MASTERTHESIS
“The (desired) role of evidence of subjective intent in establishing an abuse of dominance in parallel trading cases in the light of its evidentiary value and legal certainty”
Name: Bernadette van der Blij
Mastertrack: European Competition Law and Regulation Supervisor: Or Brook
Abstract
The role of intent in abuse of dominance cases is controversial. There is both discussion about what its role actually is and about what its role should be. Clarification on this point is beneficial for legal certainty and moreover may help to evaluate the approach taken by the EU institutions. This thesis is specifically concerned with evidence of subjective intent: evidence that is found in internal documents of undertakings stating the aim or the strategy to exclude competitors, engage in parallel trade or do something else abusive. At first glance it seems that the European Commission and the EU Courts only see a limited role for intent, by continuously referring to abuse as an objective concept and stating that intent evidence may only play a supportive role in finding an abuse. However, it is claimed that not all case law supports this contention and this characterisation of intent is disputed in literature as well, as some hold that mere subjective intent actually has been sufficient to establish abuses.
This thesis looks particularly at the use of evidence of subjective intent to prove intent in parallel trading cases. In such cases an abuse of dominance may be found when an undertaking engages in conduct aimed at restricting parallel trade. Protecting parallel trade and thus taking measures of enforcement against those who jeopardise it, has always been high on the Commission’s priority list. It is concluded that reliance on evidence of subjective intent is inconsistent at first glance, but that for both the Commission and the Courts a line of development is detectable. Moreover, at some point both the Commission and the Courts have sought recourse to evidence of subjective intent to find an abuse. Especially in AstraZeneca the Commission relied to a large extent on such evidence, something which was not followed by the Courts. Rather, the General Court steers away from using intent especially as a factor constituting abuse, bringing the case law back in line with the rhetoric of abuse as an objective concept.
The inconsistent decisional practice and jurisprudence raises questions regarding the desirability of subjective intent evidence. This thesis assesses the desirable role of evidence of subjective intent in parallel trading cases in light of its evidentiary value and legal certainty. It claims that an approach where subjective intent evidence plays a central role is problematic, mainly because mere intentions have no intrinsic value to classify a certain conduct as abusive. Over-reliance on evidence of subjective intent may therefore lead to over-enforcement and jeopardise legal certainty, as it provides no guidance for undertakings
regarding what is allowed on the market. Moreover, it is not in line with guidance instruments of the Commission and the rhetoric that abuse is an objective concept.
Table of contents
1. Introduction p. 4
1.1. Research question and definitions p. 4
1.1.1. Objective and subjective intent p. 4
1.1.2. Parallel trade p. 5
1.1.3. Benchmarks: evidentiary value and legal certainty p. 7
1.1.3.1. Evidentiary value p. 8
1.1.3.2. Legal Certainty p. 9
1.2. Methodology and structure p. 9
Chapter 2. Intent in EU abuse of dominance cases in general p. 12
2.1. Legislation and guidance p. 12
2.2. Jurisprudence p. 13
2.2.1. Predatory pricing p. 13
2.2.2. Other cases p. 14
2.3. Academic interpretations of what the law is p. 16
2.4. Conclusions: the role of intent in abuse of dominance cases in general p. 17
Chapter 3. The role of subjective intent evidence in parallel trading cases p. 19
3.1. The Commission p. 19
3.1.1. Minimal role for intent in early cases p. 19
3.1.2. Increasing role for intent over time p. 20
3.1.3. Reliance on evidence of subjective intent is introduced p. 22
3.1.4. Conclusion regarding the Commission’s approach p. 23
3.2. The Courts p. 24
3.2.1. Inconsistent use of intent in early cases p. 24
3.2.2. A general line develop p. 25
3.2.3. Court steers away from using intent p. 27
3.2.4. Conclusion of the Court’s approach p. 28
3.3. Conclusion: the role of evidence of subjective intent in parallel trading casesp. 28
Chapter 4. The desired role of evidence of subjective intent in parallel trading p. 31 cases
4.1. Evidentiary value p. 31
4.1.1. Proof of evidentiary value in case law p. 31
4.1.2. Fundamental problems with evidentiary value p. 32
4.1.3. Conclusion evidentiary value p. 33
4.2. Legal certainty p. 33
4.3. Conclusion: desired role of evidence of subjective intent p. 34
5. Conclusion p. 36
1. Introduction
This thesis assesses the (desired) role of evidence of subjective intent in establishing an abuse of dominance in parallel trading cases. The role of intent under Article 102 TFEU is controversial and there is both discussion on what the role of intent actually is and what its role should be. The goal of this thesis is therefore dual: it first describes the role of evidence of subjective intent in parallel trading cases and subsequently answers the normative question what its role should be in the light of the evidentiary value of such evidence and legal certainty. This introduction first explains the main elements of the research question and then discusses the methodology and provides the structure.
1.1. Research question and definitions
My research question is the following: what is the role of evidence of subjective intent in establishing an abuse of dominance in parallel trading cases and what should its role be in light of its evidentiary value and legal certainty?
1.1.1. Objective en subjective intent
As shown in this paragraph, intent has been used in various ways in literature and it is therefore important to be precise about the sort of intent that is discussed. A first important distinction is that between objective and subjective intent. Objective intent refers to intent that can be inferred from objective conduct of an undertaking (e.g. price-setting), other objective (economic) factors (e.g. other prices on the market) and the circumstances of the case.1 Authors do not agree whether the exercise of establishing objective intent is still an inquiry into an undertaking’s ‘state of mind’.2 On the one hand it is argued that recourse to objective factors is sought to establish that an undertaking had a certain state of mind, so to establish a subjective aspect on the basis of objective factors.3 On the other hand it is argued that intent is objectified to such an extent that an undertaking’s subjective state becomes wholly irrelevant.
1 Maria João Melícias, ‘The Use and Abuse of Intent Evidence in Antitrust Analysis’ [2010] 33(4) WC 569, 571.
2 Katharine Kemp, ‘A Unifying Standard for Monopolization: Objective Anticompetitive Purpose’ [2017] 39 HJIL 113, 117; Pinar Akman, ‘The role of intent in the EU case law on abuse of dominance’ [2014] 39(3) ELR 316, 318.
In this sense the goal of relying on objective factors is to strip the subjective aspect from the test for abuse completely.4
Subjective intent definitely does require an investigation into such ‘state of mind’. This thesis will be concerned with intent as a subjective element aimed at anticompetitive foreclosure. This subjective element refers to the motive guiding the undertaking to engage in certain conduct and its awareness of this conduct and its consequence of foreclosing competitors.5 In other words, by engaging in a certain way the undertaking knowingly aims to foreclose competition. Such intent is shown by internal documents (e.g. strategic plans, minutes of meetings, e-mails etc.) and statements of directors about their intention or motivation to engage in certain conduct.6 Such evidence of subjective intent is the core element of this thesis. It will be concerned with the role of such evidence in establishing an abuse and thus not in other stages of the enforcement process (e.g. the fining-stage).
