OPINION OF ADVOCATE GENERAL
SAUGMANDSGAARD ØE
delivered on 9 September 2020 (1)
Cases C‑152/19 P and C‑165/19 P
Deutsche Telekom AG (C‑152/19 P),
Slovak Telekom a.s. (C‑165/19 P)
v
European Commission
(Appeal – Competition – Abuse of dominant position – Slovak market for broadband services – Conditions laid down by the incumbent operator for unbundled access by other operators to the local loop – Decision finding an infringement of Article 102 TFEU and Article 54 of the EEA Agreement – Access obligation imposed by the regulatory framework – Bronner case-law – Inapplicability – Imputability of a subsidiary’s conduct to the parent company – Concept of an ‘economic unit’ – Decisive influence – Actual exercise of such influence – Body of consistent evidence)
I. Introduction
1. These cases ask the Court to clarify once again the scope of the judgment in Bronner (2) within the regulatory landscape of Article 102 TFEU. That case concerned the refusal by a dominant undertaking to make infrastructure which it owned available to competitors.
2. In essence, the applicants, Deutsche Telekom AG (‘DT’) in Case C‑152/19 P and Slovak Telekom, a.s. (‘ST’) in Case C‑165/19 P, suggest that the conditions established in paragraph 41 of that judgment, in particular the requirement that the service be indispensable, should be applied to implicit refusals to grant access which follow not from an explicit refusal on the part of the dominant undertaking, but from unfair contract terms.
3. For the reasons I shall set out below, I propose that the Court reject the concept of an ‘implicit refusal to grant access’ and emphasise the limited scope of the judgment in Bronner. To my mind, the judgment in Bronner is and must remain a special case within the regulatory landscape of Article 102 TFEU.
4. I shall also propose that the Court dismiss the second and third grounds of appeal raised by DT in Case C‑152/19 P. An examination of those grounds of appeal will enable the Court to reiterate the principles relating to the imputability of conduct engaged in by a subsidiary (ST) to its parent company (DT), on the understanding that the latter company’s shareholding in its subsidiary is too small to be covered by the presumption established in Akzo Nobel. (3)
II. The factual and regulatory background to the disputes
5. The factual background to the disputes was set out in paragraphs 1 to 11 of the judgment of the General Court in Deutsche Telekom v Commission (‘the judgment in DT’) (4) and in paragraphs 1 to 11 of the judgment of the General Court in Slovak Telekom v Commission (‘the judgment in ST’). (5) It may be summarised as follows.
6. DT and ST are the incumbent telecommunications operators in Germany and Slovakia respectively. As from 4 August 2000 and throughout the period covered by the decision at issue, that is to say, from 12 August 2005 to 31 December 2010, DT had a 51% shareholding in ST.
7. As regards the supply of internet access, the local loop is the physical twisted metallic pair circuit (also called ‘the line’) that connects the network termination point at the subscriber’s premises to the main distribution frame or any other equivalent facility in the fixed public telephone network.
8. Unbundled access to the local loop allows new entrants, usually called ‘alternative operators’, to use the existing telecommunications infrastructure belonging to those incumbent operators in order to offer various services to end users, in competition with the incumbent operators.
9. Local loop unbundling was organised at EU level, inter alia, by Regulation (EC) No 2887/2000 (6) and Directive 2002/21/EC. (7)
10. In essence, that regulatory framework required the operator ‘with significant market power’, as identified by the national regulatory authority, to grant alternative operators unbundled access to its local loop and to related services under transparent, fair and non-discriminatory conditions, and to maintain an updated reference offer for such unbundled access.
11. Following an analysis of its national market, on 8 March 2005, the Slovak telecommunications regulatory authority adopted a decision designating ST as an operator with significant power on the wholesale market for unbundled access to the local loop within the meaning of Regulation No 2887/2000. That decision, which ST challenged, was confirmed by the Chairman of that authority on 14 June 2005.
12. In implementation of that decision, ST published its reference unbundling offer on 12 August 2005. That offer, which was amended on nine occasions between that date and the end of 2010, set out the contractual and technical conditions for unbundled access to ST’s local loop.
13. ST’s offer covered 75.7% of Slovak households and all local loops that could be used to transmit a broadband signal. However, during the period between 2005 and 2010, access to only very few of ST’s local loops was unbundled, as from 18 December 2009, and these were used by a single alternative operator to provide retail broadband services to undertakings.
III. The decision at issue
14. On 15 October 2014, the European Commission adopted a decision penalising DT and ST for infringement of Article 102 TFEU and Article 54 of the EEA Agreement (‘the decision at issue’) on the Slovak broadband services market. (8)
15. In the decision at issue, the Commission found that the undertaking formed by DT and ST had committed a single and continuous infringement of Article 102 TFEU and Article 54 of the EEA Agreement, during the period between 12 August 2005 and 31 December 2010, in connection with the conditions under which ST offered unbundled access to its local loop.
16. More specifically, the infringement found by the Commission consisted of the following practices:
– withholding from alternative operators network information necessary for the unbundling of local loops;
– reducing the scope of ST’s obligations regarding unbundled local loops;
– setting unfair terms and conditions in ST’s reference unbundling offer as regards collocation, qualification, forecasting, repairs and bank guarantees; and
– applying unfair tariffs which do not allow an equally efficient competitor relying on wholesale access to ST’s unbundled loops to replicate the retail broadband services offered by ST without incurring a loss.
17. The Commission imposed a fine of EUR 38 838 000 jointly and severally on DT and ST, and a fine of EUR 31 070 000 on DT.
IV. The procedures before the General Court and the judgments under appeal
A. The judgment in DT
18. In support of its action before the General Court, DT relied on five pleas in law alleging, respectively:
– errors of law and of fact in the application of Article 102 TFEU as regards ST’s abusive conduct, and a breach of the rights of defence;
– errors of law and of fact as regards the duration of ST’s abusive conduct;
– errors of law and of fact in the imputation of ST’s abusive conduct to DT, in so far as the Commission has not proved that DT did indeed exercise a decisive influence over ST;
– misinterpretation of the concept of ‘undertaking’ within the meaning of EU law and breach of the principle that the penalty must be specific to the offender, and failure to state reasons; and
– errors in the calculation of the amount of the fine imposed on DT and ST.
