which the inclusion of recoupment to the test of predation would depend. Accordingly, these factors are all –we believe- within the context of EU Law, and therefore, the content of the study follows this approach. 5. Contents The content of the study will be developed in four chapters. In a first chapter we will analyze the current treatment of predatory pricing under EU Competition Law. This includes the construction of a concept of predatory pricing, the description of the current assessment used when ascertaining it, including the applicable cost benchmarks, all according to the terms laid down by the ECJ. To this end, a description of the predatory pricing landmark rulings of the EU Judiciary is also included. Finally, an analysis of recoupment as a concept and an introductory description of the current treatment of recoupment under EU Competition Law are included also in the first chapter. The second chapter deals directly with the recoupment factor and the reform to the application of Article 102 TFEU. The chapter describes important elements of the mentioned reform with a special emphasis on the Guidance of the Commission. Then, the most important characteristics of the Guidance are described, focusing on the elements of said document related to predatory pricing and the recoupment factor. The chapter also includes a description of the recoupment assessment conducted by the Commission in the France Télécom case, coupled with conclusions as to the type of recoupment screen the Commission may conduct in future cases. The chapter ends with an analysis of the Post Danmark ruling (ECJ) and its possible connection and influence in future cases of predatory pricing. The third chapter develops two closely related subjects: consumer welfare as an objective of EU Competition Law and the Opinion of AG Mazák about recoupment in the France Télécom case. First the concept of consumer welfare is addressed in the context of the EU case law, comparing it to the concept provided to it by the Commission in the Guidance. Then an analysis of consumer welfare as an objective of EU Competition Law is included, with the aim of determining if it is a valid objective and the major aim of EU Competition Law when dealing with abusive behaviors, and specifically with predatory pricing. The subject will be analyzed due to the existence of EU case law that may pose uncertainty as to the pursued objectives of EU Competition Law when tackling predatory pricing. The chapter finishes with a des iptio a d a al sis of the AG s Opi io . The final chapter is dedicated to the conclusions of the study. 19 Brooke Group Ltd v Brown &Williamson Tobacco Corp 509 US 209 (1993). 9 10 Chapter I: The current treatment of Predatory Pricing under EU Competition Law This chapter provides the reader with a study of the definition and relevant elements of predatory pricing, and an explanation of the treatment given by EU Competition Law to this conduct. The definition of predatory pricing as an anticompetitive conduct has been analyzed by several scholars in and outside the EU.20 However, for purposes of this thesis, the construction of the analysis relies mainly on European Competition Law publications.21 This goes in line with the fact this study is related directly to EU Competition Law and a potential treatment of predatory pricing by the EU Competition Authorities and Courts. We will also analyze the factors involved in the feasibility of predatory pricing, among which we will find the recoupment of losses. At this point, an explanation of the importance of recoupment will be done. Also, a discussion will be made as to other problems 22 that may rise from the methods for testing recoupment within an assessment for predation. For a proper understanding and complete analysis of the subject, a description of the cost assessment used by the European Judiciary and the Commission is included. The cost assessment is based on the so called cost benchmarks, which are part of the assessment of predatory pricing. Finally, we include a description of the most important European case law on predatory pricing and of the current treatment of recoupment of losses under EU Competition Law. This as well as the rest of the content of this chapter is necessary in order to have a complete understanding of the current treatment of predation. Once we do this, in the following chapters we can focus our attention on the recoupment factor within the test of predation, the specific subject of this study. 20 See inter alia, Areeda and Turner (1975), p. 697, Scott Hemphill (2001), p. 1581, Farrell and Katz (2005), p. 203, D.G. Goyder (2003), p. 286, and Van Bael & Bellis (2009), p. 934. 21 As to other specific subjects, this limitation, i.e. only consulting EU authors, does not apply as we will also consult other sources outside the EU. 22 These problems are mentioned by Scott Hemphill (2001) and may be produced when structural features of the market are not considered and as a result, according to his findings, real predatory conducts may have been acquitted under the current assessment for predation of the US Supreme Court. Infra, in this chapter we will study the importance of recoupment and will describe some of the findings of Hemphill from his study about the calculation of recoupment of losses under US Law. 11 3. The concept of predatory pricing23 Providing a generally accepted concept of predatory pricing is not a simple task, mainly for two reasons: i) The numerous and extensive discussions mentioned by Competition Law authors 24 that question the very existence of predatory pricing as a valid concept or anticompetitive conduct, 25 and also related to this reason, ii) the several components and factors that intervene in conforming a predation conduct, in the absence of which we could be facing a perfectly competitive action.26 The divergence of opinions regarding predatory pricing, related mostly to whether predation can be considered an anticompetitive conduct produce in many cases27 non-convergent and even conflicting concepts. Due to this, we will provide a generally accepted definition of predatory pricing within the framework of EU Competition Law, and then we will analyze in a more detailed manner the elements and characteristics of predatory pricing according to EU Competition Law. Predatory pricing in the definition given by Niels, Jenkins, and Kavanagh (Niels et al), involves a dominant company deliberately incurring short-term losses to eliminate competitors or to prevent entry into the market, in order to charge monopolistic prices in the long term .28 We adhere to said definition. 23 For clarity purposes we will make reference both to predation and predatory pricing as the same term. Even though predation may refer to a more general concept, in most of the cases it is directly related to prices. We, in this particular aspect, take the view of Bishop and Walker when they state that Predation is said to occur where aggressive competition, normally in the form of low prices, either induces firms to exit the market or deters new firms from entering (emphasis added). (2002), p. 218. 24 E.g., Farrell and Katz indicate that: Economists and courts have often been skeptical of claims of predation for several reasons. First, consumers often gain from the intense rivalry during the predatory phase, even if one of the rivals has predatory intent. Second, many have questioned whether rivals will remain weakened in the recoupment stage. If rivals can bounce back, then predation will be unprofitable and consumers would suffer no second-stage welfare losses even if a firm irrationally engaged in predation. Lastly, it may be hard to distinguish harmful predation from beneficial competition. (2005), p. 204. 25 As pointed out by Jones and Sufrin: Some economists have argued that it is hardly ever a rational business st ateg a d that is e a e. This ie as fa ousl adopted Bo k: it see s u ise…to o st u t ules a out a phenomenon that probably does not exist or which, should it exist in very rare cases, the courts would have grave diffi ult i disti guishi g f o o petiti e p i e eha io . (2011), p. 392. See also Buttigieg: Indeed some commentators go to the extent of advocating the non-intervention of antitrust law in all cases of alleged predatory price-cutting on the grounds that it is doubtful whether this would ever be profitable to the predator. (2009), p 181. 26 The problem of defining predation and the several conflicting and even confusing definitions of the conduct is exemplified by Bishop and Walker in the following text: Various definitions of predatory behaviour have been advanced in the academic literature. Ordover and Willing define predatory actions as those that are unprofitable without the exit of a competitor that it causes. Evans and Schmalensee define predatory acts as those that are rational only if they chasten or eliminate competition. Fisher suggests that predation involves the deliberate sacrifice of profits in order to gain or protect a dominant position. Viscusi, Vernon and Harrington hold that p edato eha iou is eha iou al ulated to e lude f o the a ket a e uall o o e effi ie t o petito . These various definitions differ. For example, whereas the UK OFT, Ordover and Willing and Viscusi et al. all require the e it of a o petito , E a s a d “ h ale see e ui e o l that o petitio is haste ed , hilst Fishe fo uses on the acquisition or maintenance of a dominant position. This makes it difficult to distinguish between predatory behaviour and normal competition. (2002), p. 219. 27 Such as the ones mentioned in the previous footnotes. 28 (2011), p. 198. 12 However, following the same authors, this sole definition makes it difficult to distinguish predation from fierce but healthy price competition. Therefore, further elaboration on the concept is required. Predatory pricing involves a process carried on during a time period by a dominant undertaking. This process is conducted in stages: Predatory pricing can be conceptualized as a three-stage process… In the first stage you have a monopolist (or a dominant company that is happy to tolerate some smaller rivals in the market). The second stage is where the dominant company engages in below- cost pricing against a new or existing competitor. In the third and final stage, the rivals have been forced out of the market and the dominant company can reap monopoly profits again.29 Therefore, a more complete conceptualization of predation involves a dominant undertaking engaging in below-cost pricing (losses) during a period of time, so that when its competitor or competitors are taken out of the market (market pre-emption), it can charge monopolistic or above competition prices (profiting again).30 It follows form the previous definition that predatory pricing entails not only the intention and corresponding actions to pre-empt the market or prevent new entrants in the market, but also a well formed strategy that results in an increase of revenues. It must be clear though, that in practice the prosecution of a predatory practice occurs at the early stages of the conduct and before the whole anticompetitive effects are displayed, such as the market pre-emption and the fore coming fixation of anticompetitive prices. This of course happens for practical and even obvious reasons, for it could become pointless 31 to prosecute unlawful actions only until the totality of their harmful effects are displayed, if otherwise they can be legally identified and corrected 29 Niels et al (2011), p. 199. This definition is very similar to the one used by the Commission in its Guidance, with a slight difference that comes up when the Commission uses the phrase ith a ie , which seems to entail a reasonable expectation (para. 70) of increase in market power and not a verifiable effect of the conduct. We will discuss these peculiarities of the enforcement priorities of the Commission later in this chapter and in Chapter II. The Guidance (para. 63) defines predation as follows: …that a dominant undertaking engages in predatory conduct by deliberately incurring losses o fo egoi g p ofits i the sho t te efe ed to he eafte as sa ifi e , so as to fo e lose o e likel to foreclose one or more of its actual or potential competitors with a view to strengthening or maintaining its market power, thereby causing consumer harm (emphasis added). 31 The issue about the convenience or not of an early intervention by the Competition Authorities was addressed by Emil Paulis in his speech at the Lisbon Conference on Competition and Economics, when he justified an early intervention in the case of predation because consumers need protection against a long-term harm: A simple example is predation. In this case, there is no harm to consumers in the short term. Consumers are charged very favourable prices. The harm comes much later when exclusion has taken place and when the company engages in recoupment. So, there are a number of situations where the Competition Authority has to protect the consumer interests, not just in the short term. There would be a fallacy if we were to look only at the short-term effects. This approach, in my view, largely closes the gap between those who take a more structuralist approach or a more competitive process-type of approach and the approach which we now take on market outcome/consumer harm effects in relation to the application of Article 82. Mateus, Abel M. and Moreira, Teresa (2010), p. 163. 30 13 beforehand.32 This assertion should not be confused with an ex-ante control against predation, as the conduct of predation is prosecuted, corrected, and sanctioned until it has been verified. This fact will also have a strong impact o the effo ts a d a tio s fo de onstrating predation, when not only a mere conduct but also proof of its effects may be required within the standards, methods, and means of a specific legal system. For example, the following extract from the Report on Predatory Pricing33 shows only some of the differences regarding the effects that must be proven to show predation in various legal systems. Some of these effects are not verifiable as actual facts, but as likely to happen. Such would be the case of the likelihood of lesse i g o petitio , following the Canadian response and the possibility of exclusion of competitors , present in the Korean example. Whether effects must be demonstrated. Twenty-o e espo di g age ies a s e ed es. Fo e a ple, the I ish espo se stated that the i pa ts o a ket e t a d o su e s ould al a s e dealt ith as ha to o su e s ust e sho fo a a use to e esta lished. The Canadian response stated that a e ui ed ele e t of p edato p i i g is that o petitio has been, is being, or is likely to be lessened or prevented substantially. The espo se f o Ko ea stated that fo p edato p i i g to e a a use, it has to e p o e whether an exclusion of competitors can be caused, in addition to the fact that the price was set at a level lower than the o al t ade p i e. The ‘ussia espo se stated that u de its la o e i g p edato p i i g of o odities, [t]he o odit p i e is ot e og ized monopolistically low if its establishment has not resulted in restriction of competition because of reduction of the number of economic entities [.] … (Underscore added) 4. Factors involved in the feasibility of predation The analysis of the factors involved in the feasibility of predation is of primary importance. First of all, because by setting their explanation, the distinction between acceptable commercial practices and predatory pricing becomes clearer, hence the identification of really unlawful conducts becomes easier too. In practice this is relevant as it prevents competition authorities from wrongly prosecuting and condemning lawful activities and therefore, focus their efforts in real predation cases; this is also meaningful in regards to consumer welfare as consumers would not be deprived from enjoying the possible benefits of lawful commercial conducts or strategies, protected from wrong prosecutions and/or judgment. As explained by Niels et al: A false positive- i.e., labeling a practice predatory when in fact it is not- leads to welfare losses, as prices are kept too high and inefficient suppliers are sheltered from price competition. On the 32 The matter has been explained with clarifying analogies by the N. Kroes, cited by Petit, Nicolas: We will not wait until [foreclosure] effects have manifested themselves. If we wait until rivals are forced to leave the market then we have two serious problems. First, you cannot resuscitate a corpse. No matter how effective the regulatory intervention, if it only happens after exit has occurred, then the damage to the market may be permanent. Second, such intervention will completely miss many examples of consumer harm that weaken competitors, but do not kill them. Competitors may be wounded, confined to a small corner of the market, but not killed. Leaving these cases to one side is a recipe for serious under-enforcement. (2009), p. 9. 33 (2008), p. 23. 14 other hand, a false negative- allowing prices that are in fact predatory- results in some short-run allocative inefficiency, as well as the social cost of monopoly in the longer run. 34 Secondly, because among these factors lies the recoupment of losses, which is the specific subject (within the generic subject of predatory pricing) we are dealing with. Despite the different degrees of importance it has in the jurisdictions worldwide, recoupment is generally recognized at least as a relevant factor in examining conducts suspected of predatory pricing. EU Competition Law is not an exemption: In EU competition law, the feasibility of recouping losses is now also recognized as a relevant factor in predatory pricing cases, although it had previously been rejected and is still not given the same de isi e i po ta e as it is i the U“A… 35 Having said that, we believe that a responsible legal academic analysis of predatory pricing must contain an analysis of the factors involved in its feasibility, which we will set hereafter. For order purposes the recoupment factor will be addressed at the end of this section. 2.1. The structural features of the market(s) We will subdivide this factor in two: The multi-market factor and the market power factor. 2.1.1. The multi-market factor: According to Jones and Sufrin,36 predatory pricing is more feasible when the undertakings involved operate in multiple markets. Their explanation contains three specific reasons: a) If operating only in one market, other possibilities such as absorbing or accommodating the incumbent´s competitor are more feasible than predation (to attain the same goals). b) When operating only in one market, the incumbent undertaking faces the problem of the ousted o petito s possi le dispose of its assets to other competitors, making it more difficult to recover from the losses resulting from the price-cutting stage. This would be due to its inability to off-set the losses on the market with the profits on other markets. c) Related to the effects of predation on the predator´s reputation,37 a multi-market operating firm may find predatory pricing attractive as a means to establishing a reputation for aggressive reaction to competition. 34 (2011), p. 201. Niels et al (2011), p. 202. 36 Predatory pricing is often considered to be feasible only where firms operate multi-market because if the firm operates in only one market it is more rational for it to absorb the new entrant (by merger or take-over, insofar as that is permitted by the competition authorities), or to accommodate it, rather than incur greater losses by u de utti g… The e is also the p o le that i e iti g the e t a t a dispose of its assets to other competitor still on the market. Where a firm is multi-market, however, it may be able to off-set the losses on one market from the profits on another. Moreover, a firm which establishes a reputation for aggressive reaction to competition in one market may deter entrants into others, so predation in one market may protect several others (2011), p. 393. 37 The p edato s eputatio is a othe fa to to o side , a al zed late in this section. 35 15 2.1.2. The market power factor: The market power factor is related to the p edato s degree of dominance.38 A successful predation conduct requires a sufficient market power during all the stages of predation. Also, closely related to this, the financial position of both the predator and the victims is an important factor, as the ability to sustain predation (predator) and oppose it (victims) depends greatly on their access to funds. 39 In Bishop s a d Walke s ords, having access to greater financial resources than your rivals is often important.40 2.2. The time factor and the amount of losses In general, the longer the period of low pricing is, the larger the amount of losses, and the lesser the possibilities that predation will be profitable. Although time is not the only factor determining the amount of losses a predator incurs, it is yet very important. If stage 2 drags on, the dominant company keeps building up losses and the prospect of ever recouping these losses diminishes.41 The amount of losses incurred by the dominant undertaking is a decisive factor, independently if caused by long periods of low pricing or not, which is also –but not the only one- a factor to take into account. This is how Bishop and Walker explain it: …even if a predating firm can raise prices above the competitive price level after excluding a rival, this may not be enough to make predation a profitable strategy. This will depend on how great the losses are that the predator incurs in the short run, how long the short run is, how soon and by how much above the competitive price level the predator can raise prices after excluding a rival and for how long. 2.3. Barriers to enter and re-enter the market It is also commonly accepted42 as a condition for predation, the necessary existence of barriers to enter and re-enter the market for the dominant undertaking to be able, both to raise prices and make excess profits, once its competitors exit the market. Otherwise, it is argued, although one entrant is knocked out by the low prices others will enter when the incumbent raises them again. The incumbent will never be able to enjoy the fruits of predation. 43 38 According to Niels et al, Successful predatory pricing requires having market power now (during the predation period) and in the future (during the recoupment period). As to current market power, the dominant company must expand output in order to depress the overall market price and put pressure on its rivals. To have a strong impact on market price, it needs a substantial market share from the start. (2011), p. 204. 39 Bishop and Walker provide us with a further explanation of this factor, stating that predation occurs on the assumption that the predating firm can sustain short- run losses, but the predated firm cannot. Thus, it is less likely for predation to happen if the target competitors are able to finance short-run losses due to the expectation of future profitability. This expectation lies in the fact that a predated firm may consider to stay in the market, knowing that the predator cannot sustain losses forever. (2002), p. 221. 40 (2002), p. 221. 41 Niels et al (2011), p. 203. 42 Inter alia, Bishop and Walker (2002), p. 220, Niels et al (2011), p. 204, and Jones and Sufrin (2011), p. 392. 43 Jones and Sufrin (2011), p. 392. 16 2.4. Effects of predation on the predator’s reputatio The p edato s eputatio i a d outside the a ket can create an entry barrier, in prejudice of future competitors. The doctrine suggests that this effect not only could affect competition by preventing entrants into the market, but that it may be a source of recouping the losses incurred by charging below-cost prices.44 The importance of the effects of reputation itself is stressed both by Bishop and Walker and Niels et al, who provide examples of several cases where the public knowledge of aggressive competitive strategies -including predatory pricing-45 prevented competitors from entering specific markets. Fo the spe ifi ase of p edatio the i po ta e of the p edato s eputatio is e e plified and Walker in the following description of the General Foods case: Bishop A possible example of aggressive responses to entry designed to create a reputation for predatory behavior is provided by General Foods in the United States in the 1970s. General Foods was the producer of Maxwell House Coffee. At the time General Foods had a market share in the eastern states of a out pe e t. Whe Folge s, a a d ased i the este states, t ied to enter various eastern markets General Foods responded with sharp price decreases in those a kets he e Folge s had e te ed. This st ateg su essfull dis ou aged Folge s f o further entry in the eastern states.46 2.5. The recoupment of losses We previously stated that recoupment is recognized by EU Competition Law as a relevant factor for the feasibility of predatory pricing.47 We also mentioned that the degree of importance of this factor to determine the feasibility of predation changes from one legislation to another. As a matter of fact, in legal systems such as the US, recoupment is not only a factor considered for the feasibility of predation, The point is stressed by Niels et al when they comment that: The i u e t s i te est is the efo e to t to sig al to all entrants that it is strong. Imagine that the incumbent is a multi-product firm. Predating in one market can bring wider benefits if it establishes a reputation as a strong incumbent that will always fight predation across its other key markets. Recoupment is no longer just about future profits for the product that is subject to predation, but also about protecting profits earned in other markets in which the incumbent is dominant. (2011), p. 206. 45 (2002), pages 223 and 224; where the authors mention inter alia the cases of General Foods, and the US airline industries in the 1990s. In both cases pricing below cost conducts where identified, however the General Foods case was dismissed by the FTC and the other example refers to a consultative document issued by the US Department of Transportation. 46 (2002), p. 224. 47 In the France Télécom ruling –which we will further analyze in this chapter- the ECJ explained the degree of importance of recoupment within the assessment of predation, after however holding that proof of recoupment is not a necessary precondition for predation: The interpretation does not, of course, preclude the Commission from finding such a possibility of recoupment of losses to be a relevant factor in assessing whether or not the practice concerned is abusive, in that it may, for example where prices lower than average variable costs are applied, assist in excluding economic justifications other than the elimination of a competitor, or, where prices below average total costs but above average variable costs are applied, assist in establishing that a plan to eliminate a competitor exists. Case T- 340/03 France Télécom v. Commission, [2009] ECR I-2369, para. 111. 44 17 but one element required to ascertain predation,48 that is, the possibility of recoupment of losses must be proven to establish the existence of predatory pricing. We will illustrate this point with the differences between EU Law and US Law, due to the continuous comparison (at least from an academic standpoint) 49 between both systems. In regard to these differences, some authors consider that the gap between both systems in the recoupment factor is narrow, as in recent cases the Commission has taken the position that the proof of recoupment of losses is implied in the establishment of dominance, 50 a sine qua non pre-requirement for every exclusionary abuse case. By doing this, the Commission rendered a major degree of importance to the factor itself than simply excluding it from the predation test, even if recoupment is not required to be proven separately. However, there are still some important differences between the legal systems, as Patrick Rey illustrates, as follows: Some experts say that recoupment is the gap between the US and the EU, and thus the gap is not that large because recoupment is basically dealt with when establishing dominance. There may e so e t uth i this state e t, ut he e a e looki g fo a s e a io f o a e o o ist s standpoint, what we would like to see is, for example, a scenario where the target is financially weak, so this is not only about entry barriers. Those financial problems, and the characteristics of the target, would be part of the recoupment story. Since those problems are not going to be particularly dealt with at the first separate stage of dominance, I think that there is still some discrepancy between the two approaches on both sides of the Atlantic. 51 After having stressed the fact that over the recoupment factor there are important differences in both the EU and US legal systems, we will first explain what is meant by recoupment of losses and why is it generally considered –at least- as an important factor in the feasibility of predation. Then, later on this chapter when we analyze the benchmark rulings of the European Judiciary on predation, we will explain As pointed out by Gurpegui Ballesteros and Szarka (2009): A da ge ous p o a ilit of e oup e t is a p erequisite to prove predatory pricing under US antitrust law. 49 To give only one of the various possible examples, we can infer the importance given to the US legal system as a whole, in the reference made to it by Patrick Rey in the following remark: I had the pleasure of coordinating the group of experts that produced the EAGCP report in the summer 2005 on exclusionary abuses. The staff discussion paper of September also focused on the same exclusionary abuses, which is moreover the type of abuse usually addressed in other parts of the world, including on the other side of the Atlantic; it is fair to say that there is quite a consensus on this type of abuse, at least on the broad principles. (emphasis added) Mateus and Moreira (2010), p. 191. 50 In their article, Gurpegui Ballesteros and Sza ka e tio A) Mazák s Opi io i the France Télécom case, indicating that he eje ted the Co issio s li e of argument that in Europe and under Article 82 of the EC Treaty, recoupment of losses is implied by the assessment and establishment of do i a e… They further point out that AG Mazák considered that the establishment of dominance is often based on historical market conditions while proof of the possibility to recoup losses is an ex-ante and forward-looking assessment of future market conditio s. Mo eo e , a al ost e a t ep odu tio of the Co issio s ite io a e fou d i pa ag aph of the DG Competition Discussion paper on the application of Article 82 of the Treaty to exclusionary abuses (2005), which however, as we will later find out, it is not the exact wording used in the final product, the Guidance. In what is important for our study, paragraph 122 of the Discussion Paper expresses: As dominance is already established this normally means that entry barriers are sufficiently high to presume the possibility to recoup. The Commission does therefore not consider it is necessary to provide further separate proof of recoupment in order to find an abuse. 51 Mateus and Moreira (2010), p. 196. 48 18 the treatment that the Commission and EU Courts give to recoupment, and we will deal in depth with this particular subject in the remaining chapters. 2.5.1. Recoupment as part of the definition of predatory pricing: When defining predatory pricing we previously indicated that viewed as a three-stage process, during the third and final stage of predation the incumbent is able to raise prices at a profitable/anticompetitive level. Many of the authors we consulted consider that in this third stage the incumbent must be able to off-set any losses incurred during the second stage. For example, Bishop and Walker indicate that a final stage of predatory pricing the predating firm is then able to raise prices above the competitive level and therefore make profits over the long run sufficient to outweigh any short run losses.52 Such is also the view of Eugene Buttigieg, who considers that a strategy of predation must encompass a time period long enough to recoup losses and increase profits. Following his approach, only once this is a o plished p edatio e o es a successful o e ial st ateg fo the i u e t u de taking. The statement is summarized in the following text: Predatory pricing should not be viewed as a static phenomenon merely entailing a short-term (below-cost) pricing but as a long-run strategy followed by the market monopolization of the incumbent for a period sufficient to permit it to recoup and increase its revenues; alleged predatory schemes not displaying such long-run properties should hence be of no antitrust concern. 53 Recoupment is precisely this ability to off-set or outweigh the losses incurred during the below-cost period of predation.54 Most of the consulted literature indeed considers recoupment as part of the definition of predation, such as the following statement provided by Jones and Sufrin: 55 The concept of predatory pricing rests on the assumption that the predator sacrifices short-term profits for future gains. It hinges on the possibility that the predator can recoup its losses, i.e., that short-term loss of profitability is more than compensated for by long-run profitability when, afte the o petito s e it, the u de taki g a aise p i es to o opol le el. However, when Massimo Motta defines predatory pricing, it seems that the recoupment factor is not part of the concept, at least as to what he calls the elements for the identification of predatory behaviour in practice. In his definition, Motta considers that identifying the existence of sufficient market power -in addition to the below cost pricing period- is sufficient to identify predation. In his own words: 52 (2002), p. 220. Buttigieg, Eugene, Competition Law: Safeguarding the Consumer Interest, p. 177. 54 In the words of Niels et al …the recoupment test comes down to measuring whether the losses in stage 2 are outweighed by the subsequent profits in stage 3. They add: In this sense, predation strategies are similar to any other financial investment decision –predation is the initial investment followed by payback once market power is achieved or restored- a d this alls fo a fi a ial a al sis… (2011), p. 202. 55 (2011), p. 400. 53 19 Predatory pricing therefore occurs when a firm sets prices at a level that implies the sacrifice of profits in the short-run. This definition, for the moment still vague, contains the two main elements for the identification of predatory behaviour in practice: first, the existence of a shortterm loss; second, the existence of enough market power by the predator so that it can reasonably expect to be able to raise prices so as to increase profits in the long-run once a rival (or more rivals) has been driven out of the market.56,57 2.5.2. The importance of recoupment: Most of those who view recoupment as part of the concept of predation consider that in the absence of recoupment, we are dealing with a non-anticompetitive conduct. If we take this position, such as the US legal system does, the degree of importance of recoupment is enormous, for the simple reason that without it predation is a lawful conduct. Indeed, in the view of the US Supreme Court in the Brooke Group ase, i the a se e o e oup e t predatory pricing produces lower aggregate prices in the market, and consumer welfare is e ha ed. 58 Notwithstanding the last statement, the allowance of other degrees of importance to recoupment does not necessarily mean that is has been excluded from the concept of predation. In this vein, as we indicated in previous lines, some positions within the EU Competition Law consider that recoupment is present in other elements of the predation test; such is the case of the Co issio s view towards 59 dominance in the France Télécom case. Furthermore, other sources consider that recoupment is not merely an arithmetic calculation, but that it can be ascertained through other factors, for example, the reputation effects of predation. There are even those who fully support the proof of recoupment as part of the assessment of predation, but propose a new optimal and purified recoupment screen: narrow and deep. This is the view of Scott Hemphill, who inter alia criticizes the current standards of the US Supreme Court in the assessment for recoupment in predation cases, affirming that the current recoupment doctrine is valuable as a means to dismiss all claims;60 he also stands up against what he calls a conduct-based screen present in the actual 56 (2005), p. 412. We may reasonably argue that with this statement Motta was not defining predation itself but the elements for its assessment. All the same, the exclusion of recoupment from his definition of predatory pricing is confirmed when later in his contribution he proposes his test for predation. His proposal indeed excludes recoupment as part of the test: This assessment should be in any case done from an ex ante, rather than ex post, vantage point, and the firm could not defend itself showing that after all it did not manage to recoup losses. In other words, the predator should not benefit from the fact that it has miscalculated the opportunity to recover losses, o that the p e has tu ed out to e a toughe ookie tha i itiall thought, o that p o pt a ti-trust action has led to conclusion of the predatory episode before it could achieve its aim. Likewise, if the market power and price above cost tiers of the test are both satisfied, one should not admit as a defence the fact that predation has not been effective and exclusion has not happened. (2005), p. 450. 57 As we will see later i this hapte , a si ila o di g as Motta s is used the Co issio i its Guida e. 58 Brooke Group Ltd v Brown &Williamson Tobacco Corp 509 US 209 (1993). 59 Case C-202/07P, France Télécom v. Commission [2009] ECR I-2369 60 This seems to be his position when he refers to, in his own words, the extreme skepticism of Brooke Group, e sh i ed i the u de l i g s o se sus a o g o e tato s : that p edato p i i g is a ely tried, and e e o e a el su essful . (2001), p. 1600. It is important to note that Scott Hemphill, expert in economics and Competition Law, considers predatory pricing not only as possible, but also as underestimated by the US courts: However, economic theory and empirical work over the past twenty years suggest that predatory pricing occurs 20 test, in comparison to a more convenient structured-based approach. As Hemphill points out in his conclusions, he proposes a basically new e oup e t s ee –narrow in the sense that it excludes conduct, deep in that it includes a robust set of structural factors-with existing court doctrine as described i the B ook G oup opi io . He phill s a alysis and conclusions propose that when ascertaining recoupment, the courts refrain from applying a merely arithmetic test (conduct- based screen) that compares losses with likely profits, and instead, they look at the structural features of the market61 where the predation conduct took place (structured-based test). In his article he explains the distinction between the two different criteria as follows: To understand whether recoupment is likely to succeed-that is, whether an investment in belowcost pricing will pay off for a would-be predator-there are two kinds of inquiries a court could undertake. It could look at industry structure, asking whether there are barriers to entry that will allow the predator to profit from its ill-gotten monopoly once the competitor has been eliminated (or co-opted). Alternatively, a court could focus on the conduct of the predator. A si ple, pe haps si plisti , a to a al ze o du t ould e to add up the p edato s losses i the period of predation, add up the likely profits after the period of predation has ended, and see which is larger. 62 The efe e e to He phill s fi di gs shed a light o e the o ple it a d a iet of pe spe ti es that arise from this single factor (recoupment). Based on his conclusions it seems clear that even if proof of recoupment is required, its sole assessment is problematic and may have produced false negatives in the past (in the way it has been conducted by the US Judiciary). He phill s findings are useful as they pose questions related to the convenience of applying this rather p o le ati additio al ele e t to the predation assessment. All the more, this question grows in importance when within the present status quo of EU Competition Law, the Commission could be considering recoupment for granted in the dominance test, which entails indeed the analysis of structural elements of the relevant market. 63 Due to this, one could even speculate that the u e t Co issio s ie could have at least considered so e of He phill s o lusio s or similar approaches. However, we may also analyze another perspective, such as considering most of the recent academic literature which envisages recoupment as part of the definition of predation. In line with this position, more often, and is more frequently successful when it does occur, than courts have typically assumed. (2001). P. 1583. 61 When Hemphill mentions the additional structural factors related to recoupment, it is interesting that he, as Niels et al, seems to consider the reputation of the predator in other market as a source of recouping losses: But economic theory and empirical research over the past twenty years suggests additional structural factors that might affect the likelihood of recoupment. In particular, information asymmetries and linkages across markets might make recoupment more likely, by allowing a predator to build a reputation for predation and/or to send misleading signals about demand and cost in a market. (2001), p. 1589. 62 (2001), p. 1586. 63 Jones and Sufrin (2011) enumerate the standard factors usually considered relevant in the appraisal of dominance, as the following: The market share of the leading firms, variability of market shares, existence of substitute products, barriers to entry, barriers to expansion, existence of spare capacity, and the nature of competitive interaction in the market. p. 289. 21 the requirement of proof of recoupment of losses as an independent element of the assessment for predation would result more appealing. Further analysis of the recoupment factor and possible elements that may suggest a leaning towards this approach will be done in the following chapters. 3. Predatory pricing among the conducts forbidden by Article 102 TFEU Predatory pricing as an anticompetitive and unlawful practice is one of the conducts prohibited by Art. 102 TFEU.64 Article 102 prohibits the abuse of a dominant position (of undertakings) within the Internal Market or a substantial part of it, in so far as it affects trade between Member States. The concept of abuse of dominance has been formulated by the European Judiciary, as the Treaty only provides a number of examples of prohibited abusive conducts. The concept of abuse given by the ECJ in the Hoffmann-La Roche case is the following: … a o je ti e o ept elati g to the eha io of a u de taki g… hi h th ough e ou se to methods different from those which condition normal competition in products and 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.65 3.1. Predation as an exclusionary conduct The traditional66 categorization of the abusive conducts under the meaning of Article 102 TFEU is as exploitative and exclusionary abuses. The difference between exclusionary and exploitative abuses is explained as follows: Abuses are often classified as either exclusionary or exploitative, though the Commission regards the disti tio as ot of g eat i po ta e. A e lusio a a use is o e of those… he e its effects are primarily on the structure of the market, for example by weakening competitors through raising their costs, refusing to deal with them or denying access to essential facilities. By contrast, exploitative abuses are aimed directly at consumers, for example by imposing excessive prices or unreasonable terms and conditions; of course a successful use of exclusionary abuses 64 The text of Article 102 TFEU provides: Any abuse of one or more undertakings of a dominant position within the internal market or a substantial part of it shall be prohibited as incompatible with the internal market in so far as it may affect trade between Member States. Such abuse may, in particular, consist in: a) directly or indirectly imposing unfair purchase or selling prices or other unfair trading conditions; (b) limiting production, markets or technical development to the prejudice of consumers; (c) applying dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage; (d) making the conclusion of contracts subject to acceptance by the other parties of supplementary obligations which, by their nature or according to commercial usage, have no connection with the subject of such contracts. 65 Case 85/76 Hoffmann-La Roche v Commission [1979] ECR 461, para. 91. 