Language of document : ECLI:EU:C:2008:520

OPINION OF ADVOCATE GENERAL

Mazák

delivered on 25 September 2008 (1)

Case C‑202/07 P

France Télécom SA

v

Commission of the European Communities

(Appeal – Competition – Abuse of a dominant position (Article 82 EC) – Market for services in high-speed internet access (ADSL) – Predatory pricing – Alignment of prices with those of competitors – Recoupment of losses suffered)





1.        In the present appeal the Court is called upon to deal with alleged abuse of a dominant position in the form of predatory pricing, which belongs to the category of exclusionary, or ‘eliminatory’, practices of dominant undertakings. (2) The Court’s previous case-law in the area of predatory pricing is essentially Akzo (3) and Tetra Pak II. (4)

2.        At the origin of the present appeal is a Commission Decision of 16 July 2003, (5) in which the Commission found that Wanadoo Interactive SA had infringed Article 82 EC by charging for its eXtense and Wanadoo ADSL (6) services (‘the services in question’) predatory prices that did not enable it to cover its variable costs until August 2001 or to cover its full costs from August 2001 onwards, as part of a plan to pre-empt the market in high-speed internet access during a key phase in its development. (7) The Commission ordered it to bring the infringement to an end and to pay a fine of EUR 10.35 million. (8)

3.        The contested decision was upheld by the judgment of the Court of First Instance of 30 January 2007 in Case T‑340/03 (9) (‘the judgment under appeal’). In its present appeal, France Télécom SA (‘FT’ or ‘the appellant’), formerly Wanadoo Interactive SA (‘WIN’), asks for the judgment under appeal to be set aside.

I –  Background to the dispute

4.        The factual background to the dispute was summarised by the Court of First Instance in paragraphs 1 to 11 of the judgment under appeal:

‘1      In the context of the development of high-speed internet access, the Commission decided, in July 1999, to launch a sectoral inquiry within the [EU] [(10)], which focused in particular on the provision of local loop access services and use of the residential local loop. Against this background, in the light of the information gathered the Commission decided to take a close look at the prices which [WIN] charged its residential customers in France for high-speed internet access. To this end, it launched proceedings [ex officio] in September 2001.

2      At the material time, WIN was part of the France Télécom group, 99.9% of its capital being held by Wanadoo SA. [FT’s] shareholding in Wanadoo fluctuated between 70 and 72.2% during the period at issue. The group formed by Wanadoo and its subsidiaries … encompassed all [FT’s] internet activities and its telephone directory business ... WIN covered the operational and technical aspects of internet access services in France, including ADSL … services.

3      The Commission sent WIN a first statement of objections on 19 December 2001 … and a supplementary statement of objections on 9 August 2002 …, to which WIN replied on 4 March and 23 October 2002 respectively.

4      On 16 January 2003, the Commission sent WIN a letter setting out facts … giving it access to the file which served as the basis for drafting that letter. WIN … had access to the file on 23 and 27 January 2003. By letter of 26 February 2003, WIN asked the Commission to clarify a number of aspects of the letter on the facts. The Commission replied by letter of 28 February 2003 and WIN then submitted a document in reply to the letter on the facts on 4 March 2003.

5      By [the contested decision], the Commission found that [WIN had infringed Article 82 EC, (11)] ordered it to bring the infringement to an end … and imposed a fine on it of EUR 10.35 million ...

6      The decision defines the relevant market as the French market for high-speed internet access for residential customers. The products with which the infringement is concerned are internet access services based on ADSL technology (Wanadoo ADSL and eXtense).

7      According to the decision, in the case of Wanadoo ADSL, at the material time, the customer had to pay a monthly subscription to [FT] for supplying the service, renting the ADSL modem from [FT], together with a subscription to WIN as its internet service provider (“ISP”). In the case of the eXtense service, the modem was bought by the user who paid only a monthly subscription to WIN corresponding to the service supplied by [FT] and the flat-rate unlimited internet access.

8      Having looked into various elements, including the market shares (recitals 211 to 222 in the preamble to the decision) and the effects of the “link-up” with [FT] (recitals 223 to 228), the Commission concludes that WIN occupied a dominant position on the relevant market. It then concentrates on showing that the below-cost pricing applied by WIN formed part of a deliberate strategy of predation aimed at “pre-empting” the market and thereby constitutes an abuse of a dominant position within the meaning of Article 82 EC (recital 254).

9      The decision fixes the starting date of the infringement as 1 March 2001 and the end as 15 October 2002, the date on which the remedy put forward by [FT] in March 2002 entered into force. The variable costs were not covered by the prices charged from March to August 2001 and the full costs were not covered from August 2001 (Article 1 of the decision, see paragraph 5 above).

10      That decision was notified to WIN on 23 July 2003 …

11      Following a merger on 1 September 2004, [FT] succeeded to the rights of WIN.’

II –  The proceedings before the Court of First Instance and the judgment under appeal

5.        By application lodged at the Registry of the Court of First Instance on 2 October 2003, WIN brought an action for the annulment of the contested decision.

6.        In the judgment under appeal, the Court of First Instance dismissed the action.

7.        In support of its claim for annulment of the contested decision, WIN had made four pleas in law, alleging (i) breach of the rights of the defence and of essential procedural requirements, (ii) failure to state reasons, (iii) breach of the principle that penalties must be specific to the offender, and (iv) breach of Article 82 EC. (12)

8.        Only the second part of the fourth plea in law (concerning the breach of Article 82 EC) – dealing with the question of abuse – is at issue in this appeal. (13) In that part, WIN maintained in essence that the Commission had infringed Article 82 EC in that it had not established to the requisite legal standard that WIN had abused its dominant position by charging predatory prices for the services in question from March 2001 to October 2002 (‘the period at issue’).

9.        That allegation consisted of two groups of complaints, concerning: (i) the recovery of costs test, and (ii) the Commission’s application of the test of predation.

10.      First of all, it would appear that the parties agree in substance on the following points:

–        the services in question are subscriptions for internet access and are therefore provided over a period of time; and the average lifetime of a subscription is 48 months;

–        every subscription produces recurring monthly revenue;

–        the cost structure consists of, on the one hand, ‘subscriber acquisition costs’ or ‘conquest costs’ such as advertising and promotional expenditure, ‘starter packs’ for customers, etc. (which belong to the category of non-recurrent costs, since a company has to bear them only once in order to secure a subscription contract with a client) and, on the other, ‘production costs’ and ‘network costs’ (which belong to the category of recurrent costs in that they are related to the provision of the subscription service and have to be borne by the company each month; they can be either variable costs, such as costs relating to access to FT’s network which the latter charges for on the basis of the number of subscribers, or fixed costs, such as costs related to the running of the company).

11.      In order to determine the recovery of costs, the Commission essentially compared monthly revenues with the sum of recurrent monthly costs and the part of non-recurrent costs, which it spread over a period of 48 months, that was on that basis attributable to one month. In particular, the Commission calculated the average of the resulting ratio for four consecutive periods: from 1 January to 31 July 2001, from 1 August to 15 October 2001, from 15 October 2001 to 15 February 2002, from 15 February to 15 October 2002.