19. By the judgment in DT, the General Court partially annulled the decision at issue. It then fixed the amount of the fine that DT is jointly and severally required to pay at EUR 38 061 963 and the amount of the fine that DT alone is required to pay at EUR 19 030 981. It dismissed DT’s action as to the remainder.
B. The judgment in ST
20. In support of its action before the General Court, ST relied on five pleas in law alleging, respectively:
– manifest errors of law and assessment in the application of Article 102 TFEU;
– infringement of its rights of defence as regards the practice resulting in margin squeeze;
– errors in the finding of margin squeeze;
– manifest errors of law and assessment by the Commission in finding that DT and ST were part of a single undertaking and were both liable for the infringement at issue;
– in the alternative, errors in the determination of the amount of the fine.
21. By the judgment in ST, the General Court partially annulled the decision at issue. It then fixed the amount of the fine that ST is jointly and severally required to pay at EUR 38 061 963. It dismissed ST’s action as to the remainder.
V. The appeals before the Court of Justice
A. The appeal lodged by DT against the judgment in DT
22. In support of its appeal against the judgment in DT in Case C‑152/19 P, DT raises four grounds of appeal alleging, respectively:
– misinterpretation and misapplication of the legal principle that, in order for a refusal of access to constitute an infringement of Article 102 TFEU, the access sought must be indispensable to activity on a downstream market;
– misinterpretation and misapplication of the legal principle that a parent company must have actually exercised a decisive influence over its subsidiary in order for an infringement of Article 102 TFEU committed by the subsidiary to be imputable to the parent company;
– misapplication of the legal principle that the subsidiary must have applied, in all material aspects, the instructions given to it by the parent company in order for an infringement of Article 102 TFEU committed by the subsidiary to be imputable to the parent company; and
– infringement of the right to be heard during the administrative procedure.
23. In addition, DT asks the Court of Justice to admit, if necessary, a ground of appeal raised by ST in Case C‑165/19 P, the subject matter of which it claims to be identical to that of the third limb of the first plea in law raised by DT before the General Court, namely the framework used for calculating the long-term average marginal costs taken as a basis for the finding of abusive margin squeeze.
24. By way of conclusion to its appeal, DT claims that the Court should:
– set aside the judgment under appeal in so far as it dismissed its action;
– annul the decision at issue, in whole or in part, in so far as it concerns DT, and, in the alternative, annul or reduce the fines imposed on it;
– in the alternative, refer the case back to the General Court for reconsideration; and
– order the Commission to pay all the costs arising from the present proceedings and the proceedings before the General Court.
25. The Commission contends that the appeal should be dismissed and DT ordered to pay the costs.
B. The appeal lodged by ST against the judgment in ST
26. In support of its appeal against the judgment in ST in Case C‑165/19 P, ST raises three grounds of appeal alleging, respectively:
– errors of law in the finding of abuse, within the meaning of Article 102 TFEU, consisting in a refusal to contract;
– a breach of the rights of defence in the assessment of margin squeeze; and
– errors of law in the assessment of the existence of margin squeeze.
27. In addition, ST asks the Court of Justice to admit, if necessary, a ground of appeal raised by DT in Case C‑152/19 P, the subject matter of which it claims to be identical to that of the fourth plea in law raised by ST before the General Court, namely the finding by the Commission that DT and ST formed a single undertaking and were both liable for the alleged infringement by ST.
28. By way of conclusion to its appeal, ST claims that the Court should:
– set aside the judgment under appeal, in whole or in part;
– annul the decision at issue, in whole or in part;
– in the alternative, annul or further reduce the fine imposed on it; and
– order the Commission to pay the costs of the present proceedings and of the proceedings before the General Court.
29. The Commission contends that the Court should dismiss the appeal and order ST to pay the costs.
VI. The proceedings before the Court
30. In Case C‑152/19 P, DT lodged its appeal against the judgment in DT on 21 February 2019. The Commission submitted written observations.
31. In Case C‑165/19 P, ST lodged its appeal against the judgment in ST on 22 February 2019. The Commission submitted written observations.
32. The Commission, DT and ST attended the hearing, common to both cases, held on 17 June 2020 in order to present oral argument.
VII. Analysis
33. In accordance with the Court’s request, this Opinion will focus on the first three grounds of appeal raised by DT in Case C‑152/19 P and on the first ground of appeal raised by ST in Case C‑165/19 P.
A. The first ground of appeal raised by DT and the first ground of appeal raised by ST
34. The first ground of appeal raised by DT and the first ground of appeal raised by ST both allege errors of law committed by the General Court in relation to the indispensability condition laid down in the judgment in Bronner for the purposes of assessing the existence of an abusive practice within the meaning of Article 102 TFEU.
35. Since those two grounds of appeal are, in common with the relevant parts of the judgment in DT (paragraphs 86 to 116) and the judgment in ST (paragraphs 92 to 154), largely the same, they can usefully be addressed together.
36. Before I embark on an examination of the arguments put forward by DT and ST, I think it useful to recall the content of the practices in question.
37. It is apparent from paragraphs 92 to 94 of the judgment in DT, and from paragraphs 113 and 114 of the judgment in ST, that DT and ST did not contest the existence of the conduct found to exist by the Commission in Section VII of the decision at issue (‘the practices at issue’), namely:
– withholding from alternative operators information about ST’s network that was necessary for the unbundling of that operator’s local loop;
– the reduction by ST of its unbundling obligations under the applicable regulatory framework; and
– the setting by ST of several unfair terms and conditions in its unbundling reference offer.
38. Those facts were not contested before the General Court and must therefore be considered to have been definitively established for the purposes of the present proceedings.
1. Summary of the arguments put forward by DT and ST
39. DT and ST submit, in essence, that the General Court wrongly held that, in order to classify the practices at issue as a ‘single and continuous infringement’ of Article 102 TFEU, the Commission was not required to prove that access to the local loop was indispensable, within the meaning of the judgment in Bronner, to the ability of competing suppliers to carry on their business on the mass retail market, given the existence of a regulatory access obligation.