1.1.2. Parallel trade
Parallel trade occurs where – after a manufacturer has put a good on the market in one country – this good is subsequently traded to a second country by another party.7 The reasons for parallel trade are usually price differences between the countries, making it worthwhile for a parallel trader to import goods from a lower-priced country, to a higher priced country. For the manufacturer this however means that it may sell less of its good in the second country, since part of the demand is already fulfilled by the parallel imports. The manufacturer may therefore want to impede such parallel trade.8
Traditionally parallel trade enjoys significant protection from EU institutions and impediments to it continuously draw the attention of EU enforcers, especially since they may
4 Kemp (n 2) 117.
5 See for a similar use of intent: Pier Luigi Parcu and Maria Luisa Stasi ‘The role of intent in the assessment of conduct under Article 102 TFEU’ in Pier Luigi Parcu, Girogio Monti and Marco Botta (eds), Abuse of
Dominance in EU Competition Law (Elgar 2017), 15.
6 This is the general consensus in literature, see for example Geoffrey A. Manne & E. Marcellus Williamson, ‘Hot Docs Vs. Cold Economics: The Use and Misuse of Business Documents in Antitrust Enforcement and Adjudication’ [2005] 47:609 ALR 609; Akman (n 2) 326; Melícias (n 1) 582-4. The ECJ however has held in Case C- 202/07 P France Télécom [2009] ECR I-02369, para 98 that internal documents may be seen as objective. This is further discussed under section 2.2.1..
7 Christopher Stothers, Parallel Trade in Europe: Intellectual Property, Competition and Regulatory Law (Hart Publishing 2007) 2.
jeopardize European market integration.9 When the EU competition law regime was introduced, furthering the internal market was seen as one of the important objects it should serve.10 This explains the protective stance EU institutions take towards parallel trade. Most cases have been brought under Article 101 TFEU and the European Court of Justice (‘ECJ’) has held that agreements aimed at limiting parallel trade constitute a by object infringement of Article 101 TFEU.11 This clearly shows the Courts (General Court (‘GC’) and the ECJ) have a strong negative stance regarding the issue. This harsh condemnation of impediments to parallel trade is criticised in literature. It is argued that this condemnation lacks a solid economic foundation and that parallel trade may actually have a welfare-reducing effect, amongst other things because it lowers the incentive to innovate for the undertaking producing the parallel traded goods.12
The EU institutions however have been going after those impeding parallel trade continuously, also under Article 102 TFEU. An abuse of a dominant position in this context may be found when a dominant undertaking prevents or puts in place obstacles to parallel trade. Such abuses already caught the eye of the European Commission (‘Commission’) early in the history of the European market integration.13 Recent enforcement has focussed strongly
on the pharmaceutical sector14 and most academic contributions that have been written about parallel trade under Article 102 TFEU concern pharma.15 The latest act of enforcement took place on 30 November 2017 when the Commission issued a statement of objections to beer brewer AB InBev for pursuing a deliberate strategy to prevent supermarkets and wholesalers from parallel trading certain brands of beer.16
9 Lazaros Grigoriadis, ‘The Application of EU Competition Law in the Pharmaceutical Sector: The Case of Parallel Trade’ [2014] 25(1) EBLR 141; Case C-468/06 Sot. Lélos kai Sia [2008] ECR I-07139, para 37. 10 For an interesting and recent article regarding competition law and market integration, albeit with a focus on Article 101 TFEU see: Pablo Ibáñez Colomo, ‘Article 101 TFEU and Market Integration’ [2016] 12(4) JCLE 749.
11 Joined cases C-501/06 P, C-513/04 P, C-515/06 P and C-519/06 P GlaxoSmithKline Services and Others v
Commission and Others [2009] ECR I-09291, para 60.
12 Nicolas Petit, ‘The Economics of Parallel Trade – Iconoclast Views on a Dogma of EU Competition Law’ [2010] available online 4-5, 8.
13 For example Case 26-75 General Motors v Commission [1975] ECR 01367; Case 27/76 United Brands v
Commission [1978] ECR 00207; Case 226/84 British Leyland v Commission [1986] ECR 03263.
14 Case C-53/03 Syfait and others [2005] ECR I-04609; C-468/06 Sot. Lélos (n 9); Case C-457/10 P
AstraZeneca v Commission [2012] published in electronic Reports of Cases.
15 For example Grigoriadis (n 9); Panos Kanavos and Joan Costa-Font, ‘Pharmaceutical parallel trade in Europe: stakeholder and competition effects’ [2005] 20(44) EP 751; some articles concern parallel trade in general: Petit (n 12).
16 European Commission Press Release, ‘Antitrust: Commission sends Statement of Objections to AB InBev for preventing cheaper imports of beer into Belgium’ (30 November 2017) http://europa.eu/rapid/press-release_IP-17-5041_en.htm accessed 29 May 2018.
Parallel trade cases are particularly interesting to analyse for various reasons, other than that it is clearly high on the priority list of the Commission. To start, obstruction of parallel import is a unique abuse in the sense that it does not necessarily consist of one type of conduct. Commentators have tried to categorize the cases, for example by making a distinction between cases including refusal to supply17, resale restrictions18 and misuse of regulatory procedures.19 Others have discussed parallel trade issues under the miscellaneous category of abuses that are harmful to the single market or that hinder inter-state trade.20 It seems that specifically the intent to impede import is what makes an act abusive or what ties various acts together as abusive. Since intent seems to be such an important element in establishing such an abuse, it is interesting and important to assess how such intent may be proven.
All in all, parallel import cases provide a unique and interesting basis to assess the role of evidence of subjective intent, something which has not been done before specifically in relation to parallel trade. A study like this will both provide clarity on the issue for undertakings on the market and enable scholars to ask normative questions about the approach that institutions are taking. Seeing the recent activity of the Commission, it is definitely the right time to ask these questions now.
1.1.3. Benchmarks: evidentiary value and legal certainty
Academic literature on intent also focusses on the question what the role of intent should be, next to the discussion what its role is.21 Unfortunately, the EU institution have not (yet)
engaged in this normative debate on what the desired role of intent is, and some authors have called for more guidance on this issue.22 The last part of the thesis discusses the desired role of evidence of subjective intent in parallel trading cases on the basis of two benchmarks: the evidentiary value of such evidence and legal certainty. The choice for these benchmarks is further explained in the methodology section. It is relevant to engage in such normative
17 C-468/06 Sot. Lélos (n 9); Eurofix-Bauco v Hilti (Case IV/30.787 and 31.448) Commission Decision 88/138/EEC [1988] OJ L 65/19.
18 27/76 United Brands v Commission (n 12); Case 102/77 Hoffmann-La Roche v Centrafarm [1978] ECR 1139. 19 26-75 General Motors v Commission (n 12); 226/84 British Leyland v Commission (n 12); C-457/10 P
AstraZeneca v Commission (n 13). This distinction was made in Stothers (n 7) 243-67.
20 Richard Whish and David Bailey, Competition Law (8th edn, Oxford University Press 2015) 752-3, 808; Alison Jones and Brenda Sufrin, EU Competition Law – Text, Cases, and Materials (5th edn, Oxford University Press 2014) 584-90.
21 For articles in favour of using (some sort of) intent see amongst others: Kemp (n 2); Marina Lao ‘Reclaiming a Role for Intent Evidence in Monopolization Analysis’ [2004] 54(1) AULR 151. For articles against using intent see for example: Akman (n 2); Manne & Williamson (n 6)
questions, since it will be clear from the first chapters that the institutions take a different approach. It is therefore interesting to evaluate the approaches in light of the benchmarks in order to say something useful about a ‘right’ approach.