66 About this traditional categorization Ioannis Lianos comments that -in the Guidance- the Commission still sticks to it. As we will see in this section, he also points out that the Guidance introduces a new categorization that locates predatory pricing among the price-related exclusionary abuses, within what he refers to as the price/nonprice dichotomy: The Commission Guidance on its enforcement priorities is not advancing a full effects-based approach in the enforcement or Article 82. The Commission observes that its enforcement priorities will focus on those t pes of o du t that a e ost ha ful to o su e s , a d it e ludes form the scope of this Guidance o du t hi h is di e tl e ploitati e of o su e s thus ai tai i g the t aditio al disti tio et ee exploitative and exclusionary abuses under Article 82. (2009), p. 44. 22 may well lead on to subsequent use of exploitative ones, or they may be used simultaneously and in combination. 67 Under this categorization exploitative abuses are those where the abuse directly exploits suppliers or customers and exclusionary abuses are those where the abuse does not directly harm the dominant fi s t adi g pa t e s o usto e s, ut o etheless esults i a lesse i g of o petitio . 68 Such classification is still kept by the Guidance, which specifically in paragraphs 6 and 7 states that the enforcement priorities of the Commission will focus on exclusionary conducts, however recognizing that [ ]o du t hi h is di e tl e ploitati e of o su e s, … is also lia le to i f i ge A ti le . 69 Later in the Guidance the Commission sets specific considerations (paras. 23 to 27)70 for what it classifies as price-based exclusionary conduct. According to the Guidance, when it refers to Predation in paragraphs 63 to 74, it relates the assessment of this conduct to said considerations, i.e., inter alia, to the cost benchmarks of paragraph 26. Following this specification of the Guidance, Ioannis Lianos describes a new classification introduced by it, the so-called price and non-price exclusionary abuses: The most important new classification introduced by the Commission Guidance is the distinction between price- and non price-related exclusionary abuses. These two categories permeate all others, in the sense that the legal standards are profoundly different depending on the characterization of a restriction as being related to price or non-price exclusionary conduct. 71 Therefore, within the categorization set by the Guidance, predatory pricing may be classified as a pricebased exclusionary conduct. Accordingly, the subtitle given to paragraphs 23 to 27 of the Guidance reads: C. Pri e- ased e lusio ar o du t . Then, the first sentence or paragraph 23 indicates that: The o side atio s i pa ag aphs to appl to p i e- ased e lusio a o du t. 72 Although the intention of an undertaking incurring in predation is to pre-empt the market, thus, affecting directly its competitor or competitors by excluding them from the market; the legal prohibition against predation tries to protect the consumer from the long-run effects of predatory pricing. As we previously explained, once the market is pre-empted the incumbent undertaking will seek to off-set the losses suffered during the below-cost pricing period by setting monopoly prices. Therefore, even though predation is not classified as an exploitative conduct because its direct effects are not aimed at consumers, in the long-run a successful predation strategy will have negative effects on consumers, who will suffer from anticompetitive prices. 73 67 Goyder, D.G. (2003), p. 283. Van Bael & Bellis (2005), p. 904. 69 Guidance (2009), paragraph. 7. 70 We will analyze such considerations, later in this chapter -when we refer to the Cost Benchmarks- and in Chapter II. 71 (2009), p. 45. 72 Guidance (2009), p. 23. 73 The effects on consumers during the stages of predation are described by E. Buttigieg, as follows: Here (predatory pricing), unlike excessive pricing policies, the target is not the exploitation of the consumer but the elimination of competition from the market; and yet though the consumer might feel that the sudden very low 68 23 4. The cost assessment for predation We will deal in this section with the so-called cost benchmarks used by the Commission and the European Judiciary when conducting the assessment for predation. As many other competition law subjects, the definition of the cost benchmarks is an economics-related topic. As such, the application of one or other cost benchmark must be justified by economic theory. However, even within economists several discussions have risen as to which are the correct parameters to apply. Emil Paulis provides us with an example of such discussions: Then, in predation there is a debate about the cost benchmark. Should we go for the avoidable cost approach or should we go for a different approach which would be also to take account of certain predation scenarios above avoidable cost, possibly up to the long-run incremental cost level? This is not a trivial question because in the case of a dominant firm the incumbent may be able, over a certain period of time, to price at a level of average avoidable cost. It can sustain that level of pricing for some time and can thereby keep out of the market other competitors who are new entrants who may not yet have reached that level of efficiency, but who would be in the range between avoidable cost and long-run incremental cost, therefore including the rightly allocated fixed costs. 74 The discussion in the example provided by Paulis is only one amongst many. As interesting as these discussions can get, we consider that they lie outside the scope of this thesis. Therefore, we will limit our study to the description of the cost benchmarks applied by the EU Judiciary and the Commission, and the role they play within the assessment for predation in EU Competition Law. When explaining this, we will make reference to the relevant EU case law in predatory pricing cases, i.e., the AKZO,75 Tetra Pak II,76 and France Telecom77 rulings with the aim of providing the reader with a view on the current cost assessment for predation. 4.1. The purpose for the application of cost benchmarks We mentioned earlier that one of the problems with defining predation as an unlawful conduct is distinguishing it from a competitive conduct. Moderate price cuts benefit consumers and foster competition, a tuall the Co petitio La s o je ti e of o su e elfa e e o passes the idea of lower prices for the benefit of consumers. I the o ds of Ei ik Oste ud, [t]he objective of Article 102 TFEU is to protect competition as a means of furthering economic efficiency, in the interests of consumer. The provision aims to protect competition as a means of providing lower prices and better quality products and services, to the e efit of o su e s. 78 bargain prices offered by the dominant undertaking is a bonanza, in the long term even he stands to suffer as the undertaking through this spate of price-cutting will consolidate its dominance in the market at the expense of equally if not more efficient firms which lack its financial strength, with the resultant loss in consumer choice and a return to monopoly prices. (2009), p. 171. 74 Cited by Mateus, Abel M. and Moreira, Teresa (2010), p. 165. 75 Case C-62/86, AKZO Chemie BV v. Commission [1991] ECR I-3359 76 Case C-333/94P, Tetra Pak International SA v. Commission [1996] ECR I-5951 77 Case C-202/07P, France Télécom v. Commission [2009] ECR I-2369 78 (2010), p. 31. 24 Most importantly, price cutting could be the result of efficiencies reached by undertakings through lawful production and business strategies. Actually, the aim of the Commission when setting forth its enforcement priorities with regard to exclusionary conducts is to promote, rather than affect competition on the merits. This is indeed clearly stated by the Commission in the Guidance when defining its purpose.79 If a too harsh policy against predation is enforced we face the risk of discouraging firms from carrying on healthy price competition, thus, an objective reference and legal certainty is required as to what would constitute an unlawful price cut in order to avoid such mistakes. As explained by Buttigieg: This is a matter which is relevant to the consumer as a too rigid approach by the antitrust authorities whereby dominant undertakings are not allowed to meet competition through moderate price-cutting or an equivocal approach which breeds uncertainty would dissuade such firms from legitimately competing on price so that they would refrain from any price cuts, thus further exacerbating the high-price environment in the concentrated market and suffocating the little competition still prevailing. After all, it is the essence of competition that firms should compete for customers by reducing price.80 Therefore, the cost benchmarks provide the actors within a market (undertakings, competitors, Competition Authorities, etc.) with a degree of legal certainty as to the parameters used to ascertain an unlawful price cut, helping one distinguishing et ee situatio s i hi h the do i a t o pa s 81 pricing would force the exit of as-efficient rivals, and situations in which that pricing would exclude only less efficient rivals.82 In connection to the last assertion, we must clarify that in the Guidance, even though the Commission makes clear that it will focus its prosecuting efforts mainly against a conduct which has already been or Paragraph 6 of the Guidance (2009) reads as follows: The e phasis of the Co issio s e fo e e t a ti it i relation to exclusionary conduct is safeguarding the competitive process in the internal market and ensuring that undertakings which hold a dominant position do not exclude their competitors by other means than competing on the merits of the products or services they provide. In doing so the Commission is mindful that what really matters is protecting an effective competitive process and not simply protecting competitors. This may well mean that competitors who deliver less to consumers in terms of price, choice, quality and innovation will leave the market. 80 (2009), p. 172. 81 Explained by Niels et al (2011), the as-efficient competitor principle seeks to draw a line between conduct that simply reflects competition on the merits and exclusionary conduct that is harmful to consumers. Conduct that involves the dominant company cutting its prices down to, but not below, its own costs is beneficial to consumers and will only tend to exclude rivals that are less efficient -i.e., rivals with higher costs. P. 186. We shall however consider later in this chapter that according to settled EU case law, pricing above AVC and below ATC can be considered predation if it forms part of a plan of a dominant undertaking to pre-empt the market. In the same vein, the Commission indicates –in reference to price-based exclusionary conduct- in the Guidance (para. 23) that it will normally only intervene where the conduct concerned has already been or is capable of hampering competition from competitors which are considered to be as efficient as the dominant undertaking. 82 Niels et al (2011), p. 189. 79 25 is capable of hampering competition from as efficient competitors; the Commission could also be interested in prosecuting cases were less efficient competitors may have or would be affected. 83 As we will explain in the following section, according to settled case law of the EU Judiciary, the cost benchmarks may be regarded as rigid or definite determinants for predation. However, even though the cost benchmarks provide with a degree of legal certainty, the current tendency in regards to exclusionary conducts is to shift to an effects-based approach rather than keeping a form-based approach.84 Furthermore, economic research has pointed out examples of cases when pricing below certain benchmarks (such as average variable cost) may be economically justified.85 This means that in an actual investigation and eventual judicial proceedings of predation, objective justifications would apply to what a cost benchmark –in principle- shows to be a predatory conduct. 86 For now, we will illustrate the idea that prices below the established cost benchmarks in some case may be justified, through the following two specific examples provided by Kavanagh et al, after they criticize the rigid application of cost benchmarks: Yet the risk is that pricing below average variable cost by dominant firms is outlawed per se, without any consideration of whether such pricing: a) actually has negative effects on competition, which is not always the case –for example, pricing below average variable cost du i g o e o th o l … is a gua l not sufficient to infer an anti-competitive effect; b) may be justified on efficiency grounds -for example, if there are strong network effects, pricing below average variable cost may be required to gain critical mass, even in the absence of any competitors.87 83 (2009), paras 23 and 24. Paragraph 24 expresses that: However, the Commission recognizes that in certain circumstances a less efficient competitor may also exert a constraint which should be taken into account when considering whether particular price-based conduct leads to anti-competitive foreclosure. 84 The effects-based approach in regards to exclusionary abuses is said to be present in the Guidance of the Commission. We will further analyze this point in Chapter II of this study. 85 Buttigieg mentions the following example and then discusses one of the negative effects that may stem out of a mechanical application of a cost benchmark, such as AVC: In a study carried out by the Organization for Economic Cooperation and Development (OECD) in 1989 it was shown that prices below average variable costs may be economically justified: for instance upon the launch of a new product or in the case of perishable or technically obsolescent or old-fashioned goods. Again, prices in markets suffering from excess capacity may fall below average variable cost without being predatory or a firm that is closing down or having cash flow problems during a slackening of business would offer low prices possibly below cost. So prices below average variable cost might not al a s e i du ed a o gful i te t to eli i ate… o e s o petito s as the Cou t is i a ia l assu i g a d when it does not even have that effe t o e o side atio should e gi e to the defe da t s a gu e ts; a mechanical application can only lead to the antithesis of the objectives of competition policy, one of which is price competition. (2009), p. 176. 86 An example of a defence opposed in a predation case -analyzed by the ECJ in the France Télécom ruling- is the meeting competition defence, used by the defendant during the appellate process. According to the defendant Wanadoo Interactive had the right to align its prices on those charged by its competitor. The ECJ ruled, inter alia, that in regards to dominant undertakings this is not an absolute right and that it could not be justified when its purpose is to strengthen that dominant position and abuse it. Case C-202/07P, France Télécom v. Commission [2009] ECR I-2369, cited by Jones and Sufrin (2011), p. 399. For a more in depth discussion on the Meeting Competition Defence, consult Jones and Sufrin (2011), p. 399. 87 Kavanagh, Marshall and Niels (Kavanagh et al) (2009), p. 5. 26 4.2. The cost measures Before we explain how the EU Courts and the Commission apply the cost benchmarks within the assessment for predation, we will provide a definition of the main cost measures. Despite the ECJ briefly defines some of the benchmarks, e.g., average variable cost (AVC) and average total cost (ATC),88 we will adhere almost totally to the definitions given by Charles P. Shimer.89 This is due to practical reasons as his article presents clear and simple explanations with examples; adequate for a legal-focused study. As for the benchmarks used by the Commission (average avoidable costs (AAC) and long-run average incremental cost (LRAIC)) we will adhere to the definitions provided by the Guidance.         Total cost: The total cost of production. 90 A firm's economic costs can be divided into two categories: fixed and variable. Fixed costs are costs facing a firm that do not vary with output and cannot change during the period of time under consideration. These costs would continue even if the firm produced no output at all. Examples of fixed costs include the firm's cost of plant and equipment, management services, interest payment on bonded debt, property taxes, and depreciation. Variable costs are those costs that are linked directly to the level of output and can be changed during the time under consideration. Examples of variable costs include the cost of materials used in production, direct labor, most indirect labor, fuel, and any other per unit fees such as licensure and royalties. Average total cost (ATC) is the total cost of producing the product divided by the number of units produced. Total cost is the sum of all fixed and variable costs that are required to operate the entire production process. Average variable cost (AVC) is the sum of all variable costs incurred in producing the product divided by the number of units produced. Marginal cost is the increase in total cost, or the incremental cost, resulting from production of the last unit of output. 91 Average avoidable cost (AAC): [I]s the average of the costs that could have been avoided if the company had not produced a discrete amount of (extra) output, in this case the amount allegedly the subject of abusive conduct. 92 Long-run average incremental cost (LRAIC): [I]s the average of all the (variable and fixed) costs that a company incurs to produce a particular product.93 88 See infra in the following section when we mention the cost benchmarks set the by ECJ in the AKZO case. , p. . Despite the fa t “hi e s a ti le is f o , the o epts i luded do ot lie apa t o o fli t with more recent contributions such as Jo es s a d “uf i s , p. . 90 Jones and Sufrin (2011), p. 389. 91 Shimer (1981), p. 471. 92 Guidance, para. 26. Regarding the AAC the Guidance continues stating that: In most cases, AAC and the average variable cost (AVC) will be the same, as it is often only variable costs that can be avoided. 93 Guidance, para. 26. See footnote number 2. Regarding the LRAIC the Guidance continues stating that: LRAIC and average total cost (ATC) are good proxies for each other, and are the same in the case of single product undertakings. If multi-product undertakings have economies of scope, LRAIC would be below ATC for each individual product, as true common costs are not taken into account in LRAIC. In case of multiple products, any costs that could have been avoided by not producing a particular product or range are not considered to be 89 27 4.3. The cost assessment of the European Courts To understand the cost assessment applied by the European Judiciary we must provide an explanation of the Areeda/Turner test, since the Court s assess e t is si ila to the Areeda/Turner test when ascertaining a conduct of price predation, with one clear exception which we will also explain in this section. The test proposed by Phillip Areeda and Donald F. Turner was published in 1975 in the Harvard Law Review. 94 The findings and conclusions of these authors significantly impacted the treatment of predation by the US Courts, which adopted their test and applied it to the fore coming cases of predatory pricing. Uslay, Malhotra, and Allvine state that the test is responsible for the shift from a populist e a of antitrust law in the US, to the actual scenario where the possibilities for plaintiffs to prevail in predation cases is virtually not viable: The Areeda-Turner rule (1975) provided relatively straightforward guidelines for evaluating predatory pricing cases and minimized the plaintiff-friendly judgments (false positives) that see ed to f ust ate a of the stakeholde s du i g the populist e a. “i e A eeda a d Tu e (1975), the courts have generally presumed that predatory pricing is rare, and average variable cost (AVC) is a reasonable measure with some exceptions mostly at the state level. During the first five years following Areeda-Turner (1975), there were no successful plaintiffs; however, the (approximately) seventeen per cent plaintiff success ratio in the 1980s was considered to be a reasonable equilibrium… This equilibrium was destroyed in 1993, when the interpretation of the rule for the Brooke Group v. Brown & Williamson Tobacco case substantially raised the bar for the plaintiffs to su i e i ou ts… In this decision, the Supreme Court required not only proof of below cost pricing, but also of recoupment of losses incurred during predation. 95 According to the Areeda/Turner test a price above reasonably anticipated AVC96 should be conclusively presumed lawful and a price below reasonably anticipated AVC should be conclusively presumed unlawful.97 When proposing this cost benchmark Areeda and Turner originally considered using marginal-cost, however, they finally proposed the use of a per se standard of average variable cost (AVC) as a substitute for marginal cost.98 common costs. In situations where common costs are significant, they may have to be taken into account when assessing the ability to foreclose equally efficient competitors. 94 Areeda and Turner (1975). 95 (2008), p. 69. 96 AVC stands for average variable cost, explained by Niels et al in the following text: AVC is the sum of all the costs that vary with the quantity of a particular good, divided by the total quantity. Typical costs that vary with changes in output are materials, fuel and labour. The AVC is usually applied in a short-term context, taking production capacity as given and excluding all fixed costs. But it can also have a time dimension –as time passes, a higher proportio of a o pa s fi ed osts te ds to become variable, making AVC an increasing function of time. As such, using AVC requires (sometimes difficult) determinations of whether a particular cost is fixed or variable. (2011), p. 191. 97 Jones and Sufrin (2011), p. 394. 28 The ECJ s ost assess e t fo predatory pricing has not changed since the AKZO case, where the Commission first considered predatory pricing.99 The case concerned a dispute between AKZO and its competitor Engineering and Chemical Supplies (ECS), related to the organic peroxides market in the European Economic Community (now the EU).100 The Commission found that AKZO incurred in a a paig of lo p i es fo flou additi es that e efited spe ifi EC“ s usto e s, hile ai tai i g su sta tiall highe p i es to o pa a le u e s ho e e al ead AK)O s egula usto e s. 101 AK)O s ai as to se u e EC“ s ithd a al f o the market for organic peroxides for the plastics application. Such aim was proven, inter alia, through documented threats made by AKZO to ECS during meetings held between officers of both firms.102 While the case involved selective price cutting due to the fact that low prices were targeted to specific companies and some of EC“ s usto e s, it also i ol ed p edato p i i g he those p i es fell ithi the cost benchmarks and conditions set forth by the Court. 103 The Cou t s ost assess e t i AKZO, which has also been applied in the same manner by the General Court,104 was repeated and confirmed both in the Tetra Pak II and France Télécom cases. The Court considers that prices below AVC are presumed to be predatory, and prices above AVC but below ATC105 are not presumed predatory but are predatory if they are proved to be part of a plan to eliminate a competitor.106 In AKZO the ECJ ruled: Prices below average variable costs (that is to say, those which vary depending on the quantities produced) by means of which a dominant undertaking seeks to eliminate a competitor must be regarded as abusive. A dominant undertaking has no interest in applying such prices except that of eliminating competitors so as to enable it subsequently to raise its prices by taking advantage of its monopolistic position, since each sale generates a loss, namely 98 Uslay et al (2008), p. 69. In the words or Areeda and Turner: We have concluded above that marginal-cost pricing by a monopolist should be tolerated even though losses could be minimized or profits increased at a lower output and higher price, for the reasons, among others, that marginal-cost pricing leads to a proper resource allocation and is consistent with competition on the merits. Neither reason obtains when the monopolist prices below marginal cost. Areeda and Turner (1975), p. 712. For a more in-depth analysis of the Areeda/Turner test and the cost benchmarks, the reader can consult the article written by Uslay et al (2008), which describes the impact of their study on the prosecution of predatory pricing cases in the US. 99 According to Jones and Sufrin (2011), p. 395. 100 Case C-62/86, AKZO Chemie BV v. Commission [1991] ECR I, pa a. : The Co issio s de isio o tested before de EU Judiciary) found that the incumbent had infringed Article 86 EEC (now Article 102 TFEU) by pursuing against ECS a course of conduct intended to damage its business and/or to secure its withdrawal from the EEC organic peroxide market. 101 Case C-62/86, AKZO Chemie BV v. Commission [1991] ECR I-3359, para. 9. 102 Case C-62/86, AKZO Chemie BV v. Commission [1991] ECR I-3359, para. 9 and 76 et seq. 103 Case C-62/86, AKZO Chemie BV v. Commission [1991] ECR I-3359, paras. 103 and 109, also paras. 110 et seq. 104 In the words of Jones and Sufrin: The AK)O test has ee su je ted to u h iti is … It as eaffi ed i Tetra Pak II and followed and applied by the General Court in France Télécom, a judgment confirmed by the ECJ. (2011), p. 397. 105 ATC stands for average total costs, which include both variable and fixed costs. 106 Jones and Sufrin (2011), p. 397. 29 the total amount of the fixed costs (that is to say, those which remain constant regardless of the quantities produced) and, at least, part of the variable costs relating to the unit produced. Moreover, prices below average total costs, that is to say, fixed costs plus variable costs, but above average variable cost, must be regarded as abusive if they are determined as part of a plan for eliminating a competitor. Such prices can drive from the market undertakings which are perhaps as efficient as the dominant undertakings but which, because of their smaller financial resources, are incapable of withstanding the competition waged against them (emphasis added).107 As the reader can easily i fe , the ai diffe e e et ee the A eeda/Tu e test a d the EU Cou ts test is that while the former does not condemn pricing above AVC, the latter considers that pricing above AVC but below ATC can be abusive if they are part of a plan to eliminate competitors. According to Buttigieg, the AKZO test is therefore a combined cost- and strategy- based one.108 He explains the difference between the tests saying that within the meaning of the Areeda/Turner test hi h u like the AKZO test involves a purely cost/price analysis without any consideration given to intent or strategy: a price will be deemed predatory only if it is below average variable cost; if it is above average variable cost even if below average total cost it will be presumed not to be predato . 109 According to Eirik Osterud, by requiring proof of the intention to pre-empt the market in the cases of prices below ATC but above AVC, predatory pricing would be treated by the Court as a conduct that requires proving exclusionary intention. Whereas prices below AVC are under the presumption of predation only because of this same reason (a cost based examination), those below ATC but above AVC are illegal only if proved to be part of a plan to eliminate a competitor (examination of the conduct, as an additional element): Tests that require exclusionary intention are based on the idea that when dominant firms engage in certain types of conduct with an intention to exclude competitors from the market, the effect is liable to be restrictive of competition. … The presumption of competitive harm is based on a combination of three factors: the existence of a dominant position; the engagement in certain forms of conduct; and the intent to exclude competitors from the market. 110 Finally with regard to the additional intent-based assessment required to those prices below ATC (but above AVC) it is interesting to note, as Osterud points out, that: By adding that, in order to establish a 107 Case C-62/86, AKZO Chemie BV v. Commission [1991] ECR I-3359, para. 71 and 72. (2009), p. 173. 109 (2009), p. 173. 110 (2010), p. 156. In a very interesting analysis of this type of conducts (those that require an exclusionary intention proof, category in which he also includes the so called vexatious litigation), the author continues des i i g the ad a tages a d d a a ks of the Cou t s pa ti ula ie . I hat see s to e a d a a k he claims: Instead of assessing the effects of the conduct directly, intent-based tests require an investigation into the do i a t fi s easo s fo e gagi g i the o du t i uestio . The test is a si plified su stitute fo a o e comprehensive market analysis. 108 30 violation of Article 102 TFEU, such prices must be part of a plan for eliminating a competitor, the Court raised the threshold for finding an abuse compared to the test for prices below AVC. 111 The last statement is important because it showcases particular characteristics of the Cou t s app oa h towards predation. First of all, by separating itself on this point from the Areeda/Turner test, the ECJ may have preferred a more conservative treatment, in order to prevent an under-inclusive application of Article 102.112 Secondly, probably considering that by doing this it could produce a deterring effect of healthy price cutting practices, it added the intention requirement for prices between AVC and ATC. I su , the Cou t s u e t assess e t o tai s a fo ased ite ia i ases he e p i es a e elo AVC, and a form plus intention criteria applied for those cases when prices are between AVC and ATC. In both of these cases the effects on the market are not considered as a leading factor. As stated by Osterud, the test does not require a more comprehensive a al sis of the p a ti e s a tual or likely effects in the prevailing market conditions. The test 113 is based on the inference that if the object pursued is to restrict competition, the conduct will also be lia le to ha e su h a effe t. 114 Precisely, this last statement is evident in the France Télécom ruling, where the GC indicated: As regards the conditions for the application of Article 82 EC and the distinction between the object and effect of the abuse, it should be pointed out that, for the purposes of applying that article, showing an anti-competitive object and an anti-competitive effect may, in some cases, be one and the same thing. If it is shown that the object pursued by the conduct of an undertaking in a dominant position is to restrict competition, that conduct will also be liable to have such an effect.115 4.4. The Co issio ’s ost assess e t in the Guidance The Co issio s cost assessment for predatory pricing is described in the Guidance. We will limit this section to a more descriptive rather than analytical discussion of the cost benchmarks and the cost assessment applied by the Commission, since we will dedicate Chapter II to a more in-depth analysis of the Guida e a d the p o ess a d pu pose ehi d it a d the Co issio s app oa h to a ds predation thereafter. The cost benchmarks set by the Commission are described in paragraph 26 of the Guidance, which indicates that: The cost benchmarks that the Commission is likely to use are average avoidable cost (AAC) and long-run average incremental cost (LRAIC). Failure to cover AAC indicates that the dominant undertaking is sacrificing profits in the short term and that an equally efficient competitor 111 (2010), p. 133. A o di g to Oste ud s fi di gs: The Court was concerned that dominant forms could successfully engage in predatory pricing strategies even if their price were above AVC. By accepting that prices above AVC could be abusive, the ECJ reduced the risk of an under-inclusive application of Article 102 TFEU, which could allow anticompetitive practices to go unpunished. (2010), p. 134. 113 In reference to pricing between AVC and ATC. 114 (2010), p. 132. 115 Case T-340/03, France Télécom v. Commission [2007], Paragraph 195. 112 31 cannot serve the targeted customers without incurring a loss. LRAIC is usually above AAC because, in contrast to AAC (which only includes fixed costs if incurred during the period under examination), LRAIC includes product specific fixed costs made before the period in which allegedly abusive conduct took place. Failure to cover LRAIC indicates that the dominant undertaking is not recovering all the (attributable) fixed costs of producing the good or service in question and that an equally efficient competitor could be foreclosed from the market. The Commission then explains, in paragraphs 63 et seq, the specific assessment for predation and how it will apply the cost benchmarks to conducts suspected to be predatory. For the Commission in the Guidance, the equivalent to the predation stage where the incumbent incurs in losses by pricing below cost116 is denominated as sa ifi e , a concept described in the Guidance as follows:  The action by which a dominant undertaking deliberately incurs in losses or foregoes profits in the short term.117 The definition of sacrifice is accompanied with the intention or purpose sought as its effect: so as to foreclose or be likely to foreclose one or more of its actual or potential competitors with a view to strengthening or maintaining its market power, thereby causing consumer harm. 118 In reference to the application of the cost benchmarks, the Guidance states that: Pricing below AAC will …i ost ases e ie ed the Co issio as a lea i di atio of sa ifi e. 119 Additionally, the Guida e poi ts out that u de the Co issio s ie only pricing below LRAIC is capable of foreclosing 120 as efficient competitors form the market. Thus, the Commission sets this last parameter (pricing below LRAIC) in combination with available sufficient reliable data, to trigger the application of the equally efficient competitor analysis to determine whether a conduct is capable of harming consumers.121 116 Following our previous definition of predatory pricing of a strategy completed in stages. Paragraph 63 of the Guidance indicates: ...the Commission will generally intervene where there is evidence showing that a dominant undertaking engages in predatory conduct by deliberately incurring losses or foregoing p ofits i the sho t te efe ed to he eafte as sa ifi e . 118 Guidance (2009), para. 63 119 Guidance (2009), para. 64. 120 Guidance (2009), para. 67. 121 Osterud explains this last reference in the Guidance to LRAIC and compares the benchmark with ATC: The Commission will apply an equally efficient competitor analysis to determine whether the conduct is capable of harming consumers. The guidance document states that, normally, only pricing below LRAIC is capable of foreclosing as efficient competitors from the market. To compare this to an ATC benchmark, in the case of an undertaking producing multiple products with common costs, LRAIC would be below ATC, because common costs are not included in LRAIC. (2010), p. 136. 117 32 5. The treatment of recoupment of losses by the European Courts In the Introduction we mentioned that under EU Law recoupment of losses is not part of the assessment for predatory pricing. This fact stems from the case law of the European Judiciary which in two cases, Tetra Pak II and France Télécom, indicated that EU Law does not require proof of the possibility of recoupment.122 5.1.Tetra Pak II In Tetra Pak II the Commission held that Tetra Pak was a dominant undertaking within the markets for aseptic machines and cartons intended for the packaging of liquid foods in the former European Economic Community (now EU), and that it had abused its dominant position from at least 1976 until 1991, both on those markets and on the markets in non-aseptic machines and cartons. 123 Among the abusive conducts, the Commission found that Tetra Pak was charging predatory prices for non-aseptic cartons in Italy (specifically Tetra Rex cartons) at a level designed to oust its competitors by resorting to cross-financing of its products, using its dominant position on the aseptic market. 124 The GC upheld the Co issio s fi di gs a o di g to hi h Tetra Pak had priced below AVC from 1976 125 to 1981. Furthermore, it deemed that there was enough evidence of a plan to eliminate Elopak (its competitor) from the Italian market, because: a) Tetra Pak, which did not manufacture Tetra Rex cartons in Italy from 1976 to 1980, imported them in order to resell them in that country at prices lower by 10 to 34% than their purchase prices; b) Tetra Pak resold in Italy at prices lower by 17 to 29% than their purchase price Rex cartons imported from Sweden; c) The prices of Tetra Rex cartons sold in Italy were lower by 20% at least and often by 50% than the prices applied in other Member States; and d) The Commission filed epo ts of Tet a Pak Italia a s Boa d of Di e to s Tet a Pak s su sidia i Ital referring to the need to make major sacrifices in the area of prices and supply terms in order to fight o petitio …126 In the proceedings before the GC, Tetra Pak alleged that sales at a loss are eliminatory only where the undertaking in question has a reasonable prospect of subsequently recouping losses so incurred. The firm supported its argument on the judgment of the US Supreme Court in the aforementioned Brooke Group v Brown & Williamson Tobacco case. 127 The GC rejected this argument holding that the Co issio s fi di gs suffi ed p i i g elo AVC a d that it as ot e essa to de o st ate specifically that the unde taki g… had a easo a le p ospe t of e oupi g losses.128 122 Despite the fact the ECJ recognized a degree of importance to recoupment as a factor to consider, as we indicated supra, proof of recoupment as an independent element is not required for ascertaining predatory pricing under EU Law. 123 Case T-83/91 – Tetra Pak v. Commission, [1994] ECR II-755, para. 1. 124 Case T-83/91 – Tetra Pak v. Commission, [1994] ECR II-755, paras. 16, 20 (3), and 146. 125 Case T-83/91 – Tetra Pak v. Commission, [1994] ECR II-755, para. 150. 126 Case T-83/91 – Tetra Pak v. Commission, [1994] ECR II-755, paras. 150 and 151. 127 Case T-83/91 – Tetra Pak v. Commission, [1994] ECR II-755, para. 143. 128 Case T-83/91 – Tetra Pak v. Commission, [1994] ECR II-755, para. 150. 33 On the appeal against the GC s ruling the ECJ completely upheld the contested judgment and in regards to the recoupment element, stated that: Furthermore, it would not be appropriate, in the circumstances of the present case, to require in addition proof that Tetra Pak had a realistic chance of recouping its losses. It must be possible to penalize predatory pricing whenever there is a risk that competitors will be eliminated.129 As Jones and Sufrin explained, The question of whether a possibility or likelihood of recoupment is part of the test for predatory pricing was raised in the leading case of Tetra Pak II and answered in the negative –at least in the circumstances of that case, where the principles laid down in AKZO were confirmed and developed. 130 From the text of the ruling we stress the fact that the ECJ, when connecting the facts to those in AKZO, restated a rigid consideration towards prices below AVC when the judges indicated that prices below average variable costs must always be considered abusive. 131 The Cou t o ti ued sa i g that In such a ase, the e is o o ei a le e o o i pu pose othe tha the eli i atio of a o petito … 132 When reading the judgment it is easy to conclude that the ECJ gave more importance to the intent present in Tet a Pak s conduct, rather than to the assessment the Commission would have had to carry on to prove the possibility of recoupment. Or one could say that the ECJ rendered more importance to intent, over the effects of predation. As Jones and Sufrin point out, this –as the same wording of the ruling suggests- may have been due to the special facts of that case: Those circumstances included the fact that Tetra Pak had a quasi-monopoly and that the alleged predation was on a market distinct from the dominated one (so that Tetra Pak could crosssubsidize). Further, there was clear evidence from the data uncovered by the Commission that Tetra Pak was pursuing a deliberate strategy of eliminating competitors. 133 However, we will see that when analyzing the ruling of the ECJ in the France Télécom case it is evident that the special circumstances of the Tetra Pak II case made no difference as to the recoupment of losses situation, because in the France Télécom case the ECJ ruled out the proof of recoupment in a generic and clear way. Lastly, we observe that even though in the Tetra Pak II ruling the ECJ did not state exactly the same reasoning when denying the recoupment requirement, when AG Ruiz-Jarabo rendered his opinion to the Court, he also gave more importance to the intent of the incumbent when incurring in predation. He did so, rather than rendering relevance to the effects the conduct may have had: recouping losses is the result sought by the dominant undertaking, but predatory pricing is in itself anticompetitive, regardless of whether it achieves that aim. 134 129 Case C-333/94P, Tetra Pak International SA v. Commission [1996] ECR I-5951, para. 44 Jones and Sufrin (2011), p. 401. 131 Case C-333/94P, Tetra Pak International SA v. Commission [1996] ECR I-5951, para. 41. 132 Case C-333/94P, Tetra Pak International SA v. Commission [1996] ECR I-5951, para. 41. 133 (2011), p. 402. 134 Opinion of AG Ruiz-Jarabo in Case C-333/94 P, Tetra Pak v. Commission, para. 78 in fine. 130 34 Buttigieg criticizes such a reasoning by comparing it with the definition of abuse of dominance o st u ted the Cou t, hi h o t a to the AG s affirmation refers to the effect of the abuse and not to the intent of the undertaking that performs it, specifically that effect (mentioned in the Hoffmann-La Roche ruling) is hindering the maintenance of the degree of competition still existing in the market or the growth of that competition: 135 But in the definition of abuse given by the Court itself in Hoffmann-La Roche the emphasis in on the objective nature of the concept and the yardstick is effect not intent. 136 5.2. France Télécom The predation conduct in this case occurred in the period between March 1, 2001 and October 15, 2002.137 During these years Wanadoo Interactive SA (WIN), a subsidiary of the France Télécom Group abused its dominant position in the French market for high-speed internet access for residential customers.138 The products for which it was found to charge predatory prices were internet access services based on ADSL (asymmetric digital subscriber line) technology, denominated Wa adoo AD“L a d eXte se . 139 The Commission found that WIN charged predatory prices for said services that did not enable it to cover its variable costs until August 2001 or to cover its full costs from August 2001 onwards, as part of a plan to pre-empt the market in high-speed internet access during a key phase in its development.140 Whe the Co issio s de isio as o tested efo e the GC, it upheld it in its entirety. 141 The GC repeated the assessment of the ECJ in AKZO and Tetra Pak II as to the presumption of predation when prices were below AVC. The same assessment of those cases was applied for prices between AVC and ATC, indeed the GC found that the Commission furnished solid and consistent evidence as to the existence of a plan of predation for the entire infringement period. In this line, the Commission filed a ote of WIN s st ategi a age e t i di ati g that: The high-speed and ADSL market will, for the next few years, continue to be conquest-driven, the strategic objective being to gain a dominant position in terms of market share, the period of profitability only coming later. 142 With regard to the recoupment factor, WIN alleged that it is entirely separate from the test of predation and the Commission must provide evidence of it. Also that if an undertaking in a dominant position cannot reasonably expect to reduce long-te o petitio ith a ie to e oupi g its losses … it is ot rational for that undertaking to engage in a policy of predatory pricing. WIN supported its arguments, 135 See supra in this chapter, the section denominated: Predatory pricing among the conducts forbidden by Article 102 TFEU. 136 (2009), p. 180. Case T-340/03, France Télécom v Commission, [2007] ECR II-117, para. 9. 138 Case T-340/03, France Télécom v Commission, [2007] ECR II-117, para. 2. 139 Case T-340/03, France Télécom v Commission, [2007] ECR II-117, para. 6. 140 Case T-340/03, France Télécom v Commission, [2007] ECR II-117, para. 5. 141 Case T-340/03, France Télécom v Commission, [2007] ECR II-117, para. 284. 142 Case T-340/03, France Télécom v Commission, [2007] ECR II-117, para. 215. 137 35 inter alia, by all the economic and legal literature as well as by a number of courts and competition authorities, including those of the United States. 143 The GC eje ted WIN s a gu e ts i di ati g that fo oth situatio s p i es elo AVC a d et ee AVC and ATC) it was not necessary to establish in addition (to the evidence filed by the Commission) proof that WIN had a realistic chance of recouping its losses. 