A –    The complaints in relation to the recovery of costs test

12.      The Court of First Instance noted (14) that ‘[i]n applying this method, the Commission took the view that the prices applied by WIN did not enable it to recover its variable costs until August 2001 or its full costs from January 2001 to October 2002 …, there being no doubt that, considering the level of recovery of variable costs, the full costs were not recovered by August 2001’.

13.      In addressing WIN’s objection to the Commission’s choice of method of calculating the rate of recovery of costs, the Court of First Instance recalled as a preliminary point that, since such a choice entailed a complex economic assessment on the part of the Commission, the Commission had to be afforded a broad discretion. The Court of First Instance’s review was therefore limited to verifying whether the relevant rules on procedure and on the statement of reasons had been complied with, whether the facts had been accurately stated and whether there had been any manifest error of appraisal or a misuse of powers. (15)

14.      Referring, next, to Akzo and Tetra Pak II, (16) the Court of First Instance pointed out that ‘first, prices below average variable costs give grounds for assuming that a pricing practice is eliminatory and that, if the prices are below average total costs but above average variable costs, those prices must be regarded as abusive if they are determined as part of a plan for eliminating a competitor’. (17)

15.      The Court of First Instance stated that in the present case the Commission had based the contested decision on an analysis of the actual recovery of adjusted costs. In accordance with the principle of depreciation of assets, the Commission had spread the costs of acquiring customers over 48 months. On that basis, it had made a separate assessment of adjusted variable costs and adjusted full costs. (18)

16.      Further, the Court of First Instance rejected WIN’s allegations that the Commission had applied a test of static recovery – without taking into account the adjustment of recurrent monthly costs over time, which would have been considerably more unfavourable to WIN. The Court stated that, in respect of each period of infringement investigated and of all subscribers, the Commission had integrated the successive reductions in tariffs occurring during the period at issue. The Commission had even structured its analysis according to those reductions. (19)

17.      Moreover, the Court of First Instance held that the Commission was entitled to consider that the revenue and costs applicable after October 2002 – and, hence, after the infringement – could not be relevant for the purposes of assessing the rate of recovery of costs during the period investigated. (20)

18.      Finally, the Court of First Instance observed that, even if WIN were to prove that the method which it advocated (that is to say, the method of discounted cash flows) was appropriate in some respects, that would be insufficient to prove that the method used by the Commission was unlawful. (21)

19.      In addition, WIN maintained that the Commission had erred in its application of its own method of calculation, particularly in its calculations of the fixed and variable costs.

20.      In that respect, the Court of First Instance held that, independently of the admissibility of that plea, the fact that the Commission, in exercising its discretion, accepted that a rate of recovery of variable costs of 99.7% did not amount to an infringement could not require it to take the same approach to a rate of recovery of full costs of 98 or 99%, as the case may be. That plea therefore had to be rejected as ineffective. (22)

B –    The complaints relating to the test of predation

21.      The Court of First Instance also dismissed WIN’s complaints relating to the test of predation applied by the Commission.

22.      First of all, it rejected WIN’s arguments that an operator had the right to align its prices in good faith with those previously charged by a competitor even if they were below cost for the undertaking concerned.

23.      Having noted that neither the Commission’s practice nor the case-law of the Community courts recognised that the right of a dominant undertaking to align its prices with those of its competitors was absolute, the Court of First Instance pointed out that undertakings in a dominant position had special obligations imposed on them and could be deprived of the right to adopt a course of conduct or take measures which were not in themselves abuses and which would even be unobjectionable if adopted or taken by non-dominant undertakings. (23)

24.      The Court of First Instance concluded that ‘WIN cannot therefore rely on an absolute right to align its prices on those of its competitors in order to justify its conduct. Even if alignment of prices by a dominant undertaking on those of its competitors is not in itself abusive or objectionable, it might become so where it is aimed not only at protecting its interests but also at strengthening and abusing its dominant position.’ (24)

25.      Secondly, the Court of First Instance decided not to uphold WIN’s complaint based on the absence of a plan of predation and reduction in competition.

26.      According to WIN, the Commission had committed a serious infringement of Article 82 EC by concluding that such a plan had existed. In fact, such a strategy of predation could in no way be considered to be rational in the market conditions prevailing at the time, particularly in view of the low barriers to entry. (25)

27.      Referring to Community case-law, the Court of First Instance recalled that, in the case of predatory pricing, if prices were below average total costs, the existence of a plan to eliminate competition had to be proved and intention to eliminate competition had to be established on the basis of sound and consistent evidence. (26) It then stated that the statements referred to by the Commission, which were contained in internal company documents, were indicative of a plan of predation and were reinforced by other evidence. The Court of First Instance held that, in its application, WIN had made merely vague assertions which did not allow the Court to give a ruling, and it therefore rejected them. It concluded that the Commission had provided solid and consistent evidence as to the existence of a plan of predation for the entire infringement period. (27)

28.      Thirdly, WIN submitted that the recoupment of losses was entirely separate from the test of predation and the Commission had to provide evidence of it. The Commission had therefore seriously erred in law in maintaining that it was not necessary to prove recoupment of losses. Furthermore, WIN submitted the Commission had committed a manifest error of assessment, coupled with an error of law, by considering that it had provided evidence that recoupment of losses was possible. (28)

29.      The Court of First Instance held that, in accordance with Akzo and Tetra Pak II, the Commission was right to take the view that proof of recoupment of losses was not a precondition to making a finding of predatory pricing. In line with Community case-law, the Commission was able to regard prices below average variable costs as abusive. In that case, the eliminatory nature of such pricing was presumed. In relation to full costs, the Commission had also to provide evidence that WIN’s predatory pricing formed part of a plan to ‘pre-empt’ the market. In neither of those two situations was it necessary to prove, in addition, that WIN had a realistic chance of recouping its losses. (29)

III –  Forms of order sought

30.      FT claims that the Court should:

–        set aside the judgment under appeal; and accordingly:

–        either refer the case back to the Court of First Instance for a fresh decision;

–        or give final judgment and annul the contested decision, by granting the forms of order sought in the application filed by the appellant at first instance;

–        order the Commission to pay the costs.

31.      The Commission contends that the Court should:

–        dismiss the appeal;

–        order the appellant to pay the costs.

IV –  The appeal

32.      The appellant puts forward seven grounds in support of its appeal.

A –    The first ground of appeal, alleging failure to state reasons

33.      The first ground of appeal consists of two complaints. By its first complaint, the appellant claims that the Court of First Instance failed to comply with its duty to provide adequate reasons in respect of the requirement that the possibility to recoup must be proven.

34.      By its second complaint under the first ground of appeal, the appellant claims that the Court of First Instance failed to comply with its duty to provide reasons as regards the right to align prices with those of competitors, which the Court of First Instance ruled out without explanation.