40. In the interests of clarity, I shall follow the structure of the first ground of appeal raised by ST, which is divided into five limbs.
41. In the first limb of its first ground of appeal, ST argues, first, that, in paragraphs 151 and 152 of the judgment in ST, the General Court wrongly concluded that the conditions laid down in the judgment in Bronner for the purposes of applying Article 102 TFEU do not apply where there is an ex ante regulatory access obligation. That finding fails to take account of the fact that the ex post control under Article 102 TFEU is fundamentally different from the ex ante regulatory controls carried out by the Slovak telecommunications regulatory authority. (9)
42. ST argues, secondly, that, in paragraph 121 of the judgment in ST, the General Court wrongly held that it was not necessary for the Commission to verify whether the ‘indispensability’ condition laid down in the judgment in Bronner was satisfied, on the ground that an ex ante regulation had already recognised the ‘need’ for access to the applicant’s local loop. The assessment of ‘need’ under the regulatory framework is fundamentally different from the assessment of ‘indispensability’ under Article 102 TFEU.
43. Similarly, DT argues that, in paragraph 101 of the judgment in DT, the General Court wrongly held that the regulatory access obligation replaces the indispensability of access within the meaning of the judgment in Bronner. A regulatory access obligation imposed ex ante and the requirement of indispensability within the meaning of the judgment in Bronner, which is examined ex post, serve fundamentally different purposes.
44. DT further criticises the reference in paragraph 97 of the judgment in DT to the judgment in Deutsche Telekom v Commission, (10) given that the latter judgment is not concerned with the relationship between the regulatory access obligation and indispensability within the meaning of the judgment in Bronner.
45. By the second limb, ST states that, in paragraphs 126 and 127 of the judgment in ST, the General Court wrongly inferred from the judgment in TeliaSonera Sverige, (11) that the conditions laid down in Bronner were not applicable. ST notes that paragraphs 55 to 58 of the judgment in TeliaSonera Sverige were concerned with a margin squeezing practice, whereas it is accused of refusing to deal with alternative operators. In ST’s opinion, such a refusal must be assessed in the light of the case-law on refusals to deal, of which the judgment in Bronner is one.
46. DT put forward a similar argument to support an allegation that there was an error of law in paragraph 109 of the judgment in DT.
47. By the third limb, ST claims that the General Court committed an error of law in paragraphs 138 and 139 of the judgment in ST in holding that the judgment of the General Court in Clearstream v Commission (12) was not relevant for the purposes of assessing ST’s conduct, on the ground that, in that case, there was no regulatory obligation to provide the service at issue and the dominant undertaking had not developed its dominant market position on the basis of a statutory monopoly.
48. By the fourth limb, ST argues that, in paragraphs 133 and 134 of the judgment in ST, the General Court wrongly confirmed that, in order to be classified as abusive within the meaning of Article 102 TFEU, an explicit or categorical refusal to contract must satisfy the strict conditions laid down in the judgment in Bronner, whereas those conditions are not applicable in the case of an implicit refusal to contract. The outcome of the position thus adopted by the General Court is that more serious conduct (an explicit refusal to contract) is treated more favourably than less serious conduct (an implicit refusal to contract). According to ST, the General Court’s judgment is also vitiated by a failure to state reasons in that regard.
49. DT put forward a similar argument in connection with paragraph 111 of the judgment in DT, criticising the different treatment afforded to an explicit refusal to grant access such as that at issue in the judgment in Bronner, on the one hand, and an implicit refusal to grant access such as that at issue in the present case, on the other.
50. By the fifth and last limb, ST submits that, in paragraphs 153 and 154 of the judgment in ST, the General Court wrongly considered that the former State monopoly held by ST constituted a legal basis for not applying the conditions laid down in Bronner. ST states that the only judgment cited by the General Court in that regard, namely the judgment in Post Danmark, (13) provides no support for that position. ST goes on to say that the judgment in Bronner calls for an assessment of indispensability at the time when the alleged abuse was committed, with the result that the existence of a statutory monopoly in the past is irrelevant.
2. Response to the arguments put forward by DT and ST
51. All of the arguments put forward by DT and ST are based on a single premiss, namely that the practices at issue could not be found to be abusive without assessing indispensability within the meaning of the judgment in Bronner.
52. In other words, if the judgment in Bronner is not a relevant authority for assessing whether those practices are abusive, all of DT’s and ST’s arguments will have to be rejected as unfounded or invalid.
53. I am satisfied that the judgment in Bronner is not relevant in this case, for the reasons set out below.
54. More generally, this case represents an opportunity for the Court to clarify the scope of the judgment in Bronner, its scope having been the subject of numerous questions at the hearing.
55. In essence, I shall demonstrate below that the judgment in Bronner is a special case within the regulatory landscape of Article 102 TFEU. The scope of that special case needs to be interpreted strictly in order to preserve the effectiveness of Article 102 TFEU. (14) In other words, the principle is that the conditions laid down in Bronner are not applicable for the purpose of assessing the existence of an infringement of Article 102 TFEU.
(a) The scenario envisaged and the conditions laid down in the judgment in Bronner
56. The scenario envisaged by the Court in the judgment in Bronner was clearly identified in paragraph 37 of that judgment: in substance, the Court examined whether ‘the refusal by the owner of the only nationwide home-delivery scheme in the territory of a Member State, which uses that scheme to distribute its own daily newspapers, to allow the publisher of a rival daily newspaper access to it’ could be classified as an ‘abusive practice’ within the meaning of Article 102 TFEU.
57. In other words, the scenario envisaged in the judgment in Bronner is that of a refusal by a dominant undertaking to make infrastructure which it owns – in that instance, a home-delivery scheme – available to one or more competing undertakings. For the sake of simplicity, I shall, in the remainder of this Opinion, use the expression ‘refusal to make available’ in reference to that scenario.
58. This question of principle is not fundamentally different from that concerning the limits that may be attached, under Article 102 TFEU, to the exercise of the exclusive rights enjoyed by the owner of an intellectual property right. This explains the multiple references to the judgment in RTE and ITP v Commission, known as ‘the judgment in Magill’, (15) in the judgement in Bronner.