1.1.3.1. Evidentiary value
The first benchmark is the evidentiary value of evidence of subjective intent. It focusses on the question to what extent such evidence is indicative for the abusive nature of certain conduct and if such evidence is helpful is establishing an abuse. The core question in literature is whether such evidence can disambiguate ambiguous conduct and thus have value in proving an abuse and enforcing competition rules. Authors in favour of using subjective intent argue that it can make the ‘ambiguous unambiguous’, e.g. it can be used where it is unclear if something is merely fierce competition on the merits or actual abusive conduct. Intent can be a tool to distinguish between the two and to predict possible effects.23 This argument is especially made in situations where economic evidence comes up inconclusive.24 Following the same logic, a test of intent can then also be a check against over-enforcement by making sure that fierce competition is not mistaken for prohibited conduct.25
Opponents of this theory argue that instead of rendering the ambiguous unambiguous, subjective evidence is very ambiguous in itself, even more so than the economic evidence it tries to clarify.26 The very essence of competition is that competitors do not like each other and moreover statements allegedly proving anti-competitive intent may be taken out of context.27 Therefore, subjective intent evidence has no value to establish abuses and reliance on intent evidence may lead to over-enforcement.28 In cases where intent evidence is clearly there, reliance on such evidence could by an ‘easy way out’ for the enforcer, leading to false positives.29
23 Richard A. Posner, ‘Antitrust in the New Economy’ 68 ALJ 925-943 [2001]; Akman (n 2) 467. 24 Manne & Williamson (n 6) 41; Colm O’Grady, ‘The Role of Exclusionary Intent in the Enforcement of Article 102 TFEU’ 37(4) WC 459. O’Grady (n 24) 470; Melícias (n 1) 577-8; Lao (n 21) 192-5.
25 O’Grady (n 24) 481.
26 Manne & Williamson (n 6) 41-3; Maurice E. Stucke, ‘Is Intent Relevant?’ [2012] 8 JLEP 801, 818; Melícias (n 1) 578-9.
27 Kemp (n 2) 205-7; Bavasso A, ‘The role of intent under Article 82 EC: from “flushing the turkeys” to “spotting lionesses in Regent’s Park”’ [2005] 26(11) ECLR 616; Melícias (n 1) 577.
28 Melícias (n 1) 576.
This thesis assesses both sides of the argument in relation to parallel trading cases and eventually sides with the opponents in holding that evidence of subjective intent has little evidentiary value.
1.1.3.2. Legal certainty
Secondly, there is the issue of legal certainty: does reliance on subjective evidence undermine legal certainty in itself and/or does the current approach regarding subjective intent evidence undermine legal certainty? Authors answering this question in the affirmative point to the fact that it is not at all clear in what cases intent evidence will be used and that recourse to such evidence is fundamentally arbitrary in nature.30 They also point at the inherent paradox of using subjective notions such as intent, where the Courts have made clear that the finding of an abuse should be based on objective factors31 and the fact that use of intent is not in line with the effects-based approach used by the Courts.32
Proponents of using an intent or object-based tests do not necessarily argue that such a test improves legal certainty, but rather hold that other approaches – such as an effects-based approach – are more prone to create legal uncertainty.33 Moreover, they downplay the contention that intent evidence would be used arbitrarily, amongst other things by pointing out that this is an issue with other means of evidence as well.34
This thesis shows that from the parallel trading cases it becomes clear that sole reliance on such evidence indeed creates uncertainty.
1.2. Methodology and structure
The second chapter performs a literature law review of the role of evidence of subjective intent in abuse of dominance enforcement in general. The goal is to describe the law as it stands from an internal perspective. It assesses if a general rule can be established or if there is unclarity e.g. a gap that needs to be filled. It does this first by looking at legislation, guidance instruments and case law. Subsequently, – in order to complete the picture – it
30 Richard A. Posner, Antitrust Law (2nd edn, University of Chicago Press 2001), 214 31 Akman (n 2) 336.
32 Bavasso (n 27) 623; Thomas Janssens, ‘The Commission Guidance on Predation: A Cautious Step In the Right Direction?’ [2009] 1 GCP available online 12; Akman (n 2) 331.
33 Kemp (n 2) 154; Melicias (n 1) 179. 34 O’Grady (n 24) 467.
assesses how other scholars have interpreted the role of intent. Including interpretations of scholars is relevant since the law in the books e.g. the law as the institutions propose it to be in legal instruments and guidance, is different from the law in action e.g. what actually happens in cases before the Courts.35 To be clear, this is still descriptive. It is an inquiry into what the law is, but it gives a complete picture of the law by looking beyond what is merely put forth in standard formulations by EU institutions. This chapter concludes that in general the role of intent is unclear and inconsistent: there is thus a gap that needs to be filled and this thesis fills this gap for parallel trading cases in particular.
The second chapter serves as the basis for going into parallel trade cases. The goal of the third chapter is to describe the role of subjective intent evidence in parallel trading cases by taking a doctrinal approach. It looks into all Commission decisions and case law of the Courts regarding parallel trade under Article 102 TFEU in a chronological manner and assesses whether subjective intent played a role. Since there is a limited amount of parallel trading cases it is possible to include all. It separately discusses the Commission’s approach and the Court’s approach, as they differ. The assessment into the role of intent has two stages. Firstly, it looks if there is a reference to intent in the cases at all, either being subjective, objective or undefined. This is not only limited to the word ‘intent’ but may include any reference to an intentional element, such as a ‘plan to eliminate competition’ or an ‘overall strategy’. The second stage establishes how this intent is subsequently proven and in particular if it is proven by relying (solely) on subjective intent evidence. It is concluded that also in parallel trading cases subjective intent has played an inconsistent role, but that a general line of development can be detected for the Commission and the Courts separately.
As it becomes clear that different approaches have been taken by the Commission and the Courts, it is interesting to shed some light on what a ‘right approach’ might look like. The fourth chapter therefore assesses the desirable role of evidence of subjective intent in parallel trading cases on the basis of two benchmarks: evidentiary value and legal certainty. It thus takes a normative approach. The choice for the benchmarks is based on the academic literature regarding the desirable role of intent and they are – more importantly – intrinsically relevant. First of all, evidence simply is what enforcers rely upon to decide whether conduct is abusive or not. Therefore questions asking whether certain evidence should be used in light of
35 Commentators have for example found that intent has played a hidden role in many cases. Duncan Sinclair, ‘Abuse of dominance at a crossroads – potential effect, object and appreciability under Article 82 EC’ [2004] 25(8) ECLR 491, 491; Bavasso (n 27) 616; Akman (n 2) 321.
its value are automatically relevant to assess whether such evidence should be used at all. It is concluded that evidence of subjective intent has little intrinsic evidentiary value and therefore such evidence should – at the most – play a supporting role in finding an abuse. Secondly, a certain degree of legal certainty is important in competition law because undertakings need to be able to assess what they can and cannot do on the market. Too much unclarity may restrict competition as undertakings may refrain from conduct that is perfectly legal just to be on the safe side or may lead to illegal behaviour simply because undertakings do not know what the boundaries are. The conclusion is that the parallel trading cases show that recourse to evidence of subjective intent harms legal certainty, since it provides no guidance to undertakings, is arbitrarily applied and is moreover not in line with the rhetoric that abuse is an objective concept.