144 The Cou t estated the ECJ s ase la rendering particular importance to the similarities with the facts in Tetra Pak II.145 In appeal before the ECJ, the ruling was upheld146 and in connection to recoupment, the ECJ denied its proof as a requirement to establish predation. In line with the GC s ite ia the ECJ i di ated: Moreover, the lack of any possibility of recoupment of losses is not sufficient to prevent the undertaking concerned reinforcing its dominant position, in particular, following the withdrawal from the market of one or a number of its competitors, so that the degree of competition existing on the market, already weakened precisely because of the presence of the undertaking concerned, is further reduced and customers suffer loss as a result of the limitation of the choices available to them. 147 In specific reference to the treatment of recoupment of losses by the ECJ in the France Télécom ruling, it is important to highlight three specific situations: i. AG Mazák e o e ded to the ECJ i his Opi io to uphold the appella t s a gu e ts elated to recoupment, however the ECJ confirmed the GC s judgment and denied that possibility of recoupment must be proven: In AG Mazák s o o ds: In such a case, where there is no possibility of recouping losses, consumers and their interests should, in principle, not be harmed.148 ii. In line with the last point, by rejecting the proof of recoupment, the ECJ ruled out any doubt that could have risen from the Tetra Pak II ruling, as it clearly stated that under EU Law, proof of the possibility of recoupment is not required; instead, the ECJ again preferred its reasoning according to which prices below AVC suffice to determine a predation conduct and that prices above AVC but below ATC are considered predatory if an intention to eliminate can be shown.149 143 Case T-340/03, France Télécom v Commission, [2007] ECR II-117, paras. 220 and 221. Case T-340/03, France Télécom v Commission, [2007] ECR II-117, paras. 227 and 228. 145 Case T-340/03, France Télécom v Commission, [2007] ECR II-117, para. 224 to 230. 146 Case C-202/07P, France Télécom v. Commission [2009] ECR I-2369, p. 120. 147 Case C-202/07P, France Télécom v. Commission [2009] ECR I-2369, para. 112. 148 Opinion of AG Mazak in Case C- 202/07 P, France Télécom v Commission, para. 74. 149 The Court indicated: Accordingly, contrary to what the appellant claims, it does not follow from the case-law of the Court that proof of the possibility of recoupment of losses suffered by the application, by an undertaking in a dominant position, of prices lower than a certain level of costs constitutes a necessary precondition to establishing that such a pricing policy is abusive. In particular, the Court has taken the opportunity to dispense with such proof in circumstances where the eliminatory intent of the undertaking at issue could be presumed in view of that u de taki g s appli atio of p i es lo e that a e age a ia le osts… Case C-202/07P, France Télécom v. Commission [2009] ECR I-2369, para. 110. 144 36 iii. As we pointed out before when we discussed recoupment as a factor for the feasibility of predation, the Court gave certain degree of importance to recoupment within the assessment of predation, specifically when excluding economic justifications for below AVC pricing or assist in establishing an eliminatory plan.150 According to the exact wording of the ruling, recoupment could be a relevant factor in assessing whether o ot the p a ti e o e ed is a usi e… however the Court continued saying that the lack of any possibility or recoupment of losses is not sufficient to prevent the undertaking concerned reinforcing its dominant position. Whe the Cou t e tio ed this last easo it e e plified sa i g that this reinforcement of dominance could be followed by the withdrawal of competitors from the market, weakening the existing degree of competition, and therefore the customers would suffer loss as a result of the limitation of the choices available to them. 151 Following the ruling, the Court deems that even if recoupment of losses is not possible, -in cases of predation- damage to consumers is still produced. This damage (according to the Court and which it denominates as loss ) is materialized in the market preemption itself; i.e., in the elimination of competitors from the market, regardless of the fact that the incumbent may or may not recover its losses. It is therefore apparent f o the Cou t s ie that the outcome of the market pre-emption is a limitation of choices suffered by the consumer. In the following chapters we will compare and balance this reasoning of the Court with other criteria, which –contrarily- defends proof of recoupment of losses as part of the predation test. We will do this with the aim of finding out which of the two approaches in reality promotes the consumer welfare objective, should this objective be followed. 6. Recoupment of losses in the Guidance We o lude this se tio i di ati g that i alig e t ith the Cou ts e tio ed uli gs, the Commission in the Guidance apparently does not place recoupment of losses as part of its assessment for predatory pricing. By the contrary, it expressly discards this possibility by saying that: Identifying consumer harm is not a mechanical calculation of profits and losses, and proof of overall profits is not required.152 Instead, the Guidance prefers other means to prove consumer harm in the investigated conduct and it states the following: Likely consumer harm may be demonstrated by assessing the likely foreclosure effect of the conduct, combined with consideration of other factors, such as entry barriers. In this context, the Commission will also consider possibilities of re-entry. 153 However, in the Guidance the Commission does not discard the importance of recoupment as a factor to consider, but positions it in what seems to be more an integrating part of the overall intentions of the conduct, not as an independent relevant factor. Therefore, recoupment would not be treated as an objective element rising out of the assessment for predation.154 Furthermore, the Guidance does not 150 Jones and Sufrin (2011), p. 403. Case C-202/07P, France Télécom v. Commission [2009] ECR I-2369, para. 111. 152 Guidance, para 71. 153 Guidance, para. 71. 154 The Guidance (2009), para. 70, reads as follows: Generally speaking, consumers are likely to be harmed if the dominant undertaking can reasonably expect its market power after the predatory conduct comes to an end to be 151 37 expressly mention recoupment of losses, changing the terminology to a more generic reasonable e pe ta of g eate a ket po e o e efit f o sa ifi e . Explained by Jones and Sufrin: The Guida e Pape does ot a tuall e tio the o d e oup e t . The Co issio , indicating when it will intervene, does nevertheless say that consumers are most likely to be harmed if the dominant undertaking can reasonably expect an increase in market power after its conduct has ended. An increase in market power, however, does not necessarily denote an ability to recoup, which will depend on the specific circumstances. The Commission will intervene not only if prices can be increased again, but also if the conduct could prevent or delay a lowering of prices.155 We ill t to u a el the Co the following chapter. issio s ie of the recoupment factor when we analyze the Guidance in 7. Final Comments We have learnt that predatory pricing is an unlawful conduct. This is a reality notwithstanding the complexity of constructing its definition, the various factors to consider towards its feasibility, and the no less important positions that question its very existence or possibility as a business strategy. As a result predatory pricing is penalized by EU Competition Law. When ascertaining predation there are several factors involved in its feasibility as a successful business strategy. According to the economic theory such factors may determine that predation is not possible under certain circumstances. Also, these factors may determine that -in some cases- when undertakings carry on below cost pricing, competition and consumer welfare are not negatively affected. Thus, identifying predation and distinguishing it from perfectly competitive conduct is of major importance when applying EU Competition Law. We also analyzed and described the cost benchmarks applied by the Commission and the European Courts when conducting the assessment for predation. As we pointed out, in the current state of affairs these benchmarks are decisive and the most important elements to consider in deciding whether or not an illegal conduct has taken place. On the other hand, the ECJ has established in settled case law that proof of recoupment of losses is not part of the assessment for predation. In regards to the possibility of recoupment of losses, both the ECJ and the Commission in its Guidance recognize its importance within the assessment of predatory pricing. However, as a decisive difference between their reasoning and AG Mazák s Opi io , su h i po ta e is ot as high as e og izing it as an independent element (that must be determined) within the assessment for predation. Now that we have gone through what we ill all a a o ie of the t eat e t of p edatio u de EU Competition Law, in the following chapters we can focus our attention in the new effects based approach toward exclusionary conducts of abuse of a dominant position, and the role that proof of greater than it would have been had the undertaking not engaged in that conduct in the first place, that is to say, if the undertaking is likely to be in a position to benefit from the sacrifice. 155 (2011), p. 404. 38 recoupment of losses may have within this new approach. By doing so, we can verify whether or not consumer welfare is at the core of such approach, at least as it has been implemented. The same applies when analyzing if by raising the standard for predation, consumer welfare may be enhanced. In the next chapter we will start covering these subjects through the analysis of the Co issio s Guidance and a more in-depth study of the current approach of the Commission towards recoupment of losses in cases of predation. We will carry out this assignment, trying to understand the effects based approach that in principle underpins the Guidance, and therefore, also the actions of the Commission when tackling predation. 39 Chapter II: The Guidance of the Commission and the recoupment factor In this chapter we will analyze the place and role of recoupment within the Com issio s u e t approach towards predatory pricing. When conducting this analysis we must consider that one of the aims of the study is ascertaining if consumer welfare could be enhanced in the potential case of the inclusion of recoupment in the predation test. If eventually the test for predatory pricing will change, there are certain elements that have paved the way for such an outcome. This is the case of the new trend towards an effects-based approach in the interpretation and application of Article 102 TFEU and the rationale behind it. Considering what we just mentioned, the content of this chapter will focus on two specific subjects: i) the effects-based approach permeating the recent reform applied to the enforcement priorities to exclusionary abuses, and ii) the treatment of recoupment by the Commission, in the context of the results brought about so far by the reform: The Guidance. As we will see in the text following this introduction, it is commonly accepted that an effects-based approach is present in the reform conducted on the application of Article 102 TFEU. 156 By analyzing this concept we will try to discover, if that is the case, if the protection of consumer welfare is an aim of such approach. If indeed a link between the new approach and consumer welfare exists, and additionally recoupment may play a role within the effects-based approach; the findings of this chapter will significantly have a say in our study. Then, when we focus on the recoupment factor, we will consider the application of an assessment of recoupment in the context of the Guidance. For this purpose we will do an analysis which includes the recoupment assessment practiced by the Commission in the France Télécom case, the contents of the Guidance, and different criteria about this subject from EU Competition Law authors. Our aim is to verify whether or not the Guidance leaves a room for a recoupment assessment in future cases. At the end of the chapter we will study the recent ruling of the ECJ in Post Danmark,157 a case of below cost-pricing in the form of selective price cutting. 158 The judgment, rendered as a response to a Preliminary Ruling concerning the interpretation of Article 102 TFEU, may be viewed as a clear sign of current and future embracement by the Court of an effects-based approach in the assessment of 156 Inter alia, this is the view of Kavanagh et al (2009), p. 1-17, Lianos (2009), p. 19-49, Gormsen (2010), p. 113-174, and Korah (2011), p. 8-24. 157 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012]. 158 When we explained the facts of AKZO in Chapter I, we indicated that that case also involved selective price cutting. Such circumstance is recognized by the Guidance (2009) in paragraph 72, which indicates that: It may be easier for the dominant undertaking to engage in predatory conduct if it selectively targets specific customers with low prices, as this will limit the losses incurred by the dominant undertaking. When studying the facts of this case we will see that although the existence of predatory pricing was early discarded by the NCA in Post Danmark, the case involves an exclusionary conduct related to prices, in which the assessment of an exclusionary intent is also conducted when the prices are between AVC and ATC, as established by the Court in AKZO. We will discover that indeed, in dealing with this assessment is where the Court moved towards a more effects-based approach in Post Danmark. 40 exclusionary abuses related to prices. Most importantly, it may be a clear sign of a gradual recognition a d a epta e, the ECJ, of the Co issio s appli atio of said e app oa h. 1. The process towards the Guidance The Guidance of the Commission is due to the reform on the application and interpretation of Article 102 TFEU, towards a more effects-based approach in matters related to abusive conducts.159 These kinds of reforms not only affected abusive unilateral behavior but were indeed applied through an entire p o ess o du ted si e the s. It ega est u tu i g the old app oa h to e ti al est ai ts, followed by the Guidelines on the application of article 101(3) TFEU (former 81(3) EC), and finally took place over the merger control regime.160 The beginning of the reform to the abusive conduct regime took longer than the ones we just referred to. So did happen with the process itself: It took the Commission about three years, from its first verifiable effort, to publish the Guidance.161 This fact is the consequence of certain legal difficulties and internal discrepancies within the Commission. 162 In the specific field that takes our attention (abusive behavior), 163 there are two important documents that precede the publication of the Guidance and are placed in the middle of the reform. The first one is the Report on an Economic Approach to Article 82 (the Report) prepared by the Economic Advisory Group for Competition Policy (EAGCP), published by the Commission in July, 2005. 164 The second one is the Discussion Paper on the Application of Article 82 of the Treaty to Exclusionary Abuses (the DP), prepared by DG Competition and published later on the same year. 165 159 Mosso (2011), p. 16. Mosso (2011), p. 13-17. We invite the reader to consult the author in case there is interest in studying the legal acts emerging from the previous reforms, as well as the most relevant characteristics that encompass them. The scope of this section of the study is limited to an outline of the reform leading to the publication of the Guidance referred to Article 102 TFEU only. 161 When describing the process, Mosso stated: The Commission issued a discussion paper in 2005. Yet it took three years to adopt the Enforcement Communication (namely the Guidance). (2011), p. 17. 162 As described by Valentine Korah: Although the Commission has been able radically to change its attitude to effi ie i e ti al a a ge e ts i the s, it has had o e diffi ult i ha gi g the interpretation of Article 82. It has no power to adopt group exemptions under Article 82, and officials were not agreed that access to markets was desirable only if the new entrants were likely to be as efficient as the incumbent(s). Korah (2011), p. 16. 160 163 When referring to the conducts prohibited by Article 102 TFEU, we will indistinctly name them as abusive behavior, abusive conduct, unilateral abusive conduct, abuse of dominance, or abuse of a dominant position. The reader ought also to remember that this is a generic term that encompasses both exclusionary and exploitative abuses. However, as we already indicated in Chapter I, the Guidance scope focuses on exclusionary abuses, and as we will see, with special importance on certain types amongst them. Moreover, the subject of this study, predatory pricing, pertains to the category of price-related exclusionary abuses. 164 As we mentioned in the Introduction of the study, in his Opinion to the ECJ, AG Mazák partially supported his arguments on the Report. He pleaded in favor of the inclusion of the recoupment factor in the test for predation applied in the France Télécom case. 165 Ezrachi (2009), p. 52. 41 The EAGCP is a discussion forum composed by academics with a recognized reputation in the field of industrial organization in Europe. Its main purpose is to support DG Competition in improving the economic reasoning in competition policy analysis. 166 DG Competition (DG Comp) is one of the directorate generals of the Commission in charge of designing competition policy, while the Commission is in charge of applying it (competition policy) by fact-finding, taking action against the infringement of the law, and imposing penalties.167 Our aim is not to outline or focus our attention in the contents of each of these two foregoing documents, due to the fact that this section is dedicated to the Guidance. The reason lies, as we will later explain, in the nature of the latter, which takes priority over the former two. However, for our purposes, said contributions are important for two specific reasons: i) both documents (the Papers) plea and have at their core the idea of changing the interpretation and application of Article 102 TFEU, 168 from a form-based to an effects-based approach (founded on economics). The aim of such approach is the protection of consumer welfare through the identification of the effects of abusive conducts which negatively affect consumers; 169 and, ii) the input provided by both documents influenced the preparation and publication of the Guidance as a final p odu t . To illustrate what we just mentioned, as to the first point, namely the purpose within the Papers, in first pla e the EAGCP s ‘epo t su a izes its o te t i the follo i g li es: This report argues in favour of an economics-based approach to Article 82, in a way similar to the reform of Article 81 and merger control. In particular, we support an effects-based rather than a form-based approach to competition policy. Such an approach focuses on the presence of anti-competitive effects that harm consumers, and is based on the examination of each specific case, based on sound economics and grounded on facts. 170 With noteworthy coincidences, in regards to the relevance granted to consumer welfare and the effects of the abusive conducts, in its introduction the DP indicates: With regard to exclusionary abuses the objective of Article 82 is the protection of competition on the market as a means of enhancing consumer welfare and of ensuring an efficient allocation of resources. Effective competition brings benefits to consumers, such as low prices, high quality products, a wide selection of goods and services, and innovation. Competition and market integration serve these ends since the creation and preservation of an open single market 166 http://ec.europa.eu/dgs/competition/economist/eagcp.html, consulted on November 16, 2012. Gormsen (2010), p. 15. 168 However the reader must not get confused in identifying the Papers as equal, even though their general objective centers in the reform applied to the interpretation of Article 102 TFEU. In this vein Ariel Ezrachi explains that: The publication of the Consultation Paper coincided with the publication of a Discussion Paper on the Appli atio of A ti le of the T eat to E lusio a A uses… Although this Pape ight ha e been seen as a continuation of the EAGCP Consultation Paper, it was more restrictive in nature. Unlike the EAGCP Consultation Paper, the Discussion Paper generally preserved the analysis of dominance as it existed in the Community case law. One could nevertheless identify hints of an economic approach woven into the Discussion Paper. (2009), p. 52. 169 The Report (2005), p. 2. 170 The Report (2005), p. 2. 167 42 promotes an efficient allocation of resources throughout the Community for the benefit of consumers. In applying Article 82, the Commission will adopt an approach which is based on the likely effects on the market. 171 As to the second point, although the content of the Guidance does no fully embrace the contents, recommendations, and specific points addressed by the Papers, it does follow some of their innovative contents and embraces the new approach proposed by them. 172 In the words of Ariel Ezrachi, The Guida e la s do the Co issio s ode app oa h to A ti le ases a d deli e s se e al of the 173 innovative elements which were aired in the preceding Papers. As to the reason why our attention is on the Guidance rather than on the Papers, we must clarify that due to its nature the Guidance takes precedence over the other documents and in comparison to them, it can be regarded as the one that acquired legal form as an instrument of soft law. 174 The Papers are valuable contributions to the reform we have been referring to. As we explained supra, the Report contains a synthesis constructed by recognized academics with basic recommendations, analyses, and examples justifying a shift in the application and interpretation of Article 102 TFEU. As to the DP, it is a staff paper, produced for consultation purposes with the aim of producing substantive Guidelines referred to Article 102 TFEU. 175 However, the substantive guidelines at the end were not issued, being the Guidance the result of the process, at least to this day. As we will find out in the following section, as opposed to the wider content of the DP, the scope of the Guidance is restricted by an enforcement priorities-related content. 171 DP (2005), para. 4. As for the DP, we will address some of the reasons that lead the Commission to finally set a Guidance with only enforcement priorities rather than substantive guidelines, which would have been similar to the DP. As for the Report, it was not drafted in the form of Guidelines. The Report is an academic contribution with recommendations set –in many cases- through practical examples, which illustrate the way in which an effectsbased approach could be implemented. Later in this Chapter, when dealing with the description of an effectsbased approach we will make use of some of the definitions and examples provided by the Report. 173 (2009), p. 53. 174 The Guidance pertains to the category of soft law instruments, recognized for la i g do ules of o du t hose legal status is u lea o u e tai . “oft la i st u e ts a e defi ed as: rules of conduct that are laid down in instruments which have not been attributed legally binding force as such, but nevertheless may have certain i di e t legal effe ts, a d that a e ai ed at a d a p odu e p a ti al effe ts. Senden (2004), p. 1. 175 In the view of Liza L. Gormsen, the relevance of the DP goes a little bit further, considering that it can be e de ed as o tai i g the DG Co p offi ials thoughts a out A ti le TFEU: The Discussion Paper was never published as substantive guidelines, so it could be considered a document which is no longer relevant. However, in analyzing the Prokent/Tomra decision, DG COMP officials refer to de Discussion Paper to emphasize that the de isio took e o o i a al sis i to o side atio i the spi it of the Dis ussio Pape . … Despite the publication of the Guidance Paper, which is an official notice from the Commission, the Discussion Paper is still an important document for understanding what some DG COMP officials are thinking on Article 82. (2010), p. 16. 172 43 2. Some relevant characteristics of the Guidance As opposed to the Papers, the Guidance is a Communication of the Commission published in the Official Journal of the European Union. 176 The text of the Guidance provides a useful insight for the understanding of some of the characteristics of this instrument which are relevant for the purposes of our study. In this section, we will focus on three important characteristics of the Guidance: 2.1. The Guida e’s li ited s ope The first sentence of paragraph 2 of the Guidance provides that it sets out the enforcement priorities that ill guide the Co issio s a tio i appl i g A ti le to e lusio a o du t do i a t 177 undertakings. The cited text, as well as the very same title of the Guidance, shows its limited scope, which is due to the specific subject it encompasses: enforcement priorities. Mosso explains the reason behind this reality in the following lines: First of all, unlike all the other guidelines I have been talking about, the Enforcement Communication (the Guidance) is not an interpretative notice but an enforcement priorities paper. Why? Because apparently in abuse control, the European courts have not yet fully embraced the effects-based approach as in other areas. There are some judgments of the courts that can clearly be read with this perspective but there are some others that are still too much form- ased. … Agai st this a kg ou d, the Co issio took the positio that, among all the potential cases involving practices that the European courts had condemned as exclusionary abuses pursuant to Article 82 EC/102 TFEU, it will choose to focus its scarce enforcement resources on only some of them. Which ones? The ones that result in anti-competitive effects and really harm the consumer. This is what is meant by enforcement priorities (emphasis added). 178 There are other explanations for the limited scope of the Guidance, the ones we consulted converge ith Mosso s i the se se that the Guida e e p esses o l the Co issio s p io ities. 179 However, for our purposes the transcribed words are of key importance. First of all, because at the time this study is being written, Mosso is the Acting Director for Policy and Strategy at the Competition Directorate 176 The publication in the Official Journal is dated February 24, 2009. Guidance (2009), para. 2. 178 (2011), p. 17. 179 E.g., Gormsen states that [i]t is clear that the Commission can adopt priority guidelines as long as they are within the limits of the law, supporting this assertion inter alia in the limited resources the Commission has. Then A iel Ez a hi goes a little it fu the i justif i g the sh s ope of the Guida e i the Co issio s desi e to avoid any conflict with the case law of the EU Courts: Fo o the fa t e ai s that the Co issio s e thi ki g is su ogated to the ou ts i te p etatio of A ti le . … It is ot ha e that the do u e t as ot e titled Guideli es ut athe Guida e o the Co issio s E fo e e t P io ities in Applying Article 82 EC Treaty to A usi e E lusio a Co du t Do i a t U de taki gs . The title e hoes the h id atu e of the do u e t a d the potential tension between some of the new concepts and the case law of the European courts. (2009), p. 56. 177 44 General of the European Commission.180 For that reason we render the information he provides at least as a proxy to the Co issio s ie s.181 I this ei , th ee of Mosso s asse tio s at h ou atte tio :    The European Judiciary has not fully reached an effects-based approach in regards to exclusionary abuses, or at least to an extent that would have enabled the preparation of substantive Guidelines. The fact that the Commission prioritizes its efforts only on some exclusionary conducts. The rationale behind this selection: Those exclusionary conducts are regarded as the ones that result in anti-competitive effects and are really consumer harming. First of all, as Mosso points out, there is still a lack of determination from the EU Judi ia s end, as to whether it will fully embrace an effects-based approach in unilateral behavior. This assertion is supported e.g., by Ariel Ezrachi, who inter alia presents the F a e Tele o s ruling as an example of a lack of commitment from the Courts to assume an effects-based approach. He points out; as we did in Chapter I, that [t]he Court rejected arguments concerning the need to prove the effects of eliminating competition.182 This was expressed by the GC in paragraph 195 or its judgment, when indicating that for the purposes of applying that article, showing an anti-competitive object and an anti-competitive effect may, in some cases, be one and the same thing.183 In the same case, we also indicated previously in Chapte I that oth EU Cou ts epeated the fo a al sis eated the Cou t i AKZO, as to the presumptions related to the cost benchmarks (prices below AVC are presumed illegal). Ezrachi accepts that the Court may have not yet had sufficient time to develop the new approach, due to the fact that not every case provides relevant opportunities for change. 184 In his opinion, a possible outcome may be the conception of two strings of EU Cou t s ju isp ude e, depe di g o hi h 185 approach is preferred by the Court. If this statement becomes reality, part of the case law of the Courts would line up with a more form-based approach and another part would embrace the new approach. This interesting outcome, which we will refer to in the next chapter, could be related to the objective of Competition Law that the Courts may want to benefit or render as a priority. Ez a hi s view a diffe f o Mosso s appa e tl o e opti isti thi ki g, which envisages the present stage of case law as a transitional stage towards a complete shift to an effects-based 180 The reader can consult Mosso s CV in the following website: http://www.concurrences.com/auteur/estevamosso?lang=en, consulted on November 16, 2012. 181 Additionally, during the same conference where Mosso gave his speech, Svend Albaek, also an officer (Senior Economist) of the Commission who participated more actively in the preparation of the Guidance, fully supported Mosso s opi io . , p. . 182 (2009), p. 58. 183 Case T-340/03, France Télécom v. Commission [2007], Paragraph 195. 184 In his own words: Again, it is important to acknowledge that only a short period of time has elapsed since the beginning of the discussion on the effects-based approach. As mentioned above, not every case provides the European courts with a relevant opportunity to advance an effects-based approach. Further, even if the courts were inspired by the effects-based approach, they might nevertheless hesitate to adopt novel approaches prior to their maduration over time. (2009), p. 59. 185 (2009), p. 60. 45 approach. 186 There is another interesting position related directly to predatory pricing: Wolfgang Wurmnest is optimistic as Mosso. He too exa i es the Cou t s uli g i France Télécom, but points out certain traits of change. We will later comment more on his study, but for now will focus on the following: Wurmnest refers to the assessment of the Court regarding prices below AVC. His conclusion from the wording of the judgment is that the Court is changing its rigid criterion according to which pricing below AVC is presumed abusive. As we mentioned in Chapter I, the Court traditionally considers that there is no rational explanation for pricing below AVC other than the intention to pre-empt the market: For example, instead of reiterating that such prices are always abusive, the Court reasoned that p i es elo AVC a e to e o side ed p i a fa ie a usi e i as u h as, i appl i g su h p i es, an undertaking in a dominant position is presumed to pursue no other economic objective save that of eli i ati g its o petito s . This reasoning reads like an endorsement of the refined approach proposed by the Commission in the Enforcement Communication as it opens the door for the incumbent to rebut the abuse-presumption by showing a legitimate reason why its below cost pricing campaign is a rational business strategy (emphasis added).187 On the basis of the p e ious slight ha ge i the Cou t s ite io that Wurmnest identifies, he concludes that it can be assumed that the ECJ will not oppose a more refined approach when testing for anticompetitive pricing abuses.188 We ill detail o e a out Wu est s fi di gs late i this hapte . From the former considerations we take as valuable for our study that, as noted by the mentioned academics, if it is true that the case law of the EU Judiciary has not entirely developed an effects-based approach; it is also true that such development is in progress. Opportunities and initiatives for change have been noticed, either if the outcome is the creation of two strings of jurisprudence, as Ezrachi suggests, or a full endorsement of an effects-based approach. We now consider the other two reasons stated by Mosso, namely, i) that in the Guidance the Commission prioritized its efforts only on some exclusionary conducts, and ii) that those exclusionary conducts are the ones resulting in anticompetitive effects and are really consumer harming. If we do that, we must emphasize that one of the conducts contemplated by the Guidance is predation, as we already described in Chapter I. Predatory pricing is indeed a priority for the Commission –follo i g Mosso s ie - due to its potential harmful effects on consumer welfare. Accordi gl , Mosso s des iptio of the ti id steps take the Co issio e efit ou stud . Fi stl , because according to him, in the exercise performed for the preparation of the Guidance, the Commission identified and focused its efforts on exclusionary abuses where the Judiciary was showing signs of adopting an effects-based approach.189 Secondly, because among those conducts is predatory 186 This is a personal interpretation out of Mosso s o ds, hi h e t a s i ed sup a. Wurmnest (2011), p. 140. 188 Wurmnest (2011), p. 141. 189 The Commission s app oa h follo s its o ligatio of ot depa ti g f o the EU Cou ts ase la . The li ited s ope of the Guida e ould espo d the to the Co issio s i te tio ot to isk o t adi tio ith the EU 187 46 pricing, which is the subject we analyze, with special attention on the recoupment factor. And thirdly, because the Commission identifies predation as harmful to consumer welfare, therefore, when tackling predatory pricing the Commission will be seeking consumer welfare as an objective. We find here the link between the enforcement of predatory pricing and consumer welfare, in the context of the enforcement priorities of the Commission. 2.2. The Guidance and consumer welfare: the concept of anticompetitive foreclosure As we already studied when dealing with the scope of the Guidance, the rationale behind the identification of the exclusionary conducts where the Commission sets its enforcement priorities, is consumer welfare. Following this objective, paragraph 5 of the Guidance indicates: In applying Article 82 to exclusionary conduct by dominant undertakings, the Commission will focus on those types of conduct that are most harmful to consumers. Consumers benefit from competition through lower prices, better quality and a wider choice of new or improved goods and services. The Commission, therefore, will direct its enforcement to ensuring that markets function properly and that consumers benefit from the efficiency and productivity which result from effective competition between undertakings. 190 We also mentioned that the exclusionary abuses contemplated in the document are regarded by the Commission as the ones that result in anti-competitive effects and really harm the consumer. 191 This general idea expressed by Mosso, referred to the promotion of consumer welfare, is expressed and repeated in the text of the Guidance: Of relevance to our study, it is important to emphasize that the objective consumer welfare starts in the Guidance with the transcribed text of paragraph 5, which sets forth the purposes of the Guidance, continues inter alia when it refers to the assessment of dominance,192 and it is still present when it refers to the particular exclusionary abuses contemplated in the text. This also applies to predatory pricing as we explained in the previous chapter, because consumer harm is acknowledged by the Guidance as an effect of predation.193 The Commission in the Guidance includes the concept of anticompetitive foreclosure as a trigger for taking action. This concept is relevant because it focus the attention of the Commission on conducts with an adverse effect on consumer welfare, and clarifies that the negative effects to consumers can jurisprudence. As Liza L. Gormsen points out: The Co u it Cou ts ha e ade lea i se e al judg e ts that guidelines and notices cannot depart from case law. Some argue that guidelines are only valid if they do not conflict ith the ou ts i te p etatio of the t eat p o isio s. (2010), p. 159. 190 Guidance (2009), para. 5. 191 Highlighted in this chapter he dis ussi g the Guida e s li ited s ope. 192 See paragraphs 10 and 11 of the Guidance (2009), in particular importance paragraph 11 in fine, which expresses: I this Co u i atio , the e p essio i ease p i es i ludes the po e to ai tai p i es above the competitive level and is used as shorthand for the various ways in which the parameters of competition –such as prices, output, innovation, the variety or quality of goods or services- can be influenced to the advantage of the dominant undertaking and to the detriment of consumers. (emphasis added) 193 See the Guidance (2009), para. 63, specifically when it indicates at the end: …the e ausi g o su e ha . 47 happen at the intermediate level, at final level, or at both levels. 194 Accordingly, in paragraph 19, the Guidance indicates: The ai of the Co issio s e fo e e t a ti it i elatio to e lusio ary conduct is to ensure that dominant undertakings do not impair effective competition by foreclosing their competitors in an anti-competitive way, thus having an adverse impact on consumer welfare... In this do u e t the te a ti-competitive foreclosure is used to des i e a situatio he e effe ti e 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 to the detriment of consumers… The Co issio will address such anti-competitive foreclosure either at the intermediate level or at the level of final consumers, or at both levels195 (emphasis added). The concept of anticompetitive foreclosure is located at what seems to be an initial stage of the assessment of the Commission, after a specific type of apparent abusive conduct is identified. 196 Then, there are specific criteria applicable for the types of exclusionary conducts contemplated by the Guidance. For example, the considerations in paragraphs 23 to 27 will apply to price-based exclusionary conduct.197 Then, paragraphs 63 to 74 contain specific considerations for cases of predatory pricing. Therefore, as for the specific case of predation, the Enforcement Communication (The Guidance) has to be read in conjunction with the more general sections on anti-competitive foreclosure, price-based exclusionary conduct, and objective justifications.198 Regarding in particular the concept and test for determining the existence of an anticompetitive foreclosure, as it reads from the text of the Guidance and is explained by Nicolas Petit 199 and Damien Geradin, the defi itio suggests that a t o-stage test will be relied upon to assess whether a given o du t is a ti o petiti e. 200 The two stages entail the finding of foreclosure and the finding of consumer harm. This is of major importance because it adds up another condition to the only finding of foreclosure or elimination of competitors, i.e., the finding of consumer harm as an effect of the conduct. Thus, 194 See footnote No. 2, para. 19 of the Guidance (2009). Guidance (2009), para. 19. 196 In a practical manner applied to predatory pricing, Rosenblatt et al explain the assessment as follows: For predatory pricing, the Commission first proves a so- alled sa ifi e i.e., losses or foregoing profits in the short te , a d the a isk of a ti-co petiti e fo e losu e of the a tual o pote tial o petito s. “a ifi e depe ds o a cost analysis of the average avoidable cost (AAC) pricing (i.e., costs which a firm would avoid incurring by ceasing a pa ti ula a ti it hi h gi es the Co issio a lea i di atio of sa ifi e . Ne e theless, the Guida e Pape demands that the Commission may also investigate whether the conduct led, in the short term, to net revenues lower than could have been expected from a reasonable alternative conduct (i.e., whether it incurred a loss that it ould ha e a oided , a d use di e t e ide e that lea l sho [s] a p edato st ateg . (2012). 197 Guidance (2009), para. 23. 198 Wurmnest (2011), p. 121. 199 (2009), p. 4. 200 (2010), p. 41. 195 48 the reference to consumer welfare is important as it suggests that a conduct that would merely affe t the st u tu e of o petitio , fo i sta e, eli i ati g less effi ie t o petito s but that would have no effect on prices or on the quality of products, or innovation, and thus would not harm consumers, would not lead to enforcement action by the Commission under Article 102. It is thus the presence of (likely) consumer harm that will trigger the intervention of the Commission (emphasis added).201 This goes in line with the statement included by the Commission in paragraph 6 of the Guidance, where it advised that what really matters is protecting an effective competitive process and not simply protecting competitors. In the same paragraph, the Commission accepted as a possible outcome that competitors who deliver less to consumers in terms of price, choice, quality and innovation will leave the market.202 Based on the text of the Guidance, the concept of anticompetitive foreclosure entails an assessment of the likely effects of the conduct of the undertaking in question, and within those effects there must be a negative effect on consumer welfare. This, of course, also means that the likelihood of effects, rather than the finding of intention is part of the definition. Therefore, [t]he Enforcement Communication seems to correct this shortcoming by tightening the standards of proof for showing that the conduct is capable of anti-competitive foreclosure.203 When making this last assertion, Wurmnest refers in a generic way to EU case law where the anticompetitive effects were sometimes inferred from the fact that the dominant undertaking held a high market share and acted with anti-competitive intent. According to him, the Guidance and the test for anticompetitive foreclosure tend to correct this practice. In cases of predation this has a huge importance, specifically in regards to the AKZO test for prices between AVC and ATC. In such cases, even though the Court avoided creating a presumption of illegality, predatory pricing can be proved if there is enough evidence of an intention to pre-empt the market, as we explained in Chapter I. Later on this chapter we will focus our attention in the further analysis established by the Guidance that pertains to the specific recoupment factor in cases of predatory pricing. Additionally, in Chapter III we will provide a definition of the concept of consumer welfare and the validity of this apparently teleological objective under EU Competition Law, taking into consideration that, as we have already learnt, it is a pivotal element in the enforcement priorities of the Commission. 2.3. The binding force of the Guidance: Is the Commission bound by the Guidance? In line with the fact that it is a source of soft law, paragraph 3 of Guidance indicates that it is ot intended to constitute a statement of the law and is without prejudice to the interpretation of Article 82 by the Court of Justice or the Court of First Insta e of the Eu opea Co u ities. 204 In addition to this, 201 Geradin (2010), p. 42. Guidance (2009), para. 6. 203 Wurmnest (2011), p. 133. 204 Guidance (2009), p. 3. 202 49 a d p o a l e ide i g the Co issio s i te tio to keep a a ple lee a , the te t of the Guida e is full of terms such as generally, normally, or in principle, with direct influence on the actions of the Commission when conducting its enforcement priorities.205 This fact is criticized by Valentine Korah as a possi le p odu e of u e tai t : This detracts from the predictability given by the guidance and enables the Commission to change its mind in a s that a lead to u e tai t . 206 As we noted before, the Commission is not allowed to depart in its decisions from the case law of the EU Judiciary. Therefore, probably obeying to this restriction the Commission clarified that its interpretation of Article 102 is without prejudice to the interpretation that may be given by the ECJ or the GC.207 In o e tio to this, fo Ni olas Petit the Guida e s effo ts fo the i ple e tatio of a effe ts-based approach could go beyond the la e ide tia e ui e e ts set up by the EU Courts. 