59. In paragraph 41 of the judgment in Bronner, the Court laid down several conditions that must be satisfied in order for a refusal to make available to constitute an abusive practice within the meaning of Article 102 TFEU. In the words used by the Court, it is necessary for that purpose ‘not only that the refusal of the service comprised in home delivery be likely to eliminate all competition in the daily newspaper market on the part of the person requesting the service and that such refusal be incapable of being objectively justified, but also that the service itself be indispensable to carrying on that person’s business, inasmuch as there is no actual or potential substitute in existence for that home-delivery scheme’.
60. From a reading of paragraph 41 of the judgment in Bronner, I infer that three conditions must be satisfied in order for a refusal to make available to be capable of being classified as ‘abusive’ (‘the conditions laid down in Bronner’):
– the refusal to make available must be liable to eliminate all competition on the relevant market from the competing undertaking;
– that refusal must not be objectively justified;
– the infrastructure in question must be indispensable to the ability of the competing undertaking to carry on its business, in the sense that there is no actual or potential substitute.
(b) The implications of the present case for competition policy within the European Union
61. According to DT and ST, the practices at issue may be classified as ‘abusive’ within the meaning of Article 102 TFEU only if the conditions laid down in Bronner are cumulatively satisfied. The Commission, on the other hand, argues that the Bronner case-law is not applicable to such conduct.
62. I should like to draw attention at this stage to the implications of the present case, which extend far beyond the dispute between the parties itself.
63. The conditions laid down in Bronner subject the finding of an abusive practice to a particularly high legal standard. They represent, to some extent, a ‘peak’ in the regulatory landscape of Article 102 TFEU.
64. It follows logically that any extension of the scope of the Bronner case-law entails a reduction in the effectiveness of Article 102 TFEU and, at the same time, a weakening of the Commission’s power to combat abusive practices. In practice, the Commission will be required to adduce significantly stronger evidence to establish the existence of abusive practices. Correlatively, undertakings in a dominant position will enjoy an increased margin for manoeuvre, inasmuch as their conduct will henceforth be punishable only if all of the conditions laid down in Bronner are satisfied.
65. More specifically, any extension of the Bronner case-law translates into the prohibition only of ‘super abuse’ of a dominant position, that is to say, abusive practices satisfying the conditions laid down in Bronner. Conversely, any practice by an undertaking in a dominant position that falls into one of the following three categories will no longer be penalised:
– it does not eliminate all competition on the relevant market from the competing undertaking (the opposite to the first condition laid down in Bronner);
– it is objectively justified (the opposite to the second condition laid down in Bronner); or
– it does not concern goods or services that are indispensable to the ability of the competing undertaking to carry on its business (the opposite of the third condition laid down in Bronner).
(c) The rationale behind the conditions laid down in Bronner
66. The implications of the present case having been set out, it falls now to look at the rationale behind the conditions laid down in Bronner, which are not set out as such in the wording of Article 102 TFEU.
67. Why has the Court set a higher legal standard for assessing the abusive nature of a refusal to make available, when other practices by dominant undertakings – such as the setting of unfair prices, (16) margin squeeze (17) or other unfair contract terms (18) – are examined without ever applying the conditions laid down in Bronner?
68. In my view, the answer to that question is clearly set out in the Opinion delivered by Advocate General Jacobs in Bronner. (19) In essence, there is a fundamental difference between, on the one hand, penalising the terms of an agreement, such as the price agreed, on the ground that they favour an undertaking which, because of its dominant position, is not subject to market discipline, and, on the other hand, penalising a refusal to make available. Penalising a refusal to make available, which amounts to requiring an undertaking to conclude an agreement, is significantly more detrimental to the freedom to conduct business.
69. That difference in nature is what justifies the higher legal standard established in the judgment in Bronner. It is also the rationale behind the ‘essential facilities’ doctrine in United States competition law, which Advocate General Jacobs had described in detail in point 45 et seq. of his Opinion in Bronner. He also clarified that difference in nature by drawing attention to the existence of a balancing of interests on two fronts.
70. The first balancing of interests is that between fundamental rights and free competition.
71. In point 56 of his Opinion, Advocate General Jacobs thus explained that ‘the right to choose one’s trading partners and freely to dispose of one’s property are generally recognised principles in the laws of the Member States, in some cases with constitutional status. Incursions on those rights require careful justification’.
72. Since then, Articles 16 and 17 of the Charter of Fundamental Rights of the European Union have enshrined the freedom to conduct a business, which includes the freedom of contract, (20) and the right to property, respectively.
73. The obligation that may be imposed on an undertaking in a dominant position under Article 102 TFEU to make the infrastructure it owns available to competing undertakings entails a serious and specific infringement of that undertaking’s freedom of contract and right to property.
74. It is because of that serious and specific infringement of fundamental rights that the Court has rightly imposed additional conditions on the application of Article 102 TFEU in such circumstances. In doing so, the Court balanced the – more serious – infringement of the fundamental rights of an undertaking in a dominant position consisting in the obligation to make its property available, on the one hand, with the – stricter – conditions for the application of Article 102 TFEU in such circumstances, namely the conditions laid down in Bronner, on the other hand.
75. The second balancing of interests is that between short-term and long-term benefits for competition and, ultimately, consumers.
76. In point 57 of his Opinion, Advocate General Jacobs states in that regard that ‘the justification in terms of competition policy for interfering with a dominant undertaking’s freedom to contract often requires a careful balancing of conflicting considerations. In the long term it is generally pro-competitive and in the interest of consumers to allow a company to retain for its own use facilities which it has developed for the purpose of its business. … Moreover, the incentive for a dominant undertaking to invest in efficient facilities would be reduced if its competitors were, upon request, able to share the benefits’.
77. In point 62 of his Opinion, Advocate General Jacobs expresses similar views with regard to the refusal to license intellectual property rights: ‘where such exclusive rights are granted for a limited period, that in itself involves a balancing of the interest in free competition with that of providing an incentive for research and development and for creativity. It is therefore with good reason that the Court has held that the refusal to license does not in itself, in the absence of other factors, constitute an abuse’.