2. Intent in EU abuse of dominance cases in general
This chapter performs a literature law review of the role of evidence of subjective intent in abuse of dominance cases in general. It starts by assessing EU primary and secondary legal instruments and jurisprudence of EU courts and seeks whether a general line can be established. For this purpose it goes into academic literature interpreting the role of intent. It concludes that the role of subjective intent is unclear. It is either very case specific or applied in an ad hoc and inconsistent manner.
2.1. Legislation and guidance
There is no reference to intent or any other expression of a subjective element in the EU treaties or any secondary source of EU law. Article 102 TFEU is silent on the issue and none of the secondary law sources relating to competition law – most importantly Regulation 1/2003 – refer to intent.36 From this it seems that intent does not play a role in the substantive test to establish an abuse. Intent is only mentioned in relation to the fining stage, as fines may be imposed on undertakings if they ‘intentionally or negligently’ abused their dominance.37 Intent may be an aggravating factor when fixing the amount of the fine.38 However, this thesis concerns the role of intent in the substantive test, so no further attention is paid to fining. Since the primary and secondary legislation provide no answer, the next step is to look at soft law instruments such as guidance by the Commission. A general reference to a subjective element is made in the Commission’s 102-Guidance, where ‘direct evidence of any
exclusionary strategy’ is listed as a relevant aspect into assessing whether allegedly abusive
conduct is likely to lead to anti-competitive foreclosure.39 According to the Commission such evidence can be helpful in interpreting the conduct of the undertaking concerned. This
36 Consolidated Version of the Treaty on the Functioning of the European Union art. 102, 2008 OJ C 115/47; Council Regulation (EC) No 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, 2002 OJ L 1.
37 Regulation 1/2003 (n 36) art. 23(2).
38 Regulation 1/2003 (n 36) art. 23(2); Bavasso (n 27) 8; O’Grady (n 24) 482-3.
39 Communication from the Commission: Guidance on its enforcement priorities in applying Article 82 of the EC Treaty to abusive exclusionary conduct by dominant undertakings, 2009 OJ C 45, para 20.
wording implies that intent is not a constitutive element as such, but can ‘guide’ the Commission towards finding an abuse. This general reference applies to all abuses.
The guidelines also discuss intent in predatory pricing in particular.40 The test is whether a dominant undertaking is deliberately incurring losses or foregoing profits (the ‘sacrifice’) in order to foreclosure competition. The Commission holds that it is possible in some cases to rely upon internal documents showing a predatory strategy in order to show this sacrifice.41 Possible foreclosure then still has to be shown and here the Commission does not explicitly refer to subjective intent.
Any reference to intent is thus lacking in primary and secondary EU law sources and guidance instruments only provide either very general statements regarding intent or abuse-specific tests. It is therefore necessary to look whether the Commission’s decisional practice and jurisprudence of the EU Courts tell more.
2.2. Jurisprudence
The Courts have not been consistent in their rulings regarding intent.42 A good starting point to explore the case law is Hoffman-la Roche where the Court explained that the concept of abuse is objective.43 From this case it seems that there is no room for a subjective element in establishing an abuse. However other cases show a different approach. These can be separated into two categories: firstly predatory pricing cases, in which intent plays an explicit role, and secondly all other cases, in which intent is applied in an ad hoc manner.
2.2.1. Predatory pricing
Predatory pricing is the only category of cases in which intent is explicitly laid down in the standard abuse-test. The Commission in AKZO found that intent may be required to establish an infringement, where low pricing could be explained in various ways. The Commission moreover held that evidence of such intent was to be found in internal documentation ‘pointing to a scheme to damage competitors’.44
40 ibid para 66. 41 ibid para 69.
42 Parcu and Stasi (n 5) 18.
43 Case C-85/76 Hoffmann-La Roche v Commission [1979] ECR 00461 para 91.
The basic test was subsequently laid down by the ECJ in AKZO.45 Prices below average variable costs (AVC) can be regarded as abusive and prices below average total costs (ATC) but above AVC must be regarded as abusive if they are determined as part of a plan for eliminating a competitor’.46 In France Telecom the ECJ explained that the first part of the test assumes intention where prices are below AVC and that this is enough to find an abuse.47 For pricing above AVC but below ATC actual intent evidence is needed.48 In subsequent cases the Courts have referred to intent in various ways.49 That such intention can be proven by (mere) reliance on evidence of subjective intent is confirmed in France Telecom, where such evidence was enough to establish an abuse.50 The ECJ in this case refers to an undertaking’s internal documents showing intent as an ‘objective factor’, arguably to square it with the rhetoric that abuse is an objective concept.51 This however goes completely against the strongly established characterisation in literature of such evidence as subjective.52
2.2.2. Other cases
In other cases intent has been used less consistently. This is for example visible in other pricing-related cases, where intent has been used to different extents. In Compagnie Maritime
Belge the Court found that the undertaking’s selective price cuts intended to eliminate other
competitors from the market and were therefore abusive, thus relying on intent entirely.53
Irish Sugar also concerned selective pricing but here the GC held that a plan to eliminate
competition is only one of the circumstances on which an abuse can be found.54 In the selective pricing case Post Danmark no evidence relating to intention could be discovered, so an abuse could not be established on that basis.55 In Tomra the ECJ again held that intent is only one of the factors to take into account to establish an abuse.56
45 Case C-62/86 AKZO v Commission [1991] ECR I-3439, para’s 71-2. 46 ibid para 72.
47 C-202/07 P France Télécom (n 6) para 109.
48 Robert O’Donoghue and Jorge Padilla, The Law and Economics of Article 102 TFEU (1st edn, Hart Publishing 2006) 246.
49 See for a short overview: Jonathan Faull and Ali Nikpay, The EU Law of Competition (3 edn, Oxford University Press 2014), point 4.308.
50 C-202/07 P France Télécom (n 6) para 109. 51 ibid para 98.
52 See for example: Manne & Williamson (n 6) 609; Akman (n 2) 326; Melícias (n 1) 582-4.
53 Case C-395/96 P Compagnie Maritime Belge Transports and Others v Commission [2000] ECR I-01365; Parcu and Stasi (n 5) 23; O’Grady (n 24) 480.
54 Case T-228/97 Irish Sugar v Commission [1999] ECR II-02969, para 114 GC.
55 Case C-23/14 Post Danmark [2015] published in electronic Reports of Cases, para 29.
Also in non-pricing abuses intent has played an inconsistent and ad hoc role. The Commission held in AstraZeneca that otherwise permissible conduct may be abusive if it is part of a plan to eliminate competition.57 Also the earlier case ITT Promedia shows that otherwise legitimate conduct may be abusive if it is part of a plan to eliminate competition.58 Any conduct with anti-competitive purpose may thus be abusive according to this statement.59 This enables the institutions to apply such an intent-based approach in any case, however this does not happen consistently.