208 According to him, the already discussed text of paragraph 195 of the France Télécom ruling (GC) is a clear example of a dismissal from the need for an effects-based approach under Article 102 TFEU. 209 Moreover, for Petit, the outlined text of paragraph 3, as well as statements such the one in paragraph 22, 210 o stitute optout lauses , which may allow the Commission to disapply the effects-based approach.211 However, we must also consider that previously the ECJ has ruled, in reference to the faculty of the Commission to depart from its Guidelines (in a generic way), that such action is not unrestricted and therefore must be motivated. This obligation particularly applies to individual cases where the principles of legitimate expectations and equal treatment may be harmed. The Court ruled as indicated in Dansk Rorindustri,212 he e it as uestio ed hethe o ot the appli atio of o e of the Co issio s Guidelines for the calculation of fines was legal, in the case of infringements committed before such Guidelines were adopted. In that case, the Court ruled as follows, referring in general to the Guidelines prepared and published by the Commission: The Court has already held, in a judgment concerning internal measures adopted by the administration, that although those measures may not be regarded as rules of law which the administration is always bound to observe, they nevertheless form rules of practice from which the administration may not depart in an individual case without giving reasons that are o pati le ith the p i iple of e ual t eat e t. … I adopti g su h ules of o du t a d announcing by publishing them that they will henceforth apply to the cases to which they relate, the institution in question imposes a limit on the exercise of its discretion and cannot depart 205 E.g., paragraphs 10,11,14, 20, and 23. (2011), p. 20. 207 Ezrachi (2009), p. 55. 208 (2009), p. 16. 209 In his own words: …the Commission may thus stray away from the effects-based approach where it can establish that the impugned conduct has an anticompetitive object, in particular, because the dominant firm intended to foreclose rivals. (2009), p. 16. 210 Paragraph 22 of the Guidance (2009), in what interests us, reads as follows: There may be circumstances where it is not necessary for the Commission to carry out a detailed assessment before concluding that the conduct in question is likely to result in consumer harm. If it appears that the conduct can only raise obstacles to competition and that it creates no efficiencies, its anti-competitive effect may be inferred. 211 (2009), p. 16. 212 Case C-189/02 P, Dansk Rorindustri and others v. Commission, [2005] E.C.R. I-5425, paras. 206-211. 206 50 from those rules under pain of being found, where appropriate, to be in breach of the general principles of law, such as equal treatment or the protection of legitimate expectations. It cannot therefore be precluded that, on certain conditions and depending on their content, such rules of conduct, which are of general application, may produce legal effects. 213 We notice that probably the Commission tried to avoid the creation of legitimate expectations when it d afted the te t of pa ag aph of the Guida e. Additio all , that follo i g Mosso s o ds, 214 it sought a safe spot when it stated that the Guidance application is without prejudice to the interpretations of the EU Courts. However, we cannot exclude the possibility of a conflict between case law and the Guidance. In such a case, considering the ruling of the Court in Dansk Rorindustri, the Commission would have to make a decision between following the letter of the Guidance or, to the contrary, abide itself to the opposing case law of the Courts, which may put at stake its ability to move forward on its reform objectives to (the application of) Article 102. As Damien Geradin states, this appears to be -in the end- a su je ti e de isio of the Co issio s staff. If that is the case, room for future change and advance towards the new approach would be feasible; despite the fact his prediction is rather negative: This would not be the first time the ECJ would abandon old case law in favour of a new analytical framework. While this approach by the Commission would be courageous and may garner support among reformists, it is most unlikely to be appealing to case teams and the Co issio s Legal “e i e, hi h a e e tai l o e i te ested i i i g thei ases by producing appeal proof decisions than in reforming the law. 215 When we analyze the ruling of the ECJ in Post Danmark, we will see how this decision may be an example of a welcoming approach of the Court to assessments of the Commission related to the application of different cost benchmarks than the ones pre- established by the Court. Also, we will app e iate ho so e of the out o es of this ase a e ased o the Guida e s o te ts, i deed i an effects-based approach. By contrast, in regard to recoupment we will learn that although the Commission attempted to conduct a recoupment test in France Télécom, this initiative had no apparent influence in the decisions of the Judiciary. This fact may be due to the lack of importance, or confusing role, we believe the same Commission rendered to its own analysis. Anyhow, the Courts appeared not to care too much about the referred factor, at least to a degree that would have allowed them to judge about the merits of the recoupment test that the Commission conducted. 213 Case C-189/02 P, Dansk Rorindustri and others v. Commission, [2005] E.C.R. I-5425, paras. 209 and 211. Look at the previous section about the Guida e s li ited s ope. 215 (2010), p. 51. 214 51 3. What is an effects-based approach? The reform on the application and interpretation of Article 102 TFEU is based on a shift from a formbased to an effects-based approach founded as well on an economics-based approach. In the following lines, we will try to explain the meaning of such a shift. We consider that a practical way of doing so is by establishing the difference between one approach and the other and by doing this, come up with the most important characteristics of an effects-based approach. We clarify first that an effects-based approach entails a sound economic assessment.216 That is why talking about an assessment based on the effects of an abusive conduct, implies that such assessment is based on economic analyses. We found that effects-based and economic-based is used indistinctively in most of the consulted literature. 217 We will use the term effects-based approach which should be understood as implying an economic assessment when conducted. The Report of the EAGCP analyzed the benefits of an effects-based approach compared to the traditional approach. 218 Its insights are useful for the understanding of the differences between both concepts. According to the Report, a form based approach is characterized by the categorization of conducts. This categorization will be problematic due to the fact that in many instances various practices can serve the same purpose, or produce the same effects on the market. 219 This would produce inconsistencies in the assessment of the conducts, ith so e p a tices possibly enjoying a relatively o e le ie t attitude e.g., e ause of diffe e t sta da ds . 220 In practice, the categorization of conducts ith diffe e t a d igid sta da ds may facilitate exclusion, or induce the dominant firm to adopt alternative e lusio a ethods, hi h a ell i fli t a highe ost o o su e s. 221 This basically means that a dominant undertaking may pursue and produce the same harmful effects by using other conducts that circumvent preconceived rigid standards. Also, the dominant undertaking could abstain from performing conducts that benefit consumers and are indeed competitive, based on the risk of falling within certain categorization of conducts. As the Report exemplifies, this may happen in cases of predation: For example, in the context of predation, tight rules against predatory pricing may lead the predator to offer better terms in other dimensions. It may be less likely that these better terms are passed on to final consumers. Final consumers then will no longer benefit from low prices in the short-run, and yet this alternative strategy may still have similar exclusionary effects in the long-run.222 As opposed to the form-based approach, an effects-based approach would start by assessing the effects of the anticompetitive conduct or conducts, in order to establish if these are compatible with Article 102 216 The Report (2005), p. 5. E.g., The Report (2005), p. 5, Hildebrand (2009), p. 363, Mosso (2011), p. 13, Petit (2009), p. 3. 218 The Report (2005), p. 5-8. 219 The Report (2005), p. 5. 220 The Report (2005), p. 6. 221 The Report (2005), p. 6. 222 The Report (2005), p. 6. 217 52 TFEU. Adopting such an effects-based approach would ensure that these various practices are treated o siste tl he the a e adopted fo the sa e pu pose. 223 In the words of Ioannis Lianos, he deciding whether the conduct of a dominant firm is compatible with competition law, the starting point should not be to classify the specific practices as a pre-existing behavioural category, but to perform a concrete analysis of the effe ts of the o du t o o su e s. 224 At this point, we could argue that the effects-based approach, as described by the Report, would leave no room to possible advantages of a categorization of conducts, such as legal certainty. 225 However, the Report clarifies that an effects-based approach indeed allows a per se rule, a trait of a categorization system. Despite this, if we analyze the example provided by the Report which defends the existence of per se rules, it seems that it is still not in favor of a categorization of conducts, at least as it has been adopted in the Guidance: In an effects-based approach a per se ule fo fi a ial p edatio , fo e a ple, ould e he e e ou ha e a fi a iall st o g i u e t/fi a iall eak e t a t, the i cumbent a ot i est i losses e o d a e tai th eshold to e defi ed ith efe e e to possi le p a ti es , ; i o t ast, a fo -based per se app oa h to p edato p i i g ould p es i e that the incumbent cannot lower its price below a certain th eshold. That is, e e he implemented as a per se rule, an effects-based approach includes more than a single practice (low pricing, high advertising, and so forth) and requires a richer description of the circumstances (financial conditions of the incumbent and the competitor) than a form-based approach.226 It seems that a middle point, such as the one the Commission opted for in the Guidance, is defended by Ioannis Lianos in the following lines: An effects-based approach does not reject the need for categorical thinking. The identification of a theory of consumer harm involves a process of categorization of the practice as falling within the ou da ies of a spe ifi theo … “pe ifi effe ts ill follo f o the lassifi atio of the practice as falling within a specific antitrust theory. 227 Even if a certain degree of categorization is kept, one of the main ideas within an effects-based app oa h is that it e og izes that most of the types of conduct covered by Article 82 can have both anticompetitive and pro- o petiti e effe ts a d the efo e a ot al a s e legal o illegal. 228 One can infer that categorization would not be a problem as long as the assessment to determine the illegality of a conduct continues lying on its actual or potential effects. This seems to be also the position taken by the Commission, which, by the way, is also see i gl i li e ith the ‘epo t s ite ia o e i g the negative effects of a too rigid categorization present in the form-based approach. In line with this, Mosso indicated that: 223 The Report (2005), p. 6. Lianos (2009), p. 20. 225 About certain merits of a form-based approach, Nicolas Petit comments the following: Whilst the forms-based approach presents some –underestimated- merits for both firms (legal certainty) and competition authorities (reduction of enforcement costs), its shortcomings are almost certainly more significant. (2009), p. 2. 226 The Report (2005), p. 6. 227 Lianos (2009), p. 21. 228 Hildebrand (2009), p. 363. 224 53 A forth improvement coming with the effects-based approach is that different types of behavior are dealt with by the same approach, an approach which focuses on effects and not on form. It is important to avoid forcing companies to stick to certain business practices for legal reasons even though there are more efficient ways to conduct business which would bring benefits to consumers. Focusing on effects will allow companies to choose the business practice they consider best for competition. 229 In regards to the benefits of an effects-based approach, it appears to be a common stance that it benefits the risk of over-dete e e, a d as a esult, p o otes the fi s a tio s hi h a e lassified within the concept of competition on the merits. It has ee o e ted, fo e a ple that [i]t reduces the risk of over-regulation and of chilling competition on the merits and anticompetitive unilateral a tio . 230 This idea is suppo ted Petit, he sa s that [u]nder a forms-based approach, competition enforcers run the risk of forbidding courses of conduct that have no anticompetitive –or that wield procompetitive- effects, thereby committing Type-I errors (i.e. false convictions). Also by Finell, who comments that [t]he move towards a more effects-based assessment, coherent with the reforms made ith e ti al a d ho izo tal est ai ts, has fi st of all edu ed the isk of o e dete e e. 231 To sum up, the idea behind an effects-based approach in dealing with exclusionary abusive behavior is that when applying it, a verification of competitive harm is required. 232 Thus, i de idi g to i g a ase, the o petitio autho it should the efo e fo us o ide tif i g the o petiti e ha of o e … The account should be both based on sound economic analysis and g ou ded o fa ts. 233 To this end, the Guidance of the Commission would be in line with the approach, considering the previous description of the concept of anticompetitive foreclosure. Accordingly, provided that it establishes that a finding of unlawful exclusionary conduct relies on proof that such conduct leads to actual or likely anticompetitive foreclosure. 4. Recoupment in the Guidance Based on our discussion in Chapter I, we noticed that, despite some changes towards the degree of importance of recoupment rendered both by the ECJ and the Commission in the Guidance, still recoupment is not an independent element of the assessment for predation; as opposed, for example, to the U“ “up e e Cou t s uli g i the Brooke Group case. If we take into consideration the most recent ruling of the ECJ where the issue was raised in a predatory pricing case, it is in principle understood that the Courts did not shift drastically to an effects-based approach in the France Télécom case.234 As stated by Kavanagh et al, [o]ne problem is that the France Télécom judgment from the CFI, 229 Mosso (2011), p. 20. Ezrachi (2009), p. 53. 231 Finell (2009), p. 286. 232 Petit (2009), p. 2. 233 The Report (2005), p. 13. 234 As we explained before in this chapter (The Guidance limited scope) Wurmnest considers slight changes in the Cou t s app oa h. 230 54 which came out after the Dis ussio Pape towards a more effects- ased app oa h…235 as pu lished, does ot follo the Co issio s o e However, one must take into account the characteristics of the reform applied to Article 102 and to the Guidance itself, to appreciate if together they open the possibility of a further change in the assessment of predation, specifically in relation to recoupment. To do this analysis, we will first go back to the Co issio s de isio i the F a e Télé o s case and study the importance given to the recoupment factor and the actual analysis of the Commission in this respect. 4.1. Recoupment: The Commission´s approach in France Télécom In France Télécom the ECJ considered recoupment as a relevant factor in assessing whether or not the practice concerned is abusive.236 The Court then provided two examples in which recoupment could be useful: …fo e a ple he e p i es lo e tha a e age a ia le osts a e applied, assist in excluding economic justifications other than the elimination of a competitor, or, where prices below average total costs but above average variable costs are applied, assist in establishing that a pla to eli i ate a o petito e ists. 237 In other words, the Court indicated that a recoupment analysis could help the Commission in proving that there are no justifications for pricing below AVC other than the elimination of a competitor. Then, in cases of pricing above AVC but below ATC, recoupment could help prove that a plan to eliminate a competitor exists. Even though -as we already recognized- the Court rejected recoupment as a requirement to ascertain predation,238 it may have made significant progress towards the allowance of objective justifications, at least in regards to the first example provided in the judgment. For Wurmnest, ad itti g that a recoupment analysis may help to weed out economic justifications for pricing below AVC, the Court seems to have silently corrected its overbroad statement in Tetra Pak II that prices below AVC are always a usi e if set a u de taki g i a do i a t positio . Therefore, one could affirm that interpreting the assertion of the Court a contrario, in the absence of recoupment the door could be open to justify a pricing conduct below AVC. In other words, the Court was acknowledging the existence of economic justifications for such conduct, a possibility which is contemplated in the Guidance of the Commission under the title of Objective necessity and efficiencies.239 235 (2009), p. 17. Case T- 340/03 France Télécom v. Commission, [2009] ECR I-2369, para. 111. 237 Case T- 340/03 France Télécom v. Commission, [2009] ECR I-2369, para. 111. 238 For Wurmnest, [t]his ould e i te p eted as if the ECJ eje ted the st i t e oup e t test hi h is ased o a quantification of gains and losses . (2011), p. 120 239 Guidance (2009), paras. 28-31. 236 55 The last reference is not the only remarkable circumstance of the case in connection to recoupment. Also in France Télécom, the Commission –for the first time-240 expressly reviewed the likelihood of recoupment and found it plausible on the following basis: the st u tu e of the a ket a d the 241 asso iated e e ue p ospe ts. We must avoid getting confused with such a statement: the Commission did not conduct a calculation of the fore coming profits of the incumbent undertaking and compared them with the losses suffered during the predation period. The exercise would have been an arithmetic calculation of losses against likely profits or a conduct-based screen (in the words of Hemphill).242 This is also known as a strict recoupment test, which puts a higher evidentiary burden on the competition authority or a private plaintiff.243 As indicated, the Commission analyzed the structure of the market, and thereby considered recoupment as plausible. The following are the main characteristics of the analysis conducted by the Commission: i. The Commission first acknowledged that recoupment is not a precondition for predation under EU law. 244 Inter alia, it e p essed: The Commission accordingly considers that proof of recoupment of losses is not a precondition for a fi di g of a use th ough p edato p i i g. ii. Immediately after, it clarified the type of assessment it undertook: …it must be pointed out that the recoupment of losses is rendered plausible in the present case by the structure of the market and the associated e e ue p ospe ts .245 iii. The Commission included two major components in its analysis: the analysis of entry barriers and the analysis of entry costs to the relevant market. 246 iv. In line with the last point, when conducting the analysis, the Commission focused its attention in four specific aspects: 247 a. The strategic barriers related to the effects of the link-up with the France Télécom group, together with the many synergies resulting therefrom; b. the disincentives to mobility on the part of existing subscribers; c. the cost of acquiring sufficient notoriety in a mass market; and d. the cost of rolling out alternatives to France Télécom's wholesale offering. v. Additionally, the Commission considered that the rebuilding and increasing of Wanadoo Interactive's margins (based on the projections provided by the firm applicable as of 2003) were evidence of the likelihood of recoupment in the long-run.248 240 According to Wurmnest (2011), p. 110. Commission Decision, Wanadoo Interactive, Case No. COMP/38.233, para. 336. 242 (2001), p. 1586. 243 Wurmnest (2011), p. 110. 244 Commission Decision, Wanadoo Interactive, Case No. COMP/38.233, paras. 332-335. 245 Commission Decision, Wanadoo Interactive, Case No. COMP/38.233, para. 336. 246 Commission Decision, Wanadoo Interactive, Case No. COMP/38.233, paras. 337 and 338. 247 Commission Decision, Wanadoo Interactive, Case No. COMP/38.233, paras. 339-364. 241 56 When considering the mentioned four aspects, the Commission reached the following conclusions, which in its view evidenced the existence of significant barriers to enter the market, most of them related to the high costs a new competitor would have to cope with when entering the relevant market: The strategic barriers related to the link-up with the France Télécom group, and synergies therefrom: The Commission determined that due to the linkage with France Télécom (WIN was the internet arm of the group), WIN enjoyed commercial and deployment facilities, as well as financial support that put it in a higher stance, compared to its competitors. 249 With regards to synergies, WIN enjoyed clear advantages in comparison to its competitors, as a consequence of its relation with France Télécom. Such relation guaranteed advantages due to the well-known name of the group, benefits from its know-how, and benefits such as technical, commercial, and logistical and financial support from the group. 250 Disincentives to mobility: The Commission identified important disincentives to mobility fo WIN s customers, which entailed that any new entrants would have had to focus mostly only on new customers to increase their market shares. For example, some of the disincentives determining the low customer turnover in the high-speed internet services were the long commitment periods of WIN contracts, and important dissuasive (additional) costs related to moving to the same services provided by other firms.251 The cost of acquiring sufficient notoriety in a mass market: The Commission indicated that even though the barriers to enter the market and bear the costs required for acquiring sufficient notoriety where not insurmountable, it would be very difficult to cope with them, even for firms with high notoriety already in the low-speed market for internet.252 248 Commission Decision, Wanadoo Interactive, Case No. COMP/38.233, paras. 365 and 366. Co pa ed to the g oups a ki g up WIN s o petito s, o e of the ould p ete d to p o ide thei subsidiaries with technical backing and backing in terms of commercial network in France on such a decisive scale as those provided by France Télécom to Wanadoo I te a ti e. Commission Decision, Wanadoo Interactive, Case No. COMP/38.233, para. 228. 250 A o di g to the Co issio , all these e e app e ia le ad a tages that o t i uted to WIN s do i a t position. Commission Decision, Wanadoo Interactive, Case No. COMP/38.233, paras. 223-246. 251 Commission Decision, Wanadoo Interactive, Case No. COMP/38.233, paras. 340- 345. 252 The high expenses, for example, of WIN in advertisement and promotion, would operate as dissuasive costs for the potential entrants. The Commission concluded in this regard that: It is, therefore, not so much the nature and unitary amount per new subscriber of the advertising and promotion expenditure as the overall volume of the commitments needed to acquire a critical subscriber base that turns advertising and promotion expenditure into a potentially dissuasive entry cost. Even companies with a firm foothold in the low-speed Internet access market may hesitate to embark upon advertising and promotion campaigns on the same scale as Wanadoo Interactive. In this o te t, the effo t e ui ed of o petito s i o de to e o e o e tha e e a kg ou d oise has to e see as a barrier to entry and to re-entry for new entrants of the Mangoosta type, or as a significant entry cost for operators already present on the low-speed market, both of high promote a high degree of market concentration. The foreseeable difficulty for competitors to acquire a critical size on a mass market enabled Wanadoo to envisage as perfectly plausible a market sufficiently concentrated to allow a recoupment of losses. Commission Decision, Wanadoo Interactive, Case No. COMP/38.233, para. 358. 249 57 The cost of rolling out alternatives to France Télécom's wholesale offering: In the recently liberalized a ket, WIN s o petito s had the optio to uild thei o et o ks a d i f ast u tu e e ui ed to provide the respective high-speed internet services (product market) in which WIN was found to predate. This would have allowed competitors to provide better quality services at a lower price. However, the Commission pointed out that the high costs of doing so created a barrier to enter the market.253 The Co issio s de ision is dated July 16, 2003. As the reader may notice, although the reform to Article 82 EEC (now Article 102 TFEU) had not yet formally begun, the Commission made a step forward and, upon the mentioned analysis reached the conclusion that recoupment was plausible. As we pointed out, the analysis is based on the particularities of the market structure. 254 The analysis also reminds us of He phill s des iptio of his p oposed a o a d deep s ee fo recoupment: It could look at industry structure, asking whether there are barriers to entry that will allow the predator to profit from its illgotten monopoly once the competitor has been eliminated (or co-opted . 255 Despite this, the test does not seem close to the dangerous possibility of recoupment of losses threshold of the Brooke Group test. Additionally one must remember that in France Télécom the European Courts did not go far enough into the discussion of whether or not recoupment had been assessed correctly by the Commission. The courts just avoided entering into this rather thorny ground and stated that e oup e t s proof was not required. We will discuss this particular situation later in this chapter, to see that neither the GC, nor the ECJ elaborated further on the subject because they sided with their previous rulings, indicating that recoupment was not required within the predation s assess e t. We will also further discuss the Co issio s a gu e ts elated to its o recoupment test. It is also important to notice that, as we ill stud i the e t hapte , the GC s uli g elated to recoupment justified AG Mazák s Opi io e o e di g the ee a i atio of the GC s judg e t. Fo ou stud s sake, it is also oti ea le that the Cou t s e og itio of a deg ee of i po ta e to recoupment may have been triggered by the preceding Commissio s assess e t, although the e as no opinion from the Courts as to the merits of the assessment in question. Most importantly, the analysis made by the Commission was based on the likely effects on the structure of the market, and could have meant that initial steps were taken into an effects-based approach related to the particular abusive conduct. 253 The cost and time needed to set up an ADSL network comparable to that of France Télécom are such that they limit considerably the scope for competing service providers to propose an economic model that is more attractive to the consumer. They must therefore also be interpreted as barriers to entry. Commission Decision, Wanadoo Interactive, Case No. COMP/38.233, paras. 359- 364. 254 According to Wurmnest, The st u tu al e oup e t-test intends to assess the chances of success of the predatory pricing strategy (and therefore its plausibility) by a closer inspection of the structural factors of the market, such as the strong market position of the predator vis-à-vis its prey, excess capacity, high barriers to entry a d the de elop e t of a ket sha es du i g the p edatio a paig . (2011), p. 110. 255 However in the screen proposed by Hemphill other elements are considered, such as a closer look at strategic decisions and the chances of success of the pricing policy to show that the pricing can cause social harm. (2001), p. 1581- 1612. 58 4.2. Recoupment a d the Guida e’s o je tive Considering that in the reform of Article 102 TFEU and, particularly in the Guidance, the main aim is consumer welfare, it is necessary to establish if proof of recoupment has a relevant role to play within the Guida e s ai o je ti e. A o di g to Go se s ie , p oof of recoupment indeed follows such an aim: Whe the ai is p o oti g o su e elfa e p oof of likel e oup e t ould e e ui ed. 256 Contrarily, when rendering more importance to economic freedom (as another aim), it would only be necessary to establish the likelihood to pre-empt the market, thus protecting the structure of competition: In this case, there is no need to examine whether the dominant undertaking is able to recoup; predatory pricing is found to be abusive where there is a risk of eliminating competition. 257 Gormsen also stresses the fa t that AG Mazák s Opinion shares her view, indicating that Mazák s Opinion concludes that the Commission cannot rely on the presumption of abuse where there is a risk that competition will be eliminated in cases where the Commission decides that consumer welfare must take p io it o e e o o i f eedo . 258 In Chapter III we will make a more extended analysis of Go se s fi di gs elated to o su e elfa e a d of Mazák s Opi io . For now, it is important to highlight the reasons provided by the author according to which recoupment has a consequential link with consumer welfare and thereby with the purposes of the Guidance. Papandropoulos shares Gor se s view. Both concur in terms of the causal link between ability to recoup and consumer harm, in conjunction with the other factors that determine predation: A o di g to e o o i theo , a edi le p edato sto should i lude th ee ele e ts: a a p ofit sa ifi e…; the fo e losu e of i als; the possi ilit of e oup e t. A se t a theo of harm that covers all these criteria, a case of predation would be difficult to make. For example, a predatory strategy that entails profit sacrifice and rival foreclosure but does not lead to e oup e t ould ot ha o su e s a d eed ot e pu sued. 259 Following the previous arguments, we can state that in the absence of recoupment the ability of the incumbent to harm consumers is discarded, as the monopoly level pricing is thereby not feasible. If this argument is correct, there is a link between the requirement of recoupment with consumer welfare, which makes se se a d has a logi al o e tio also ith the Guida e, o side i g the Guida e s main objective. 4.3. The Commission may have conducted a dominance test in France Télécom When des i i g He phill s positio in the previous chapter, we mentioned that there is a discussion concerning the correct manner in which the recoupment assessment should be conducted. Hemphill argues if favor of granting greater importance to the structure of the market and less importance to the foreseeable profits of the incumbent firm. We described these factors in Chapter I, and Wurmnest reminds us of them: Fa to s to e o side ed a e, i te alia, ha ges of a ket sha es, the du atio of 256 (2010), p. 146. Gormsen (2010), p. 142. 258 (2010), p. 146. 259 Papandropoulos (2010), p. 426. 257 59 the lo p i i g a paig , apa it est ai ts o a ie s to e t . 260 He also underscores the fact that a strict recoupment test ignores other important factors such as the reputation effects of predation, which may well determine an economic success for the incumbent firm: Fi a ial gai s ste i g fo reputation effects in other markets –which are usually not quantifiable- a e left out of the a al sis. The idea of a st u tu al e oup e t-test 261 reminds us of the theory of the Commission in France Télécom, according to which recoupment is implied in the dominance test. The ECJ summarized the Co issio s a gu e t as follo s: The mere existence of such a position is sufficient to determine that e oup e t of losses is possi le. 262 However, there are important arguments against the position taken by the Commission. According to Patrick Rey, the dominance test falls short in comparison to a proper recoupment test.263 Indeed, we believe that this argument is correct, as the likelihood of recoupment in certain cases could not depend necessarily on the dominance position of the predator. For example, as Papandropoulos points out, there are factors that could determine, once the market is pre-empted, that the predator (dominant) cannot exert enough market power to recoup losses. Moreover, there may be also cases where the predator did not enjoy a dominant position but its tough reputation could provide tools for a successful recoupment: However, relying on dominance to assess recoupment may lead to over-enforcement (i.e., concluding that recoupment is likely when it is not necessarily the case). Indeed, there are many instances in which recoupment is not likely despite the pre-predation dominant position of the alleged predator. What matter is the increased scope for exercising market power once the predatory strategy has successfully led to exit. For example, rivals may be led to exit but their assets may remain in the market, or if a rival offering a distant substitute has been excluded, the scope for exercising market power would be limited. Moreover, successful recoupment can occur despite the absence of substantial market power ex-ante. Indeed, predation aimed at building a reputation may be implemented in small markets where the predator does not have significant market power (but holds some advantage relative to existing competitors such as excess capacity to implement a predatory strategy). Market power can then be exercised on other markets where the tough reputation becomes a barrier to entry. 264 Furthermore, the assumption of recoupment after a positive dominance test clearly restricts the possibilities of defense of the incumbent firm, as the application of Article 102 TFEU already implies holding a dominant position. Indeed, if proof of dominance is sufficient to establish likelihood of recoupment, dominant undertakings have no other defence than proving that they are not dominant when trying to oppose a recoupment assessment. Therefore, the correlation between dominance and e oup e t is ot to e take fo g a ted, ithout the isk of u fai l affe ti g the defe da ts ights. 260 In a similar way to Hemphill, the author proposes a screen based not on the calculation of losses and profits, but on what he calls a st u tu al e oup e t test . Wurmnest (2011), p. 139. 261 Wurmnest (2011), p. 139. 262 Case C-202/07P, France Télécom v. Commission [2009] ECR I-2369, para. 102. 263 Mateus and Moreira (2010), p. 196. 264 Papandropoulos (2010), p. 427. 60 As pointed out by Gormsen, assuming recoupment because dominance has been proved would be paradoxical, as Article 82 does not even come into consideration unless the undertaking is dominant. 265 When analyzing the judgments of the GC and of the ECJ in France Télécom, we can infer the possibility of a a iatio i the Co issio s a gu e ts as to the recoupment assessment it conducted. We believe that the Commission could have never been convinced of the accuracy or deepness of its recoupment assessment, and as a consequence it finally preferred to keep the argument that dominance implies ability to recoup, an argument that was raised until the appeal before the ECJ took place. We reach this conclusion based on the following:     The Commission never accepted that recoupment was a requirement to establish predation. It announced this conviction since the moment it issued its decision against WIN.266 When WIN argued (appeal before the GC) that the Commission had committed a manifest error of la , a d a e o of assess e t by considering that it had furnished evidence of a possibility to e oup losses ; the judgment indicates that the Commission first argued that recoupment was not a precondition to making a finding of predatory pricing. Only as an alternative, it argued that e oup e t in the present case is plausible by the structure of the market and the related revenues hi h thus a e e pe ted i the futu e. 267 After the GC s judgment was challenged before the ECJ, this last Court ruled expressing that the Commission continued first denying its obligation under EU law to prove recoupment. However, it seems that it then modified the rest of its arguments by raising the correlation between dominance and e oup e t: A o di g to the Co issio , u like the app oa h u de U ited “tates law, the analysis of abuse within the meaning of Article 82 EC presupposes that the undertaking concerned is in a dominant position. The mere existence of such a position is sufficient to determine that e oup e t of losses is possi le. 268 Apparently, only as a last resource it kept the argument of the possibility of recoupment, with some changes in comparison to the one raised before the GC: Fi all , i the p ese t ase, the e po e tial g o th of the a ket o e ed ade su h e oup e t likel . 269 Based on this analysis, the possibility of recoupment was kept by the Commission as a last resort and the grounds upon which recoupment was argued even changed. This could have been because the Commission preferred to side with the case law of the Courts, thus, saving the opportunity to discard the requirement in future cases where recoupment could have been harder to assess. Another reason may be that the Commission was conscious that the assessment was not correctly done, thus would have been easily rebutted by the defendant. It could have happened also that the Commission was finally convinced that the analysis it had conducted was closer to a dominance test than to a duly conducted structural recoupment test. 265 Gormsen (2010), p. 144. Commission Decision, Wanadoo Interactive, Case No. COMP/38.233, paras. 332-335. 267 Case T-340/03, France Télécom v. Commission [2007], paras. 222 and 223. 268 Case C-202/07P, France Télécom v. Commission [2009] ECR I-2369, para. 102. 269 Case C-202/07P, France Télécom v. Commission [2009] ECR I-2369, para. 102. 266 61 The main question that remains without a clear answer is: Why did the Commission perform a recoupment test if it was convinced that e oup e t s proof was not necessary? Then, even if the Commission decided to align itself to the case law of the Courts, according to which recoupment is not required, why did it finally defend the correlation dominance-recoupment, when it had already performed a separate recoupment test? 4.4. The Guidance opens a room for recoupment Despite the last considerations about the Co issio s recoupment test in France Télécom, the sole fact that the Commission decided to conduct a recoupment test is a valuable input for our study. It may not have been the most adequate assessment, 270 but could have opened the possibility for future assessments, and even for the convincement of its need in the years to come. This last point is especially important considering that by 2003, the Commission may have not been convinced of the need of a recoupment test applied through an effects-based approach, due to the fact the reform to the abusive condu t s egi e had ot e e –at least formally- begun. However it went on and performed the assessment. Now that the Guidance exists as a result of this reform, it may exert the required pressure for a major adjustment on the predation assessment; despite the fact the Commission already warned that a e ha i al al ulatio of p ofits a d losses, a d p oof of o e all p ofits is ot e ui ed. 271 According to the last assertion, which is part of the text of the Guidance referring to predation, the Commission does not abide itself to a strict recoupment test. However, this does not discard the possibility of its application in future cases, even though it seems more likely that the Guidance leaves o e oo fo a test ased o the st u tu e of the a ket: Likely consumer harm may be demonstrated by assessing the likely foreclosure effect of the conduct, combined with consideration of other factors, such as entry barriers. In this context, the Commission will also consider possibilities of re-e t . 272 According to Wurmnest, the Commission did well by disregarding a strict recoupment test, and even though in his conclusions he envisages more room for future changes related to objective justifications and efficiencies, he does conclude that the Co issio see s to e committed to a more structural recoupment-test hi h looks at the a ket setti g a d st ategi o te t of the i u e t s 273 eha iou . Kavanagh, Marshall, and Niels (Kavanagh et al) have a similar approach that envisages the inclusion of recoupment due to the Guidance's influence, however they lean more towards a recoupment test similar 270 Our intention is not to evaluate the recoupment assessment conducted by the Commission, considering that the Courts did not do it, and that from the reading of the judgments in France Télécom, the Commission never defended, neither provided arguments about the merits of the assessment. Moreover, we have not found academic sources that substantially support a thesis in favor or against it. We base our analysis on the mentioned arguments provided by the Commission at the judicial stage and depicted in the rulings both of the GC and the ECJ. 271 Guidance (2009), p. 71. 272 Guidance (2009), p. 71. 273 Wurmnest (2011), p. 141. 62 to the one in the US. 274 They infer from paragraphs 70 and 71 of the Guidance 275 a shift to a ds o side atio of e oup e t 276 that would eventually bring EU policy closer to the US. It is also significant that in their conclusions they render importance to the effort of the Courts in allowing such a change. Indeed a change could have taken place in the France Télécom case if the Court would have liste ed to Mazák s Opi io : Yet the European courts may well be the ultimate drivers of change. Of great significance is the Advocate-Ge e al s Opi io i the F a e Télé o appeal u e tl efo e the ECJ, to the effe t that recoupment (which itself is more in line with an effects-based approach to abuse of dominance) is indeed an important factor when assessing predatory pricing cases. 277 All i all, if e o e t the Co issio s atte pt to use a recoupment assessment in France Télécom with the reform to the application of Article 102 and in particular, with the text of the Guidance, we may conclude that a recoupment requirement is indeed likely to happen. It could be more likely that such a requirement follows a European recipe, distant from the one applied in the US, and therefore different from the one expected by Kavanagh et al. Following the assessment used in France Télécom, even if it resembled more a dominance test, it may signal a future test that considers more the structure of the market, rather than an arithmetic calculation of losses and future revenues. 5. The Post Danmark case In Post Danmark, 278 the ECJ s ruling is related to the interpretation of Article 102 TFEU in a case of selective price cutting. The decision is the response of the Court to a Reference for a Preliminary Ruling u de A ti le TFEU f o the Højeste et De a k s “up e e Cou t . We will first summarize the facts of the case and the questions referred to the ECJ, afterwards we will provide the reader with an outline of the most important aspects of the ruling. We will make special emphasis on the points relating to this thesis. 5.1. The facts The main proceedings involve a dispute between Post Danmark (PD) and the Danish Competition Council (DCC), the Konkurrenceradet. At that time, PD and its competitor (Forbruger-Kontakt) were the two largest undertakings in the unaddressed mail sector in Denmark, which was already entirely liberalized. The relevant market was defined as the distribution of unaddressed mail in Denmark. 279 This is effe ti el a e oup e t test, as applied i the U“A ut eje ted i the EU- both in existing case law, and in the Discussion Paper. Such a test should form an inherent part of any assessment of alleged predatory prices, and the u de of p oof should ot e essa il fall o the defe da t alo e. (2009), p. 17. 275 We direct the reader to Chapter I where we described the treatment of recoupment in the Guidance. 276 (2009), p. 15. 277 (2009), p. 17. Their contribution was written when the appeal was pending before the ECJ, whose judgment in regards to recoupment as a i depe de t e ui e e t fo the p edatio test o fi ed the p e ious Cou t s criteria in Tetra Pak II. 278 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012]. 279 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para. 3. 