35.      For the sake of convenience, I shall deal with the above complaints in reverse order.

1.      Right to align prices with those of competitors

a)      Principal arguments of the parties

36.      The appellant objects to the fact that the Court of First Instance merely stated at paragraph 187 of the judgment under appeal that ‘[e]ven if alignment of prices by a dominant undertaking on those of its competitors is not in itself abusive or objectionable, it might become so where it is aimed not only at protecting its interests but also at strengthening and abusing its dominant position’ (emphasis added). (30) The appellant submits that the above statement is purely hypothetical and that the Court of First Instance failed to specify in any way whether, in the present case, WIN had intended to strengthen or abuse its dominant position.

37.      The Commission submits that before the Court of First Instance the appellant pleaded only an ‘alignment exception’ or a ‘fundamental right to align’, available to all undertakings, whether dominant or not, even where the prices of its competitors mean that it adopts prices that are below its costs. The Commission therefore contends that the Court of First Instance was right to rule out the existence of such an absolute right.

b)      Assessment

38.      In paragraphs 176 to 182 of the judgment under appeal, the Court of First Instance dealt with the question of WIN’s right to align its prices with those of its competitors. The Court of First Instance found that it was not possible to assert that a dominant undertaking has an absolute right to align its prices with those of competitors. That position had been recognised by the Commission in its previous practice and in the relevant case-law.

39.      That conclusion seems to me to be correct.

40.      However, in paragraph 185 of the judgment under appeal, the Court of First Instance made a general reference to established case-law which provides that, although the fact that an undertaking is in a dominant position cannot deprive it of the right to protect its own commercial interests if they are attacked and such an undertaking must be allowed the right to take such reasonable steps as it deems appropriate to protect those interests, such behaviour cannot be countenanced if its actual purpose is to strengthen this dominant position and abuse it. (31)

41.      Then, at paragraph 186 of the judgment under appeal, the Court of First Instance recalled its own case-law, in which it stated that it follows from the nature of the obligations imposed by Article 82 EC that, in specific circumstances, undertakings in a dominant position may be deprived of the right to adopt a course of conduct or take measures which are not in themselves abuses and which would even be unobjectionable if adopted or taken by non-dominant undertakings. (32)

42.      The Court of First Instance concluded by stating: ‘WIN cannot therefore rely on an absolute right to align its prices on those of its competitors in order to justify its conduct. Even if alignment of prices by a dominant undertaking on those of its competitors is not in itself abusive or objectionable, it might become so where it is aimed not only at protecting its interests but also at strengthening and abusing its dominant position.’ (33)

43.      The Court of First Instance appears to have dealt with the right of a dominant undertaking to align its prices with those of competitors in a general fashion, without, however, applying it to the facts of the particular case before it.

44.      In other words, the Court of First Instance did not specifically analyse whether or not WIN aligned its prices in order to strengthen and abuse its dominant position. In my view, the Court of First Instance completely failed to address that question in the instant case.

45.      In that regard, it is necessary to examine whether the Court of First Instance satisfied the formal obligation to state reasons imposed by Article 36 of the Statute of the Court of Justice, applicable to the Court of First Instance by virtue of the first paragraph of Article 53 thereof. Article 36 provides that judgments shall state the reasons on which they are based. (34)

46.      Advocate General Léger observed in his Opinion in Acerinox that ‘[i]t may be considered that the statement of the reasons on which a judgment is based must clearly and unequivocally disclose the Court of First Instance’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’. (35)

47.      While the Commission argued in its pleadings before the Court of First Instance that a dominant undertaking should not be permitted to align its prices where the costs of the service in question would not be recovered by the dominant undertaking, (36) it is apparent that the Court of First Instance’s answer on that issue is somewhat more subtle. Indeed, it would seem from the wording used in paragraph 187 of the judgment under appeal that the Court of First Instance was attempting to leave the door open for a ‘meeting competition defence’ in future cases, where a dominant undertaking prices below costs.

48.      Having referred to the United Brands line of case-law (37) and the case-law pertaining to the special obligations imposed on undertakings in a dominant position, the Court of First Instance, in answer to its own question in paragraph 184 of the judgment under appeal, adopted a formulation of what constitutes permissible alignment with prices of competitors that is slightly different to the test postulated by the Commission and applied in the contested decision, as described in paragraphs 176 and 183 of the judgment under appeal.

49.      The Court of First Instance should therefore have assessed whether or not that (new) formulation applied to the facts in the instant case, something which it manifestly did not do.

50.      Instead, I agree with the appellant’s argument that the Court of First Instance’s statement at paragraph 187 of the judgment under appeal (38) is purely hypothetical and prima facie appears to be incomplete. In this connection, I cannot accept the Commission’s argument that that statement is ‘completed’ (and/or the new formulation of what constitutes permissible alignment is applied to the facts in the instant case) in paragraphs 199 to 218 of the judgment under appeal, where the Court of First Instance held that the Commission had established the necessary elements to prove predatory pricing below average full costs. First, that argument was proffered too late in the procedure to be admissible, that is to say, it was raised by the Commission only at the hearing and was as such new vis-à-vis the written pleadings. Moreover, in my view, that argument is based on paragraphs of the judgment under appeal which do not relate to paragraph 187.

51.      Finally, I consider the lack of reasoning on the part of the Court of First Instance and the deficiency in its line of argument, as noted above, to be all the more serious as it would appear that this is the first case in which the defence of alignment has been directly addressed by the Community courts in a situation such as that in the present case.

52.      It follows that the Court of First Instance failed in its obligation to state reasons and furthermore erred in law.

53.      Accordingly, the second complaint under the first ground of appeal should be upheld.

2.      Proof of (the possibility of) recoupment of losses

a)      Principal arguments of the parties

54.      The appellant essentially maintains that in Tetra Pak II the Court held that in the circumstances of that particular case it was not appropriate to require, in addition, proof that a dominant undertaking had a realistic chance of recouping its losses. (39) In the judgment under appeal, the Court of First Instance decided to turn that statement into a general rule, but failed to provide reasons for doing so. The appellant adds that the circumstances in Tetra Pak II were totally different from those at issue here, since Tetra Pak II held a so-called super-dominant position in a mature market. Finally, Community case-law has never stated that it is unnecessary to prove the possibility of recoupment of losses.

55.      The Commission refers in its rejoinder to its response to the seventh ground of appeal and in substance contends that Community case-law does not indicate that proof of recoupment of losses is required. In the alternative, the Commission contends that the contested decision did in fact examine in a very detailed manner whether recoupment was likely in the present case and concluded that it was.

b)      Assessment

56.      In paragraph 226 of the judgment under appeal, the Court of First Instance cited paragraph 44 of Tetra Pak II, which states that ‘it would not be appropriate, in the circumstances of the present case, to require in addition proof that Tetra Pak had a realistic chance of recouping its losses. It must be possible to penalise predatory pricing whenever there is a risk that competitors will be eliminated. The Court of First Instance found, at paragraphs 151 and 191 of its judgment [in Tetra Pak II], that there was such a risk in [that] case. The aim pursued, which is to maintain undistorted competition, rules out waiting until such a strategy leads to the actual elimination of competitors.’ The Court of First Instance then held, in paragraphs 227 and 228 of the judgment under appeal, that ‘… the Commission was therefore able to regard as abusive prices below average variable costs. In [such a situation], the eliminatory nature of such pricing is presumed … In relation to full costs, the Commission had also to provide evidence that WIN’s predatory pricing formed part of a plan to “pre-empt” the market. In the two situations, it was not necessary to establish in addition proof that WIN had a realistic chance of recouping its losses … The Commission was therefore right to take the view that proof of recoupment of losses was not a precondition to making a finding of predatory pricing.’ (40)

57.      As has already been stated above in respect of the complaint relating to the right to align prices, the statement of the reasons on which a judgment of the Court of First Instance is based must clearly and unequivocally disclose that court’s reasoning, 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. (41)

58.      In my view, the Court of First Instance failed to explain why it considered that proof of the possibility of recoupment of losses was not necessary in the light of the specific facts of the instant case.