Complexes of conduct may also become abusive upon finding an underlying anti-competitive purpose. In Orange Polska the Commission argued that the undertaking developed a strategy to limit competition and indicated that these acts separately were not harmful, but that they together formed a pattern of abusive conduct (a single and continuous infringement).60 To prove this strategy, the Commission relied on internal documents.61 The Commission followed the same path in Slovak Telekom. It found that the undertaking pursued ‘an abusive
strategy to shut out competitors’ consisting of multiple elements together forming a single and
continuous infringement.62 Something similar may happen in the current investigations into AB InBev. In a press release the Commission held that: ‘AB InBev has abused this dominant
market position by pursuing a deliberate strategy to prevent supermarkets and wholesalers from [parallel trading]’. It is however unclear in these cases to what extent the Commission
solely uses references to intent evidence as a convincing rhetoric e.g. to build a storyline supporting the finding of an abuse, or whether such evidence actually is decisive for establishing the abuse.
From this section it becomes clear that also in jurisprudence the role of intent is unclear. Only in predatory pricing there seems to be consistency, as intent forms part of the standard test to find an abuse. However, this is very case-specific and therefore does not tell much about other cases. In all other cases, intent has played an ad hoc role, making it unclear when the Commission and Courts will use it in their assessment. Especially the notion that any conduct may become abusive when the underlying purpose is anticompetitive, enables to Commission
57 AstraZeneca (Case COMP/A.37.507/F3) Commission Decision 857/EC [2006] OJ L 332/24, para 327. 58 Case T-111/96 ITT Promedia v Commission [1998] ECR II-02937, para 30.
59 Parcu and Stasi (n 5) 43; Akman (n 2) 8-9.
60 Telekomunikacja Polska (Case COMP/39.525) Commission Decision 22 June 2011, para’s 148-64 and 707-711, 713.
61 ibid 148-155.
and Courts to have recourse to intent whenever they see fit, but not necessarily in a consistent manner.
2.3 Academic interpretations of what the law is
Authors have sought to detect a general line in the jurisprudence in order to determine what the law is. This section shows that there is discussion amongst scholars on this point, which proves the general claim that the role of (subjective) intent is unclear.
It is generally held that the concept of abuse is in reality not as objective as the Courts hold it to be.63 Some argue that intent serves as a proxy for effects, meaning that actual effects do not need to be shown if intent can be proven.64 If this argument is followed through, this would mean that intent – and maybe even subjective intent – could be enough to find an abuse. The argument that mere subjective intent is sufficient, is contested. It has been argued that neither from the treaty, nor from the judgment of the Courts such a thing can be inferred and that intent can never be sufficient of its own.65
This has not withheld commentators from arguing that mere subjective intent is enough to find an abuse, especially since the Courts wrongly equate intent with object.66 The GC held in
France Telecom that ‘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’.67 It subsequently held that if a dominant undertaking’s object is to ‘oust a competitor’ this may be an abuse even if the effect is not achieved.68 In this sense object is equated with intent, which in its turn can be proven by relying solely on internal documents.69 Relying merely on subjective intent and equating intent with object (and thus with effect) would definitely be at odds with the Commission’s nuanced approach in the Guidance, where proof of intent is only one factor that contributes to finding an abuse.70 This approach has also
63 Melícias (n 1) 580-1; Pinar Akman ‘The Tests of Illegality Under Articles 101 and 102 TFEU’ [2016] 61(1) TAB 84, 86-7.
64 Akman (n 2) 86; Melícias (n 1) 585; Parcu and Stasi (n 5) 18-25. The notion that effects do not need to be shown is not novel: in British Airways it was established that conduct can be abusive if capable of having restrictive effects. Case T-219/99 British Airways v Commission [2003] ECR II-05917, para 293.
65 O’Grady (n 24) 462-4. 66 Akman (n 2) 333-4.
67 Case T-340/03 France Télécom v Commission [2007] ECR II-00107, para 195. 68 ibid para 196.
69 Faull and Nickpay (n 49) point 4.308. 70 102-Guidance (n 36) para 20.
been confirmed by the Courts in for example Tomra. Mere reliance on subjective intent would especially be problematic in cases where intent is used to catch or paste together conduct that in itself is not considered abusive, but becomes so upon showing that it is part of a bigger exclusionary plan.71 If in such cases object is equated with effects and intent (such as an exclusionary strategy), which in its turn may be proven by evidence of subjective intent as found in internal documents, then solely on the basis of such evidence an abuse can be found. Intent then becomes a vehicle to censure exclusionary behaviour that is not caught by established abuses and as a glue that sticks together non-abusive conduct as abusive.72
Again, this theory that subjective intent evidence may play a sole decisive role in abuse of dominance cases is controversial. This clearly shows that there is unclarity regarding what the law is.
2.4. Conclusion: the role of subjective intent in abuse of dominance cases in general
To conclude, it is clear that only in predatory pricing cases intent plays an explicit standard role in constituting an abuse. In some of these cases evidence of subjective intent was enough to establish an abuse. In other cases intent has played an inconsistent role, as the Commission incidentally seeks recourse to intent. For example, it has held that (complexes of) conduct that would normally be legitimate may become abusive when the underlying purpose is exclusionary.
The subsequent question is if such intent or purpose can be proven by relying on internal documents showing intent. The Commission in its guidance and the Courts in their cases have held that for establishing an abuse subjective intent evidence is merely one of the factors that may play a role. This means that sole subjective intent is never enough to find an abuse, but that it may definitely help the Commission in finding an abuse. This contention has been heavily disputed by those who argue that mere subjective intent actually has been constitutive of abuses in the past, especially since effect is equated with object, which may be proven by mere reliance on evidence of subjective intent. It is interesting to see if this is true for parallel trading cases.
71 T-111/96 ITT Promedia (n 58); AstraZeneca (Case COMP/A.37.507/F3) (n 57); Telekomunikacja Polska (Case COMP/39.525) (n57), Slovak Telekom (Case AT.39523) (n 60).
All in all, it seems hard to establish a clear general line regarding the use of subjective intent evidence in abuse of dominance cases in general, since such evidence is used inconsistently and its actual role is disputed. A gap in the law is thus identifiable at this point: there is no real clarity regarding the role of subjective intent. The next chapter fills this gap for parallel trading cases.
3. The role of subjective intent evidence in parallel trading cases
The previous chapter established that intent has played an inconsistent role in abuse of dominance cases in general. It concluded that – according to the Commission and Courts – sole subjective intent can never by itself be constitutive of an abuse, which is disputed in literature. There is thus a clear gap in the law regarding the question what the role of subjective intent actually is. This chapter fills this gap for parallel trading cases, by researching for all these cases what role subjective intent actually played. As explained in the introduction this is relevant since tackling impediments to parallel trade is high on the priority list of the Commission and there seems to be an inherent intentional element in obstructing parallel trade. It is therefore important that the role of subjective intent in such cases is clear. At first sight, the Commission and Courts have inconsistently used subjective intent in parallel trading cases. In some cases it only plays a role in the objective justification stage,73 whereas in other cases intent plays a role in establishing an abuse.74 Moreover, in some cases intent plays a decisive role in constituting an abuse,75 where in others it is held that it may only be one of the factors to take into account.76 A last point of inconsistency is that in some cases intent is proven by objective factors,77 while in others recourse is sought to evidence of subjective intent.78 Even though there are many inconsistencies at first sight, this chapter shows that there is a line detectable for both the Commission and the Court as the case law develops.