274 63 Additionally, PD enjoyed a monopoly in the delivery of addressed letters and parcels not exceeding a certain weight. For that reason PD had a network covering the entire national territory which it also used for the distribution of unaddressed mail. 280 Until 2004, the following supermarket undertakings were major customers of Forbruger-Kontakt: Superbest, Spar, and Coop groups. At the end of 2003, PD concluded its contracts with these undertakings for the distribution of their unaddressed mail. As of January 2004 the distribution contracts entered into force.281 The Konkurrentcerådet (Danish competition council), following Forbruger-Ko takt s o plai t, issued two decisions dated September 29, 2004 and November 24, 2004. Both decisions were later upheld by the Kondurrenceankenævnet (competition appeals tribunal), which is a Court of second and final instance (regarding the National Competition Authority or NCA), before the decisions were also contested to the Danish Judiciary:282 First decision (September 29, 2004): The NCA held that PD had abused its dominant position by p a ti i g a ta geted poli of edu tio s desig ed to e su e its usto e s lo alt : i ot putti g the o a e ual footi g i te s of ates a d e ates, a d ii ha gi g its o petito s fo e customers rates different from those it charged its own pre-existing customers without being able to justify those significant differences in its rate and rebate conditions by considerations relating to its costs.283 Second decision (November 24, 2004): The NCA found that it was not possible to establish that PD had intentionally sought to eliminate competition and that, accordingly, it had not committed predatory pricing.284 Both decisions were contested by PD to the Judiciary inasmuch they concerned an abuse of a dominant position by means of selectively low prices applied to SuperBest, Spar, and Coop. In a first judicial decision the Østre Landsret (eastern regional court) upheld the decisions insofar as they found that PD had abused its dominant position by pursuing, in 2003 and 2004, a pricing policy for former customers of its competitor different from its policy for its pre-existing customers, without being able to justify that difference on cost-related grounds.285 Finally, PD ought a appeal agai st the Øst e La ds et s uli g efore the Højesteret, and in particular it alleged that a o di g to the ECJ s uli g i AKZO, the prices offered to the Coop group may be considered to amount to abuse only if an intention to drive a competitor from the market was established. The NCA opposed PD alleging that an intention to drive a competitor out of the market is 280 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para. 4. Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para. 6. 282 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], paras. 8, 13, and 14. 283 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], paras 8 and 14. 284 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], paras. 13 and 14. 285 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], paras. 15 and 16. 281 64 not absolutely necessary in order for a practice of selective prices, lower than average total costs but higher than average incremental costs, to amount to an abuse of dominance. 286 In this same line, the questions posed to the ECJ are related to the contract executed with the Coop group (Coop). An important fact is that before the latter concluded the distribution contract with PD it also conducted negotiations with Forbruger-Kontakt. The offers of the two undertakings to Coop were o pa a le i te s of p i e; PD s ei g o l a gi all lo e . 287 The referring Danish court (Højesteret) indicated to the ECJ that it had been demonstrated that the price offered by PD to Coop did not allow PD to cover its average total costs, but did allow it to cover its average incremental costs (AIC).288 5.2. The questions referred to the ECJ The ECJ interpreted the questions posed by the Danish court as follows: What are the circumstances in which a policy, pursued by a dominant undertaking, of charging low prices to certain former customers of a competitor must be considered to amount to an exclusionary abuse?, and in particular, whether the finding of such an abuse may be based on the mere fact that the price charged to a single customer by the dominant undertaking is lower than the average total costs attributed to the business activity concerned, but higher than the average incremental costs pertaining to the latter?289 5.3. Main aspects of the ruling 5.3.1. Consumer welfare: In its response the Court provides a general explanation of the scope of Article 102 TFEU, focused on the o ept of p ote tio f o o du ts p odu i g o su e ha . It i di ated that the T eat s p o isio (Art. 102) covers not only those practices that directly harm consumers but also those practices that ause o su e s ha th ough thei i pa t o o petitio . The Cou t ei fo ed this asse tio indicating that it is in such sense that the expressio e lusio a a use ust be understood.290 The Court also acknowledged that the departure or marginalization of less efficient competitors is a logical consequence of competition on the merits. It characterized the less efficient competitors as those who are less attractive to consumers from the point of view of -inter alia- price choice, quality, or innovation. Most importantly, the Court indicated that the effects of an unlawful exclusionary conduct are a detriment to consumers: Article 82 EC applies, in particular, to the conduct of a dominant undertaking 286 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para 17. Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para 7. 288 The Danish competition authorities applied this cost benchmark (AIC) instead of AK)O s stemming AVC. The Cou t a epted the easu e as a p o fo AVC, hi h is also A eeda a d Tu e s hose p o to a gi al cost. See Chapter I where we discussed on the costs assessment of the European Courts. 289 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], paras. 18 and 19. 290 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para. 20. 287 65 that, …, has the effe t, to the detriment of consumers, of hindering the maintenance of the degree of competition existing in the market or the growth of that competition (emphasis added).291 5.3.2. Co petitio o the e its a d use of the te as-effi ie t o petito s : The Court restated its settled case law in the sense that having a dominant position is not unlawful under EU Competition Law. 292 Additionally, it clarified that the purpose of Article 102 is not to prevent an undertaking from acquiring, on its own merits, the dominant position on the market. In this vein, it explained that Article 102 does not seek that less efficient competitors than the dominant undertaking remain in the market. 293 It completed this idea expressly saying that not every exclusionary effect is necessarily detrimental to competition.294 Of importance to comment that the case law in which the Court founded this contents of the ruling is TeliaSonera Sverige,295 a recent ruling (2011) of the ECJ recognized as innovating with regards to the application of an effects-based approach, and particularly, the application of the aseffi ie t o petito test, a d the e og itio of the do i a t u de taki gs ight to o je ti el justif abusive conducts. The Court synthetized the aforementioned criteria in direct application to price-related exclusionary conduct, indicating that Article 102 prohibits a dominant firm from adopting pricing practices that have an exclusionary effect on competitors considered to be as efficient as it is itself and strengthening its dominant position by using methods other than those that are part of competition on the merits. 296 5.3.3. The cost-benchmark (AIC) and aspects directly related to price-based exclusionary conducts: In this specific field, the Court first restated its ruling in AKZO,297 according to which prices below AVC must, in principle, be regarded as abusive inasmuch as, by charging such prices a dominant undertaking pursues no economic purpose but driving out its competitors. Moreover, also from AKZO, prices below ATC but above AVC must be regarded as abusive if found as part of a plan to eliminate a competitor. 298 As noted in the summary of the facts, the ECJ agreed to the use of AIC as a substitute for the AK)O s benchmark of AVC. This fact is remarkable and follows the Opinion of AG Mengozzi, who bases its recommendations in previous criteria of the Commission in a case related also to the mail delivery business sector: Deutsche Post.299 According to the AG, using AIC is more appropriate when dealing with 291 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para. 24. Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para. 21. The Court mentioned: Case 322/81 Nederlandsche Banden-Industrie-Michelin v Commission [1983] ECR 3461, para. 57 and Joined Cases C-395/96P Compagnie Maritime Belge Transports and Others v Commission [2000] ECR I-1365, para. 37. 293 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para. 21. 294 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para. 22. 295 Case C- 52/09, TeliaSonera Sverige [2011] ECR I-0000. 296 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para. 25. 297 See Chapter I, when dealing with the costs assessment of the European Courts. 298 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para. 27. 299 Commission Decision, Deutsche Post AG, Case No. COMP/35.141. In this case the Commission found that Deutsche Post had abused its dominant position by offering the services in question at prices below its incremental 292 66 dominant undertakings entrusted with a task of general economic interest, due to the fact that in many cases (as in PD) there is existing infrastructure in place for the services provided, serving both the liberalized and the not liberalized activities. 300 The AG determined that a comparison of prices with the variable costs of a dominant undertaking entrusted with a task of general economic interest (public service or universal service) was inappropriate. This comparison could mean overestimating losses because the u de taki g s task of ge e al e o o i i te est e tails highe osts tha those of its competitors for the part of its business operations accounted for in the market open to competition. Conversely, to take the variable costs of the dominant undertaking as the only criterion could also lead to an overestimate of its costs, if it operates with high fixed costs (the cost of using its network is an example) and small variable costs. 301 The AG concluded that in this specific sector of the economy AIC is a good proxy to the AVC, and the ECJ embraced such conclusion for the same reasons provided by the AG: 302 In the main proceedings, as in the Deutsche Post case cited above, the average incremental costs, like the average variable costs referred to in paragraphs 71 and 72 of the AKZO v Commission judgment, represent, therefore, the lowest limit or minimum below which the price of the undertaking in the dominant position would be predatory. 303 After regarding the AIC as a proper benchmark, the Court estimated that the prices charged by PD to Coop cannot be considered to amount to an exclusionary abuse simply because the prices charged to a single customer were lower than the ATC, but higher than the AIC, both attributed to the concerned activity.304 Then the Court suppo ted its de isio o the e tio ed as-effi ie t o petito test , indicating that: to the e te t that a do i a t u de taki g sets its p i es at a le el o e i g the g eat ulk of the costs attributable to the supply of the goods or services in question, it will, as a general rule, be costs, thereby it had offered predatory prices. The Commission defined those costs as the costs arising solely from the provision of a specific service, which depend on the quantity supplied and which would disappear if the service were discontinued, which implies that the common fixed costs, which do not depend on the provision of a single service, are not included in the incremental costs. The case is related to State Aid under the context of Article 107 TFEU. The Co issio s decision was annulled by the GC because it failed to prove that the measures provided amounted to state aid according to Article 107 TFEU. Lastly in appeal, the ECJ upheld the GC s uli g. ‘eference: Case C-399/08 P Commission v. Deutsche Post. 300 PD enjoyed a monopoly in the addressed mail sector and its operations in the liberalized sector of unaddressed mail benefited from the network used for the non-liberalized sector. In the terms used by the AG, the use of AIC instead of AVC appeared to be explained by the coexistence, within PD, of, on the one hand, reserved and nonreserved operations the carrying on of which, in either case, is subject to universal service obligations and, on the other hand, purely commercial operations which take place on the liberalized market for unaddressed mail. Opinion of AG Mengozzi in Case C-209/10 Post Danmark A/S v Konkurrenceradet, para. 33. 301 Opinion of AG Mengozzi in Case C‑209/10 Post Danmark A/S v Konkurrencerådet, para. 34. 302 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], paras. 31 to 34. 303 Opinion of AG Mengozzi in Case C‑209/10 Post Danmark A/S v Konkurrencerådet, para. 38. 304 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para. 37. 67 possible for a competitor as efficient as that undertaking to compete with those prices without suffe i g losses that a e u sustai a le i the lo g te . 305 5.3.4. Effects of the conduct: The ECJ noticed that the pricing policy of PD did not produce anticompetitive effects. This means that ot o l the Da ish autho ities did ot fi d e ide e of a PD s pla to p e-empt the market or in other way affect its competitor (anticompetitive object), but that the ECJ went further and took notice of the present state of the market, according to the facts provided by the Danish Court. Consequently the Cou t poi ted out that PD s o petito a aged to ai tai its dist i utio et o k despite losi g the volume of mail related to the three customers involved. Additionally, in 2007 Forbruger-Kontakt managed to regain Coop and Spar groups as customers.306 5.3.5. Objective justifications and efficiencies: Lastly, the ECJ reminded the Danish Court that it is open to a dominant undertaking to provide justification for behavior that is liable to be caught by the prohibition under Article 82 EC. Such possibility would allow the undertaking to demonstrate that its conduct is objectively necessary, or that the exclusionary effects produced may be counterbalanced or outweighed by advantages in terms of efficiency that also benefit consumers. 307 The Court outlined the requirements to prove, and therefore establish the mentioned gains in efficiency, in very similar terms to paragraph 30 of the Guidance. It also made clear that it is on the concerned u de taki g s shoulde to sho the effi ie gai s. The e ui e e ts ead as follo s: i the effi ie ies are likely to result from the conduct in question, ii) the efficiencies counteract any likely negative effects on competition and consumer welfare in the affected markets, iii) the gains have been, or are likely to be brought about as a result of the conduct, iv) the conduct is necessary for the achievement of the gains in efficiency, and, v) the conduct does not eliminate effective competition, by removing all or most existing sources of actual or potential competition.308 5.3.6. Operative part of the ruling: The ECJ responded to the Danish Court that Article 102: … ust e i te p eted as ea i g that a poli hi h a do i a t undertaking charges low prices to certain major customers of a competitor may not be considered to amount to an exclusionary abuse merely because the price that undertaking charges one of those customers is lower than the average total costs attributed to the activity concerned, but higher than the average incremental costs pertaining to that activity, as estimated in the procedure giving rise to the case in the main proceedings. In order to assess the existence of anti-competitive effects in circumstances such as those of that case, it is necessary to consider whether that pricing policy, 305 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para. 38. Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para. 39. 307 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], paras. 40 and 41. 308 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para. 42. 306 68 without objective justification, produces an actual or likely exclusionary effect, to the detriment of o petitio a d, the e , of o su e s i te ests. 309 The decision on the costs was left to the national court. 5.4. The Guidance and Post Danmark If we want to relate the PD ruling with the Guidance we have to be careful and avoid confusions. The main reason for this assertion was given when reviewing the scope of the Guidance: we learnt that in substance it follows the evolution of the case law of the European Courts in respect to the effects-based approach. Therefore, we cannot allow ourselves to say without first conducting a careful analysis that, if any, the innovations of the ruling, or signs of movement of the Court towards an effects-based approach have been adopted from the Guidance. Just to mention an example, if we pick one of the subjects mentioned supra, namely the as effi ie t o petito test a d t a k its oots i the Eu opea ase law, we can go as back as AKZO, a case where the Court applied such an assessment.310 AKZO was ruled by the ECJ in the year 1991, when the reform to the exclusionary conducts regime had not yet seen the light. Consequently, we cannot affirm without incurring in an e o that the Cou t s alte ati e to the as efficient competitor test (in PD) derives from the Guidance. However, we can try to connect this part of the ruling with the context of the case, and the rest of the aspects we have outlined before, to underscore traits of the ruling that make sense or could be in line with the purpose, objectives, and contents of the Guidance. We will hereon focus our attention on possible outstanding points of the ruling that could be related to the topic we are developing. Then, we will try to establish a connection with those points and the Guidance of the Commission. This connection, as we said, does not mean that we are assuring that the outcomes of the ruling are a direct result of the Guidance, but at least, that they do make sense or are in line with it, without prejudice about the possibility the Commission or the Guidance may have influenced the decision to a certain extent. Let us take a look at the comments about the ruling made by Nicolas Petit. According to him the most remarkable aspects of the ruling are the following: The judgment dissipates the uncertainty generated by Compagnie Maritime Belge in clarifying that selective price cuts are presumably legal when prices are above average incremental costs. But this is not all. The Court makes very explicit –and this is right in my opinion– that dominant firms can compete on the merits even if this forces rivals off the market (§22). In so doing, it recognizes that not all foreclosure is u la ful, ut o l that anticompetitive foreclosure atte s u de A ti le TFEU. Last, ut ot least, the judg e t upholds the u a ed A ti le 309 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para. 44. In paragraph 72, in fine, of the AKZO ruling, when referring to pricing below AVC, the Court stated: Such prices can drive from the market undertakings which are perhaps as efficient as the dominant undertaking but which, because of their smaller financial resources, are incapable of withstanding the competition waged against them. Case C-62/86, AKZO Chemie BV v. Commission [1991] ECR I-3359. 310 69 §42).311 TFEU defe se that the Co issio had plugged i § of its Guidance Paper (see About his first finding, although selective price cutting is not related to our topic, it is relevant when it takes the form of predation, as set forth by the Guidance in paragraph 72: It may be easier for the dominant undertaking to engage in predatory conduct if it selectively targets specific customers with low prices, as this will limit the losses incurred by the dominant undertaking. 312 We will not elaborate too much about this point, but it is important to notice that the Court is somehow aligning the standard of proof of selective price cutting with the standard applied to predatory pricing when prices are between AVC (in this case AIC) and ATC. This seems to be the case because the ruling eliminates any presumption of illegality when prices are between these cost levels. As it has been the rule applied to predatory pricing since AKZO, prices between these two cost benchmarks require an additional proof of intention to be deemed as unlawful. The Court, similarly, but in the selective price cutting case of PD, leaves in any case the burden of proof to the Commission (in this case to the Danish NCA) requiring it to prove the effects of the possible exclusionary conduct. In a simple way, in selective price cutting the fact that prices are between AIC and ATC makes no difference, insofar as no proof of anticompetitive effects is brought to the case. We find this to be very clear in paragraph 44 of the ruling, which expresses the answer of the ECJ to the Danish Court: …a poli hi h a do i a t u de taki g ha ges lo p i es to e tai ajo usto e s of a competitor may not be considered to amount to an exclusionary abuse merely because the price that undertaking charges one of those customers is lower than the average total costs attributed to the activity concerned, but higher than the average incremental costs pertaining to that activity... In order to assess the existence of anti-competitive effects in circumstances such as those of that case, it is necessary to consider whether that pricing policy, without objective justification, produces an actual or likely exclusionary effect, to the detriment of competition a d, there , of o su ers’ i terests. e phasis added 313 In relation to predation, when we said that the Court aligned the standard of proof for selective price cutting, is due to the fact that in predatory pricing there is no presumption of unlawfulness when prices are between these cost levels (AVC and ATC), as opposed to the case when prices are below AVC. However, in PD the Court goes even further and makes a substantive difference in regards to predation, in comparison to selective price cutting. Whilst in predatory pricing the Commission would have to 311 (2012). The website s pu li atio used as a source is a blog that seems to be duly prepared and published by the referred author (N. Petit). 312 Guidance (2009), para. 72. 313 In view of the particularities of the case, namely that the Danish Court made clear that no evidence of a strategy to pre-empt the market was found, the remark made by Rosenblatt et al makes sense: …the judg e t depa ts from a central need to show anticompetitive intent and underlines the greater importance of assessing the real isks of fo e losi g a as-effi ie t o petito . While this eadi g is i li e ith u e t Co issio p a ti e, this ele e t of the Cou t s judg e t a e due la gel to the o di g of the p eli i a uestio itself, hi h explicitly ruled out the presence of an anti- o petiti e st ateg . (2012). 70 prove that there is an intention to pre-empt the market, 314 in PD the Court required proof of the exclusionary effect, harming competition and, thereby the interests of consumers. We do not believe that the Court made a general statement applicable to all price-related exclusionary abuses, because in paragraph 44 of the ruling the Court clarified that its analysis pertains to cases where the dominant firm charges low prices to certain major customers of a competitor, a trait of selective price cutting abusive conducts. Additionally, even though the Court made several references to AKZO,315 in paragraph 27 of PD it clarified that the issue (probably referring specifically to the cost assessment) in the AKZO test was to determine whether an undertaking had practiced predatory pricing.316 Therefore, we believe the Court was careful not to be interpreted as changing the AKZO rule for predatory pricing conducts or to create a general rule applicable to all price-related exclusionary conducts, even though it seemed to render true importance to AKZO when deciding in PD. 317 Ho e e , this does t ea that i a future case of predatory pricing, due to the similarities of the conducts, and –what is more- of their possible effects, the Court could adopt the same approach taken in PD. When the Court, in paragraph 44, requires the proof of actual or likely exclusionary effect, to the det i e t of o petitio a d, the e , of o su e s i te ests, e fi d a li k ith the Guida e i reference to the concept of anticompetitive foreclosure, namely with the requirement of the ECJ to establish the anticompetitive effects of the conduct, which would indeed be such, if they have the effect of harming consumers. This is losel elated to Petit s remaining remarks, i.e., the ones about the anticompetitive foreclosure test and objective justifications and efficiencies, which are more generic and we believe they may be applicable to predatory pricing. When the Court held that not all foreclosure is unlawful, it did so by providing meaning to the term competition on the merits. In paragraph 22 of PD the ECJ stated that competing on the merits could lead to the departure from the market or the marginalization of competitors that are less efficient and therefore less attractive to consumers in regard to prices, quality, choice, etc. 318 In this way, it recognized that not all foreclosure implies consumer harm, and as a result, that not all foreclosure is anticompetitive. At this point we believe it is fair to recognize a remarkable convergence between the ECJ s uli g a d the Guida e, pa ti ula l , agai ith the o ept of anticompetitive foreclosure, and with paragraph 6 of the Guidance, in what refers to the priorities of the Commission: the Commission is 314 According to the Court in AKZO, Tetra Pak II, and France Télécom. See Chapter I, when we analyzed the costs assessment of the European Courts, specifically the analysis of Osterud s remarks about the proof of intention when pricing between AVC and ATC. 315 Inter alia, in paragraphs 24, 25, 27, 28, and 31. 316 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para. 27. 317 A out the Cou t s e tio of AK)O i the PD uli g , i the same sense Rosenblatt, Armengod and Scordamaglia-Tousis (Rosenblatt et al) hold that: In determining whether an undertaking has practiced predatory pricing, the ECJ, in a remarkably short judgment, reiterated the AKZO rule that prices below the AVC are in principle a usi e a d p i es elo ATC ut a o e AVC a e a usi e if the a e pa t of a pla fo eli i ati g a o petito . The Court thus confirmed that the lawfulness of a low-price policy depends on a cost-based analysis as well as on the undertaking s st ateg . (2012). 318 According to Rouseva and Marquis, in PD the ECJ p o ided a othe useful la ifi atio , atte pti g fo the fi st ti e to gi e o te t to the oto iousl slippe te o petitio o the e its , (2012). 71 mindful that what really matters is protecting an effective competitive process and not simply protecting competitors.319 At the same time the Court, with a logical link with the last remark, and with particular reference to price-based exclusionary conduct, recognized in PD that the enforcement priorities should focus on those conducts with exclusionary effects on those competitors which are as efficient as the incumbent. This follows the idea of the Guidance according to which the Commission will normally only intervene where the conduct concerned has already been or is capable of hampering competition from competitors which are considered to be as efficient as the dominant undertaking. 320 The ruling shows a clear statement in this sense, expressing that Article 102 prohibits the incumbent from, among other things, adopting pricing practices that have an exclusionary effect on competitors considered to be as efficient as it is itself and strengthening its dominant position by using methods other than those that are part of competition on the merits. 321 Finally, but not less importantly, as we already noticed and so does Petit, the ruling is relevant and helpful in clarifying what can be considered objective justifications and efficiencies. Also, the ruling makes clear, with regard to efficiency claims that the burden of proof lies on the dominant undertaking. As Petit also points out, there is a resemblance between the ruling (paragraphs 42 and 43) and paragraph 30 of the Guidance. 322 More precisely, paragraph 42 of the ruling indeed endorses almost completely the text of paragraph 30 of the Guidance: except for the absence of the examples and clarifications provided by the Guidance, the text of the ruling is basically an exact copy of the requirements set forth by the Guidance. In application to predatory pricing, it is important to highlight that the Guidance holds that in general it is considered unlikely that predatory conduct will create efficiencies, 323 although the Commission will consider efficiency claims if the conditions of paragraph 30 are met. It follows from the introduction made by the Court in paragraph 41 to the subject of objective justifications and efficiency claims, that the requirements of paragraph 42 are of general application to price-related abusive conducts. We believe that in this regard the judgment brings about a safer harbor both to the Commission and to an 319 Guidance (2009), para. 6. Furthermore, for Rouseva and Marquis the ruling goes even further than a mere coincidental approach between the Court and the Commission, creating a binding criteria when dealing with the enforcement priorities for exclusionary conducts: By the same token, the judgment ratifies the principle that, while foreclosure can certainly be harmful to consumers, it need not be. Since consumer harm is not an inevitable consequence of competitor exclusion, it should be held analytically distinct. Furthermore, and crucially, as a result of the judgment in Post Danmark, this basis for the application of Article 102 in exclusionary conduct cases is not e el a ite io dete i i g the e fo e e t p io ities of the Co issio ; it is i di g la . (2012). 320 Guidance (2009), para. 23. Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para. 25. 322 As the analysis of objective justifications and efficiencies lies outside the scope of the thesis, we limit ourselves to underscore the resemblances between PD and the Guidance. The reader can however consult a more o p ehe si e stud of the su je ts i ‘ouse a a d Ma uis s a ti le: http://jeclap.oxfordjournals.org/content/early/2012/10/25/jeclap.lps059.full?keytype=ref&ijkey=dN3TqBTAFFbve qz#content-block. 323 Guidance (2009), para. 74. 321 72 incumbent firm when it comes to assess and afterwards approve or deny an efficiency claim (for the Commission) and to formulate such a claim (for the firm). Such benefit would apply also to predatory pricing, because as we noticed, the Guidance does not eliminate the possibility for the creation of efficiencies through predatory conducts. In the same sense of paragraph 74 of the Guidance, we believe the statement of Rosenblatt et al about PD to be true: …the ase does e og ize that p i i g st ategies a e o pli ated a d ust e assessed i detail efo e ei g o de ed. 324 Thus, despite being unlikely for predation to create efficiencies (text of paragraph 74), the door for the assessment of a claim to this end is open and even clarified by PD, provided that the requirements of paragraph 30 of the Guidance are fulfilled. Of direct application to predatory pricing, we relate the last lines with the remarks made by Wurmnest when interpreting the France Télécom ruling. We learnt that according to him the Court showed an open door for objective justifications when prices are below AVC, in his own words: by showing a legitimate reason why its below cost pricing campaign is a rational business strategy. PD may confirm such a conclusion by expressly recognizing the possibility for objective justifications and efficiencies. 6. Final Comments Our intention in this chapter was not limited to construct a descriptive analysis of the Guidance, but to additionally relate the Guidance to practical examples in which it has been, or may be, put in action in the future. Ultimately, we wanted to know if the contents of the Guidance provide enough tools for a future inclusion of a recoupment test in predatory pricing enforcement, in line with the objectives of EU Competition Law. This exercise had to take into account the EU Courts and its case law, which provides i po ta t tools to ou stud , a o g the , e ide e of the Cou ts a epta e o eje tio of the Guidance and its approach. Our findings are significant. The limited scope of the Guidance includes predatory pricing among the enforcement priorities of the Commission. This means that the Commission, following the EU case law decided to adopt an effectsbased approach when dealing with predation. Additionally, this means that in line with its consumer welfare objective, the Commission identified predatory pricing as a conduct whose effects are, indeed, consumer harming. The Commission and EU Competition Law academics have identified a gradual adoption of an effects based approach by the EU Courts, when ruling on cases of exclusionary conduct. In connection to predation, Wurmnest identifies two remarkable situations related to France Télécom: i) A possible opening for economic justifications against the presumption of unlawfulness when pricing below AVC, and ii) Opportunities for a future assessment of recoupment based on the structural characteristics of the market. Both possibilities may also find support in the Guidance, which is a sign of convergence between it and the European case law. The ruling in Post Danmark also provides clear evidence of a gradual association between the approach of the Court and the approach of the Guidance. Moreover, the ECJ shows in the ruling an embracement 324 (2012). 73 of major components of the Guidance, such as consumer welfare as a main objective, the anticompetitive foreclosure assessment, the concept of competition on the merits, which recognizes the possibility of departure of less efficient competitors from the market, and the objective justifications and efficiency claims. Even though the case is not directly related to predatory pricing and evidently not to recoupment, the ruling will probably have repercussions on future cases of predation. All the more, the adoption of the Court of an effects-based approach when dealing with prices between AVC and ATC is absolutely remarkable, if it turns out to be also applicable to cases of predatory pricing. This will have to be determined by the Court in future cases, as it is not yet even clear if this extract of the ruling will apply to all cases of selective price cutting. As far as our reading of the PD ruling concerns, the Court has not made a general statement in this regard, and as a consequence, it has to be clarified in future cases. As to the recoupment factor, both the Commission and the text of the Guidance show possibilities of a future adoption of a recoupment test in predation. However, case law has not yet shown signs of the adoption of recoupment as part of the test for predation, although there is recognition of its importance (France Télécom) in different terms. The initiative of the Commission in France Télécom and the text of the Guidance thereafter, in conjunction may provide for a future recoupment test in cases of predation. However, if this were to happen, we envisage the creation of a European recipe based on the structure of the market and, most likely, different from the dangerous possibility of recoupment of the Brooke Group ruling. 74 Chapter III: Consumer welfare as an objective of EU Competition Law and Advocate Ge e al Mazák s Opinion about recoupment As shown in the title, our final chapter includes two specific and easily identified subjects which are closely related to the aims of our study. In the last chapter we learnt that according to the Guidance, in its enforcement priorities the Commission seeks enhancing consumer welfare as an ultimate goal. As Mosso pointed out, the scope of the Guidance is focused on those conducts which are more harmful to consumers. Additionally, when studying the concept of anticompetitive foreclosure, we learnt that consumer welfare is at its core. Accordingly, the effects-based approach that characterizes the reform to Article 102 TFEU also seeks the enhancement of consumer welfare. Based on these findings we argued that there is a link between the letter of the Guidance and consumer welfare. Considering the text of the Guidance, its effects-based approach could have been adopted with the aim of enhancing consumer welfare. Likewise, it is likely that in future cases of predatory pricing the Commission will adhere itself to the Guidance and to the stipulations, standards, and other applicable considerations set forth therein, which should in principle follow the consumer welfare goal. With this, despite the fact that in practice the effects-based approach could have not been fully embraced yet, it is likely that the Guidance will allow further gradual change in the Commissio s assess e t of exclusionary abuses, which could allow the inclusion of recoupment in the test for predation. This could be possible at least in what refers to the limited scope of the Guidance, which includes predatory pricing. All the same, considering that the purpose of our study refers to a potential change both in the Co issio s a d the EU Cou ts assess e t of p edato p i i g, e elie e that fu the a al sis is required i ega d to the Judi ia s e d. Without departing from the scope of the thesis we will consider in the first section the definition of consumer welfare and the possibility of considering it as the ultimate objective of Article 102 TFEU. Because this analysis may perfectly be the single subject for another thesis, we will focus our attention on the main conclusions of consulted EU Competition Law literature. Our aim is limited to determining if a consumer welfare standard, in regards to predation, has a valid role to play under EU case law. In a second and final section we ill a al ze AG Mazák s Opi io ega di g recoupment. The AG s Opinion is fundamental for our study not only due to its dissonant status with the current predation test, but because it was in principle prepared arguing in favor of an effects-based approach325 and thus, 325 E.g., such is the implicit reading of the attorneys at Van Bael & Bellis in their comments on the Opinion: The Ad o ate Ge e al s opi io is pa ti ula l i te esti g i sofa as it o luded that it is e essa u de A ti le EC to de o st ate the possi ilit of e oup e t i a p edato p i i g ase. “u h a dete i atio … ould help ensure greater consistency on this issue with other jurisdictions, including the U.S. and UK. In addition, requiring a plausible theory of recoupment would also appear to be more consistent with an effects-based approach to Article 82 EC (although, interestingly, the Commission remains opposed to the use of such an analysis). Van Bael & Bellis (2008), p. 11. Additionally, we have previously mentioned that inter alia, the AG supported his Opinion about 75 it is supposed to be in line with the criteria and contents that inspired the reform to Article 102 TFEU. As we indicated in the introduction of the thesis, the Opinion, alongside with the reform to the application of Article 102, stands up as a sign of a potential change in the predation test that includes recoupment as an additional requirement. An analysis of the Opinion implies a study of the whether its view of recoupment is consistent with EU Competition law and therefore of the degree of relevance and i flue e that Mazak s positio a ha e i futu e ases. I so doi g, e ill des i e the Opi io i what is connected to recoupment and also the basis chosen by the AG when reaching his conclusions. We opted for the inclusion of these two resting subjects in one chapter because both of them are closely related to the judicial phase of the application of EU Competition Law, in contrast with the Co issio s e fo e e t p io ities the Guida e , hi h efe hiefl to the Co issio s tasks and actions in an earlier procedural stage. Moreover, while by analyzing the consumer welfare objective we pretend to understand the role of such an aim with regard to the application of Article 102 in cases or p edatio , he deali g ith the AG s Opi ion we will consider an individual case where the Court was faced with the challenge of changing its current assessment, a proposition which was supposedly based on a consumer welfare standard. Therefore, from a general to specific perspective, the chapter will cover two subjects closely related to our study in connection to the judicial review of predation cases. 1. Consumer welfare as the main objective when enforcing Article 102 against predatory pricing In the first section we will provide the reader with a definition of consumer welfare, which we consider to be merely a clarifying tool or an academic duty, as we have been using the term without defining it yet. Due to the limited scope of our study, we will restrict ourselves to outline a concept of consumer welfare as it is understood under EU Competition Law. We will avoid questioning the substance or accuracy of the definition or comparing it to other standards such as total welfare or producer welfare,326 an analysis which lies outside our study. What really matters in this section is examining the validity of consumer welfare as a goal when enforcing Article 102 TFEU, in cases of predation. Thus, if consumer welfare, as positioned by the Commission in the Guidance is indeed a valid goal under EU Competition Law, then we could conclude that an embracement by the EU Judiciary of an effects-based approach focused on identifying actual or likely consumer harm, will be more likely. By doing this, we can identify how close or how ready the Commission and the Courts could be to including recoupment as a part of the assessment for predation, insofar as doing it would be consistent with the consumer welfare goal. To this end we will adhere to the findings of Liza Lovdahl Gormsen in her contribution titled: A Principled Approach to Abuse of Dominance in European Competition Law. A substantial part of her book is dedicated to ascertaining the place and importance of consumer welfare as an objective of EU recoupment on the EAGCP Report, which argues in favor of an effects-based approach when applying Article 102 TFEU, as we studied in Chapter II. 326 Total elfa e is the su of o su e elfa e a d p odu e elfa e . P odu e elfa e u de stood as p odu e su plus efe s to the su of all p ofits ade by producers in an industry. Akman (2009), p. 2. 76 Competition Law in the specific field of abuse of dominance, and most importantly, she addresses the subject of whether or not recoupment is required when a consumer welfare objective is pursued. 1.1. Definition of consumer welfare In economics, consumer welfare is often defined as consumer surplus. Consumer surplus is the difference between the amount the consumer would be willing to pay for the product and the actual price of the product. 327 Accordingly, consumer welfare relates with the deadweight loss to society, which is the loss to consumers who would have purchased the product (at a price higher than the marginal cost) but have switched to a less desired substitute as a result of monopolistic pricing, with the loss to consumers who have continued to purchase the product at the monopolistic price, and with the loss of consumers who decided not to purchase the product at all. 328 When identified with allocative efficiency, consumer welfare is understood as a more comprehensive a d oade te hi h efe s to the elfa e of all o su e s i so iet : To othe s, i ludi g many f o the Chi ago “ hool, o su e elfa e is a u h oade o ept. The elie e the a tit ust la s should e applied i a a that a i izes so iet s ealth as a hole –or to use their language, that p ote ts allo ati e effi ie .329 On the other hand, allocative inefficiency exists, or which is the same, allocative efficiency is harmed when producers agree directly or indirectly to increase price and reduce output. This is the case when dealing with cases of infringement of Article 101 TFEU. As for Article 102 TFEU, the inefficiency would be produced when producers unilaterally increase prices and reduce p odu tio . Producers engaging in such conduct expect to obtain more surplus from consumers than would be possible if competition prevailed, as the shift su plus f o o su e s to the sel es. 330 As a matter of fact, as pointed out by Thomas Rosch, a debate over the differences of content between the te s o su e elfa e a d allo ati e effi ie a e atego ized as e el a ade i , with no real world impact because there is very little difference between the two standards. 