59.      The Court of First Instance cited a judgment which unambiguously stated that it was not appropriate to require proof that a dominant undertaking had a realistic chance of recouping its losses in the circumstances of that particular case. I consider that the Court of First Instance should not have simply turned that statement, which was clearly based on the specific facts of Tetra Pak II, into a general rule. In doing so without any explanation whatsoever, the Court of First Instance has manifestly failed to fulfil its obligation to state adequate reasons.

60.      I therefore conclude that the Court of First Instance committed a double error. First, its incorrect reading and application of Tetra Pak II constitutes use of an incorrect rule. Secondly, as explained above, the Court of First Instance has manifestly failed to fulfil its obligation to state adequate reasons.

61.      I may add that, in addition, the Court of First Instance failed to answer precisely the argument raised by WIN. WIN had not argued that the Commission had to prove the ‘actual’ recoupment of losses. Rather it had submitted that the Commission should show that WIN could ‘anticipate the possibility’ of recouping its losses.

62.      In fact, to the extent that the failure by the Court of First Instance to consider certain parts of the appellant’s arguments affects the outcome of the proceedings and so adversely affects its interests, it may also constitute a breach of the right to be heard. (42)

63.      The first complaint under the first ground of appeal should therefore also be upheld.

64.      It follows from the above considerations that the Court of First Instance breached its duty to provide reasons coupled with an error of law and the judgment under appeal should therefore be set aside.

B –    The first complaint under the seventh ground of appeal, alleging infringement of Article 82 EC – recoupment of losses – possibility of recoupment of losses as a pre-requisite for finding predatory pricing

65.      The seventh ground of appeal consists of two complaints. By its first complaint, which I shall address here, the appellant alleges that the Court of First Instance infringed Article 82 EC in holding that proving the possibility of recoupment of losses was not a pre-requisite for finding predatory pricing.

1.      Principal arguments of the parties

66.      The appellant submits that Community case-law requires proof of a possibility of recouping losses, without which predation is inconceivable, since without the expectation of recoupment predatory pricing does not constitute rational economic behaviour. The appellant maintains that that view is shared by numerous national courts and competition authorities as well as by a large number of academic writers.

67.      The Commission contends that Community case-law does not require separate proof of the possibility of recouping losses. Contrary to the approach of US antitrust law, the analysis of abuse under Article 82 EC presupposes the existence of a dominant position (see Hoffmann-La Roche v Commission (43)), which is sufficient in itself to conclude that recoupment of costs is possible.

2.      Assessment

68.      After referring to Tetra Pak II, the Court of First Instance concludes at paragraph 228 of the judgment under appeal that the Commission was right to take the view that proof of recoupment of losses was not a precondition to making a finding of predatory pricing.

69.      I consider that the Court of First Instance’s, and for that matter the Commission’s, interpretation of the Court’s case-law is incorrect. In my view that case-law requires the possibility of recoupment of losses to be proven.

70.      I consider that, by using the qualifying words ‘in the circumstances of the present case’ in Tetra Pak II, the Court clearly intended to avoid making a general statement (44) that would render it unnecessary to prove the possibility of recoupment in future predatory pricing cases. (45)

71.      Indeed, as Advocate General Fennelly rightly pointed out in his Opinion in Compagnie maritime belge transports and Others v Commission, (46) ‘[t]he Court would not appear to have gone as far as Advocate General Ruiz-Jarabo Colomer, who had recommended [(47)] that the Court “should not lay down the prospect of recouping losses as a new prerequisite for establishing the existence of predatory pricing contrary to Article [82 EC]”, inter alia, because, in [Advocate General Ruiz-Jarabo Colomer’s] view, “recouping losses is the result sought by the dominant undertaking, but predatory pricing is in itself anti-competitive, regardless of whether it achieves that aim”.’

72.      Moreover, in using such qualifying words, the Court clearly departed from the categorical declaration in the Court of First Instance’s judgment in Tetra Pak II that ‘it is not necessary to demonstrate specifically that the undertaking in question had a reasonable prospect of recouping losses so incurred.’ (48)

73.      In my view, it also follows from Akzo (49) and Hoffmann-La Roche (50) that proof of the possibility of recoupment is required in order to find predation pursuant to Article 82 EC. (51) Unless there is a possibility of recoupment, the dominant undertaking is probably engaged in normal competition. (52)

74.      In such a case, where there is no possibility of recouping losses, consumers and their interests should, in principle, not be harmed. I may note here that I share the view of Advocate General Jacobs who in his Opinion in Oscar Bronner stated that ‘the primary purpose of Article [82 EC] is to prevent distortion of competition – and in particular to safeguard the interests of consumers – rather than to protect the position of particular competitors’. (53)

75.      Finally, I will take this opportunity to point out that, apart from the above case-law, the importance of proof of the possibility of recoupment was brought to the fore inter alia by the Economic Advisory Group on Competition Policy (EAGCP), (54) the Organisation for Economic Co-operation and Development (55) and the European Regulators Group. (56)

76.      In my concluding remarks on the question of proof of the possibility of recoupment I would like to point out that I am not convinced by the Commission’s line of argument that in Europe and under Article 82 EC recoupment is implied by the dominance, not least because the determination of dominance is often based on historical market conditions, whilst as I explained above proof of the possibility of recoupment is inherently ex ante and forward-looking, assessing the market structure as it will be in the future. (57)

77.      From the foregoing considerations it follows that the first complaint under the seventh ground of appeal should be upheld.

78.      In view of the above considerations alone, in my view, the judgment under appeal should be set aside. Moreover, I consider that the issues of the proof of the possibility of recoupment and of the right of alignment are so central to the present case that this action requires re-examination at first instance.

C –    The second ground of appeal, alleging infringement of Article 82 EC – alignment of prices with those of competitors

79.      By its second ground of appeal the appellant claims that the Court of First Instance infringed Article 82 EC by denying WIN the right to align its prices, in good faith, with those of its competitors.