This chapter proceeds by separately discussing the approach taken regarding subjective intent by the Commission and the Courts and eventually provides an integral conclusion. Table 1 (page 30) provides a useful overview of the role of intent in the cases discussed.
3.1. The Commission
73 E.g. Chiquita (Case IV/26699) Commission Decision 76/553/EEC [1976] OJ L 95/1, 14. 74 E.g. Case 226/84 British Leyland (n 13).
75 E.g. Case 226/84 British Leyland (n 13); Case 102/77 Hoffmann-La Roche (n 17); C-468/06 Sot. Lélos (n 9);
AstraZeneca (Case COMP/A.37.507/F3) (n 57).
76 E.g. Case T-321/05 AstraZeneca v Commission [2010] ECR II-02805.
77 E.g. General Motors Continental (IV/28.851) Commission Decision 75/75/EEC [1975] OJ 29/14. 78 E.g. Case 226/84 British Leyland (n 13); AstraZeneca (Case COMP/A.37.507/F3) (n 57).
3.1.1. Minimal role for intent in early cases
In early cases the Commission relies solely on objective factors to establish abuses and the role for intent is very minimal. In General Motors the Commission sees no role for intent and bases itself entirely upon objective factors. The case concerned mandatory type-approval of cars. General Motors (GM) was the only undertaking authorised to grant such approvals for its cars. The abuse consisted of charging excessive prices for approvals for parallel imported cars, with the effect of impeding parallel imports. According to the Commission GM ‘intentionally’ abused its dominant position, however it becomes clear from the rest of the decision that this relates to the fining stage and not to the actual finding of the abuse.79 The Commission establishes the abuse entirely upon objective factors, such as price differences, and focusses solely on the excessive pricing rather than a possible underlying purpose to curb parallel imports.
In United Brands the absence of intent is taken into account, but only in assessing whether the conduct is objectively justified. This intent is inferred from objective circumstances. One of the abuses regarded a clause prohibiting distributers to resell unripe bananas, which amounted to an export prohibition.80 This enabled United Brands Company (UBC) to maintain a system of price differentiation, which may have hindered parallel trade to higher-priced countries.81 UBC argued that the clause was objectively necessary and that it never intended to restrict competition.82 The Commission concludes that ‘[t]he fact that UBC's sales conditions
originally included an export prohibition can be explained only by its intention to forbid ripeners to resell across the frontiers of the Member States’.83 It is clear that the Commission rejected the objective justification as the underlying reason was not to ensure quality, but to prohibit export. However, the Commission already established the abuse itself on the basis of objective findings, without any reference to (subjective) intent.84
All in all, intent plays a very limited role in these early cases. It is only used in the objective justification stage (if at all) and is moreover objectively established. There is thus no recourse to subjective evidence at all.
79 General Motors Continental (IV/28.851) (n 77) art. 1, para’s 9, 11. 80 27/76 United Brands (n 12), para’s 130-62.
81 ibid at 252-3.
82 Chiquita (Case IV/26699) (n 73) 14; 27/76 United Brands (n 12), at 252 and para’s 142-4. 83 27/76 United Brands (n 16) at 253.
3.1.2. Increasing role for intent over time
Starting with Hoffman-LaRoche v Centrafarm, the role of intent starts to develop to one where the intent to impede parallel trade can be a decisive factor in establishing an abuse. One of the questions in this reference for preliminary ruling asked whether a prohibition of manufacturer Hoffman-La Roche (HLR) to resell repackaged parallel imported goods amounted to an abuse even if it could not be proven that the effect of maintaining price differences was intended.85 HLR wanted to prohibit this in order to protect its trademark. The Commission holds that otherwise legitimate exercise of a right may be abusive where it serves another objective than exercising the right itself. Such an ‘illegitimate’ objective is for example distribution control.86 The Commission does not clarify how presence (or absence) of such an objective may be proven and if evidence of subjective intent would play a role. But, it is clear that underlying purpose of conduct can be a decisive factor in finding an abuse. Also in British Leyland the Commission relies on the undertaking’s underlying purpose to restrict parallel trade, but how this underlying purpose is proven does not become clear from the short decision. This case concerned type-approval of cars and British Leyland (BL) held the monopoly for issuing these approvals for its vehicles. The Commission found that BL was abusing its dominant position in various ways: illegitimate refusal to issue certificates of conformity for imported cars, deliberate non-renewal of type-approval certificates for parallel imported cars and charging excessive prices for certificates of conformity for such cars.87 The Commission makes multiple references to internal documents, such as minutes of meetings and business records to prove that the facts took place.88 Intent also plays a role in the objective justification stage as BL argues that there are alternative reasons explaining the conduct. The Commission however holds that BL’s behaviour cannot be justified on these (or any) objective grounds, since it is clear that there was a desire to impede reimports. Moreover it holds that, BL’s justification arguments might have been credible ‘had its decision not been
taken at a time when the Dealer Council was urging BL to restrict the flow of reimported Metros [the cars] into the United Kingdom’.89 Again, the underlying purpose of impeding parallel import guided the Commission in finding an abuse, but the exact role of intent is
85 Case 102/77 Hoffmann-La Roche (n 17) 1143. 86 ibid at 1159-60.
87 BL (Case IV/30.615) Commission Decision 84/379/EEC [1984] OJ L 207/11, art. 1. 88 ibid para’s 14-8.
unclear. Moreover, it is not clear whether this underlying purpose may be proven by evidence of subjective intent.
This approach continues in Sot. Lélos, where the Commission holds that the aim of parallel trade may render conduct abusive, but does not clarify how such an aim is proven e.g. if this can be done by relying on evidence of subjective intent. The issue was that GlaxoSmithKline (GSK) started to withhold certain orders in order to prevent parallel export. The ECJ was asked whether a pharmaceutical company abuses its dominant position if it refuses to meet certain orders from wholesalers because these wholesalers are involved in parallel exports.90 Under general case law refusal to supply is not in itself an abuse, but only if there are additional circumstances rendering the conduct abusive.91 The question was if the intent to limit parallel trade is such a special circumstance. It was not disputed that the aim of GSK was to limit parallel exports.92 The Commission held that a refusal to supply with the aim of limiting parallel trade is in principle abusive, rejecting GSK’s argument that the decisive factor for abuse depend on the (objective) economic and regulatory context.93
The cases discussed in this section show a similar approach: the Commission sees intent to impede parallel trade as a decisive factor rendering conduct abusive. However, it is not clear if such intent has to be proven by objective factors or if recourse can be taken to evidence of subjective intent as found in internal documents.
3.1.3. Reliance on subjective intent is introduced
The role of evidence of subjective intent was clarified in AstraZeneca, which clearly shows a heavy reliance on such evidence. According to the Commission AstraZeneca (AZ) had abused its dominant position in two ways, the second of which is relevant here. In order to market medications an authorisation is needed and parallel traders can usually obtain such authorisation by reference to the one of the manufacturer.94 The alleged abusive strategy of AZ was to bring a tablet form of the medicine on the market and deregister the authorisation of the capsule version of that medicine.95 This strategy allegedly aimed to restrict parallel
90 C-468/06 Sot. Lélos kai Sia (n 9) para 28.
91 See for extensive explanation Stothers (n 7) 245-54. 92 C-468/06 Sot. Lélos kai Sia (n 9) para 36.
93 ibid para 31.