78. Thus, the imposition of a higher legal standard for the purposes of assessing whether a refusal to make available is abusive is also justified by economic considerations aimed at preserving the long-term benefits of free competition in terms of investment and creativity.
79. To sum up, this balancing of interests on two fronts, first, between fundamental rights and free competition and, secondly, between the short-term and long-term benefits of free competition, highlights the difference between penalising the terms of an agreement and penalising a refusal to make available. That difference in nature explains why, in the judgment in Bronner, the Court applies a higher legal standard for the purposes of assessing whether a refusal to make available is abusive.
(d) The spuriousness of the concept of ‘implicit refusal of access’
80. One of the arguments put forward by DT and ST in support of their claim that the conditions laid down in Bronner should be applied to the practices at issue revolves around the concept of ‘implicit refusal of access’. According to DT and ST, the Bronner case-law ought to be applied not only in the event of an explicit refusal of access as envisaged by the Court in paragraph 37 of the judgment in Bronner, but also in the case of unfair contract terms imposed by an undertaking in a dominant position which lead in fact to the same result, in other words, an implicit refusal of access.
81. I can understand the attraction of the concept of ‘implicit refusal of access’, inasmuch as certain unfair contract terms may, in certain cases, preclude the conclusion of an agreement. However, I would point out here and now that to focus artificially on the effect of certain contract terms would be to disregard the wider framework of analysis on which the judgment in Bronner is based, in particular the balancing of interests on two fronts which I have just described.
82. The strategic interest for an undertaking in a dominant position, such as that formed by DT and ST, in putting forward such arguments based on the concept of implicit refusal is obvious. As I explained in points 62 to 65 of this Opinion, extending the conditions laid down in Bronner to further practices would make it possible, at a stroke, to reduce the effectiveness of Article 102 TFEU, curb the Commission’s power, and increase the margin for manoeuvre available to undertakings in a dominant position.
83. On the other hand, I find it harder to understand why the Commission insists on using that distinction, be it in the form mentioned above or under the guise of different terminology contrasting categorical with implicit refusal of access. When asked about this at several points in the hearing, the Commission had some difficulty in explaining why the practices in question could not be classified as an ‘implicit refusal of access’.
84. In reality, that difficulty stems from the spuriousness of the very concept of ‘implicit refusal of access’. That concept, which is in no way supported by the judgment in Bronner or by the Opinion of Advocate General Jacobs in that case, has a scope the elasticity of which is potentially unlimited. By way of illustration, does the imposition of an unfair price not constitute an implicit refusal of access?
85. One might, in extremis, ask whether any abusive practice does not in some way constitute an implicit refusal of access, since any disadvantage imposed by a dominant undertaking is liable to discourage potential customers from using the goods and services it offers.
86. It is important to note, however, that the Court has never applied the conditions laid down in Bronner, or any equivalent legal criterion, to unfair contract terms. The irrelevance of the conditions laid down in Bronner is particularly striking in the case of pricing practices, which would be a perfect example of an implicit refusal of access (if such a thing existed), on account of the crucial role of price in free competition. In its very early case-law on unfair pricing, however, the Court did not use a legal criterion equivalent to the conditions laid down in Bronner. (21)
87. More recently too, the Court refrained from applying the conditions laid down in Bronner in two judgments concerning the pricing practices of copyright collecting societies, although it may reasonably be presumed that their services were indispensable to certain downstream activities. (22) The Court also dismissed the relevance of the judgment in Bronner in relation to margin squeeze, which is a specific category of abusive pricing practice, in the judgments in TeliaSonera Sverige (23) and Telefónica and Telefónica de España v Commission. (24)
88. To sum up, the Court has never applied the conditions laid down in Bronner to abusive pricing practices, even if those practices constituted perfect examples of an implicit refusal of access.
89. Consequently, a decision now to treat such practices as implicit refusals of access would effectively reverse entire swathes of case-law relating to abusive practices and put the conditions laid down in Bronner at the very heart of Article 102 TFEU. The judgment in Bronner would become the rule rather than an individual case, an outcome which would run counter to the very wording of Article 102 TFEU, the scope of which is not limited to abusive practices relating to ‘indispensable’ goods and services within the meaning of that judgment.
90. In order to limit the scope of the concept of ‘implicit refusal’, some might consider it appropriate to restrict it to the most serious abusive practices. By way of illustration, only a very unfair price would be classified as an ‘implicit refusal of access’ triggering the application of the conditions laid down in Bronner, while slightly unfair prices would continue to be ‘ordinary’ abuse.
91. It would, to my mind, be a serious mistake to go down that road. Doing so would create a substantial source of arbitrariness at the very heart of competition law, a field in which legal certainty is of paramount importance to businesses. The dividing line between implicit refusal of access and ordinary abuse cannot but be arbitrary. (25)
92. Furthermore, reclassifying the most serious abusive practices as ‘implicit refusals of access’ would lead to a situation that would be paradoxical, to say the least. In that scenario the conditions laid down in Bronner would be applied to the most serious forms of abuse (classified as ‘implicit refusals of access’), thus making them more difficult to penalise. In other words, the most serious forms of abuse (a very unfair price, for example) would be subject to a less strict scheme of law than the least serious (a slightly unfair price, for example).
93. Thus, contrary to the appellants’ line of argument as summarised in points 48 and 49 of this Opinion, it is indeed the very concept of ‘implicit refusal of access’ that would cause the most serious conduct to be treated more favourably.
94. I would recall here that the gravity of the dominant undertaking’s conduct is not a relevant criterion for the purposes of assessing whether Article 102 TFEU has been infringed, as the Commission has rightly noted. Gravity is a factor only when it comes to fixing the amount of the fine, in accordance with Article 23(3) of Regulation (EC) No 1/2003. (26)
95. In short, the rationale behind the conditions laid down in Bronner, as set out in points 66 to 79 of this Opinion, lies in the difference in nature between penalising the terms of an agreement and penalising a refusal to make available. In the light of that rationale, there is little doubt in my mind that the conditions laid down in Bronner are not intended to be applied to unfair contract terms.