1.      Principal arguments of the parties

80.      The appellant maintains that the Court of First Instance did not dispute that WIN had merely aligned its prices with those of its competitors. However, the Court of First Instance failed to draw any conclusion from that. Without explaining why, the Court of First Instance limited itself to stating that such alignment ‘might become’ abusive or objectionable. The Court of First Instance did not therefore respect the conditions for the application of Article 82 EC. The right to such alignment is enshrined in the decision-making practice of the Commission and the case-law of the Court, as well as in academic writing on competition law and in the decision-making practice of the French authorities. Furthermore, it is the only means for the appellant to remain competitive in the market. Finally, the appellant submits that, following the United Brands line of case-law, (58) the Court of First Instance was required to examine whether or not WIN’s riposte to the competition was ‘reasonable’ and ‘appropriate’.

81.      The Commission contends that the appellant alleges neither an error of law on the part of the Court of First Instance in respect of the pleas concerning WIN’s alleged right to alignment, nor any contradiction in the reasoning of the judgment under appeal. As to the arguments relating to a ‘reasonable’ and ‘appropriate’ riposte, the appellant raises them for the first time on appeal, by complaining that the Commission did not carry out such a test of proportionality. In any event, the appellant merely criticises a single paragraph of the judgment under appeal (paragraph 187). (59) The Commission submits that to prohibit alignment by a dominant undertaking when it entails below-cost prices is completely consistent with the principles underlying Article 82 EC, according to which a dominant undertaking has a ‘special responsibility’. In the alternative, the Commission contends that the contested decision concluded on the facts that WIN’s allegation that it had merely aligned itself with its competitors was not founded. The Court of First Instance did not have to rule on those facts.

2.      Assessment

82.      First of all, as I stated in point 39 of this Opinion, the Court of First Instance correctly established in paragraphs 176 to 182 of the judgment under appeal that Community case-law does not accord a dominant undertaking an absolute right to align its prices with those of competitors.

83.      In the present ground of appeal, however, the appellant does not convey what makes the above reasoning on the part of the Court of First Instance incorrect nor does the appellant argue that the Court of First Instance committed an error of law.

84.      The appellant none the less argues, just as it did before the Court of First Instance, that the (absolute) right of any operator to align its prices in good faith with those of competitors is recognised by the Commission itself in its previous decisions and in the case-law of the Court, and meets with unanimous approval in academic literature and economic analysis.

85.      By virtue of the first paragraph of Article 58 of the Statute of the Court of Justice and Article 112(1)(c) of the Rules of Procedure, an appeal must indicate precisely the contested elements of the judgment which the appellant seeks to have set aside, and also the legal arguments specifically advanced in support of the appeal. The Court of Justice has consistently held that that requirement is not satisfied by an appeal which confines itself to repeating or reproducing word for word the pleas in law and arguments previously submitted to the Court of First Instance. (60)

86.      Advocate General Geelhoed referred to the relevant case-law in that respect in his Opinion in Sumitomo Metal Industries and Nippon Steel v Commission: (61) ‘[i]n a now extensive and unequivocal body of case-law, the Court of Justice has held that the pleas submitted must refer to the decision of the Court of First Instance [(62)] and that, on pain of inadmissibility, it must be clearly indicated in which paragraphs of the decision the Court of First Instance infringed the law. [(63)] It follows from that requirement of precision, inter alia, that it cannot suffice merely to mention a plea without elaborating on it. [(64)]’

87.      Therefore the second ground of appeal should, in my view, be declared inadmissible.

88.      In order to give the appellant the benefit of the doubt, let us, however, assume that, in spite of the above doubts as to the admissibility of the present ground of appeal, it should be regarded as admissible.

89.      In my view, contrary to what the appellant seems to argue there is no contradiction in the grounds of the judgment in that the judgment under appeal does not contain any statement to the effect that ‘the Court of First Instance would not contest on the facts that WIN had merely aligned its prices to those of its competitors’.

90.      As I concluded in my assessment of the first ground of appeal above, the Court of First Instance did not apply the relevant case-law on the right of alignment to the facts of the instant case.

91.      In fact, the Court of First Instance did not examine on the facts in the instant case whether or not WIN had actually aligned its prices in good faith with those of its competitors. The problem is that that factual issue is now in dispute between the parties at the appeal stage.

92.      However, appeals before the Court are confined to points of law, which means that the Court of First Instance has sole jurisdiction to make findings in respect of the facts underlying the proceedings at first instance. In addition, parties are not allowed on appeal to offer to adduce evidence of facts not found by the Court of First Instance.

93.      In my view, it follows that the second ground of appeal is doubly inadmissible.

94.      In any event, in accordance with my conclusion in the present Opinion, the judgment under appeal should in my view be set aside and referred back to the Court of First Instance for re-examination. That would allow the Court of First Instance to make findings in relation to the factual issue of WIN’s alignment. (65)

95.      As to the question of principle, I would nevertheless take this opportunity to draw attention to what seem to me to be relevant points in this respect. First, there is the case-law relied on by the Court of First Instance at paragraphs 185 to 186 of the judgment under appeal. Secondly, the Court stated in Tetra Pak II that ‘the actual scope of the special responsibility imposed on a dominant undertaking must be considered in the light of the specific circumstances of each case which show a weakened competitive situation’. (66) Therefore, the question of alignment in a case such as this one should be examined on a case-by-case basis. And, thirdly, in my view one should allow for circumstances where a dominant undertaking is exceptionally permitted to show that its pricing below average variable cost is objectively justified. (67)

D –    The third ground of appeal, alleging infringement of Article 82 EC – distortion of the test of predation

96.      By its third ground of appeal the appellant claims that the Court of First Instance also infringed Article 82 EC in failing to find fault with the Commission’s method for calculating cost recovery, which involved a distortion of the test of predation required by the Court in Akzo. The method which the Commission used made it impossible to know whether WIN subscribers generated a profit or loss for that business during their subscription period.

1.      Principal arguments of the parties

97.      As regards variable costs, the appellant maintains that for the purposes of considering prices lower than variable costs as abusive under Article 82 EC, paragraph 71 of Akzo requires the method for calculating cost recovery to show that the products or services were sold at a loss. The method used can only be compatible with Article 82 EC if it proves that, for some subscribers, the subscription generated a loss over a period of 48 months. However, since WIN claimed in its action for annulment that, for practically all the period at issue, each individual subscriber generated a profit for WIN, the Court of First Instance could not, without breaching Article 82 EC, avoid verifying whether the Commission had established that at least some subscribers had generated an overall loss for WIN. The Court of First Instance nevertheless upheld the Commission’s approach, which did not give a complete view of the profitability of each subscription.

98.      As regards full costs, the appellant refers to its argument concerning variable costs and claims that the Court of First Instance distorted the test of predation by not verifying whether it had been proven that the full costs of subscribers had not been covered. In fact, no evidence was produced of the existence of any subscribers whose subscription had made a loss for WIN over a 48-month period.

99.      The Commission refutes those allegations by arguing, first of all, that not only is the method applied in the present case the same as the one used in the cases leading to the judgments in Akzo and Tetra Pak II – which takes the costs simply as they appear in the undertaking’s accounts – but this time the Commission even relaxed the method in a manner favourable to the appellant. For instance, the Commission neither applied a discount rate nor imputed an opportunity cost among the costs used in the calculation.