94 AstraZeneca (Case COMP/A.37.507/F3) (n 57) para’s 258-64. 95 ibid para 255.
trade since in this way parallel traders could no longer trade their version of the medicine, as the authorisation would fall through.
The Commission relies to a large extent on internal documents showing intent. It first of all states that it emerges from such documents that one of the key aims of AZ’s strategy is to obstruct parallel trade.96 The Commission then extensively discusses internal documents such as minutes, memoranda and strategy documents. These also include documents in which AZ discusses the effect of their conduct.97 In setting out the legal framework, the Commission does hold that abuse in objective concept but continues to say that an undertaking’s behaviour to protect its commercial interest ‘cannot be allowed if its actual purpose is to strengthen a
dominant position or abuse it’.98 And moreover that ‘[c]onduct that may otherwise be
permissible even on the part of a dominant undertaking may be rendered abusive if its purpose is anti-competitive, in particular if it is part of a plan to eliminate competition’.
According to the Commission anti-competitive purpose can thus render normal conduct abusive.
The question then is if this purpose can be proven by internal documents showing subjective intent and the Commission takes this route. It holds that eliminatory intent follows from three main factors: abundant documentary evidence proving the strategy; the fact that the conduct is selectively planned for certain countries; and that it fits in the long-term goal of AZ.99 For all these factors the Commission relies heavily on internal strategy documents. Also when it comes to objective justifications the Commission denies existence of justifications by referring to internal documents.100 Eventually the Commission discusses the effects of the conduct, first by holding that it does not have to establish actual effect. It then argues that – since it is clear the very reason for AZ to implement the strategy were the negative effects on parallel trade – the conduct was also capable of having such effects.101 The Commission clearly equates intent with effects and uses intent as a proxy for effects.
96 ibid para 256.
97 ibid para’s 265-326; especially regarding effect from para 307. 98 ibid para 327.
99 ibid para 789. 100 ibid para 817-47. 101 ibid para 848.
All in all the Commission in AstraZeneca relies to a significant extent on internal documents showing intent to prove the abusiveness of the conduct, its anticompetitive effects and the absence of objective justifications.
3.1.4. Conclusion regarding the Commission’s approach
To conclude it is clear that the Commission goes from a practice where it does not take into account evidence of subjective intent at all in e.g. General Motors, to relying heavily on such evidence in AstraZeneca. In between it takes gradual steps e.g. by acknowledging that conduct aimed at restricting parallel trade is in principle abusive, without saying how such aim is proven. In AstraZeneca it eventually relies extensively on evidence of subjective intent to prove the abuse, making clear that an underlying aim to impede parallel trade can be proven by mere reliance on such evidence.
3.2. The Courts
3.2.1. Inconsistent use of intent in early cases
In the early cases the Courts use intent inconsistently: in General Motors objective intent plays a role in the background but in United Brands reference to intent is entirely lacking. In General Motors the Court is guided by absence of objective anti-competitive intent in overturning the Commission’s decision finding an abuse.102 GM argued before the Court that they had no purpose of discouraging parallel import and supports this by showing that it immediately reduced prices and refunded excesses upon becoming aware of the abuse.103 GM thus argues absence of intent without referring to internal documents, but based on objective factors. The Court holds that in order to establish an abuse all factors have to be taken into account.104 It finds that the conduct of GM was not abusive, since there was an adequate explanation for the excessive price. Absence of abuse was shown by the fact that GM reduced its rates and refunded the excess.105 The Court does thus not refer to intent and bases its finding on objective factors such as the short temporal scope of the abuse and the refund. Arguably, the absence of intent does play a role in the background. It was clear for the Court – from objective factors – that GM had no intention of impeding parallel imports. Even though it was clear that the conduct had the effect of impeding parallel import, the presence of
102 Case 26-75 General Motors (n 12). 103 ibid at 1373.
104 ibid para’s 12, 15. 105 ibid para’s 21-2.
an alternative explanation prevented the finding of an abuse. This indicates that intent can play a ‘guiding’ role in finding an abuse.
Even though there seems to be a role for intent in General Motors this is not the case in
United Brands. The Court makes no reference to a subjective element at all. It simply holds –
based upon objective findings – that the obligation on a ripener not to resell green bananas is a restriction of competition.106 As regards the objective justification it held that the practice could only be justified if it did not raise obstacles that went beyond the objectives to be attained. Here however the conduct definitely went beyond such objectives, in particular by limiting and partitioning markets.107 The Court thus sees no role for intent either in establish an abuse or in finding justifications.
3.2.2. A general line develops
In subsequent cases a general approach becomes detectable. In this approach conduct may become abusive where the underlying purpose is to impede parallel trade. This starts with
Hoffman-LaRoche v Centrafarm, where the Court holds that the legitimate exercise of a right
may become abusive when it is used as an instrument to abuse a dominant position. In a rather confusing paragraph it says that if a trade-mark right is exercised lawfully, Article 102 TFEU (then 86 EC) is not breached on the sole ground that the undertaking holds a dominant position as long as the trade-mark is not used as an instrument to abuse that dominant position.108 An illegitimate underlying purpose, such as impeding parallel trade, may thus render conduct abusive. However, it is not clear how such a purpose is proven, especially if this can be done by relying on evidence of subjective intent.
The Court gives more clarity regarding such evidence in British Leyland, where the Court has both recourse to such subjective evidence and objective evidence to find two different abuses. The Court first deals with the non-renewal of the type approvals for imported cars. The Court refers to minutes of multiple meetings where BL promises its dealers to ‘make every effort to
limit re-importations’.109 The Court holds that it is clear from the minutes that the only reason for the conduct was the intention to impede re-importations.110 It concludes that by engaging
106 27/76 United Brands (n 12), para’s 157; 151-162. 107 Ibid para’s 158-9.
108 ibid para 16.
109 226/84 British Leyland (n 12) para 14. 110 ibid para 18.
in this conduct ‘with the object of creating a barrier to re-importations’ BL abused its dominant position.111 The Court thus strongly relies on this object to establish the abuse and this object has been found by referring to internal documents in which the desire to limit parallel imports was stated. When it comes to the abuse concerning the refusal to issue certificates of conformity, the Court comes to the conclusion that BL’s conduct can only objectively be explained as driven by the intention to create barriers to re-importations. It thus bases the finding of an intention on objective factors, for example from the fact that BL is reluctant to provide traders necessary information to parallel importers.112 This case thus shows intent is enough to render conduct abusive and that it is both possible to prove intent by relying on objective factors and by relying on subjective evidence found in internal documents.