96. In the light of the foregoing, it is vital, in my opinion, that the concept of ‘implicit refusal of access’ be rejected in relation to Article 102 TFEU, be this in the judgment to be given or in any other context.
(e) The inapplicability of the conditions laid down in Bronner to the practices at issue
97. Now that I have set out the implications of the present case, the rationale behind the conditions laid down in Bronner and the spuriousness of the concept of ‘implicit refusal of access’, it remains for me to examine whether the practices at issue fall within the purview of the scenario envisaged in the judgment in Bronner, as reiterated in points 56 and 57 of this Opinion.
98. That scenario consists in a refusal by a dominant undertaking to make infrastructure which it owns available to one or more competing undertakings.
99. The practices at issue, as described in point 37 of this Opinion, do not fall within the purview of that scenario, as the General Court noted, without error, in paragraphs 98 and 99 of the judgment in DT and in paragraphs 118 and 119 of the judgment in ST.
100. ST did not refuse unbundled access to the local loop which it owns, but imposed unfair conditions on undertakings wishing to access it, as the Commission rightly noted.
101. It is irrelevant in this regard whether ST was compelled to grant access to the local loop by reason of regulatory obligations. The conclusion would be the same if ST had chosen voluntarily to grant access to the local loop. The only factor that matters for the purposes of ruling out the relevance of the judgment in Bronner is that ST did not refuse access to infrastructure which it owns.
102. Contrary to what DT and ST claim, that interpretation is borne out by the judgment in TeliaSonera Sverige, (27) as the General Court rightly held in paragraphs 106 to 110 of the judgment in DT and in paragraphs 123 to 127 of the judgment in ST.
103. In paragraph 55 of the judgment in TeliaSonera Sverige, (28) the Court recalled, in essence, that the conditions laid down in Bronner, in particular, the requirement of need, were not applicable for the purposes of assessing the ‘abusive nature of conduct which consists in supplying services or selling goods on conditions which are disadvantageous or on which there might be no purchaser’.
104. Furthermore, in paragraph 58 of that judgment, the Court held that extending the scope of the judgment in Bronner to any conduct of a dominant undertaking in relation to its terms of trade would amount to ‘a requirement that … the conditions [laid down in Bronner] would in every case have to be satisfied, and that would unduly reduce the effectiveness of Article 102 TFEU’.
105. In doing so, the Court refused to endorse the view expressed by Advocate General Mazák in that case. He had argued in favour of the implicit refusal to supply proposition and the corresponding obligation to ascertain whether the inputs are indispensable, in accordance with the line of argument put forward by TeliaSonera Sverige, which the Court ultimately and rightly rejected. (29)
106. Similarly, in paragraph 96 of the judgment in Telefónica and Telefónica de España v Commission, (30) the Court recalled that a margin squeeze constitutes an independent form of abuse distinct from that of refusal to supply, to which the conditions laid down in Bronner are not applicable.
107. Thus, those two judgments confirmed the limited scope of the judgment in Bronner, which is a special case within the regulatory landscape of Article 102 TFEU.
108. Accordingly, the criticism directed by DT and ST at the references in the judgments in DT and ST to the judgments in Deutsche Telekom v Commission (31) and Post Danmark (32) and to the judgment of the General Court in Clearstream v Commission (33) must be dismissed as unfounded. For those arguments seek to criticise the reasoning on the basis of which the General Court rules out the relevance of the judgment in Bronner in the circumstances of the present cases. As I have just stated, the General Court did not commit any error of law in that regard.
109. I turn finally now to examine the last argument put forward by DT and ST, summarised in points 41 to 44 of this Opinion. That argument is directed in particular at paragraph 101 of the judgment in DT and paragraph 121 of the judgment in ST, the wording of which is identical:
‘Therefore, given that the relevant regulatory framework clearly acknowledged the need for access to [ST’s] local loop, in order to allow the emergence and development of effective competition in the Slovak market for high-speed internet services, the demonstration, by the Commission, that such access was indeed indispensable for the purposes of the last condition set out in paragraph 41 of the judgment [in Bronner], was not required’.
110. According to DT and ST, the General Court wrongly took to be equivalent, on the one hand, the examination of need carried out ex ante by the national regulatory authority under the regulatory framework, and, on the other hand, the examination of indispensability which falls to be carried by the Commission ex post pursuant to Article 102 TFEU as interpreted by the judgment in Bronner.
111. I willingly admit that, in line with the argument put forward by DT and ST, I find it difficult to regard these two types of examination as equivalent. However, their argument is ineffective inasmuch as it is based on an erroneous reading of the judgments under appeal.
112. Contrary to what DT and ST claim, the General Court did not take those two types of examination to be equivalent, but rightly held that the conditions laid down in Bronner are not applicable in the circumstances of the present cases.
113. For one thing, that reading is based on the words used in paragraph 101 of the judgment in DT and paragraph 121 of the judgment in ST, the text of which is reproduced above, which do not establish any equivalence between those two types of examination. For another, those paragraphs form part of a broader line of argument, set forth in paragraphs 97 to 105 of the judgment in DT and paragraphs 117 to 122 of the judgment in ST, at the end of which the General Court rightly concluded that the conditions laid down in Bronner are simply not applicable in those circumstances.
114. The General Court did not commit any error of law in doing so. As I recalled in point 101 of this Opinion, what matters for the purposes of ruling out the relevance of the judgment in Bronner is that ST did not refuse access to infrastructure which it owns.
115. Furthermore, in paragraphs 97 of the judgment in DT and paragraph 117 of the judgment in ST, the General Court rightly recalled that, where the legislation relating to the telecommunications sector defines the legal framework applicable to it and, in so doing, contributes to the determination of the competitive conditions under which a telecommunications undertaking carries on its business in the relevant markets, that legislation constitutes a relevant factor in the application of Article 102 TFEU to the conduct of that undertaking, whether for the purposes of defining the relevant markets, assessing the abusive nature of such conduct or setting the amount of the fines. (34)
116. In the present case, it is common ground that the regulatory framework imposed an obligation on ST to grant access, as the General Court held in paragraphs 99 and 100 of the judgment in DT and paragraphs 119 and 120 of the judgment in ST.