100. As regards the alternative method of calculation (the method of discounted cash-flows), the Court of First Instance stated that the Commission had not committed a manifest error of assessment, and the present appeal does not allege that the Court of First Instance committed an error of law in that regard. As to the necessity of taking into account the entire 48-month lifetime of a subscription, the Commission submits that, since the rate of recovery was below 100% for all the successive short periods that were examined in the contested decision (totalling approximately one and a half years), the rate of recovery could not but have been below 100% as well for the whole average duration of subscription, that is to say 48 months. In that respect, the Commission contends that a rate of recovery below 100% cannot become greater than 100% over a longer period, except in a case where the situation in the period after the infringement allows the undertaking to produce profit margins per subscriber that are considerably above a competitive level on a sustained basis.

2.      Assessment

101. The appellant does not clearly specify which parts (that is to say, which paragraphs) of the judgment under appeal are being challenged. (68)

102. In its application before the Court, the appellant argues that the Commission applied a static analysis of recovery of losses which actually amounts to adding the acquisition costs to 48 times the amount of recurrent monthly costs, as they were at the date of subscription, and comparing that total with 48 times the monthly revenues, as they were on that date.

103. In the judgment under appeal, the Court of First Instance held at paragraphs 140 and 142 respectively that ‘contrary to what WIN has alleged, the Commission did not apply a test of static recovery’ and ‘contrary to what WIN asserts, the method does not in effect add the acquisition costs to 48 times the amount of recurrent monthly costs, as they existed at the date of subscription, and compare that total with 48 times the monthly revenues, as they existed on that same date’. (69)

104. However, apart from seemingly contesting on appeal the Court of First Instance’s appreciation of facts, the appellant does not criticise the Court of First Instance for any error of assessment or for any distortion of facts in this respect and only repeats the same arguments submitted to the Court of First Instance. (70) As for the arguments concerning the method of discounted cash flows, the Court of First Instance held at paragraphs 153 to 155 of the judgment under appeal that the Commission did not commit an error of law in not choosing that method and the appellant does not argue that the Court of First Instance committed an error of law in that respect.

105. Therefore I consider that that in itself is sufficient to reject the third ground of appeal as inadmissible and, in any event, unfounded.

E –    The fourth ground of appeal, alleging that the Court of First Instance misconstrued Article 82 EC and its duty to provide reasons – costs and revenues subsequent to the period at issue

106. By its fourth ground of appeal the appellant claims that, by deciding that the costs and revenues subsequent to the period of the alleged infringement (that is to say, after 15 October 2002) should not be taken into account, the Court of First Instance misconstrued both Article 82 EC and its duty to provide reasons. The Commission wrongly concluded there had been an infringement on the basis of that temporal limitation on costs and revenues to be taken into account.

1.      Principal arguments of the parties

107. The appellant complains that the Court of First Instance upheld the Commission’s analysis, which excluded costs and revenues after 15 October 2002. The Court of First Instance could not – without contradicting itself and breaching Article 82 EC – endorse the Commission’s approach, which, on one hand, excluded from the calculation of the rate of recovery revenues and costs subsequent to the alleged infringement but which fell within the 48-month period and, on the other hand, recognised that, as regards subscriptions, the costs and revenues were legitimately spread over 48 months. In order to satisfy the requirements of Article 82 EC, the rate of recovery should be calculated only over a period of 48 months.

108. The appellant submits that by validating the Commission’s position the Court of First Instance committed a contradiction in the grounds in that the contested decision had clearly established that as of 15 October 2002 competitive conditions were ‘restored’ (71) and now the Commission admitted that it had never taken into account revenues subsequent to the alleged infringement because they would have only been realised in a situation of weakened competition.

109. According to the Commission, this ground of appeal is in reality merely an extension of the third ground of appeal and arises from confusion. There has never been a question of allowing all revenues and costs of a subscription to be spread over a period of 48 months. Rather, the Commission’s method, which was accepted by the Court of First Instance, allowed – in accordance with the principle of asset depreciation – a particular category of costs to be spread over a certain period, namely variable non-recurrent costs (‘conquest costs’ or ‘subscriber acquisition costs’) incurred at the beginning of a subscription.

110. The Commission submits that recurrent costs, however, should not be spread over a period of time. According to the contested decision, the undertaking was unable to ‘achieve a level of recovery of recurrent costs (network costs and production costs) which is sufficient to ensure that the margin between revenue and recurrent costs will, within a reasonable time, also cover the non-recurrent variable costs invested in the commercial development of the particular product’. (72) However, all the costs and revenues which arose after the alleged infringement and which the appellant seeks to have included in the calculation are recurrent.

111. The Commission further submits that it would be erroneous to include in the calculation projections of future positive margins resulting from falls in future costs which would be accessible to all competitors but which would not be accompanied by a fall in prices. Such an outcome could be achieved only in the context of weakened competition. The Commission notes that in any event the appellant’s extrapolations would not lead to a positive rate of recovery of full costs and that, even if one accepted the projections presented by the appellant of extremely high profit margins over 48 months, such margins could be realised only in an environment of weakened competition. Finally, the Commission contends that, even excluding ‘acquisition costs’, the direct cost each month to WIN of each subscriber would be greater than revenue from the subscription.

112. Finally, the Commission submits that the appellant’s (erroneous) approach consists of accepting that its subscribers generate increasing monthly losses for one and a half years (the period covered by the contested decision), but then assuming a significant positive margin for the years following the Commission’s intervention, so that the subscribers could turn out to be profitable by the end of the subscription period.

2.      Assessment

113. It would appear, prima facie, that the present ground of appeal is merely an extension of the previous ground and that the appellant repeats the argument that only a 48-month period would have allowed a calculation of the recovery of costs that meets the requirements of Article 82 EC. First of all, the present ground of appeal would appear to be based on the mistaken assumption that the Court of First Instance recognised that ‘costs and revenues are spread over 48 months’. The Court of First Instance never did that in the judgment under appeal. Instead, the Court of First Instance confirmed the Commission’s method of spreading only a category of costs, that is to say variable non-recurrent costs.

114. In any event, my examination of paragraphs 129 to 156 of the judgment under appeal has revealed nothing capable of affecting the validity of the Court of First Instance’s approach. In my view, the Court of First Instance did not commit any error of law and applied Article 82 EC correctly when it held that ‘WIN has not proved that, in using the data recorded in WIN’s accounts and correcting it in favour of WIN to take account of the particular context of the market in question, while complying with the requirements for an assessment under Article 82 EC, the Commission applied an unlawful test of recovery of costs in the present case’ and that ‘it is not apparent from the case-law that use of the method of discounted cash flows was necessary in the present case, and … WIN has not advanced any argument establishing that the Commission committed a manifest error of assessment in this regard.’ (73)

115. Finally, I would recall that, according to the Court’s rules of procedure, the subject-matter of the proceedings before the Court of First Instance may not be changed in the appeal. However, the appellant’s present ground of appeal is significantly different from the plea raised before the Court of First Instance, where the appellant argued that, by refusing to use the discounted cash-flow or net present value (NPV) method (which was, according to the appellant, the only relevant method), the Commission had committed an error of law. The appellant’s application before the Court, however, does not mention that method anymore.