So far it is clear that the Court holds that an underlying intention to impede parallel trade renders conduct abusive. This line is also confirmed in Syfait and Sot. Lélos.113 The Court
refers to General Motors and British Leyland to show that conduct that aims to restrict parallel trade has in the past constituted an abuse.114Again, it was not at issue here whether GSK had the aim to impede parallel trade, as they admitted this.115 The Court holds that the underlying purpose of restricting parallel trade may be abusive if such conduct cannot be explained otherwise. In this context, a dominant undertaking may protect its commercial interests, but not if the purpose is to strengthen or abuse a dominant position.116 The Court draws the conclusion that there can be no escape from abuse when the conduct is aimed at preventing all parallel trade.117 However, some parallel trade may be restricted if this is legitimately and proportionally done for the protection of an undertaking’s commercial interest. This is for example the case in the context of refusal to supply if the orders by the wholesaler are out of the ordinary.118 To summarize the Court will thus see conduct aimed at avoiding all parallel trade as abusive, but this may be justified when the undertaking is acting in its commercial interest.
111 ibid para 21. 112 ibid para 23.
113 The cases are linked together since they consider the same situation and referred the same question for preliminary ruling. In Syfait the Court however did not rule on the substance and therefore the same question was later asked in Sot. Lélos. C-53/03 Syfait and others (n 13) para’s 29-38.
114 ibid para 37.
115 C-468/06 Sot. Lélos kai Sia (n 9) para 36. 116 ibid para 50.
117 ibid para 66. 118 ibid para 77.
The general line in the Court’s approach thus is that conduct with the intent to impede parallel trade may in itself be abusive. Only in United Brands the Courts did not refer to intent, arguably because they did not need such an element to find the abuse. Regarding subjective intent specifically, only in British Leyland the Court relies solely on evidence of subjective intent to prove one of the abuses. From the cases discussed it is thus clear that intent may play a decisive role in finding an abuse, as intent to impede parallel trade may render conduct abusive.
3.2.3. Court steers away from using intent
In AstraZeneca the GC seems to make a turn. AZ had appealed the Commission decision, holding that the Commission has no hard evidence, but only inferences that are ‘unfairly and
erroneously’ drawn from documents.119 Moreover, AZ denies having any bad faith or
‘malevolent’ intent.120 AZ concludes its defence by arguing that the Commission has not shown effects.121 The GC overturns the decision of the Commission on the point of parallel trade in holding that the Commission has not adequately proven that the conduct was such as to restrict parallel import.122
The GC steers away from reliance on intent arguing that the overall strategy is not the issue here.123 The sole issue according to the Court is the deregistration of the medicine and whether it was ‘such as’ to prevent parallel import in the view of the objective context.124 Even though the GC acknowledges that it is quite clear from the documents that AZ intended to obstruct parallel imports, it holds that if such intention was absent this would not preclude the classification of an abuse.125 Abuse is an objective concept and a desire to protect legitimate interests does not justify competition outside the merits.126 According to the GC the question is thus whether the conduct was such as to prevent parallel trade and this conclusion must be based on objective findings. The GC then holds that the Commission failed to show the causal link between the conduct and capable effects. It especially holds that internal documents of AZ stating that such a link was there, were not sufficient evidence to prove existence of such
119 T-321/05 AstraZeneca v Commission (n 76) para 699. 120 ibid para’s 715-34. 121 ibid para’s 744-50. 122 ibid para 852. 123 ibid para’s 802-5. 124 ibid para 812. 125 ibid para’s 813-4. 126 ibid para 816.
a link.127 On this point it holds that intention may be taken into account, but that findings must first and foremost be based on objective evidence.128 The appeal before the ECJ did not go into the specific intent-issue.129
The GC thus steps away from the earlier approach that conduct aimed at restricting parallel trade may in itself be abusive, by deciding that the right test to establish an abuse is to assess whether the conduct is ‘such as’ to lead to anti-competitive effects in the objective context.
3.2.4 Conclusion of the Court’s approach
To conclude, the Court has recently stepped away from an earlier line that it had developed in its case law. With the exception of United Brands all cases refer to intent, specifically as the underlying purpose of certain conduct to impede parallel trade. According to the initial approach of the Court (before AstraZeneca), such an underlying purpose may render the conduct in principle abusive. The Court mainly relies on objective factors to prove such a purpose and only in British Leyland the Court has recourse to subjective intent evidence to do so. In AstraZeneca the Court however steps away from its approach that the underlying purpose to impede parallel trade is enough to render conduct abusive, by introducing the ‘such as’ test.
3.3. Conclusion: the role of subjective intent in parallel trading cases
Even though – as shown in the introduction of this chapter – at first sight it seems that intent has played an inconsistent role in parallel trading cases, closer investigation makes it possible to detect clear lines of development in the cases.
The Commission’s decisional practice mainly shows a development from cases where there is no reliance on intent at all (General Motors) to heavy reliance on intent in the latest case (AstraZeneca). It is clear that according to the Commission conduct is in itself abusive when the underlying purpose is to impede parallel trade. Eventually in AstraZeneca it clearly sought to prove this purpose by relying on subjective intent evidence. This seems to confirm the controversial theory explained in the first chapter that – despite the rhetoric of abuse as an objective concept – mere subjective intent may be enough to establish an abuse. Maybe this development over time can be explained by the Commission trying its luck as enforcer and
127 ibid para 848. 128 ibid para 849.
see how far it can go by relying on (subjective) intent before the Courts will intervene. And in
AstraZeneca the GC clearly did.
As for the Courts, the initial approach shows that conduct is abusive in itself when the aim is to impede parallel trade. This aim is usually proven by relying on objective factors, but in
British Leyland the Court relied merely on evidence of subjective intent to prove the abusive
nature of the conduct. However, in AstraZeneca the Court steered away from this focus on intent. It takes issue with the heavy reliance of the Commission on subjective intent evidence especially to prove possible effects and holds that intention may be taken into account, but that findings of an abuse must first and foremost be based on objective factors. Even though this does not seem in line with the earlier judgements discussed, it does fit within the rhetoric that abuse is an objective concept. The GC thus steers away from the intent-debate by introducing the ‘such-as’-test. How this test exactly works and how it fits with earlier judgments regarding intent is not clear, but what is clear from AstraZeneca is that the GC will not accept pure reliance on evidence of subjective intent. It will be interesting to see how the Commission will deal with this in its expected AB InBev decision.
4. The desired role of subjective intent evidence in parallel trading cases
In the previous chapter it was established that the Commission and Court take a different approach in using evidence of subjective intent in parallel trading cases. The Commission gradually seems to want to push through reliance on intent and particular the mere use of evidence of subjective intent to establish an abuse. The Court eventually puts a stop to this and goes back to the standard rhetoric of abuse as an objective concept, where subjective evidence may only play a supportive role. This last chapter assesses the desired role of evidence of subjective intent in light of two benchmarks: its evidentiary value and legal certainty. On this basis it evaluates the approaches taken by the institutions.
4.1. Evidentiary value
As explained in the introduction, issues of evidence are important as it concerns the very means by which an abuse is established. There is an ongoing debate in literature regarding the evidentiary value of intent-based evidence especially focussing on whether intent evidence can disambiguate the ambiguous. The core issue is to what extent such evidence is indicative of an abuse of dominance and (thus) has value in establishing an abuse. Related to this are issues of sound enforcement. Depending on one’s stance regarding evidence of intent it may either result in over-enforcement or serve as a check against it.
4.1.1. Proof of evidentiary value in case law
One way to assess if evidence of subjective intent has evidentiary value is to assess whether it has had such value in past cases. Arguments that rely on assessment of past cases are put forth by proponents of a role for (subjective) intent to proof that intent may indeed be successfully