117. It follows from the foregoing that the first ground of appeal raised by DT and the first ground of appeal raised by ST must be dismissed as unfounded.
B. The second ground of appeal raised by DT
1. Summary of the arguments put forward by DT
118. In its second ground of appeal, DT claims that the judgment in DT is vitiated by errors of law as regards the application of the principle that the parent company must actually have exercised a decisive influence over the subsidiary. DT states that the General Court rightly recalled that principle, in paragraph 230 of the judgment in DT, but nonetheless committed two types of error in applying it.
119. By the first limb of its second ground of appeal, DT claims that the General Court wrongly considered that facts which show that it is possible to exercise a decisive influence may also be used as evidence that a decisive influence was actually exercised.
120. According to DT, facts which simply show that it is possible to exercise a decisive influence cannot be used to establish that a decisive influence was actually exercised. Any other interpretation would remove the distinction between the possibility of exercising a decisive influence and the actual exercise of it, and would have the effect of unlawfully extending the presumption applicable to wholly owned subsidiaries. (35)
121. According to DT, the General Court committed that error at several points in the judgment in DT, in finding that evidence showing simply that it was possible to exercise a decisive influence proved that such influence had actually been exercised:
– in paragraph 233, with respect to the accumulation of posts within the subsidiary and the parent company;
– in paragraph 249 et seq., with respect to the presence of some of the applicant’s senior executives on ST’s board of directors;
– in paragraph 280 et seq., with respect to DT’s having provided staff to perform certain activities within ST; and
– in paragraph 294, with respect to ST’s having forwarded reports on its commercial policy.
122. By the second limb of its second ground of appeal, DT claims that, when carrying out a legal classification of the facts on which the Commission relied, the General Court wrongly applied the principle that a decisive influence must actually have been exercised.
123. Thus, in paragraphs 262, 273, 274 and 278 of the judgment under appeal, the General Court concluded that decisive influence had actually been exercised on the mere ground that it was possible to exercise such influence, but omitted to examine separately whether such influence had actually been exercised.
2. Response to the arguments put forward by DT
124. I would recall as a preliminary point that the General Court has exclusive jurisdiction to find and appraise the relevant facts and, in principle, to examine the evidence it accepts in support of those facts. Save where the clear sense of the evidence has been distorted, that appraisal does not therefore constitute a point of law which is subject as such to review by the Court of Justice. (36)
125. In the present case, DT has not claimed that the evidence examined by the General Court was distorted. Therefore, it is not for the Court of Justice, in its appellate jurisdiction, to re-examine the probative value of the factual evidence referred to by DT in its second ground of appeal.
126. In order to identify the precise scope of DT’s line of argument, I think it useful to view it within the context of the Court’s case-law on the imputability of a subsidiary’s conduct to the parent company, it being understood that the parent company’s shareholding in the subsidiary is too small to be covered by the presumption established in Akzo Nobel. (37) During the period relevant to the present cases, DT held a 51% shareholding in ST. (38)
127. According to settled case-law, the concept of ‘undertaking’ covers any entity engaged in an economic activity, regardless of its legal status and the way in which it is financed. On that point, the Court has stated that in this context the term ‘undertaking’ must be understood as designating an economic unit even if in law that economic unit consists of several natural or legal persons, and that if such an economic entity infringes the competition rules, it is for that entity, consistently with the principle of personal liability, to answer for that infringement. (39)
128. Thus, the conduct of a subsidiary may be imputed to the parent company in particular where, although having a separate legal personality, that subsidiary does not decide independently upon its own conduct on the market, but carries out, in all material aspects, the instructions given to it by the parent company, having regard in particular to economic, organisational and legal links between those two legal entities. (40)
129. That is the case because, in such a situation, the parent company and its subsidiary form a single economic unit and therefore form a single undertaking for the purposes of the case-law mentioned above. Thus, the fact that a parent company and its subsidiary constitute a single undertaking enables the Commission to address a decision imposing fines to the parent company, without having to establish the personal involvement of the latter in the infringement. (41)
130. In the context of that link between competition law and the economic concept of ‘an undertaking’, the Court has held that, in examining whether the parent company is able to exercise decisive influence over the market conduct of its subsidiary, account must be taken of all the relevant factors relating to the economic, organisational and legal links which tie the subsidiary to its parent company and, therefore, of economic reality. (42)
131. In that regard, the Court went on to say that the Commission cannot merely find that the parent company is in a position to exercise decisive influence over the conduct of its subsidiary, but must also check whether that influence was actually exercised. (43)
132. In other words, it is for the Commission to demonstrate, on the basis of factual evidence, including in particular any management power one of those entities may have over the other, that the parent company does in fact exercise a decisive influence over its subsidiary. (44)
133. As regards the form of proof, the Court has held that the exercise of decisive influence may be inferred from a body of consistent evidence, even if some of that evidence, taken in isolation, is insufficient to establish the existence of such influence. (45)
134. We come now to the line of argument put forward by DT under its second ground of appeal.
135. DT claims that the General Court wrongly held that facts which show that it is merely possible to exercise a decisive influence may also be used as evidence of the actual exercise of decisive influence.
136. In other words, DT seeks to exclude an entire category of factual evidence, namely that showing that it is possible to exercise a decisive influence, from the evidence which may be used by the Commission to establish the actual exercise of such influence.
137. That line of argument strikes me as being entirely without foundation, for at least three reasons.
138. First, such a restriction is not in any way the product of the Court’s case-law on the imputability of a subsidiary’s conduct to the parent company, which I summarised above.
139. More specifically, it follows expressly from that case-law that the actual exercise of a decisive influence may be inferred from a body of consistent evidence, even if some of that evidence, taken in isolation, is insufficient to establish the existence of such influence. (46) In that regard, the Court has not attached any restriction or criteria to the consistent evidence which the Commission may use.
140. Secondly, I can see no logical reason why the same factual evidence cannot serve to demonstrate both the possibility of a decisive influence and the actual exercise of such influence.
141. Of course, a body of consistent evidence establishing the actual exercise of a decisive influence must be more robust and more detailed than a body of evidence establishing the mere possibility of such influence. The fact nonetheless remains that the same factual evidence may legitimately be used in both contexts.