331 Put a little bit differently but showing the intrinsic correlation between the two concepts, Gormsen indicates that for 327 The Netherlands Competition Authority (NMa) (2012), p. 2, Akman (2009), p. 2. A practical example is provided in the following definition: Consumer surplus is an economic measure of consumer satisfaction, which is calculated by analyzing the difference between what consumers are willing to pay for a good or service relative to its market price. A consumer surplus occurs when the consumer is willing to pay more for a given product than the current market price. Consumers always like to feel like they are getting a good deal on the goods and services they buy and consumer surplus is simply an economic measure of this satisfaction. For example, assume a consumer goes out shopping for a CD player and he or she is willing to spend $250. When this individual finds that the player is on sale for $150, economists would say that this person has a consumer surplus of $100. http://www.investopedia.com/terms/c/consumer_surplus.asp#axzz2HbD1qMiK, consulted on January 10, 2013. 328 Dayagi-Epstein (2009), p. 75. Rosch (2006), p. 2. According to Jones and Sufrin (2011), allocative efficiency results from the fact that goods are produced i the ua tities alued so iet … is a state i hi h o e of the pla e s, selle s o u e s, ould be made better off without someone being made worse off. p. 8. 330 Gormsen (2010), p. 24. 331 Rosch (2006), p. 2. 329 77 systems pursuing a consumer welfare standard, reductions in allocative efficiency are unacceptable, because consumers suffer. 332 Apparently following such intrinsic correlation, according to DayaguiEpstei , that is ho the Co issio see s to ha e u de stood o su e elfa e i the Guida e, namely, ith o disti tio o as e ui ale t to allo ati e effi ie . 333 His conclusion derives from comparing the text of paragraph 4 of the Discussion Paper (DP),334 which separates the terms in what refers to the objectives towards exclusionary abuses, with paragraph 19 of the Guidance, which excludes said distinction. To illustrate his comparison, in relevant part, paragraph 4 of the DP reads: …with regard to exclusionary abuses the objective of Article 82 is the protection of competition in the market as a means of enhancing consumer welfare and ensuring an efficient allocation of resources. While the text of paragraph 19 of the Guidance states the following: The ai of the Co issio s e fo e e t a ti it i elatio to e lusio a o du t is to e su e that dominant undertakings do not impair effective competition by foreclosing their competitors in an anti- o petiti e a , thus ha i g a ad e se i pa t o o su e elfa e… 335 To put it in practical terms, although the Guidance does not contain a definition of consumer welfare, it does show examples of adverse and beneficial effects on consumer welfare, such as those related to prices, quality, innovation, and consumer choice. 336 A similar stance has been taken by the Dutch Competition Authority (NMa), which assures that we may interpret consumer welfare more broadly, to include quality, range and service, all of which are of course, related to the price. 337 Indeed, the Guidance indicates in paragraph 5 that consumers benefit from competition through lower prices, better quality and a wider choice of new or improved goods and services. Then, as regard to the possible adverse effects of anticompetitive exclusionary abuses, in paragraph 11 the Guidance reads: I this Co u i atio , the e p essio i ease p i es includes the power to maintain prices above the competitive level and is used as shorthand for the various ways in which the parameters of competition –such as prices, output, innovation, the variety or quality of goods or services- can be influenced to the advantage of the dominant undertaking and to the detriment of consumers.338 On the other hand, we must outline the way in which the EU Courts understand consumer welfare, so that we can identify whether or not it corresponds to the conceptualization of the Commission. In Österreichische Postsparkasse AG, the GC held that the ultimate purpose of the rules that seek to ensure 332 Gormsen (2010), p. 27. Dayagi-Espstein (2009), p. 76. 334 DP (2005), para 4. 335 Guidance (2009), para. 19. 336 Dayagi-Epstein (2009), p. 77. 337 NMa (2012), p. 2. 338 Guidance (2009), para. 11. 333 78 that competition is not distorted in the Internal Market is to increase the well-being of consumers.339 Thus, although the term consumer welfare was not used, we may interpret that consumer welfare was identified with the well-being of consumers. The Court also seems to have identified the interests of consumers as the economic interests of final customers (the purchasers of goods and services): Competition law and competition policy therefore have an undeniable impact on the specific economic interests of final customers who purchase goods or services. 340 Then, in GlaxoSmithKline the GC basically repeated itself by indicating that the objective of Article 81(1) of the EC Treaty (now Article 101(1) TFEU) is to prevent undertakings, by restricting competition between themselves or with third parties, from reducing the welfare of the final consumer of the products in question.341 Even though the reasoning of the GC was overturned by the ECJ, the latter did not depart form the apparent meaning given by the GC to consumer welfare. In that case, the GC identified preventing the reduction of welfare of final consumers as an objective of Competition Law. What the ECJ clarified in its ruling is that Article 101 aims to protect not only such interests (of consumers) but also the structure of the market, and competition as such.342 From the above mentioned rulings we may conclude that, at least as to what refers to the understanding of consumer welfare, the Commission and the Courts have very similar approaches. Going back to Post Danmark, just as it is written in paragraphs 5 and 11 of the Guidance, the Court seems to identify in a similar way the Commission does, the ways in which consumers can benefit from competition, i.e., through price, choice, quality, and innovation. As we indicated in Chapter II, in paragraph 22 of PD the Court stated that: Competition on the merits may, by definition, lead to the departure from the market or the marginalization of competitors that are less efficient and so less attractive to consumers from the point of view of, among other things, price, choice, quality or innovation (emphasis added).343 When it comes to the Guida e, as e studied i the p e ious hapte , the o ept of o su e s a appea ith a su tle diffe e e i o pa iso to the Cou ts ase la , o e spe ifi all , to the GC. I both of the two first mentioned cases (Österreichische Postsparkasse and GlaxoSmithKline), the General Court gave importance to the interests of final consumers, whereas, in the Guidance the concept is extended to intermediate producers and distributors: The o ept of o su e s e o passes all di e t or indirect users of the products affected by the conduct, including intermediate producers that use the products as an input, as well as distributors and final consumers both of the immediate product and of Joined cases T-213/01 and T-214/01 Österreichische Postspa kasse AG a d Ba k fϋ A eit u d Wi ts haft AG . EC Commission [2006] ECR II-1601, para. 115. 340 Joined cases T-213/01 and T-214/01 Öste ei his he Postspa kasse AG a d Ba k fϋ A eit u d Wi ts haft AG . EC Commission [2006] ECR II-1601, para. 115. 341 Case T-168/01 GlaxoSmithKline v. Commission [2006] ECR II-2969, p. 118. 342 Joined cases C-501/06 P et al, GlaxoSmithKline Services and Others v Commission and Others [2009], p. 62 and 63. 343 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012], para. 22. 339 79 products provided by intermediate producers. 344 However, a final clarification of the Guidance, appli a le u de e tai i u sta es, to ou ie happe s to o e lose to the Ge e al Cou t s concept of consumer: Where intermediate users are actual or potential competitors of the dominant undertaking, the assessment focuses on the effects of the conduct on users further downstream. 345 Although this clarification does not necessarily entail that in all cases where the intermediate users compete with the incumbent firm the users in the downstream market will be final consumers, at least the Guidance makes sure to lower the degree of importance when it refers to the interests of competitors. Clearly using a consumer welfare standard that focuses on final consumers, Dayagui Epstein argues that the Commission should have gone even further by accepting that when intermediate customers are not competitors of the dominant undertaking, their interests do not necessarily coincide with those of the final consumers. 346 All the same, we believe that the clarification of the Guidance goes in line with the rest of its text, including the concepts of anticompetitive foreclosure, the as-efficient competitor test, and competition on the merits, which envisages the elimination of less efficient competitors from the market, as possible. Such was indeed the reasoning of the ECJ, as studied, in the Post Danmark ruling. Before we move on to the next section we must clarify that, although there are some discussions as to the meaning of consumer welfare, 347 it seems to be enough clarity and convergence to this end, at least between the Commission and the Courts. As we will analyze in the next section, what really causes a debate is determining whether or not consumer welfare takes a prominent stance in relation to other objectives of EU Competition Law. For the purposes of our study, we will narrow such debate to an analysis of the place of consumer welfare as an objective of Competition Law in comparison to other objectives -such as economic freedom and market integration- when dealing with predatory pricing. As we will see, this is related to the theory of harm that must be used if the consumer welfare objective is applied, and therefore, also related to whether or not recoupment should be part of the assessment of predatory pricing. 1.2. Is consumer welfare the objective when dealing with predation? As we indicated supra, the ECJ in GlaxoSmithKline (GSK) i pli itl e og ized o su e elfa e as a valid objective under EU Competition Law. However, it also ruled in favor of the attainment of other objectives, in that particular case, the Court pointed out the structure of the market and competition as such, as a means to foster the Internal Market or market integration in the EU. 344 Guidance (2009), para. 19, fn 2. Guidance (2009), para. 19, fn 2. 346 Dayagui-Epstein (2009), p. 81. 347 Additionally, in this line Thomas Rosch indicates in reference to US Antitrust Law discussions: Courts and federal law enforcement officials routinely invoke consumer welfare as the guiding principle behind their application of the antitrust laws. Yet there is continuing debate over what consumer welfare means. (2006), p. 1. Then, Pinar Akman argues: The la k of o se sus o hat is ea t o su e elfa e a d the fa t that the EC Co issio does ot efe to o su e i the se se of fi al-user which is the usual understanding of the term in consumer law casts further doubt on its genuineness. (2009), p. 3. 345 80 In GSK, the General Court considered that an agreement restricting parallel trade by object was not illegal, insofar as it did not harm final consumers. The case involved a distribution contract entered into by Glaxo Wellcome SA (GW), a subsidiary of GSK, with Spanish wholesalers. The agreement made a distinction between prices charged to wholesalers in the case of domestic resale of reimbursable drugs to pharmacies or hospitals, and higher prices charged in the case of exports of medicines to any other Member States.348 Due to the special characteristics of the health sector, the GC considered that the higher prices which restricted parallel trade would not affect final consumers, as the prices of the medicines in question are shielded from the free play of supply and demand, owing to the applicable regulations and are set or controlled by the public authorities.349 Even though after appeal the holding of the GC remained unchanged, the ECJ overturned the reasoning used by the GC. The ECJ held that an agreement which restricts by object parallel trade, tending to restore the national divisions i t ade et ee Me e “tates ight e su h as to f ust ate the T eat s objective of achieving the integration of national markets through the establishment of a single market. Thus, the ECJ alled the Ge e al Cou t s atte tio a d stated that there is nothing in that provision (Article 101(1) TFEU) to indicate that only those agreements which deprive consumers of certain advantages may have an anti- o petiti e o je t… A ti le EC (101 TFEU) aims to protect not only the interests of competitors or of consumers, but also the structure of the market and, in doing so, competition as such.350 Although we may say that following the GSK uli g o su e elfa e a e ide tified as a alid objective of EU Competition Law, a problem arises when this objective is placed alongside other objectives which may also be considered important, or more important, by the EU Courts. GSK is a vivid example of this conundrum. In reference to this particular case, Akman considers that the ECJ, when confronted with the choice et ee o su e elfa e a d a ket i teg atio , p efe ed the latte over the former: Thus, the ECJ is clearer than it has ever been before: there is much more to EC competition law than consumer welfare and when the rest – like the single market imperative – clashes with consumer welfare, it is consumer welfare that gives way. 351 Even though the Court has expressed that Articles 101 and 102 seek to achieve the same aim in different levels,352 and therefore the situation would well apply and could be configured in cases of unilateral behavior, the example provided (GSK) refers to a case of agreements between undertakings, thus, it refers to an infringement of Article 101(1) TFEU. We ill o fo us ou atte tio o Go se s stud , which is connected to cases of abusive unilateral behavior, hence to Article 102 TFEU. 348 Joined cases C-501/06 et al, GlaxoSmithKline Services Unlimited v. Commission [2009], paras. 2 and 6. Joined cases C-501/06 et al, GlaxoSmithKline Services Unlimited v. Commission [2009], paras. 44-46. 350 Joined cases C-501/06 et al, GlaxoSmithKline Services Unlimited v. Commission [2009], paras. 62 and 63. 351 Akman (2009). 352 Therefore, we consider that even in a case that deals with Art. 102, the Court may perfectly rule in the same way as it did in GlaxoSmithKline. That is why we provided this practical example when introducing the subject. Case 6-72, Europemballage Corporation and Continental Company Inc. v Commission [1973], para. 25. 349 81 Gormsen identifies various objectives pursued by Article 102 TFEU. To start with the already studied case in which two objectives are considered, namely consumer welfare and market integration, as the Court did in GSK, different approaches could be taken depending on the objective which is at stake. Accordingly, if the o je ti e is a ket i teg atio , o du t eati g a a ie s to t ade that the ules on free movement are trying to dismantle cannot be allowed. If however the objective is consumer elfa e, o l o du t hi h ha s o su e s ust e p e e ted. 353 We believe that in GSK the Court did not prefer one objective over the other, but protected the objective that was at stake at the material ti e. I this se se e depa t f o Ak a s easo i g he he e tio s a lash of o je ti es, hi h e believe did not to happen, at least in that particular case. However, we believe he is right when he argues that EU Competition Law embraces much more than consumer welfare, i.e., there are other objectives besides consumer welfare. Then, besides the overarching aim of u disto ted o petitio found in Protocol 27 on the Internal Market and Competition,354 applicable in general to the Competition Law provisions under EU law, Gormsen points out economic freedom as another objective of Competition Law, in addition to the mentioned objectives of consumer welfare and market integration. 355 For the purposes of our study, what really matters here is to notice the coexistence of other objectives –alongside with consumer welfare- he deali g ith ases of u ilate al a usi e eha io . Go se s ie is sha ed othe scholars. For example, Ernst-Joa hi Mest ä ke iti izes the Co issio s positio i the Guidance, which gives more importance to consumer welfare and apparently forgets the importance of economic freedom, of great relevance under the German Competition Law tradition. 356 Then, Athanassios Skourtis, considers that the discussion of the interplay of various goals under EU Law is not new. He mentions that despite a recent shift towards economization and an apparent prevalence of the consumer welfare standard, EU Competition Law is also perceived as a means to achieve market integration. 357 For Akman … o sumer welfare is neither the starting point nor the overriding interest in European competition analysis like it is in American antitrust law. He continues saying that [t]he EC Commission puts emphasis 353 Gormsen (2010), p. 64. The provision of the protocol reads: The High Contracting Parties, considering that the internal market as set out in Article 3 of the Treaty on European Union includes a system ensuring that competition is not distorted, have agreed that: To this end, the Union shall, if necessary, take action under the provisions of the Treaties, including u de A ti le TFEU… According to Gormsen, [t]he main and overarching objective in Article 82 is undistorted competition. This is important if there is a need for balancing different conflicting objectives as undistorted competition will be the benchmark against which to balance. Undistorted competition, as compared to market integration, must be the overarching objective, as market integration is a means to an end of other objectives. (2010), p. 83. 355 Gormsen (2010), p. 59 et seq and 82. 356 Mestmäcker (2011), p. 25 et seq. The autho s positio is i ou ie adi al i hat efe s to the ala i g of importance between the two objectives. For example, he considers that the as efficient competitor test, as it encompasses the possibility of less efficient competitors to leave the market, runs counter fundamental EU law rights. His conclusion is that the Guidance, in this respect, is inapplicable under German Law: In the attempt to protect competition and not competitors the Commission ends up with the protection of market-dominant u de taki gs agai st i effi ie t o petitio . To sum up: the as-efficient-competitor test is incompatible with the protection of residual competition in markets of dominant undertakings and it violates fundamental community law rights. The test will probably not be adopted in German law. (2011), p. 51. 357 Skourtis (2012), p. 3. 354 82 o o side atio s othe tha as possible.358 o su e elfa e , like the openness of the market with as many players As Mestmäcker does, Gormsen identifies a potential conflict between consumer welfare and freedom of competition when assessing the conducts prohibited under Article 102 TFEU. 359 This means that, depending on the objective preferred either by the Commission or the Courts, the respective decisions and rulings could vary.360 I Go se s o ds: In the case of a conflict between economic freedom and consumer welfare, the Community Courts must make a choice as to whether they favour economic freedom at the expense of consumer welfare or vice versa. 361 Our view is that the situation posed by Gormsen could extend, at least in practical terms, to cases where other objectives, and not only economic freedom, could be regarded as equally or more important by the Court. This situation allows the consequence in which even if the other objectives are in the same stance as consumer welfare, a ruling of the Court could have a different outcome, compared to a situation in which consumer welfare (only) is the preferred standard. That is the case in the already discussed example of GSK: even if we consider that there was not a clash of objectives, if the ECJ would have followed only a consumer welfare standard, the ruling of the GC would have remained untouched. However, the ECJ ruled considering also market integration as an objective of EU Competition Law, in the same stance –we believe- as consumer welfare. We also believe that the situation according to which the outcomes of the cases could vary depending on the objective pursued, is indeed logical and understandable. Depending on the objective that is being followed, the effects of the harm that must be proven differs, and as a result, one single case can have different outcomes depending on what decision is taken as to WHAT or WHO to protect. Accordingly, abuse must be understood as harm to something or someone and it can mean harm to competition, harm to competitors, harm to consumers or a combination of these to name a few.362 358 Akman (2009), p. 4. The author outlines two possible situations in which the two objectives could conflict: The first is where o petitio a o gst i als is p ote ted ut does ot e efit o su e elfa e, fo e a ple, o petito s that a e not (yet) as efficient as the dominant co pa a e p ote ted i the sho t te o less effi ie t o petito s a e p ote ted. … I so e i u sta es, it is plausi le that the p ote ted a ket pla e s ill e o e effi ie t o e time. However, it is also possible that the most productive way of supplying customers will be through fewer suppliers, particularly when economies of scale are great. Depending on which of these effects is the greatest, the consumer could gain or lose from such measures. The second potential conflict situation is where an exclusion of some small medium-sized companies, which may lack economies of scale, would not harm consumer welfare if these companies were unable to guarantee consumer welfare in the form of lower prices, better quality and an effective choice. However, it ould ha the e luded o i pai ed o pa ies e o o i f eedo as the ould o longer be able to participate in the market. (2010), p. 86. 360 To show examples of such a situation, in the following lines we will discuss about the objective followed in Tetra Pak ll, which we believe favors economic freedom, as opposed to the situation proposed by Mazák, who by adding recoupment as a requirement to the predation test, argued in favor of following a consumer welfare standard. See also the examples of the possible different standards used in Van den Bergh and British Gypsum, which we included in one of the following footnotes of this chapter. 361 (2010), p. 86. 362 Akman (2008), p. 3. 359 83 Go se s a al sis of the pote tial o fli t et ee o su e elfa e a d e o o i f eedo is eloquent and as we will discover in the following lines, of direct application to the subject of our study as well.363 Let us consider, as Gormsen does, the case of France Télécom i ega d to AG Mazák s Opinio a d the ECJ s uli g. We ill deal i o e detail ith the AG s Opi io i the follo i g se tio , but will introduce the subject as an example of a practical situation related to the potential conflict between consumer welfare and economic freedom. AG Mazák recommended the re-examination of the Ge e al Cou t s uli g, spe ifi all u de s o i g that recoupment should have been proved, as it is (according to him) required by EU case law. The ECJ did ot follo Mazák s ad i e. As applied to the potential conflict between consumer welfare and economic freedom, in that particular case, the difference of criteria was due to the objective followed, respectively, by the AG in his opinion and the ECJ in its ruling. The AG followed a consumer welfare standard, whilst the Court protected economic freedom. Whether recoupment is a requirement depends on which objective is being pursued. If the Commission (or the Court) believes that economic freedom must be protected, then it is enough to protect the structure of competition. In this case, there is no need to examine whether the dominant undertaking is able to recoup; predatory pricing is found to be abusive where there is a risk of eliminating competition.364 A similar situation would have occurred in Tetra Pak II, where economic freedom prevailed over consumer welfare, and thus, the ECJ deemed recoupment not to be required. We remind the reader of the use of the phrase i the i u sta es of the p ese t ase , by the Court to justify that recoupment was not required. According to Gormsen, those circumstances meant that following an economic freedom objective, the Court already had found a risk that a competitor was going to be eliminated, so proof of recoupment was deemed unnecessary and hence rejected: The Tetra Pak II ase does ot eje t e oup e t i ge e al, ut i the i u sta es of the p ese t ase , e ause a isk that o petito s ill e eli i ated was already found in that case. This indicates that additional proof of exploitation in the form of recoupment is not necessary where evidence shows that there is a risk that a competitor will be eliminated. 365 For Mazák, passing the test of recoupment means that the dominant undertaking engaged in predation will be able to set monopolistic prices after its competitor(s) is ousted from the market. Ergo, when 363 According to Gormsen, depending on the objective, the assessment of the conduct in question may vary. In some cases, when the aim is consumer welfare, a deeper assessment may be required: Parallel trade is not the only kind of conduct capable of restricting a source of supply; predatory pricing and rebates, for example, can have a similar effect. If the regulatory aim is freedom of competition, then it is acceptable to prohibit conduct capable of restricting a source of supply. However, if the aim is consumer welfare, it needs to be examined whether the restriction harms consumers by balancing the pro- and anticompetitive effects of the conduct. The traditional view when examining exclusionary conduct, such as rebates or predatory pricing, has been to assume abuse where there is a risk of competition being eliminated. Such a presumption does not sit well with an objective of consumer welfare, because not even where effective competition is eliminated can it automatically be assumed that consumers are harmed. (2010), p. 142. 364 Gormsen (2010), p. 142. 365 Gormsen (2010), p. 143. 84 recoupment is proved, a negative effect on consumers is also proved. For him, a mere likelihood of market pre-emption is not enough, due to the fact that consumer harm is absent: When the aim is promoting consumer welfare proof of likely recoupment would be required. This however, would exclude the Commission from automatically assuming abuse where there is a risk of elimination of competition, which is a risk that arguably already exists where a company is dominant. This is acceptable if the objective is economic freedom, which goes back to o doli e alis a d its fea of a ket po e , ut ot if the goal is o su e elfa e. Mazák s Opinion concludes that the Commission cannot rely on the presumption of abuse where there is a risk competition will be eliminated in cases where the Commission decides that consumer welfare must take priority over economic freedom.366 The last assertion has been repeated in the two previous chapters, namely that recoupment is thought to be required due to the fact that in absence of it, consumers would perceive a benefit in the form of ette p i es. I deed, i Brooke Group, the “up e e Cou t s o e as e pli itl fo the elfare of 367 o su e s i the output a ket… As we indicated in Chapter I, according to the landmark US “up e e Cou t s uli g u su essful p edatio is i ge e al a oo to o su e s. 368 In line with Go se s ite io , if the Co issio a d the EU Cou ts are going to follow a consumer welfare standard, the enforcement of exclusionary abuses must be focused on determining the harmful effects on consumers, an independent (or further) assessment from the one used to determining effects on competitors. This does not mean that harmful effects on competitors cannot produce harmful effects on consumers, but when it does, the sole actual or likely effects on competitors do not entail harmful effects on consumers, the latter must also be ascertained and not presumed: In monopolization cases, better known in Europe as abuse of a dominance cases, assessing the business conduct requires an even more finely tuned assessment of consumer harm. In these cases effects on competition and effects on competitors through exclusionary practices should be clearly distinguished from cases where consumers suffer material harm as a result of increased prices or reduced output and quality. What might be unfair vis-à-vis competitors and result in foreclosure is not necessarily anti-competitive. The controversial assessment of predatory pricing and rebate systems in the EC is well-known. The need for a more economic effects based approach and sharpened evidentiary requirement of explicit consumer harm is indispensable if enforcement agencies want to reliably differentiate between anti- and pro-competitive conduct.369 It is important however, to link this fact with a striking reality of EU Competition Law, which positions it in a situation where not only consumer welfare is an objective. Therefore, as we have seen, the EU Courts are not bound by a consumer welfare goal which will determine that the sole applicable theory of harm is the harm to consumers. Therefore, when other goals of EU Competition Law are pursued by the Courts, other theories of harm can be applied, and therefore, harm to consumers may not necessarily 366 Gormsen (2010), p. 146. Rosch (2006), p. 11. 368 Brooke Group Ltd v Brown &Williamson Tobacco Corp 509 US 209 (1993). 369 K J Cseres (2007), p. 146. 367 85 need to be proved. It could be the case that, as in predatory pricing (Tetra Pak II and France Télécom), a possible negative effect on competitors (market pre-emption) suffices for a conduct to be deemed unlawful, if economic freedom is pursued. 370 It can also be the case that, as in GSK, other objectives are protected, and therefore, even if a harmful effect on consumers does not exist, the conduct infringes EU Law, with the aim of enhancing other objectives, such as market integration. In spite of this reality, the situation does not preclude the Courts from finding a final approach, i.e., deciding that one objective should prevail over the others. As we indicated, we believe that this has not yet occurred and that in a case such as GSK, the ECJ protected the objective at stake at the material time, in the same stance as consumer welfare. If we apply this assertion to what we learnt in Chapter II in regard to the Guidance and its limited scope, we have to remember that according to Mosso, the Commission targeted those conducts in which the Courts have shown signs of moving towards an effects-based approach, and really harm consumers. Those conducts include predatory pricing. If it is true that the Courts have moved towards (or are close to) an effects-based approach when dealing with predatory pricing, that does not necessarily mean that the Courts yet embrace consumer welfare as the prevalent objective or that they embrace the subsequent consumer theory of harm when dealing with p edatio . I this espe t e elie e Go se s o ds to e t ue: …a o je ti e of e o o i f eedo is concerned with the effects on the structure of competition whereas an objective of consumer welfare is concerned with the effects on consumers. Thus, adopting a methodology of effects does not necessarily mean pursuing an objective of consumer welfare.371 In other words, although the Guidance is written in terms of consumer welfare as an ultimate goal, an effects based approach can certainly be used to ascertain harmful effects not only to consumers, but also to the structure of the market or the players in the market. So, just as the Courts have to decide whether or not to move towards and effects-based approach; that decision also applies when dealing with the objectives of EU Competition Law. The Guidance itself may have acknowledged that the application of an effects-based approach can focus on different goals when it clarified, as we indicated supra when we discussed the definition of consumers, that where intermediate users are actual or potential competitors of the dominant undertaking, the assessment focuses on the effects of the conduct on users further downstream. As opposed to the text of the Guidance, if the assessment focuses on the effects of the conduct on 370 A practical example of two different standards of harm is provided by Akman when comparing two of the General Cou t s uli gs, a el Van den Bergh and British Gypsum: In British Gypsum the conclusion of exclusivity ag ee e ts i espe t of a su sta tial p opo tio of pu hases o stituted a u a epta le o sta le to e t to that market. Similarly in Van den Bergh both the EC Commission and the CFI have considered the effects of the freezer-exclusivity agreements the dominant ice-cream manufacturer entered into with retailers. The CFI held that the exclusivity clause had the effect of preventing the retailers from selli g othe a ds of i e ea … a d of preventing competing manufacturers from gaining access to the relevant market. It is noteworthy that although in Van den Bergh mention is made of demand for the other non-dominant brands of ice-cream and thus effects on o su e s a e o side ed, i British Gypsum the effe ts efe ed to a e those o the a ket hi h i this ase relate to the possible foreclosure of the market to competitors. As such, there is no mention or assessment of effe ts o o su e s hich would be effects on the consumers of plasterboard, i.e. builders/building-owners. Akman (2009), p. 7. 371 Gormsen (2010), p. 148. 86 competitors, other goals, as economic freedom may be rendered as more important. The cited clarification is therefore a result of the decision made by the Commission as to the preferred objective of consumer welfare, which is the course that the effects-based approach should take according to the Guidance. In the terms used by Dayagui-Epstein: The recognition of the need to assess the effects of the conduct under scrutiny on the interests of the final consumers when the intermediate customers are competitors of the dominant undertaking, is another indication of the importance awarded in the Commission Guidance to the consumer welfare standard. 372 Interestingly enough, it may be that the Guidance has provided stake holders in the market with safer harbors when dealing with important business related day-to-day decisions, and thus, legal certainty has been enhanced.373 This would be at least to a certain degree due to the fact that, following its text, in the Guidance the Commission leaned towards and effects-based approach that preferred consumer welfare as a goal. At least the letter of the Guidance is clear enough in those respects, even if the Co issio s de isio s as e did i the ase of France Télécom) trigger criticisms as to their lack of clarity or conflict with its own Guidelines. On the other hand, the Courts still have a long distance to cover when dealing with the objectives of EU Competition Law and deciding upon an ultimate goal, if that could be the final outcome. We consider the situation to be no less than problematic and more legal certainty to be absolutely desirable in respect to this objectives-conundrum. As pointed out by Skourtis, [a]s EU competition law and policy has been applied against the backdrop of a multiplicity of goals for considerable time, experience may prove to be beneficial in that respect. Nevertheless, also in that regard conceptual and terminological clarity of the paradigm, sound economics and law analysis and awareness that the weighing process might bear implications on the degree of justitiability and the effectiveness of institutional design, are of the essence. 374 Indeed, there would be greater legal certainty if the Courts bring about more clarity as to the objectives of EU Competition Law, and therefore, negative effects as the ones described by Akman in the following lines can be avoided: This ambiguity leads to a lack of ex ante business certainty which may result in undertakings acting overcautiously. This may in turn lead to business refraining from entering certain p a ti es that a othe ise i ease o su e elfa e . 375 So, the efforts and intentions of the Commission must be complemented by rulings which follow a consumer welfare standard, if that will be the preferred goal in the EU, at least in what refers to the scope of the Guidance. In this line, maybe the 372 Dayagui-Epstein (2009), p. 81. In the same line, the author also identifies a potential conflict between consumer welfare and economic freedom with different outcomes depending on the goal preferred: Had this approach been applied i the ase of Co e ial “ol e ts, he e a o petito s f eedo to o pete as p efe ed o e the welfare of the final consumers, the Court might have reached a different verdict. 373 Paragraph 2 of the Guidance (2009) indicates that it is intended to provide greater clarity and predictability as regards the general framework of analysis which the Commission employs in determining whether it should pursue cases concerning various forms of exclusionary conduct and to help undertakings better assess whether certain behaviour is likely to result in intervention by the Commission under Article 102. 374 Skourtis (2012), p. 5. 375 Akman (2008), p. 3. 87 ECJ has shown signs of moving forward and preferring consumer harm as a standard. That could have been the case in Post Danmark, where we could argue that consumer welfare was preferred over economic freedom, in the terms described by Gormsen. Let us remember again that in that ruling the Court expressly accepted the possibility of less efficient competitors leaving the market, in application of the as-efficient competitor test and the anticompetitive foreclosure standards. 376 Accordingly, a positive fi di g fo ou stud s ai is that i deed, o su e elfa e is a alid o je ti e he deali g ith p i erelated exclusionary abuses, which of course includes predatory pricing. It would be for the EU Courts to decide whether or not, as the Commission did in the Guidance, consumer welfare will turn out to be the leading objective. 2. AG Mazák’s Opinion in France Télécom In this section we will analyze certain parts of AG Mazák s Opinion377 in the France Télécom case. We will focus our attention on his assessment about recoupment and the reasons upon which he recommended the ECJ to uphold the appella t s WIN s a gu e ts i this ega d. 378 We will first provide the reader with an outline about Mazák s a gu e ts a d the e ill t to go deepe i his assess e t describing its main aspects and the basis used by the AG to sustain his specific recommendation (the reexamination of the action at first instance). 2.1. AG Mazák’s Opi io a out recoupment of losses After the General Court rejected the action for annulment filed by WIN agai st the Co issio s decision that condemned it for predatory pricing, the undertaking contested that ruling before the ECJ. WIN based its appeal on seven grounds. The first and seventh grounds of appeal are related to recoupment, and thus, are the ones we will consider hereafter. 376 According to Gormsen, there are also cases in which the two objectives may co- exist, such as when the protection of the competitive process would benefit consumers, that is, for example, when the prevention of the market from being pre-empted, will also prevent a dominant undertaking to become a monopoly: Economic freedom and consumer welfare only seem to co-exist where an intrinsic protection of the process of competition also benefits consumer welfare, or where an instrumental protection of the competitive process does not sacrifice economic freedom. Because of the conflict, a balancing act needs to be carried out to examine which of these two objectives should take priority over the other. Gormsen (2010), p. 111. 377 Opinion of AG Mazák in Case C- 202/07 P, France Télécom v Commission. 378 We direct the reader to Chapter I where we included a summary of the facts of the case and its final outcome the ECJ s uli g upholdi g the GC s uli g a d the efo e, the Co issio s o tested de isio . Also i Chapte II we analyzed the recoupment assessment conducted by the Commission in France Télécom. 88 2.1.1. First ground of appeal: WIN alleged that the GC did not provide the reasons for ruling that proof of recoupment was not required when assessing an alleged predatory pricing conduct. In support of this ground of appeal, WIN maintained that while in Tetra Pak II the ECJ indicated that it was only under the circumstances of that individual case that it was not appropriate to require proof that a dominant undertaking had a realistic chance of recouping its losses; the GC had turned such statement into a general rule. 379 WIN argued that the GC failed to provide the reasons for doing so. Additionally, WIN argued that the circumstances of Tetra Pak II were totally different from the case at stake, since Tetra Pak held a so-called super-dominant position in a mature market. 380 WIN also argued that the EU case law had never stated that it was unnecessary to prove the possibility of recoupment of losses.381 The AG ag eed ith WIN s a gu e t, i di ati g fi stl that i deed the GC had failed to e plai h it considered that proof or the possibility of recoupment of losses was not necessary in the light of the specific facts of the instant case. 382 By turning the specific statement that was only applicable to Tetra Pak II to a general rule without further explanation, the GC had manifestly failed to fulfill its obligation to state adequate reasons.383 The AG therefore considered that the GC made three errors. Firstly, the GC made a double error: i) it used an incorrect rule through its incorrect reading and application of Tetra Pak II to the case, and ii) it also failed to fulfill its obligations to state adequate reasons.384 In addition to the two previous errors, the AG deemed that the GC failed to answer precisely the argument raised by WIN. For Mazák, WIN had argued that the Commission must have shown that WIN could anticipate the possibility of recouping its losses, a d ot that the Co issio ust ha e p o ed the a tual recoupment of losses.385 In regard to the first ground of appeal, in what refers to recoupment, the AG concluded that the GC breached its duty to provide reasons coupled with an error of law and the judgment under appeal should therefore be set aside.386 379 Opinion of AG Mazák in Case C- 202/07 P, France Télécom v Commission, para. 54. WIN probably was showing the Court that, as opposed to that case, they were dealing now with a recently liberalized market, with different characteristics to the relevant market in Tetra Pak II. 381 Opinion of AG Mazák in Case C- 202/07 P, France Télécom v Commission, para. 54. 382 Opinion of AG Mazák in Case C- 202/07 P, France Télécom v Commission, para. 58. 383 Opinion of AG Mazák in Case C- 202/07 P, France Télécom v Commission, para. 59. 384 Opinion of AG Mazák in Case C- 202/07 P, France Télécom v Commission, para. 60. 385 Even though at this specific point the AG does not specifies the part of the ruling where the GC made such istake, p o a l he as aki g efe e e to pa ag aphs a d of the uli g, he e the GC eje ted WIN s arguments indicating that for both situations (prices below AVC and between AVC and ATC) it was not necessary to establish in addition (to the evidence filed by the Commission) proof that WIN had a realistic chance of recouping its losses (emphasis added). Case T-340/03, France Télécom v Commission, [2007] ECR II-117, paras. 227 and 228. 386 Opinion of AG Mazák in Case C- 202/07 P, France Télécom v Commission, para. 64. 380 89 2.1.2. Seventh ground of appeal: In the seventh ground of appeal WIN argued that the GC wrongly applied Article 102 TFEU in holding that proving the possibility of recoupment of losses was not a pre-requisite for finding predatory pricing. 387 For WIN, the EU case law requires proof of recoupment, without which predation is inconceivable, since without the expectation of recoupment, predatory pricing does not constitute rational economic behavior. WIN supported its argument by stating that its view was shared by numerous national courts and competition authorities as well as by a large number of academic writers.388 Although the AG s Opi io a d the uli g of the ECJ a e sile t a out the sou es i hi h WIN supported its argument, they do show that the Commission contended the argument with reference to the US Antitrust Law. We learnt in the previous chapters that the Commission argued that dominance entails recoupment: In addition, such demonstration, required by the case-law of the national courts of the United States of America, is based on a different economic logic than that of Community law. According to the Commission, unlike the approach under United States law, the analysis of abuse within the meaning of Article 82 EC presupposes that the undertaking concerned is in a dominant position.389 The AG s ie is that the GC s i te p etatio of the ase la as i o e t as he o side ed that the case law requires the possibility of recoupment of losses to be proven.390 Fi st he e tio ed the ECJ s ruling in Tetra Pak II, which he interpreted differently from the GC: when using the qualifying words i the i u sta es of the p ese t ase , the AG interpreted that the ECJ clearly intended to avoid making a general statement that would render it unnecessary to prove the possibility of recoupment in future predatory pricing cases. 