116. Consequently, the fourth ground of appeal should in my view be declared inadmissible and, in any event, unfounded.

F –    The fifth ground of appeal, alleging infringement of Article 82 EC and of the duty to give reasons – price predatory even when accompanied by a considerable reduction in the market share of the relevant undertaking

117. By its fifth ground of appeal the appellant also alleges that the Court of First Instance infringed Article 82 EC, and its duty to give reasons, by holding that a price may be predatory even when it is accompanied by a considerable reduction in the market share of the relevant undertaking. Such a price cannot be regarded as likely to lead to the exclusion of competitors.

1.      Principal arguments of the parties

118. The appellant maintains that, while the Court of First Instance recognised that WIN’s market share had fallen from August 2002, it still found that the alleged infringement had continued until 15 October 2002. However, predation presupposes a significant reduction in competition and therefore cannot occur where competition increases.

119. The Commission submits that at first instance the appellant had raised that argument only in order to dispute the existence of a dominant position and to request a reduction in the fine. It uses the argument in order to dispute the existence of an abuse for the first time on appeal.

120. As to the substance, the information at the Commission’s disposal shows that WIN’s market share constantly grew up until 31 August 2002. Any drop in its market share during the last one and a half months of the infringement could only be due to the reduction of FT’s wholesale tariffs for access to the network, which WIN – in contrast to its competitors – decided not to reflect in its prices, thereby bringing the infringement to an end on 15 October 2002. The Commission adds, for the sake of completeness, that any such reduction in market share would not bring the legality of the decision as such into question but could at the most question the duration of the infringement. That would have no impact on the amount of the fine, however, since no review thereof has been requested in the appeal.

2.      Assessment

121. As the Commission rightly states, the present appeal is limited to the Court of First Instance’s assessment of abuse of a dominant position and WIN raised the argument of an alleged reduction in its market share at first instance only in order to dispute the existence of a dominant position (74) and to request a reduction in the fine. (75)

122. According to settled case-law, to allow a party to put forward for the first time before the Court of Justice a plea in law which it has not raised before the Court of First Instance would be to allow it to bring before the Court, whose jurisdiction in appeals is limited, a case of wider ambit than that which came before the Court of First Instance. In an appeal the Court’s jurisdiction is confined to review of the findings of law on the pleas argued before the Court of First Instance. (76)

123. The appellant’s fifth ground of appeal should therefore be declared inadmissible.

G –    The sixth ground of appeal, alleged distortion of facts and infringement of Article 82 EC – alleged plan of predation

1.      Distortion of facts and evidence

124. The sixth ground of appeal consists of two complaints. By the first complaint the appellant claims that, as regards the alleged plan of predation, the Court of First Instance distorted the facts and the evidence submitted for its consideration.

a)      Principal arguments of the parties

125. The appellant maintains that the Court of First Instance distorted the evidence on which it based its analysis of whether a plan of predation existed. It relied exclusively on WIN’s documents and deduced the existence of a plan of predation from terminology used therein such as ‘pre-emption’ and ‘to pre-empt’. However, those documents merely reflected ‘rather ambitious commercial objectives’, to use the Court of First Instance’s own words. (77)

126. The Commission submits that the appellant seeks a re‑examination on appeal of a complaint which was rejected as inadmissible by the Court of First Instance, without, however, disputing the finding that it is inadmissible. Secondly, the appellant did not bring forward any argument to support the alleged distortion, and it is for the Court of First Instance alone to assess the value which should be attached to the evidence adduced before it. For the sake of completeness, the Commission maintains that the Court of First Instance did not base its analysis exclusively on documents containing the expression ‘to pre-empt’ but also on numerous other documents.

b)      Assessment

127. The appellant is seeking re-examination of a plea which was rejected by the Court of First Instance in paragraphs 204 and 205 of the judgment under appeal as inadmissible.

128. However, in the present appeal the appellant may not raise that plea afresh. He can merely challenge, on specific grounds, the Court of First Instance’s finding that the plea at first instance is inadmissible (78) and must advance arguments in that connection to establish that the Court of First Instance has erred in law. (79) In my view, the appellant has failed to challenge the finding of inadmissibility of the Court of First Instance.

129. The present complaint concerns a question of fact and offers no convincing arguments in support of an alleged distortion of facts.

130. Under Article 225 EC and the first paragraph of Article 58 of the Statute of the Court of Justice the Court has no jurisdiction to establish the facts or, in principle, to examine the evidence which the Court of First Instance accepted in support of those facts. Provided that the evidence has been properly obtained and the general principles of law and the rules of procedure in relation to the burden of proof and the taking of evidence have been observed, it is for the Court of First Instance alone to assess the value which should be attached to the evidence produced to it. (80) That appraisal does not therefore constitute, save where the clear sense of the evidence has been distorted, a point of law which is subject as such to review by the Court of Justice. (81)

131. The first complaint under the sixth ground of appeal should therefore be declared inadmissible.

2.      Infringement of Article 82 EC

132. By its second complaint the appellant claims that the Court of First Instance infringed Article 82 EC. That article requires an objectively identifiable plan to remove competitors and cannot, in any circumstances, be satisfied by a subjective test of the concept of an abuse of a dominant position.

a)      Principal arguments of the parties

133. The appellant maintains that abuse of a dominant position is an objective concept. However, the Court of First Instance based its analysis and conclusion on purely subjective factors, rather than on any objective evidence such as threats to competitors, selective pricing vis-à-vis competitors’ customers, price discrimination, outbidding, or the duration, consistency and magnitude of loss-making sales.

134. The Commission contends, first, that the element of intent in abuse is necessarily subjective and, secondly, that Community case-law does not require a plan of predation to be substantiated by the type of objective evidence suggested by the appellant.

b)      Assessment

135. It is sufficient to state that, just as in some of the previous grounds of appeal, the appellant appears to be resorting to the same arguments that it raised before the Court of First Instance.

136. Therefore the second complaint under the sixth ground of appeal should also be rejected as inadmissible.

137. As regards the plan of predation, in my view, the Court of First Instance has correctly and to the requisite legal standard applied here the Akzo and Tetra Pak II case-law.

138. Therefore, the sixth ground of appeal should be rejected as inadmissible and, in any event, unfounded.

H –    The second complaint under the seventh ground of appeal, alleging infringement of Article 82 EC – recoupment of losses – undertaking’s evidence on the impossibility of recoupment of losses

139. By its second complaint under the seventh ground of appeal the appellant alleges that the Court of First Instance infringed Article 82 EC in confusing the Commission’s evidence on the possibility of recoupment of losses with the undertaking’s evidence on the impossibility of recoupment of losses.

1.      Principal arguments of the parties

140. The appellant maintains that WIN itself provided proof that recoupment of losses was impossible. The Court of First Instance should therefore have adjudicated on the question whether the Commission was entitled to disregard proof of the impossibility of recoupment which is provided by the undertaking.