142. Thirdly, it is my view that DT’s arguments would in practice have the effect of restricting the factual evidence usable by the Commission exclusively to ‘flagrant’ (47) evidence, such as, for example, a written message containing an instruction from the parent company ordering the subsidiary to change its pricing policy.
143. However, the Commission only rarely has such flagrant evidence at its disposal. It is therefore essential, in order to ensure that the Commission is able to take effective action in the competition field, that it should be able to rely on any factual evidence, of whatever kind, it being understood that the body of such factual evidence, considered as whole, must show that a decisive influence has actually been exercised.
144. On the basis of DT’s reasoning, the usability of certain facts and evidence would, as the Commission has noted, depend on formal criteria unsuited to the economic reality in which undertakings operate.
145. It follows from the foregoing that the premiss on which DT’s second ground of appeal is based is erroneous, with the result that that ground of appeal must be dismissed in its entirety.
C. DT’s third ground of appeal
1. Summary of the arguments put forward by DT
146. By its third ground of appeal, DT submits that the judgment in DT is vitiated by errors of law as regards the application of the principle that the subsidiary must have applied, in all material aspects, the instructions given to it by the parent company.
147. According to DT, it has been settled case-law since the judgment in Imperial Chemical Industries v Commission (48) that the imputability of a subsidiary’s conduct to the parent company is subject to four cumulative conditions:
– the parent company must have been in a position to exercise a decisive influence;
– the parent company actually exercised such decisive influence;
– for that reason, the subsidiary did not decide independently upon its own conduct on the market; and
– the subsidiary applied, in all material aspects, the instructions given to it by the parent company.
148. The fourth condition, to the effect that the subsidiary must have applied, in all material aspects, the instructions given to it by the parent company, serves to verify that the decisive influence exercised by the parent company is relevant.
149. DT submits that, in this regard, the General Court simply held, first, that the fact that the subsidiary enjoys a degree of autonomy is not incompatible with its belonging to the same economic unit as its parent company (paragraph 470 of the judgment in DT) and, secondly, that ST’s general strategy on the market was defined by DT (paragraph 471 of the judgment in DT).
150. As regards the second finding, DT states that this is not supported by paragraphs 237 to 464 of the judgment in DT, to which the General Court refers in paragraph 471 of that judgment. DT argues that, in those paragraphs, the General Court listed several factors by way of evidence that DT had exercised a decisive influence over ST, without for that matter establishing the existence of any actual instructions given by DT to ST.
151. DT maintains that, consequently, the General Court was a fortiori not able to find that ST had followed instructions from DT in all material aspects. DT also claims that the judgment in DT is vitiated by a failure to state reasons in this regard.
2. Response to the arguments put forward by DT
152. DT’s third ground of appeal is flawed by the same irremediable defect as its first and second grounds of appeal, namely an erroneous premiss.
153. Contrary to what DT submits, the Court has never held that the imputability of a subsidiary’s conduct to the parent company is subject to compliance with the four conditions referred to in point 147 of this Opinion.
154. There is, in fact, only one relevant criterion in this regard, namely the existence of an economic unit, that is to say, an undertaking, formed by the parent company and the subsidiary, as the Commission has rightly argued. It is only in these circumstances that the Commission is authorised to impute the subsidiary’s conduct to the parent company or, in other words, to lift ‘the corporate veil’ between separate legal entities in order to increase the effectiveness of competition law. (49)
155. The status of the four conditions referred to by DT must be understood in the light of those principles.
156. It is my view that, at this stage in the development of its case-law, the Court has identified two evidentiary options enabling the Commission to establish the actual existence of an economic unit formed by a parent company and its subsidiary:
– first, the Commission may establish that the parent company is capable of exercising a decisive influence over the conduct of the subsidiary and, moreover, that it has actually exercised such influence; (50)
– secondly, it may prove that the subsidiary does not decide independently upon its own conduct on the market but carries out, in all material aspects, the instructions given to it by the parent company, regard being had in particular to the economic, organisational and legal links between those two legal entities. (51)
157. DT’s line of argument amounts, in essence, to a merger of those two evidentiary options, inasmuch as it requires the Commission to furnish double proof: it would have to establish both that the parent company had actually exercised a decisive influence and that the subsidiary had, in all material aspects, applied instructions from the parent company.
158. To my mind, there can be no doubt that that line of argument is entirely unfounded, from the point of view of both case-law and logic.
159. From the point of view of case-law, it does not follow from any judgment of the Court that the Commission is required to furnish such double proof.
160. From the point of view of logic, those two evidentiary options serve the same purpose, which is to establish the existence of an economic unit (or an undertaking) formed by the parent company and the subsidiary. It would therefore be redundant to require the Commission to pursue both evidentiary options at the same time. As the Commission has stated, those two evidentiary options must be regarded as equivalent.
161. In paragraph 471 of the judgment in DT, the General Court noted that, in the light of the material contained in paragraphs 237 to 464 of that judgment, which established the decisive influence which DT had actually exercised over ST, the Commission had rightly concluded that those two legal entities formed a single economic unit.
162. Consequently, contrary to what DT claims, the General Court did not commit any error of law in finding that the Commission was not required to establish, in addition, that ST had followed, in all material aspects, the instructions given to it by DT.
163. I would also observe that, according to settled case-law, (52) the statement of the reasons on which a judgment is based must clearly and unequivocally disclose the General Court’s thinking, so that the persons concerned can be apprised of the justification for the decision taken and the Court of Justice can exercise its power of review.
164. Paragraphs 237 to 473 of the judgment in DT do disclose, clearly, unequivocally and in detail, the reasons why the General Court considered that DT and ST formed a single economic unit.
165. It follows from the foregoing that DT’s third ground of appeal must also be dismissed in its entirety.
VIII. Conclusion
166. In the light of the foregoing considerations, and without prejudice to the merits of the other grounds of appeal, I propose that the Court dismiss the first three grounds of appeal raised by Deutsche Telekom AG in Case C‑152/19 P and the first ground of appeal raised by Slovak Telekom, a.s. in Case C‑165/19 P.