391 To this end, Mazák uoted AG Fe ell s Opi io i Compagnie Maritime Belge transports and Others v Commission (CMB).392 The AG stated that in Tetra Pak II the ECJ did not go as far as AG Ruiz- Jarabo had recommended, i.e., the Court not to lay down the prospect of recouping losses as a new prerequisite for establishing the existence of predatory pricing contrary to Article 102 TFEU.393 Mazák also e tio ed the ECJ s uli gs i AKZO and Hoffmann- La Roche. Although he made no further analysis about these rulings, he pointed out (in his footnotes) paragraphs 71 and 91 of the judgments, respectively, and indicated that [u]nless there is a possibility of recoupment, the dominant undertaking is 387 Opinion of AG Mazák in Case C- 202/07 P, France Télécom v Commission, para. 65. Opinion of AG Mazák in Case C- 202/07 P, France Télécom v Commission, para. 66. 389 Case C-202/07P, France Télécom v. Commission [2009] ECR I-2369, para. 102. 390 Opinion of AG Mazák in Case C- 202/07 P, France Télécom v Commission, para. 69. 391 Opinion of AG Mazák in Case C- 202/07 P, France Télécom v Commission, para. 70. 392 Opinion of AG Fennelly in Joined cases C- 395/96 P and C- 396/96 P. 393 We learnt in Chapter I that AG Ruiz- Jarabo expressed a rigid criterion against recoupment in his Opinion in Tet a Pak II: …recouping losses is the result sought by the dominant undertaking, but predatory pricing is in itself anticompetitive, regardless of whether it achieves that aim. Opinion of AG Ruiz-Jarabo in Case C-333/94 P, Tetra Pak v. Commission, para. 78. 388 90 probably engaged in normal competition. For this last assertio , he also used AG Fe source. ell s Opinion as a In the next paragraphs where Mazák dealt with the recoupment subject, he made it clear that he used a consumer welfare standard to reach his conclusions. He affirmed that where there is no possibility of recouping losses, consumers and their interests should, in principle, not be harmed. He expressly indicated that the purpose of Article 102 is safeguarding the interests of consumers rather than to protect the position of particular competitors. 394 Just as we explained it in the previous section, the recoupment e ui e e t akes se se he the pu sued goal is o su e elfa e, a d Mazák s Opinion follows this approach. If recoupment is not likely, the incumbent firm would not be able to reach monopolistic prices in the long run. Hence, this situation, coupled with the predation phase of low prices, will benefit consumers. Finally he criti ized the Co issio s li e of a gu e tatio a o di g to hi h recoupment is implied in dominance, not at least because the determination of dominance is often based on historical market o ditio s, hilst… p oof of the possi ilit of e oup e t is i he ently ex ante and forward-looking, assessing the market structure as it will be in the future. 395 Mazák recommended the ECJ to uphold the recoupment-related complaint under the seventh ground of appeal and as a consequence, the re-examination of the action at first instance. He recommended reexamination due to the importance of the recoupment issue, which he characterized as so e t al .396 2.2. Main aspects of the Opinion 2.2.1. About the first ground of appeal: In the first ground of appeal the AG chiefly dealt with the application (to France Télécom) of a statement that was apparently based on the specific circumstances of Tetra Pak II. According to Mazák, the GC turned that statement into a general rule, and without further explanation, applied it to the France Télécom case. For the AG, the error of law consisted in the use of an incorrect rule, precisely, that recoupment was not required in the predation test. The error occurred when the GC applied a statement addressed only to another case as a general rule, when i fa t it as t. Additio all , the AG poi ted out a ea h of the GC s dut to p o ide easo s e ause it failed to e plai why the particular statement of Tetra Pak II was used and applied (as a general rule) to the case. This fi st pa t of the AG s assessment is related majorly to the alleged failure of the GC to deliver a statement of reasons that clearly and unequivocally disclosed its thinking. This is a duty which is imposed to the 394 Opinion of AG Mazák in Case C- 202/07 P, France Télécom v Commission, para. 74. Opinion of AG Mazák in Case C- 202/07 P, France Télécom v Commission, para. 76. 396 Opinion of AG Mazák in Case C- 202/07 P, France Télécom v Commission, para. 78. The AG also characterized as so e t al the issue elated to the right of alignment. 395 91 Judiciary by the Statute of the Court of Justice. 397 Thus, we believe this part of the opinion to be merely elated to p o edu al ules appli a le to the Cou t s o ligatio s he p epa i g a d passi g its uli gs, and not to the recoupment factor in substance. It is when the AG assesses the next ground of appeal that he conducts an analysis of recoupment, by mentioning the case law and academic sources which he believed support its inclusion to the predation test. We will comment about these issues in the following lines. 2.2.2. About the seventh ground of appeal: The AG made a clear statement about recoupment: The case law requires recoupment to be proven. In support of such a strong statement he mentioned Tetra Pak II, AKZO, and Hoffmann-La Roche. However, we believe that his argumentation was lax and a more substantive analysis was desirable. Our criticism is basically due to a lack of explanation and detail about the sources the AG pointed out. Mazák was dealing with a highly controversial situation, which had been already debated at least in Tetra Pak II and CMB, and had detractors with several arguments, such as Ruiz-Jarobo.398 Even though Mazák provided various sources in support of his thesis, we believe such sources were not sufficiently analyzed. This was the case, e.g., when he affirmed that the case law supported recoupment as part of the predation test, and the same happened when he constructed other statements, such as the recommendation of a consumer welfare standard as a ruling benchmark. Considering what we pointed out in the first section of this chapter, namely that the consumer welfare goal is a possibility amongst other objectives, our opinion is that the AG had also to convince the Court about the convenience, according to EU Competition Law, of setting consumer welfare as a major goal when dealing with predatory pricing. 397 As was also pointed out by the AG in paragraph 46 of his opinion, in Article 36 (applicable to the GC by virtue of Article 53) the Protocol on the Statue of the Court of Justice of the EU states that Judgments shall state the reasons on which they are based. 398 In his opinion, Ruiz-Jarobo opposed to the possibility of including recoupment in the predatory pricing test. He also debated against the argument according to which the approach of the US Supreme Court had to be repeated in the EU, and finally stated four reasons against recoupment. Although as we mentioned in Chapter I, Ruiz-Jarobo gave more importance to the intent of the dominant undertaking rather than to the effects of the conduct, he also provided strong arguments, such as the complexity of certain recoupment screens and some ambiguity as to its definition: I suppo t of its a gu e ts, the appella t efe s to the ase-law of the Supreme Court of the United States, in particular to Brooke Group v Brown & Williamson Tobacco, in which it was stated that below cost prices can be predatory in nature only if the dominant undertaking had a reasonable prospect of subsequently recouping its deliberately incurred losses. The Supreme Court therefore considered that predatory pricing exists if sales are made below cost and the undertaking in question expects to recoup the losses incurred. That second requirement must be specifically proved, because the possibility of recouping the losses is the ultimate goal of the predatory pricing strategy and, were it not feasible, the practice would be one of advantage to consumers. I do not consider it desirable that the Court of Justice should lay down the prospect of recouping losses as a new prerequisite for establishing predatory pricing contrary to Article 86, for a number of reasons: -selling at a loss in order to eliminate a competitor would be suicidal if it were used by a dominant undertaking with no prospect of recouping the losses incurred; -the economic potential of the dominant undertaking and the weakening of competition on the dominated or related market will in principle ensure that losses are recouped; -proof of a prospect of recouping losses is diffi ult to defi e a d e ui es o ple a ket a al ses, as is lea f o the “up e e Cou t s o aselaw; -recouping losses is the result sought by the dominant undertaking, but predatory pricing is in itself anticompetitive, regardless of whether it achieves that aim. Opinion of AG Ruiz-Jarabo in Case C-333/94 P, Tetra Pak v. Commission, paras. 77 and 78. 92 To start, when supporting the recoupment requirement in the case law, he first indicated that the qualifying phrase i the i u sta es of the p ese t ase meant that the ECJ avoided the construction of a general rule applicable to all cases of predation. Then, he quoted Fennelly on his assertion about the ECJ departing from Ruiz-Ja a o s e o e datio . ‘uiz-Jarabo had recommended the Court not to establish recoupment as a prerequisite for predation. Apparently, Mazák interpreted the silence of the Court as to Ruiz-Ja a o s e o e datio the Cou t e e e p essl said that recoupment was not a requirement) a contrario, and concluded that it had established the recoupment requirement as part of the predatory pricing test. Then he linked this to the ph ase i the i u sta es of the p ese t ase and determined that the phrase had a qualifying character, in consideration only to the special facts of Tetra Pak II. To complete his statement about the case law requiring recoupment, he finally mentioned AKZO and Hoffmann-La Roche. However, as to these two rulings, he limited himself to mentioning paragraphs 71 and 91, respectively, in footnotes, without further explanation. At this point we must remember that Mazák affirmed that the case law had established the recoupment requirement. Therefore, the appropriate exercise to prove it was, indeed, showing and analyzing the rulings supporting the statement. To this end, even if the conclusions about Tetra Pak II and RuizJa o o s Opinion were logical and could have been convincing, we believe that the case law he used in support of his assertion had to be analyzed, unless it expressly and unequivocally endorsed recoupment in its wording, which is not the case. Paragraph 71 of AKZO reads: Prices below average variable costs (that is to say, those which vary depending on the quantities produced) by means of which a dominant undertaking seeks to eliminate a competitor must be regarded as abusive. A dominant undertaking has no interest in applying such prices except that of eliminating competitors so as to enable it subsequently to raise its prices by taking advantage of its monopolistic position, since each sale generates a loss, namely the total amount of the fixed costs (that is to say, those which remain constant regardless of the quantities produced) and, at least, part of the variable costs relating to the unit produced. 399 While paragraph 91 of Hoffman-La Roche reads: For the purpose of rejecting the finding that there has been an abuse of a dominant position the interpretation suggested by the applicant that an abuse implies that the use of the economic power bestowed by a dominant position is the means whereby the abuse has been brought about cannot be accepted. The concept of abuse is an objective concept relating to the behaviour 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.400 399 400 Case C-62/86, AKZO Chemie BV v. Commission [1991] ECR I-3359, para. 71. Case 85/76 Hoffmann-La Roche v Commission [1979] ECR 461, para. 91. 93 In line with our last remarks, none of the rulings in question contain a clear and straight statement about recoupment, not even AKZO which specifically refers to below-cost pricing. Again, we believe that in support to his strong statement, under the risk of not convincing the ECJ, Mazák needed to conduct a deeper analysis which clearly verified the reasons why the EU case law requires recoupment. As a matter of fact, when the ECJ ruled in France Télécom it supported its position against recoupment, inter alia, in the very same rulings (and same paragraphs) mentioned by Mazák. 401 Additionally, what the AG thought that the ECJ did not expressly do in Tetra Pak II, it finally did it in France Télécom: the Court constructed a rule of general application in regard to recoupment, taking it out -as a necessary precondition- from the predation test: Accordingly, contrary to what the appellant claims, it does not follow from the case-law of the Court that proof of the possibility of recoupment of losses suffered by the application, by an undertaking in a dominant position, of prices lower than a certain level of costs constitutes a necessary precondition to establishing that such a pricing policy is abusive (emphasis added).402 The situation could have been different if Mazák would have stated the exact reasons why he considered that the case law supported recoupment. For example, he could have argued that the concept of abuse is an objective concept which should be ascertained by assessing the effects of the conducts of the incumbent firm. Accordingly, decisions based only on the intention of the undertaking fall short from an appropriate analysis. So, as paragraph 91 of Hoffmann-La Roche states, this assessment entails, inter alia, determining whether or not France Télécom used methods different from those which condition normal competition in products or services, and whether or not the conducts hindered competition. If, according to his position, in the absence of recoupment there is normal competition due to the absence of negative effects on consumers, the referred case law could be interpreted as a good basis or justification for recoupment. Additionally, if in the absence of recoupment the dominant undertaking cannot reach the only interest which rationally justifies predation, then paragraph 71 of AKZO is also a good backing case law for recoupment, insofar as the production of such interest will negatively affect the aims which EU Competition Law intend to enhance. AKZO suggests that the only possible interest of a rational predator is no other that of eliminating competitors so as to enable it subsequently to raise its prices by taking advantage of its monopolistic position, since each sale ge e ates a loss… Ironically, the ECJ clearly ignored Mazák and seems to have listened more to AG Ruiz-Ja a o s recommendations, as to the exclusion of recoupment from the predation test. Another point of the opinion we believe relies only on a source that the AG did not analyze enough is the one referring to consumer welfare, which we believe Mazák regarded as the ultimate Competition Law objective. Mazák indicated: 401 402 Case C-202/07P, France Télécom v. Commission [2009] ECR I-2369, paras. 104, 109 and 110. Case C-202/07P, France Télécom v. Commission [2009] ECR I-2369, para. 110. 94 In such a case, where there is no possibility of recouping losses, consumers and their interests should, in principle, not be harmed. I may note here that I share the view of Advocate General Ja o s ho i his Opi io i Os a B o e stated that the p i a pu pose of A ti le [ EC] is to prevent distortion of competition –and in particular to safeguard the interests of consumersathe tha to p ote t the positio of pa ti ula o petito s . Again, while adhering to the view of another AG, Mazák did not indicate why in the particular case (France Télécom) consumers would have been benefited in the absence of proof of recoupment. The AG did not provide reasons to explain why he believed consumer welfare is the ultimate goal of competition, rather than protecting the position of particular competitors. 403 If we recall what we learnt in the first section of this chapter, consumer welfare is not the only objective of EU Competition Law. So, this was a critical issue that had to be developed in the Opinion if the AG deemed that consumer welfare should have been the aim preferred. Apparently Mazák took the contrary for granted, and supposed that the ECJ had already embraced the consumer welfare standard. If consumer welfare is the chosen standard then the recoupment requirement fits -to some extent easily- into such approach, however, that is not to happen if other objectives a e at stake i the Cou t s vision. If Tetra Pak II was already a clear example of the protection of economic freedom, it was reasonable to expect a similar outcome, thus, a consumer welfare standard had to be suggested with more strength and convincing reasons, and therefore, justified under EU Competition Law. As opposed to Mazák s e o e datio , the ECJ stated: Therefore, since Article 82 EC refers not only to practices which may cause damage to consumers directly, but also to those which are detrimental to them through their impact on an effective competition structure… As the Court has already stated, it follows that Article 82 EC 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 basis of quality.404 Some could argue from the text of the ruling that the Court protected economic freedom as a means to reach consumer welfare, which of course can be interpreted as a consumer welfare standard or even as a protection of two objectives, such as it happened in GSK. However, we remind the reader that according to the ECJ in paragraph 112 of France Télécom, the Court identified the harmful effects on o su e s as the loss as a result of the limitation of the choices available to them . For those in favor 403 The source used by Mazák (AG Jacobs) basically contains the same premise about consumer welfare over the protection of competitors, and only provided an additional example related to the legal standard required to prove an abuse: Thirdly, in assessing this issue it is important not to lose sight of the fact that the primary purpose of Article 86 is to prevent distortion of competition - and in particular to safeguard the interests of consumers - rather than to protect the position of particular competitors. It may therefore, for example, be unsatisfactory, in a case in which a competitor demands access to a raw material in order to be able to compete with the dominant undertaking on a downstream market in a final product, to focus solely on the latter's market power on the upstream market and conclude that its conduct in reserving to itself the downstream market is automatically an abuse. Such conduct will not have an adverse impact on consumers unless the dominant undertaking's final product is sufficiently insulated from competition to give it market power. Opinion of AG Jacobs in Case C- 7/97 [1998] ECR I-7791, para. 58. 404 Case C-202/07P, France Télécom v. Commission [2009] ECR I-2369, paras. 105 and 106.. 95 of recoupment and the consumer welfare standard it could be argued that the lack of choices would have been compensated with the lack of monopolistic prices, and therefore, the negative effects would have been neutralized. Such assessment, including whether or not the limitation of choices would allow for real harmful effect on consumers, must be done when following an effects-based approach. Therefore, an assumption is not sufficient. Moreover, the exercise has also to be conducted if there is a use of economic freedom as a means to reach consumer welfare, which could also be a reading of the Cou t s uli g. 2.2.3. Fe ell as Mazák s ai sou e: Mazák s Opi io e o e ded upholdi g a g ound of appeal according to which the GC had wrongly applied Article 102 TFEU in holding that recoupment is not a requirement for finding predation. Therefore, we believe that his main duty was to support his recommendation by establishing the reasons why Article 102 requires recoupment, and as a result, the GC had erred. Other considerations, such as the reasons for the advantages of the inclusion of recoupment and rebutting the recoupmentdominance correlation, are in our view secondary. Because Article 102 is clearly silent about recoupment, the AG used ECJ case law. He prepared his own analysis as to Tetra Pak II, which we previously outlined, and afterwards, he briefly referred to AKZO and Hoffmann-La Roche. Fennelly had already conducted the same analysis in his previous opinion in CMB, i hat efe s to the latte t o ECJ s uli gs. CMB i ol es a ase of p i e edu tio s i the fo of a p a ti e k o as fighti g ships i the maritime transport sector. Accordingly, the members of a shipping conference line (CEWAL), holding a collective dominant position, were found to infringe Article 102 by, inter alia, modifying its freight rates by departing from the tariff in force in order to offer rates the same as or less than those of the principal independent competitor for vessels sailing on the same date or neighboring dates. The aim of such p a ti e as fou d the Co issio to e the eli i atio of CEWAL s ai o petito , G&C. 405In CMB, as opposed to the cases of predatory pricing, the low prices did not go below the applicable cost benchmarks of the AKZO test, namely AVC and ATC. However, the practice was categorized as an implementation of a policy of selective price cutting with the demonstrable aim of eliminating all competition, and thus, found to be abusive.406 With regard to the costs, even though the special characteristics of the maritime transport sector determined that marginal costs were very low, as we said, the discussed prices did not go below the studied cost benchmarks. Therefore, it seems that the establishment or quantification of losses is blurry in CMB, at least in the text of the opinion. As a tangible difference to France Télécom, the text of the Opinion mentions circumstances such as sho tfall i e e ue esulti g f o the fighti g ates 407 or 405 Opinion of AG Fennelly in Joined Cases C-395/96 P and C-396/96 P, Compagnie Maritime Belge Transports and Others v. Commission [1998], paras. 1 and 7. 406 Opinion of AG Fennelly in Joined Cases C-395/96 P and C-396/96 P, Compagnie Maritime Belge Transports and Others v. Commission [1998], para. 132. 407 Opinion of AG Fennelly in Joined Cases C-395/96 P and C-396/96 P, Compagnie Maritime Belge Transports and Others v. Commission [1998], para. 122. 96 sha i g of loss of e e ue . 408 The following extract of the Opinion shows that, in the end, the prices did not go below the costs levels: There are peculiar features of certain markets such as maritime transport where costs may be an unreliable guide to the reasonableness of competitive strategies adopted by dominant firms. In the first place, once a ship has been designated to sail on a particular day, then, provided capacity is available, the cost of transporting an additional container shipped as a result of a reduced-rate offer may be close to zero. More generally, freight rates will largely be determined not by the marginal cost for the shipping line of providing the service but by the price elasticity of demand for the product shipped. 409 …as the Co issio suggests…, the e fa t that Ce al as a le to set the fighti g ates at o above cost may suggest in itself that the normal rates were substantially above cost, or, I would observe, that marginal cost was, in any event, very low. 410 These circumstances show that CMB is a case with different circumstances to predatory pricing. In fact, it does not involve predation. Nevertheless, AG Fennelly suggested that recoupment is implied in paragraph 71 of AKZO, and also in paragraph 91 of Hoffmann-La Roche, a reasoning that later Mazák repeated. The argumentation in favor of recoupment eads i Fe ell s Opi io like this: The sharing of loss of revenues prompts me to revert briefly to the possible need to establish an intention or a possibility of recoupment. The process of sharing revenue losses is in essence a form of recoupment. The strategic purpose of the fighting rates carries with it the unspoken implication that rates will not be reduced for any sailings, current or future, where that is not necessary to meet competition. Furthermore, once the competitor was eliminated, they would clearly no longer be justified. Thus, to the extent that it is necessary, I believe that the present case passes the test of recoupment. At the same time, I would say that some such requirement should be part of the test for abusively low pricing by dominant undertakings. It is implied in the fi st pa ag aph of the uotatio f o AK)O… It is i he e t i the Hoff a -La Roche test. The reason for restraining dominant undertakings from seeking to hinder the maintenance of competition by, in particular, eliminating a competitor is that they would thus be enabled to charge abusively high prices. Thus, an inefficient monopoly would be reinstated and consumers would benefit o l i the sho t u . If that esult is ot pa t of the do i a t u de taki g s strategy it is probably engaged in normal competition. Unfortunately, when in paragraphs 126 and 124 of his Opinion Fennelly makes a brief analysis of the two rulings, he does not mention why they support recoupment. Rather than that, Fennelly only transcribed the referred paragraphs of the landmark rulings. In the first case, he did that to establish the principles in respect of below-cost pricing by dominant undertakings (AKZO), and in the second case, he mentioned that the sta ti g poi t fo the dis ussio of a use of a do i a t positio is the Cou t s 408 Opinion of AG Fennelly in Joined Cases C-395/96 P and C-396/96 P, Compagnie Maritime Belge Transports and Others v. Commission [1998], para. 136. 409 Opinion of AG Fennelly in Joined Cases C-395/96 P and C-396/96 P, Compagnie Maritime Belge Transports and Others v. Commission [1998], para. 133. 410 Opinion of AG Fennelly in Joined Cases C-395/96 P and C-396/96 P, Compagnie Maritime Belge Transports and Others v. Commission [1998], para. 135. 97 judgment in Hoffmann-La Roche. As a following act he proceeded to transcribe the texts of the judgments, with no further analysis.411 It is fair to stress, ho e e , that p o a l Fe ell s assess e t i efe e e to recoupment was complete and accurate to the particular case of CMB. Indeed, the way he envisaged the members of the conference group (collective dominance had been established) would recoup any losses or loss of revenues they may had suffered is logical and understandable: at the same time, Cewal was in a position to devise a scheme of selective designation of sailings for the rate reductions. The resulting loss of revenue was shared between conference members. Both because of the selectivity of the reductions and their very large market share, the members could spread and absorb the loss of revenue. 412 In addition to that, one must take into account the fact that the group held a collective dominant position, close to a monopoly level. If that was the case, considering also that the establishment of real losses was blurry, the possibility of recoupment was easily previewed in the transcribed paragraph 136 of the Opinion. As the AG stated, [t]he p a ti e flo ed f o Ce al s u justified lai to ai tai a monopoly in the relevant market which it had sought to enforce via the Ogefrem Agreement and was incontestably designed not merely to beat competition but to eliminate the competitor. 413 As e a i fe f o Fe ell s a al sis, it a e the ase that he fo used o e o p o i g h i CMB recoupment was feasible, than in arguing that recoupment was indeed a requirement under EU Competition Law, which is precisely the task which we believe Mazák had in the case of France Télécom. The te t of Fe ell s Opi io pa ag aph that defe ds ecoupment is in a substantive part dedicated to prove that in CMB recoupment was clear. Additionally, he mentioned the two rulings of the Court and stated that if recoupment is ot pa t of the do i a t s u de taki g s st ateg it is p o a l engaged in normal competition. This last phrase was slightly changed by Mazák who indicated that unless there is a possibility of recoupment, the dominant undertaking is probably engaged in normal competition. Mazák, again, used the same support (Fennelly) when making this last assertion. All in all, we believe that Mazák missed an opportunity to convince the Court about the recoupment requirement. Our analysis does not deal directly with the techniques the AGs must apply when rendering their Opinions or the extent of the assessments they must perform. However, if we compare Mazák s Opi io to othe s, su h as Fe ell s i CMB or Ruiz-Jarabo s in Tetra Pak II, we must say that further analysis may be missing. All the more, as we said, we must consider that Mazák was dealing with a highly controversial issue that not only depended on mentioning fancy sources that support recoupment. As an example, as we will further argue in the conclusions to the thesis, the support of an a gu e t o the U“ “up e e Cou t s uli gs ould fall o d g ou d ith ega d to EU Co petitio Law, at least if we consider that the objectives of both jurisdictions do not always converge. In this vein, 411 Opinion of AG Fennelly in Joined Cases C-395/96 P and C-396/96 P, Compagnie Maritime Belge Transports and Others v. Commission [1998], paras. 126 and 124. 412 Opinion of AG Fennelly in Joined Cases C-395/96 P and C-396/96 P, Compagnie Maritime Belge Transports and Others v. Commission [1998], para. 135. 413 Opinion of AG Fennelly in Joined Cases C-395/96 P and C-396/96 P, Compagnie Maritime Belge Transports and Others v. Commission [1998], para. 135. 98 the shift towards a predation test that requires recoupment probably entails a change of paradigm, if we consider previous rulings as Tetra Pak II. This change not only has the consequence of modifying rules as the AKZO test, which relies only on the application of a cost assessment to prove predation. Moreover, the change also entails the application of an effects-based approach, and additionally, setting and focusing the assessment of abuse on a novel analysis focused on ascertaining consumer harm, if consumer welfare is chosen as a standard. As a result, the Opinion should have been prepared in recognition of such an ample, comprehensive, and no less important effect on EU Competition Law. Our opinion is that Mazák could have incurred in the sa e e o he poi ted out a out the GC s uli g he assessing the first ground of appeal: a failure to state reasons. 414 2.3. Secondary aspects of Mazák’s Opinion Although e elie e Fe he commented: ell s Opi io to e the ai sou e that suppo ted Mazák s, i pa ag aph 75 Finally, I will take this opportunity to point out that, apart from the above case-law, the importance of proof of the possibility of recoupment was brought to the fore inter alia by the Economic Advisory Group on Competition Policy (EAGCP), the Organization for Economic Cooperation and Development and the European Regulators Group. 415 We believe that he clearly rendered secondary importance to such sources, as he did not comment on them and only made a few textual transcriptions which he included in his footnotes. Probably the AG did not expect the Court to pay much attention to academic contributions and therefore, non-binding sources. Then, as we pointed out, the AG rejected the correlation dominance-recoupment defended by the Commission. He basically did that by arguing that the assessment of recoupment entails the analysis of the conditions of the market in the future, i.e., after the market pre-emption has occurred, which of course differs from the dominance test, which focuses on the present structure of the market. In a footnote he added that: …it should e app e iated that p oof of do i a e does ot e essa il i pl that the do i a t firm will also be able to recoup its losses. While proof of recoupment means that the dominant fi s o opol ould pe sist i futu e, it is i po ta t to app e iate that the loss-making and e oup e t phases do ot oi ide… The fa t that a fi as do i a t at the ti e it e gaged i 414 When referring to the duties of the AG, Article 252 TFEU points out the preparation of reasoned submissions as one of them: It shall be the duty of the Advocate-General, acting with complete impartiality and independence, to make, in open court, reasoned submissions on cases which, in accordance with the Statute of the Court of Justice on the EU, require his involvement. Additionally, when describing the Opinions of the AGs, Craig and de Búrca express the following: The AG s opi io is i te ded to o stitute i pa tial a d i depe de t ad i e, a d i p a ti e it te ds to be a comprehensive, reasoned account of the law governing all aspects of the case. It will often shed the light on an ECJ judgment that is difficult to interpret (emphasis added). (2011), p. 62. 415 Opinion of AG Mazák in Case C- 202/07 P, France Télécom v Commission, para. 75. 99 below-cost selling does not mean that it will in future be able to recover past losses by raising prices: conditions of competition may well be different in future. 416 The argument contends that both analyses pertain to different conditions as a result of different time periods and therefore, markets with possible different structures. Once a competitor(s) is ousted from the market the possibilities of recoupment take place or depend on different conditions of competition. We have analyzed very similar arguments in the two previous chapters of the thesis, namely, when we a al zed the Co issio s recoupment test conducted in its decision against WIN, in the France Télécom ase Chapte II , a d he e o e ted o Pat i k ‘e s ite io i Chapte I, hile analyzing the recoupment factor. We consider the fore going aspe ts of the opi io to e se o da , e ause Mazák s ai o je ti e as to account for the recoupment requirement under EU case law. Due to this, the analyses about the differences between recoupment and dominance or about the appropriate recoupment screen, even if convincing, would have made no difference if recoupment was not required by the Court. In any case, it appears that the Court does not believe in the correlation recoupment/dominance, because if it did, it would have stated that recoupment had already been proved when the dominance test was conducted. Contrarily, the Court ruled against recoupment being required when assessing predation. 3. Final comments Defining a main or ultimate goal of EU Competition Law is no less than problematic and rather difficult. From our point of view it is a still unresolved issue in the hands of the EU Judiciary and ultimately, of the ECJ. We studied in the case law, at least three goals of EU Competition Law that may be at stake in competition law cases: market integration, economic freedom, and consumer welfare. As opposed to the Courts, the Commission in its Guidance made its choice and pointed out consumer welfare as a major goal, although in practice, like in the France Télécom case, it appears to have opted for an economic freedom standard. The Guidance deals only with specific conducts in which the Court has shown signs of applying an effects-based approach. This may also mean that the Commission has ide tified a o su e elfa e sta da d i the Cou ts judg e ts, when ruling upon such specific conducts, including predatory pricing. Although the application of an effects-based approach may entail the use of a consumer welfare standard, it does not mean that an effects-based approach cannot be applied when using other standards, such as economic freedom. If the preferred standard is consumer welfare an effects-based approach must be used to prove consumer harm; if the preferred standard is economic freedom, an effects-based approach can be used to prove harmful effects on the structure of competition. In the last case, whether or not consumers are likely to be harmed would not need to be proved. It must also not be forgotten that even when using a consumer welfare standard only, the protection of competitors is plausible when used as an instrument to protect consumers. Therefore, an effects based approach to 416 Opinion of AG Mazák in Case C- 202/07 P, France Télécom v Commission, para. 76, note 57. 100 this end, would have to be conducted to ascertaining that the likely anticompetitive foreclosure would produce harm to consumers. Notwithstanding the existence of other goals of EU Competition Law, consumer welfare is indeed an important objective when dealing with abusive behavior, and specifically, when dealing with pricerelated conducts. From our perspective, although the Court had already recognized consumer welfare as an objective in its case law, it confirmed this aim in the Post Danmark ruling, which is directly connected to price-related exclusionary abuses. In this case, we believe the Court may have preferred the consumer welfare objective over the economic freedom objective. As to the Opinion of AG Mazák in regard to recoupment, it is clear that the AG applied a consumer welfare standard when making his assessment. He made that clear by saying that what really matters is to safeguard the interests of consumers, rather than protecting the position of particular competitors. As Gormsen pointed out, in the specific case of predation recoupment is required due to the fact that in absence of it, harmful effects on consumers are not likely to happen. Thus, as we pointed out before, if the preferred standard is consumer welfare, recoupment must be proved. Despite we believe the last assertion to be true; in his Opinion the AG did not prove to the ECJ his most important statement. According to him the case law of the ECJ requires recoupment as a prerequisite for predatory pricing. In trying to prove his point the AG relied on a previous Opinion of one of his colleagues, which neither accurately shows how the EU case law supports recoupment. Accordingly, the ECJ ignored the AG a d o fi ed the Co issio s de isio i the pa ti ula ase. It a e that the AG rendered too much importance to his sources or deemed that the Court had already adopted a consumer welfare standard, and therefore considered recoupment as a major premise and deemed unnecessary to re-state what was already established by Fennelly. We believe that a further analysis was desirable. 101 Chapter IV: Conclusions The content of the thesis is an analysis of a potential situation of relevance under EU Competition Law: the inclusion of recoupment to the predation test. The study entailed taking the pulse of what we called the ele a t ele e ts of EU Co petitio La su ou di g this issue. B doi g this, e e a i ed the feasibilit of the i lusio of a e e ui e e t to the test of p edato p i i g. “o e of these ele a t ele e ts oti ated the stud a d ithout the a legal dis ussio of recoupment could have been rather pointless, uninteresting, or devoid of academic importance. We identified a historical juncture of EU Competition Law, which could facilitate the opportunity for the addressed inclusion of the requirement. Basically we viewed this opportunity for change in situations and elements such as the reform to the application of Article 102 TFEU, the Guidance as a result of it, and the decision of the Co issio , the AG s Opi io , a d the uli gs taki g pa t i the France Télécom case. With this central task in mind, the thesis includes the analysis of three major components: 1. Predatory pricing as an unlawful conduct and its treatment under EU Competition Law. 2. The Guidance of the Commission and the effects-based approach it fosters. 3. Consumer welfare as an objective of EU Competition Law when dealing with predatory pricing. Predatory pricing is the abusive conduct on which the study was based: the recoupment factor has a direct repercussion on the assessment of this conduct. The study of predation was essential, focused mainly on the current test applied when prosecuting the abuse and passing judgments upon it. In a few words: we had to know how the Commission and the EU Courts deal with predation. The Guidance is a pivotal aspect we considered. Not only because it outlines the general framework of the work and efforts of the Commission when tackling predation, which is of great importance; but also because it is the result of the reform on the application of Article 102 TFEU. Through its analysis, we focused our efforts in examining the effects-based approach that is being adopted by the Commission, the way it was materialized in the context of the Guidance, and the possible repercussions on the assessment of predation. Additionally, we studied the objective of Competition Law fostered by the Guidance, which gives a major importance to consumer welfare. By studying consumer welfare as an objective of EU Competition Law we largely wanted to connect our considerations regarding the possible application of the Guidance to predatory pricing, with the reality of the case law of the EU Courts. In simple terms, this means that even if we could have considered that the inclusion of recoupment to the predation test was feasible under the effects-based approach of the Guidance, this statement would have been incomplete until we examined the EU Cou ts ase la . We required knowing if there is consistency between the Guidance and the EU Courts with special regard to the objectives of EU Competition Law and the assessment of predatory pricing, including recoupment. Under the mentioned panorama, the study aimed at giving reasoned answers to the following research questions: 102    After considering the current situation of EU Competition Law, is the inclusion of recoupment in the predation test feasible? If recoupment is added to the test of predation, what would be a likely recoupment screen under EU Competition Law? Under the consumer welfare standard of EU Competition Law, would recoupment of losses play an important role in enhancing consumer welfare? We reached the following conclusions: The current assessment of predatory pricing follows a form-based approach. However, both the Commission and the ECJ show signs of moving to an assessment characterized by a more effectsbased approach. When we talk about the current assessment of predatory pricing we refer to the AKZO test, which we studied in Chapter I. Under the AKZO test a rigid cost assessment determines the unlawfulness of the price-related conduct. As Buttigieg (2009) and Osterud (2010) explain it, a presumption of illegality when prices are below average variable costs (AVC) is coupled with a proof of intention when prices are between AVC and average total costs (ATC). The assessment of predation is subject to presumptions of illegality in the former case, and to proof of intention in the latter. None of these two criteria require an analysis of the effects of the conduct in the relevant market, and therefore, the harmful effects to the objectives of EU Competition Law are taken for granted when the conditions of AKZO are reached. Additionally, if the AKZO test is applied as it stands, it leaves almost no room for objective justifications or efficiencies. In the first case, when prices are below AVC the presumption of illegality is indeed strong, and from a literal interpretation, almost insurmountable. Accordingly, AKZO provides that prices below AVC, by means of which a dominant undertaking seeks to eliminate a competitor, must be regarded as abusive, and that a dominant firm has no interest in applying such prices except that of eliminating competitors. In this case the effects of the conduct are presumed and the requisite legal standard is reached by the sole fact of prices being below AVC. Then, when prices are between AVC and ATC, if an intention to eliminate a competitor is proved, the story is basically repeated. Actually the GC when applying AKZO in France Télécom, and dealing with prices between these two cost benchmarks, stated that showing and anti-competitive object and an anti-competitive effect may, in some cases, be one and the same thing. 417 As we know, the ruling was confirmed by the ECJ. This reasoning, which renders a great importance to intent rather than to effects, coupled with the formbased characteristic of the cost-related test of AKZO, leaves no room for the inclusion of the recoupment factor to the predation test. To this matter, the ECJ indicated in Tetra Pak II that it is possible to condemn predation whenever there is a risk that competitors will be eliminated and it followed Advocate General Ruiz-Jarobo, according to whom -and this is the reading we make of his criterionrecoupment would be secondary because predation is in itself anticompetitive, regardless of whether it 417 See Chapter I, when we analyzed the cost benchmarks used when ascertaining predation. 103 achieves that aim.418 Therefore, recoupment was not required because it deals with the construction of a prospective state of things in the near future, after the market has been already pre-empted. So, nothing more than proof of the risk of market pre-emption was required. In this sense, it is important to emphasize that in the three landmark rulings concerning predatory pricing, namely AKZO, Tetra Pak II, and France Télécom, the Commission provided sufficient evidence of a strategic plan to pre-empt, with clear intentions of the incumbent undertakings to succeed by ousting their competitors from the relevant market, or preventing the entrance of new competitors to it. We believe that basically the Courts were satisfied when proof of the risk of market pre-emption was reached, and proof of the intentions of the undertakings was effective to this end. Consequently, the threshold to prove an abuse of a dominant position was indeed reached, at least when prices were between AVC and ATC. When prices were below AVC, it was only required proof of such a fact, and the harmful effects of the conduct were presumed. We share the criticism of Wurmnest (2011) to this way of judging of the EU Judiciary, a subject we developed in Chapter II. He argues that when ruling as the Court did in the referred cases, the anticompetitive effects were inferred from the fact the dominant undertakings held a high market share and acted with anticompetitive intent. We share the criticism because of two specific reasons, the first of greater importance than the second: i) this way of judging opposes to the concept of abuse constructed by the ECJ in Hoffmann-La Roche (HLR), and ii) it is contrary to an effects-based approach. We referred to the first reason in Chapter I, which was used by Buttigieg (2009) to criticize AG RuizJarobo when he indicated that predatory pricing is in itself anticompetitive, regardless of whether or not the predator recoups its losses. AGs Fennelly and Mazák also cited paragraph 91 of HLR to support their view about recoupment, and in Chapter III we speculated about why under this ruling recoupment could be interpreted as part of the test for predation. The reason why we believe the EU Courts were ruling against the concept of abuse of HLR, is e ause that uli g i di ates that a use is a o je ti e o ept a o di g to hi h a do i a t u de taki g s behavior has the effect of hindering the maintenance of the degree of competition or the growth of that competition. As indicated by Buttigieg (2009), according to HLR the yardstick in ascertaining abuse is effect, not intention. Clearly the judgments of the Courts, by relying on intention to ascertain the abusive character of the conducts, exempted the Commission from conducting an assessment that included proof of the likely anticompetitive effects of the conduct. As to the second reason, namely that the way of ruling opposes an effects-based approach, we stated that it is less i po ta t. As e poi ted out i Chapte III hile stud i g Go se s fi di gs, a effects-based approach is more important when pursuing a consumer welfare aim. Moreover, even though we stated that an effects-based approach could indeed be used when economic freedom is an aim, for example; intention on the other hand could be a good proxy for abuse when the aim is economic freedom. Accordingly, we described the reasons why, as Gormsen (2010) does, we believe that at least in Tetra Pak II and in France Télécom, the EU Courts were pursuing an economic freedom aim. Thus, the second reason is more relevant when the objective pursued is consumer welfare, but not 418 See Chapter I, when we referred to Tetra Pak II. 104 when the objective is economic freedom. If economic freedom is the promoted goal the position of the competitors of the dominant undertaking in the market is protected. Then, in cases like Tetra Pak II and France Télécom, the clear and convincing proof of intention to oust them from the market could be used as the legal standard required when determining abusive behavior. However, we must emphasize that should the Courts conclusively decide to foster a consumer welfare goal, we believe the mentioned way of ruling does not serve this goal, and as a result, the reason we discuss becomes of greater relevance. If on the other hand economic freedom is also protected as a means to enhance consumer welfare as the end goal, still an effects-based approach would be required, in order to determine the likely anticompetitive effects of the conduct on the interests of consumers. We will later retake our conclusions about the consumer welfare goal in connection to the recoupment of losses. Why do we believe the Commission and the ECJ show signs of leaning towards a more effects-based approach when assessing predatory pricing? The response to this question comes from the findings of our study in Chapter II. As to the Commission, the signs can be identified mainly in the text of the Guidance and in the recoupment test conducted in the Co issio s de isio i the France Télécom case. As to the ECJ, we viewed signs of change in the Post Danmark case, hi h ould e a efull e t apolated to spe ulate a out the Cou t s assess e t i future cases of p edato p i i g, a d also i Wu est s o lusio s a out France Télécom. The te t of the Guida e lea l p o ides fo the adoptio of Co issio s assess e ts ased o a effects-based approach with consumer welfare at its core. With regard to the scope of the Guidance, we believe to be of major importance that the Commission identified those conducts which really harm consumers, and towards which the Judiciary had shown signs of moving to an effects-based approach, and for these reasons included them in the Guidance. Considering that predatory pricing is addressed by the Guidance, and therefore is characterized as consumer harming, a first conclusion from the text of the Guidance is the following: the assessment of predation by the Commission will be done using an effects-based approach in future cases, a situation that the Commission has identified –in principle- as approved by the Courts when indicating that the Judiciary has shown signs of embracing an effectsbased approach. Then, although the recoupment test that the Commission conducted in France Télécom poses serious dou ts o e i g its ele a e ithi that ase, these se ious dou ts efe o e to the Co issio s reactions during the appellate process, than to the assessment itself. As a matter of fact, the Co issio s a gu e ts hile its de isio as o tested efo e the Judi ia e e o fli ti g a d confusing: it practiced an independent recoupment test; however, alleged that it was not required, and at the end argued that recoupment was implied when WIN passed the dominance test. Notwithstanding this relevant situation, a recoupment test was conducted. Additionally, even though we will probably never know about the merits of this test from a ruling of the EU Courts, we do know that the Commission intended to perform an assessment based on the structure of the market, where the barriers to enter the market, among other elements, were analyzed. This fact is related to the prospective recoupment test that EU Competition Law may adopt in future cases, about which we will refer later, but also to the effects-based approach the Commission is adopting when assessing 105 predation. If the Commission decided to assess the likelihood of recoupment in France Télécom, it must have been for an important reason. To this argument we must add that by the time this decision was given, the reform to Article 102 had not yet begun, and therefore the application of the Guidance was not required. Thus, we believe that in future cases the sole existence of the Guidance, coupled with the progress made by the ECJ in cases like Post Danmark, could convince the Commission of the inclusion of recoupment to the assessment of predation, following also the consumer welfare standard preferred by the Commission in the Guidance. Of course, a final push towards recoupment must be made by the EU Courts, all the more, considering the rule of general application that the ECJ constructed in France Télécom, according to which recoupment is not required. This general rule must be reversed if recoupment will become part of the test. In reference to the ECJ we will first address the conclusions expressed by Wurmnest (2011) and then we will refer to PD. We also sha e Wu est s asse tio as to the situatio he elie es a o e t the shortcomings of the way of ruling of the Courts in past predation cases (AKZO, Tetra Pak II, and France Télécom): the test for anticompetitive foreclosure which in his words tightens the standards of proof for showing that the conduct is capable of anti-competitive foreclosure.419 At this point is where we believe that the Co issio s Guida e a d the ECJ i t o of the uli gs of the Cou t that e a al zed 420 may have found common ground to change the rigid AKZO test. We explain this conclusion in the following paragraphs. First of all, Wurmnest refers to the benefits of the application of the anticompetitive foreclosure test, hi h e tails ot o l a easo ed fo e ast of o petito s fo e losu e, ut as a additio al ele e t this foreclosure must be consumer harming. Therefore, according to the Guidance, not every driving of o petito s out of the a ket ill e a ti o petiti e, o l the o e that ha s o su e s i te ests. Following Petit (2009) and Geradin (2010) the assessment is made in two steps: after a possible abusive conduct has been identified, the first step entails the finding of foreclosure, and the second one entails the fi di g of o su e ha . Additio all , the Guida e states i pa ag aph that the Co issio s intervention will normally only occur when the risk of foreclosure takes part against competitors o side ed to e as-effi ie t as the do i a t u de taki g. The referred assessment, if applied, must have an influence on the AKZO test because the presumption plus intent-based test it encompasses, does not converge with the anticompetitive foreclosure test. That is why Wurmnest (2011) affirms that the practice of the Courts -according to which the anticompetitive effects of the conduct were inferred from market power and the proof of anticompetitive intent- is being corrected in the Guidance, specifically, when including the concept of anticompetitive foreclosure. This is of major importance in application of the AKZO test when prices are between AVC and ATC, due to the fact that proof of intention (only) is required for the configuration of an unlawful predation conduct. To this end, the PD ruling stands out as a major sign of change, with special inference on this part of the AKZO test. 419 See the analysis in Chapter II about the Guidance and Consumer Welfare, where we addressed the concept of anticompetitive foreclosure. 420 We refer to France Télécom and PD. 106 But before we reach to PD, we will refer first to the last of Wu est s o lusio s hi h is elated to France Télécom. We believe his conclusion could have direct influence on the AKZO test, and can be o e ted to the o lusio s d a f o the ECJ s uli g i PD. When we studied the limited relevance given to recoupment by the ECJ in France Télécom, we indicated that the Court could have opened the possibility for objective justifications when prices are below AVC. The Court indicated that in such a case (pricing below AVC) recoupment could assist in excluding economic justifications. Interpreted a contrario, this could be an indication of an open door to objective justifications when pricing below AVC, thus, in absence of recoupment, economic justifications may be plausible. We concur with Wurmnest (2011) when he states that the ECJ, by doing this, seems to have silently corrected its overbroad statement that pricing below AVC (by dominant undertakings) is always abusive. The last situation brings us to PD and clearly makes sense with our review of the ruling in Chapter II. If e o e t Wu est s o lusio s to the efe ed uli g e a ide tif i PD an application by the Court of an effects-based approach with direct repercussions on the AKZO test, which may be applicable to future cases of predation. PD could influence the AKZO test for one reason, while there is a second situation that may also affect the AKZO test. The first one is the application of the anticompetitive foreclosure test when pricing between AVC 421 and ATC, while the second situation is the endorsement of the ECJ to the possibility of objective justifications and efficiency claims when dealing with price-related exclusionary abuses. In PD the ECJ referred to the specific case of a dominant undertaking pricing between AVC and ATC, in a case of selective price cutting that did not involve predation. 422 With regard to this specific situation which in principle is regulated by the AKZO test, the ECJ clearly indicated that in order to assess the existence of anti-competitive effects, it is necessary to consider whether that pricing policy, without objective justification, produces an actual or likely exclusionary effect, to the detriment of o petitio a d, there y, of o su ers’ i terests. If we apply the ruling of the Court to a case of predation, it means that the old proof of intention rule has been changed for the proof of anticompetitive effects that harm consumers, as simple as that. Additionally, the ruling in the emphasized extract openly embraces the possibility for objective justifications, at least when prices are between AVC and ATC. As we indicated in Chapter II we do not believe that the Court made a rule of general application, extendable to predatory pricing. Nonetheless, we believe that due to the similarities of the conducts and of their likely effects, in future cases of predation or other kinds of price-related conducts, the Courts could apply this reformed AKZO rule or could turn it into a rule of general application. The application of PD to cases of predation, in our view remains uncertain until the Courts clarify this point. The Cou t s e p essed app o al to o je ti e justifi atio s a d effi ie ies i ases of p i e-related exclusionary abuses could also affect the AKZO rule in both of its scenarios: when prices are below AVC a d he p i es a e et ee AVC a d ATC. If e add the fi di gs take f o the uli g to Wu est s conclusion as to the first scenario (pricing below AVC) we could infer that the strong presumption of 421 In PD the Court agreed to use AIC as a substitute of AVC, for the sake of clarity we will refer to the original cost benchmark of AVC. 422 Predation was previously discarded by the Danish NCA, as a matter of fact, an intention to pre-empt the market was never proved in the case. That is the reason why predatory pricing did not happen, following the AKZO rule. 107 unlawfulness of AKZO may be weakened. Specifically, the factor that would weaken it is the possibility granted by the Court in PD to dominant undertakings of objectively justifying their conduct. In cases where prices are between AVC and ATC, for example, the possibility for objective justifications and efficiencies becomes even clearer if the assessment of the conduct changes. If the unlawfulness of the conduct does not depend anymore on proof of intention but on proof of an anticompetitive foreclosure, the pricing policy could be objectively justified (as opposed to the unrealistic task of objectively justifying an intention) or an efficiency defence could be available, if the conditions of the ruling are met. To sum up, if we read PD i o ju tio ith Wu est s o lusio s, o side i g also that the ECJ has adopted pa t of the Guida e s o te t i the efe ed uli g anticompetitive foreclosure and objective justifications and efficiencies) and we apply that to predation, we must say that the assessment applied to such conduct has suffered remarkable changes. The only element that limits our conclusion is the fact that PD has not created a rule of general application in regard to the major change we identified: the assessment of effects instead of proof of intention when prices are between the two cost levels. In addition to this drawback, the ECJ did not extend its ruling to the case where prices are below AVC, due to the fact that it was not questioned about it. Therefore, the application of the AKZO rule when pricing below AVC remained untouched by the Court. However, the AKZO test suffered changes when talking about the objective justifications and efficiencies, which we believe apply to all price-related exclusionary abuses. If the assessment of predation is going through a change motivated by the application of an effectsbased approach with consumer welfare as a goal, this may pave the way for the inclusion of recoupment in future cases of predation, a possibility we will address hereafter. Recoupment should be part of the definition of predatory pricing when an effects-based approach is applied with the aim of enhancing consumer welfare: We believe that the rationale behind a recoupment test in predation is no other than determining a specific effect of predation: whether or not the predator undertaking will be able to off-set the losses it suffered during the below-cost or predation stage of predatory pricing. Without recoupment the predator will not be able to raise prices to a monopolistic level, and therefore, consumers would not be harmed. That is why recoupment is linked to an effects-based approach, because it entails the assessment of a prospective scenario likely or not likely to happen in a future stage. We take the words of Jones and Sufrin (2011) used in Chapter I when they define predation: …that sho t-term loss of profitability is more than compensated for by long- u p ofita ilit he , afte the o petito s e it, the undertaking can raise prices to monopoly level. Following this view, we learnt in Chapter I that for those who believe that recoupment is part of the definition of predatory pricing, without recoupment a successful predatory strategy is not completed. Predation is therefore not successful or not anticompetitive without recoupment for those who believe in the phrase lo p i es a e the hall a k of o petitio , because without recoupment economic 108 theory423 states that the predator would not be able to succeed in its intentions to reach monopolistic prices, and thereby harm consumers. As a result, consumers would have already benefited from the low-pricing stage and without recoupment, they would not be harmed with monopolistic prices. Although there are good arguments that consider that predation is still anticompetitive even without the possibility of recoupment (Ruiz-Ja o o s Opi io , Motta , the EU Cou ts uli gs i Tetra Pak II and France Télécom), such arguments attack the conduct per se, most of them due to the possible effects of the conduct on competitors, driving them out of the market or impeding other competitors from entering the market. Therefore, these arguments do not run counter the statement according to which without recoupment the predation strategy is unsuccessful due to the impossibility of setting monopolistic prices. Conversely, they run counter the possibility of punishing the conduct only when recoupment is likely, because they deem the conduct as abusive per se, even if the strategy of the dominant undertaking is not fulfilled. So, if recoupment turns to be a requirement when applying the test of predation, it is because the conduct will be deemed as anti-competitive only when it has negative effects on prices, and thereby harms consumers. Accordingly, the premise lo p i es a e the hall a k of o petitio does not apply under EU Competition Law when other objectives, different from consumer welfare are pursued. If the goal is consumer welfare recoupment should be part of the definition of predation, and therefore it must be proved. To this end, if an effects-based approach is followed, it must be applied with the aim of promoting consumer welfare also. An effects-based approach may be used to prove effects of conducts on other targets, such as competitors, protected by other objectives of EU Competition Law. Such is the case when the goal is economic freedom, when harmful effects on competitors may be sufficient to condemn a conduct, just as it happened in Tetra Pak II and in France Télécom. On the other hand, it may also be the case that even if consumer welfare is the preferred goal, other objectives may be instrumentalized with the aim of enhancing consumer welfare. The Court apparently considers that by protecting competitors, it is also promoting consumer welfare, because consumers suffer as a result of the limitation of the choices available to them. In France Télécom when the ECJ stated that Article 102 refers not only to practices which may cause damage to consumers directly, but also to those which are detrimental to them through their impact on an effective competition structure. However, in this case, if consumer welfare is the standard an effects-based approach identifying the likely effects on consumers is required when such a situation takes place. If the Court considered that the harmful effects on consumers were the limitation of choices, a good argument in favor of recoupment can also be established. If recoupment is not proved the benefits of the failed predation strategy, with positive effects on the prices, can compensate the limitation of choices. Thus, we believe that still an effects-based approach would be required to ascertain the possible effects on consumers, even when other objectives are instrumentalized with the aim of promoting consumer welfare. 423 Among others, we studied the convergent position as to this assertion of: Jones and Sufrin, AG Mazák, Gormsen, Papandropoulos, the US Supreme Court, Bishop and Walker, and Eugene Buttigieg, only to name some of them. 109 Summing up, the particularity of EU Competition Law is that not only consumer welfare is the aim. We must remember that there is an overarching goal of undistorted competition (Protocol 27 of the Treaty of Lis o , hi h e studied i Chapte III he e a i i g Go se s fi di gs. Also a ket integration is a goal, follo i g A ti le TEU the eatio of the I te al Ma ket a d the U io s o ligatio of p o oti g e o o i , so ial a d te ito ial ohesio … a o g Me e “tates a d e a ples as the ECJ s uli g i GSK. Indeed the incorporation of recoupment to the predation test serves a consumer welfare goal, but this is not, so far, the only and preferred goal of EU Competition Law. Probably at this point is where a misconception of EU Competition Law creates the constant comparison of it with other legal systems, such as the US Antitrust Law. Therefore, when applying the reasoning of the US Supreme Court in cases like Brooke Group, the exercise does not match or does not make sense under the EU legal system, because each system can be following different goals. In spite of the last statement, the Commission in the Guidance seems to prefer a consumer welfare goal, and in doing this, it may center its efforts in the inclusion of recoupment in the test for predation. The Commission has shown more signs in favor of the adoption of a recoupment test than the Courts. A recoupment test would be based on the structure of the market and not on a mere calculation of losses and likely profits. When the Commission applied the recoupment test in France Télécom it based it on the structure of the market and the associated revenue prospects. The analysis of entry barriers and the analysis of entry costs were the two major components of the test. This recoupment test has similarities with the deep and narrow recoupment test proposed by Hemphill (2001), which seems to be the one prognosticated by Wurmnest (2011), i.e., a more structural recoupment test which looks at the market setting and st ategi o te t of the i u e t s eha io . If e i te p et the Guida e s te t i light of the aforementioned considerations, we can conclude that it allows the Commission to apply a structured-based recoupment test in future predation cases. To this end, we must consider first that the Guidance acknowledges that predatory pricing entails a sacrifice (para. 64), then, that it states that consumers are likely to be harmed if the dominant undertaking can reasonably expect its market power to be greater after the predation period comes to an end (para. 70), and most importantly, that it indicates that likely consumer harm may be demonstrated by assessing the likely foreclosure effects of the conduct, combined with consideration of other factors, such as entry barriers (para. 71). In addition to this, we can discard the conduct-based recoupment screen criticized by Hemphill (2001), acknowledging that the Guidance discards it as well. Accordingly, paragraph 71 of the Guidance indicates that the identification of consumer harm does not depend on a mechanical calculation of profits and losses, and that proof of overall profits is not required. The wording of the Guidance also discards the acceptance of the correlation between dominance and recoupment, which we believe the ECJ did not accept either in France Télécom. The Guidance is clear in that the assessment is focused on the likely effects on the market, a wording that implicitly refers to a prospective situation. The ECJ was silent about the argument raised by the Commission alleging the correlation dominance/recoupment, thus, we believe it did not sha e the Co issio s ie . A a , 110 the Guida e see s to e lea i this espe t a d st o g a gu e ts like Pat i k ‘e s , Papa d opoulos s , a d e e Mazak s, lea l sho that the recoupment assessment entails a different and separate analysis compared to dominance, and among other things, involves the assessment of a prospective situation (in the future) with changes in the market structure, due to the effects of the low-pricing stage of predation. However, we believe that any impulse of the Commission towards the application of a recoupment test could be stopped by what Petit (2009) called the the la e ide tia e ui e e ts of the EU Courts. So, even though in France Télécom we identified an initiative of the Commission in assessing recoupment, this initiative faces the risk of not becoming reality. This is due to the ruling of the ECJ in the same case, where we believe the Court created a rule of general application according to which recoupment is not part of the predation test. As a matter of fact, France Télécom is a practical and illustrative example of the dissonant approaches of the Commission and the Courts towards recoupment. While the Commission seems to have been stepping into a middle indecisive point as to the recoupment factor, the GC seems opposed to the inclusion, and the ECJ sends confusing sings as to what refers to recoupment and its importance. In the end, the ECJ discarded the requirement and constructed a rule of general application. The dissonant tone between the Courts and the Commission creates a negative environment. The Commission, although probably convinced of the advantages of a recoupment test, due to the lax requirements of the Courts may be satisfied with lax assessments as well, which would be coincident with the requirements of the Courts. It is indeed a safe spot. The drawback, at least for the Co issio s effo ts to a ds a effe ts-based approach focused on consumer welfare, would be that unless the recoupment factor is included in the test for predation, a consumer welfare goal faces the risk of not being correctly promoted. Also, the situation fosters an environment of uncertainty that affects the interests of competitors, addressees of EU Competition Law. The feasibility of a recoupment requirement for the predation test ultimately depends on the objective pursued by the EU Competition Law. We believe that the lack of convergence between the Commission and the Courts is largely the consequence of different goals promoted by different cases when applying EU Competition Law. The case law of the EU Courts is not conclusive as to what objective shall be promoted. Such was the case in GSK, if we compare the ruling of the GC with the ruling of the ECJ. Then, in PD the ECJ seems to have followed a consumer welfare standard, while in cases of predation like Tetra Pak II and France Télécom, consumer welfare could have been left behind. While the Commission in the Guidance seems to be following a consumer welfare standard, the case law of the EU Courts (in predation cases) seems to be following an economic freedom standard, or either the latter objective is being used as an instrument to promote the former. This goes i li e ith ou stud of Go se s o t i utio i Chapte III, hi h e al ead addressed in this chapter. The point is clear as to the connection between recoupment and consumer welfare when applying the predation test. However, at this point we want to make a few clarifications: 111 Although we believe that the Commission in the Guidance prefers a consumer welfare goal and that the concept of an anticompetitive foreclosure entails the examination of the likely harmful effects on consumers, the Guidance still shows that other objectives can be promoted (for instance the protection in certain cases of less-efficient competitors seem to promote economic freedom). We have to recognize that this situation may follow rulings of the Courts such as GSK, Tetra Pak II, or France Télécom. Additionally, in cases of predation, the Guidance states that the anticompetitive foreclosure test could consider other elements such as whether and how the suspected conduct reduces the likelihood that competitors will compete. The Guidance provides examples of these situations, such as engaging in predation to influence the expectations of potential entrants and thereby deter entry (para. 68). We studied similar situations when we addressed the factors involved in the feasibility of predation, such as the reputation of the dominant undertaking and the barriers to enter the market. We consider, as we believe that Hemphill (2001) does, that this factors could be included in a recoupment test based on the structure of the market. In such a way, it could be considered, for example, that part of the recoupment of the losses of the competitor consists on the fierce competition reputation, with repercussions on other markets, which will deter new entrants as a result. However, it may also happen that the Commission, even if recoupment is not feasible, considers independent factors that could reduce the likelihood that competitors will compete (such as the p edato s eputatio i othe a kets , a d ith this a d othe ele e ts that dete i e likel egati e effects on competitors, could deem the conduct as an abuse. We believe that in these cases the Commission could change the objective pursued either to the overarching undistorted competition objective or to economic freedom, a possibility that the Courts could approve, considering the case law we have studied. All in all, we highlight that, even thought we believe that the Guidance provides an adequate environment for recoupment when the goal pursued is consumer welfare, the Guidance itself may contradict the possibilities for the inclusion of the requirement in the predation test, when other Competition Law objectives take precedence over consumer welfare. The inclusion of recoupment to the predatory pricing test will most likely result in the enhancement of consumer welfare. A failed predatory strategy benefits consumers, at least in terms of price. Thus, an impossibility to recoup losses will transform into an impossibility to set monopolistic prices. Therefore, consumers will benefit from the low pricing campaign during the predation stage; coupled with the fact the dominant undertaking was not able to recoup its losses. As a result the consumers will not be negatively affected. At first sight, this seems consistent with the consumer welfare definition fixed by the GC in GSK and in Österreichische Postsparkasse AG, which refers to the well-being of final consumers of goods and se i es. It e e akes se se he o e ted fo e a ple ith the Dut h Co petitio Autho it s (2012) views, according to which consumer welfare includes quality, range and service, all of which are of course, related to price. 112 Ho e e , the last situatio ust e o side ed i the o te t of the Guida e s ie of the possi le benefits that consumers could obtain from healthy competition: lower prices, better quality, and a wider choice of new or improved goods and services (para. 5). We must also consider that according to the ECJ in France Télécom, a possible negative effect on consumers is the lack of choices resulting from the market foreclosure. Also in paragraph 22 of PD, the Court takes into account price, choice, quality or innovation, when defining elements that can influence or determine that players in the market are less attractive to consumers. In sum, the Commission and the ECJ find benefits to consumers also in other factors, and not only in price. To this matter, a good argument in favor of recoupment can be that the reasons a dominant undertaking is not able to recoup losses is probably due do the fact that the entry and cost barriers are low, or that the effects of the predation stage are not sufficient, and therefore other competitors can enter the market or other competitors can re-take what is left from the business of the salient competitor(s). As a result, not only monopolistic prices were not set, but the level of competition was kept, and this allows for benefits to consumers also in terms of choice and possibly, quality, and innovation. We stated that, in cases like France Télécom, the ECJ could have been wrong due to the fact that the loss to consumers in terms of choice could have been compensated with the benefits to consumers in terms of prices. Therefore, even though we still believe, as economic theory points out, that without recoupment a predation strategy will produce benefits to consumers; we believe that any recoupment test under EU Competition Law must consider not only price, but all the factors that benefit consumers, according to the ECJ. As a final remark, we believe that it is desirable that the Commission and the EU Courts find a conclusive common stance in regards to the preferred objective of EU Competition Law when dealing with cases of predatory pricing. The dissonant tone that we pointed out in France Télécom has an effect on the degree of legal certainty for the market players. Dominant undertakings should have a clearer view about whether or not their business strategies will be considered abusive. Competitors and other players in the market should also have a clearer view of the rules of Competition Law, either to protect their interests or to measure their current and future actions. This state of things may even be worsened if in the Guida e o e ide tifies a o su e elfa e sta da d hile the Co issio s de isio s, o the Cou ts uli gs p o ote othe o je ti es. The situatio has legal e tai t -related repercussions and could even affect the economic interests of the EU with regard to the interests of potential competitors in other regions or geographic markets. An environment of lack of legal certainty may drive businesses away from the EU, e.g., when other markets offer better conditions in this respect, probably coupled with other benefits. Consequently, this could amount to considerable losses and could affect important interests, including others beyond the ones we have studied. The situation is not simple due to the great importance of objectives such as market integration and even economic freedom, which under EU Law are recognized with no less value than consumer welfare. Nonetheless, this assertion does not mean that helping elements, such as the Guidance, cannot be clarified, sharpened, improved, or even reproduced. This should be done while seeking a common 113 ground with the EU Judiciary, where the result is efficient in providing the market players with more certainty while doing business in the EU. 114 List of cases General Court (GC):     Case T-340/03, France Télécom v. Commission [2007] Case T-83/91 – Tetra Pak , [1994] ECR II-755 Joined cases T-213/01 and T-214/01 Öste ei his he Postspa kasse AG a d Ba k fϋ A eit u d Wirtschaft AG v. EC Commission [2006] ECR II-1601 Case T-168/01 GlaxoSmithKline v. Commission [2006] ECR II-2969 Court of Justice of the European Union (ECJ):         Case C-202/07P, France Télécom v Commission, [2009] ECR I-2369 Case 85/76 Hoffmann-La Roche v Commission [1979] ECR 461 Case C-62/86, AKZO Chemie BV v. Commission [1991] ECR I-3359 Case C-209/10, Post Danmark A/S v Konkurrenceradet [2012] Case C-333/94P, Tetra Pak International SA v. Commission [1996] ECR I-5951 Case C-189/02 P, Dansk Rorindustri and others v. Commission, [2005] E.C.R. I-5425 Joined cases C-501/06 P et al, GlaxoSmithKline Services and Others v Commission and Others [2009] Case 6-72, Europemballage Corporation and Continental Company Inc. v Commission [1973] Unites States Supreme Court (US Supreme Court):  Brooke Group Ltd v Brown &Williamson Tobacco Corp 509 US 209 (1993). Consulted bibliography 1. Buttigieg, Euge e; Co petitio La : “afegua di g the Co su e I te est , Wolke s Klu e La & Business (2009). 2. Go de , D.G.; EC Co petitio La , O fo d . 3. Lia os, Kokko is a d othe s; The ‘efo of EC Co petitio La , Ne Challe ges , Klu e Law International (2010). 4. Oste ud, E ik; Ide tif i g E lusio a A uses Do i a t U de taki gs u de EU Co petitio La , Wolte s Klu e La & Busi ess . 5. Lo dahl Go se , Liza; A P i ipled App oa h to A use of Do i a e i Eu opea Co petitio La , Ca idge U i e sit P ess 6. Va Bael & Bellis, Co petitio La of the Eu opea Co u it , Wolke s Klu e La & Busi ess (2010). 7. Ez a hi, A iel; EU Co petitio La : A a al ti al Guide to the Leadi g Cases , O fo d a d Po tla d, Oregon (2010). 115 8. Hilde a d, Do is; The ‘ole of e o o i a al sis i the EC o petitio ules , Wolke s Klu e La & Business (2009). 9. Niels, Jenkins, and Kavanagh, Economics for Competition Lawyers , Oxford University Press (2011). 10. Bishop, Simon and Walker, Mike, The Economics of EC Competition Law: Concepts, application and measurement , Second Edition, Thompson, Sweet & Maxwell (2002). 11. Mateus, Abel M. and Moreira, Teresa, Competition Law and Economics, Advances in Competition Policy Enforcement in the EU and North America , Edward Elgar Publishing Limited (2010). 12. Motta, Massimo, Competition Policy, Theory and Practice , Cambridge University Press (2005). 13. Alison Jones and Brenda Sufrin, EU Competition Law , Oxford University Press (2011). 14. Carles Esteva Mosso, The more economic approach paradigm –an effects-based approach to EU Co petitio Poli , p. 11-22 in Jurgen Basedow and Wolfgag Wurmnest, Structure and Effects in EU Competition Law, Studies on Exclusionary Conduct and State Aid , Walters Klumer, Law and Business (2011). 15. Pe elope Papa d opoulos, The I ple e tatio of a Effe ts-Based Approach under Article 82: P i iples a d Appli atio , p. -433 in Ioannis Lianos and Ioannis Kokkoris, The Reform of EC Competition Law , Wolters Klumer, Law and Business (2010). 16. Da ie Ge adi , Is the Guida e Pape o the Co issio s E fo e e t P io ities i E fo i g A ti le TFEU useful? , p. -52 in Federico Etro and Ioannis Kokkoris, Competition Law and the Enforcement of Article 102 , Oxford University Press (2010). 17. A iel Ez a hi, The Co issio s Guida e o A ti le EC a d the Effe ts Based App oa h- Legal and Practical Challenges , p. 51-65 in Ariel Ezrachi, Article 82 EC: Reflections on its Recent Evolution , Hart Publishing, Oxford and Portland, Oregon (2009). 18. Orit Dayagi-Epstei , The e olutio of the otio of Co su e I te est i Light of the Modernization of A ti le EC , p. -86 in Ariel Ezrachi, Article 82 EC: Reflections on its Recent Evolution , Hart Publishing, Oxford and Portland, Oregon (2009) . 19. Ioa is Lia os, Catego i al Thi ki g i Co petitio La a d the Effe ts- ased App oa h i A ti le 82 EC , Studies of the Oxford Institute of European and Comparative Law, p. 19 et seq. Vol. 12. Oxford and Portland, Oregon (2009). 20. Joachim Mest ä ke , The de elop e t of Ge a a d Eu opea o petitio la ith spe ial efe e e to the EU Co issio s A ti le Guida e of , p. -62 in Lorenzo Federico Pace, Eu opea Co petitio La : The i pa t of the Co issio s Guida e o A ti le 2 , Edward Elgar Publishing, Inc. (2011). 21. Ernst Vale ti e Ko ah, Guida e o the Co issio s E fo e e t P io ities i appl i g A ti le to Abusive Exclusionary Conduct by Dominant Undertakings: from protecting freedom to enter a market to an efficient allo atio of esou es to i ease o su e elfa e , p. 8-24 in Lorenzo Federico Pace, Eu opea Co petitio La : The i pa t of the Co issio s Guida e o A ti le 102 , Edward Elgar Publishing, Inc. (2011). 22. Rosch, Thomas, Monopolies, Innovation, and Predatory Pricing: Observations on Some Hard Questions in the Section 2 Context , Antitrust 2010 Conference, Federal Trade Commission (2010). 23. Hemphill, Scott, The role of recoupment in predatory pricing analyses ; Stanford Law Review Vol. 53, No. 6 (Jul., 2001), pp. 1581-1612. 24. Phillip Areeda and Donald F. Turner, Predatory Pricing and Related Practices under Section 2 of the Sherman Act , Harvard Law Review, Vol. 88, No. 4 (1975), pp. 697-733, 116 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. http://www.contrib.andrew.cmu.edu/~kryukov/Predation/areeda75.pdf, consulted on November 2, 2012. Joseph Fa ell a d Mi hael L. Katz, Co petitio o p edatio : Co su e Coo di atio , “t ategi P i i g a d P i e Floo s i Net o k Ma kets , (Willey- Blackwell), The Journal of Industrial Economics, Vol. 53, No. 2 (Jun., 2005), pp. 203-231, http://www.jstor.org/stable/3569877, consulted on November 2, 2012. Usla , Malhot a, a d All i e, Predatory Pricing and Marketing Theory: Applications in Business-toBusi ess Co te t a d Be o d , Journal of Business-to-Business Marketing (2008), 13:3, 65-116, http://www.scribd.com/doc/53333162/1/AREEDA-TURNER-RULE-BENEFITS-AND-DRAWBACKS, consulted on November 7, 2012. Cha les P. “hi e , Predatory Pricing: The Retreat from the AVC Rule and the Search for a Practical alternative, B.C.L. ‘e . , http://lawdigitalcommons.bc.edu/bclr/vol22/iss3/2, consulted on November 8, 2012. Ni olas Petit, F o Fo alis to Effe ts? The Co issio s Co u i atio o E fo e e t Priorities in Applying Article EC , Antitrust Law & Policy eJournal 09/2009; DOI:10.2139/ssrn.1476082, http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1476082, consulted on November 9, 2012. Linda A.J. Sende , “oft La a d its i pli atio s fo i stitutio al ala e i the EU , Ut e ht La Review, published by Igitur (2004), http://www.utrechtlawreview.org/index.php/ulr/article/viewFile/9/9, consulted on November 19, 2012. Lis e Fi ell, The A ti le ‘efo : I easi g the i e ti es to o pl ? , Wolters Kluwer Law & Business (2009), p. 283-304. Mau i e E. “tu ke, ‘e o side i g A tit ust Goals , Boston College Law Review, Vol. 53, p. 551-629, (2012), http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1904686, consulted on November 26, 2012. Ni olas Petit, Recent Article 102 TFEU Case-la , , http://chillingcompetition.com/2012/04/24/recent-article-102-tfeu-case-law/#comments, consulted on December 5, 2012. Wessel Geu se , Post Da a k: does the ECJ take the effe ts ased app oa h fu the tha a e e price/cost-test a d does it o lige the atio al judge to appl that effe ts ased app oa h e u ? (2012), http://europeanlawblog.eu/?p=329, consulted on December 5, 2012. Ekaterina Rousseva and Mel Marquis, Hell Freezes Over: A Climate Change for Assessing E lusio a Co du t u de A ti le TFEU , http://jeclap.oxfordjournals.org/content/early/2012/10/25/jeclap.lps059.full?keytype=ref&ijkey=dN 3TqBTAFFbveqz#content-block, consulted on December 5, 2012. Howard Rosenblatt, Héctor Armengod and Andreas Scordamaglia-Tousis, Post-Danmark: Predatory Pi i g i the Eu opea U io , http://www.globalcompetitionreview.com/reviews/47/sections/162/chapters/1820/monopolisatio n/, consulted on December 5, 2012. The Nethe la ds Co petitio Autho it NMa , Co petitio E fo e e t a d Co su e Welfa e: Setting the Agenda Preamble , http://www.icn-thehague.org/page.php?id=129 (2012), consulted on January 10, 2013. 117 37. Pi a Ak a , Co su e Welfa e a d A ti le EC: P a ti e a d ‘heto i , E“‘C Ce t e fo Competition Policy and Norwich Law School, University of East Anglia (2009). 38. Pi a Ak a , The Co su e Welfa e delusio i Gla o“ ithKli e: A espo se to B u e L o s , Centre for Competition Policy (2009), http://competitionpolicy.wordpress.com/2009/11/25/the%E2%80%98consumer-welfare%E2%80%99-delusion-in-glaxo-%E2%80%93-a-response-to-brucelyons/, consulted on January 11, 2013. 39. J. Thomas Rosch, Mo opso a d the Mea i g of Co su e Welfa e : A lose look at We e haeuse , Fede al T ade Co issio . 40. Atha assios “kou tis, Is o su e elfa e the o l a fo a d? A e-appreciation of o petitio la o je ti es a te po tas i oth U“ a d EU , Ce t e fo Commercial Law and Financial Regulation (2012), http://kluwercompetitionlawblog.com/2012/08/10/is-consumerwelfare-the-only-way-forward-a-re-appreciation-of-competition-law-objectives-ante-portas-in-bothus-and-eu/, consulted on December 14, 2012. 41. Va Bael & Bellis, VBB o Co petitio La : Advocate General proposes the setting aside of the CFI s judg e t i F a e Télé o Wa adoo , Volu e , No. . 42. K J Cse es, The o t o e sies of the o su e elfa e sta da d , The Co petitio La ‘e ie , Volume 3 Issue 2 pp 121-173 (2007). 43. Report Prepared by The Unilateral Conduct Working Group for the International Competition Network, Report on Predatory Pricing , Kioto, Japan (2008). Opinions of Advocates General of the European Court of Justice      Opinion of AG Mazák in Case C- 202/07 P, France Télécom v Commission Opinion of AG Ruiz-Jarabo in Case C-333/94 P, Tetra Pak v. Commission Opinion of AG Mengozzi in Case C‑209/10 Post Danmark A/S v Konkurrencerådet Opinion of AG Fennelly in Joined cases C- 395/96 P and C- 396/96 P, Compagnie Maritime Belge Transports and Others v. Commission Opinion of AG Jacobs in Case C- 7/97 [1998] ECR I-7791, Oscar Bronner GmbH & Co. KG v Mediaprint Zeitungs- und Zeitschriftenverlag GmbH & Co. KG, Mediaprint Zeitungsvertriebsgesellschaft mbH & Co. KG and Mediaprint Anzeigengesellschaft mbH & Co. KG Decisions of the European Commission   Commission Decision, Wanadoo Interactive, Case No. COMP/38.233 Commission Decision, Deutsche Post AG, Case No. COMP/35.141 118