141. The Commission submits that the appellant raised no plea at first instance regarding the question whether the Commission was entitled to disregard such proof. In any event, the Commission maintains that the argument is rejected implicitly by paragraphs 103 to 121 and 261 to 267 of the judgment under appeal. Finally, the Commission contends that, for the sake of completeness, it had in fact analysed the possibility of recoupment in the contested decision and considered it likely in the case at issue.

2.      Assessment

142. In my view, the appellant did not raise any argument before the Court of First Instance claiming that recoupment of losses was impossible other than claiming that the Commission in the contested decision itself proved the impossibility of recoupment of losses by its assertion that ‘the conditions of competition were restored’.

143. Indeed, the obligation of the Court of First Instance to give reasons for its decisions does not go so far as to require it to respond in detail to every single argument advanced by the appellant, particularly if the argument was not sufficiently clear and precise and was not adequately supported by evidence. (82)

144. I consider that the argument that WIN advanced in its application before the Court of First Instance – to the effect that the Commission had contradicted itself in its market analysis – in order to prove that recoupment was impossible cannot be considered sufficiently clear and precise. Nor did it amount to an assertion that the Commission was not entitled to disregard proof of the impossibility of recoupment provided by the undertaking in question.

145. In any event, such proof would appear to be implicitly but necessarily rejected in paragraphs 103 to 121 and 261 to 267 of the judgment under appeal.

146. I therefore consider that the second complaint under the seventh ground of appeal should be rejected as inadmissible as a new argument or plea not raised before the Court of First Instance and, in any event, unfounded.

V –  Conclusion

147. In the light of all of the foregoing, I propose that the Court should set aside the judgment under appeal.

Furthermore, I consider it appropriate for the Court to refer the case back to the Court of First Instance for judgment, in accordance with the first paragraph of Article 61 of the Statute of the Court of Justice, and reserve the costs.


1 – Original language: English.


2 – Predatory pricing is sometimes also referred to as ‘charging below-cost prices’.


3 – Case C‑62/86 AKZO v Commission [1991] ECR I‑3359 (‘Akzo’).


4 – Case C‑333/94 P Tetra Pak v Commission [1996] ECR I‑5951 (‘Tetra Pak II’).


5 – Commission Decision of 16 July 2003 relating to a proceeding under Article [82 EC] (Case COMP/38.233 − Wanadoo Interactive) (‘the contested decision’).


6 – Asymmetric Digital Subscriber Line.


7 – Article 1 of the contested decision.


8 – Articles 2 and 4 thereof.


9France Télécom v Commission [2007] ECR II‑107.


10 –      Pursuant to the powers conferred on it by Article 12(1) of (EEC) Regulation No 17 of the Council of 6 February 1962: First Regulation implementing Articles [81] and [82] of the Treaty (OJ, English Special Edition 1959-1962, p. 87).


11 –      See point 2 of the present Opinion.


12 – In support of its action before the Court of First Instance, WIN also made alternative claims for the cancellation or a reduction of the fine, alleging: (i) breach of the principles that penalties must be specific to the offender and have a proper legal basis, (ii) absence of effects of the conduct in question, (iii) that the duration of the infringement was wrongly determined, and (iv) breach of the principle of proportionality.


13 – See paragraphs 122 to 230 of the judgment under appeal.


14 – See paragraph 138 of the judgment under appeal.


15 – See paragraph 129, recalling, to that effect, Case C‑7/95 P John Deere v Commission [1998] ECR I‑3111, paragraph 34 and the case-law cited.


16 – Cited in footnotes 3 and 4 respectively.


17 – See paragraph 130 of the judgment under appeal.


18 – See paragraph 132.


19 – See paragraphs 140 to 143.


20 – See paragraph 152.


21 – See paragraph 153.


22 – See paragraphs 165 to 169.


23 – See paragraph 186.


24 – See paragraph 187.


25 – See paragraph 188.


26 – See paragraphs 195 to 198.


27 – See paragraphs 199 to 215.


28 – See paragraphs 219 to 223.


29 – See paragraphs 224 to 229.


30 – See paragraph 187 of the judgment under appeal.


31 – Citing Case 27/76 United Brands v Commission [1978] ECR 207, paragraph 189; Case T‑65/89 BPB Industries and British Gypsum v Commission [1993] ECR II‑389, paragraph 117; and Joined Cases T‑24/93 to T‑26/93 and T‑28/93 Compagnie maritime belge transports and Others v Commission [1996] ECR II‑1201, paragraph 146.


32 – Citing Case T‑111/96 ITT Promedia v Commission [1998] ECR II‑2937, paragraph 139.


33 – See paragraph 187 of the judgment under appeal.


34 – See, in respect of the corresponding provisions of the ECSC Statute of the Court, Case C‑197/99 P Belgium v Commission [2003] ECR I‑8461, paragraph 63.


35 – Case C‑57/02 P Acerinox v Commission [2005] ECR I‑6689, point 32 of the Opinion, citing, to that effect, Case C‑259/96 P Council v De Nil and Impens [1998] ECR I‑2915, paragraphs 32 to 34; Case C‑449/98 P IECC v Commission and Others [2001] ECR I‑3875, paragraph 70; and the orders of 19 July 1995 in Case C‑149/95 P(R) Commission v Atlantic Container Line and Others [1995] ECR I‑2165, paragraph 58; of 14 October 1996 in Case C‑268/96 P(R) SCK and FNK v Commission [1996] ECR I‑4971, paragraph 52; and of 25 June 1998 in Case C‑159/98 P(R) Netherlands Antilles v Council [1998] ECR I‑4147, paragraph 70. See also Acerinox, paragraph 36 of the judgment.


36 – One could readily find fault with that argument, since the Commission does not specify which costs (variable or total, etc.) it has in mind.


37 – See United Brands v Commission, cited in footnote 31, paragraph 189; BPB Industries and British Gypsum v Commission, cited in footnote 31, paragraph 117; and Compagnie maritime belge transports and Others v Commission, cited in footnote 31, paragraph 146.


38 – Which essentially states that alignment might become objectionable where it is aimed not only at protecting the dominant undertaking’s interests but also at strengthening and abusing its dominant position.


39Tetra Pak II, cited in footnote 4, paragraph 44.


40 – See paragraphs 226 to 228 of the judgment under appeal.


41 – See the case-law cited in footnote 35.


42 – See Case C‑221/97 P Schröder and Others v Commission [1998] ECR I‑8255, paragraph 24.


43 – Case 85/76 [1979] ECR 461.


44 – For a similar interpretation, see, for instance, Whish, R., Competition Law, Butterworths, 4th edition, 2001, at p. 650; Jones, A., and Sufrin, B., EC Competition Law, Oxford University Press, 2001, at p. 342; Korah, V., ‘The Paucity of Economic Analysis in the Decision on Competition: Tetra Pak II’, Current Legal Problems, vol. 46, 1993, at p. 172; and Kon, S., and Turnbull, S., ‘Pricing and the Dominant Firm: Implications of the Competition Commission Appeal Tribunal’s Judgment in the NAPP Case’ (2003) E.C.L.R. 24(2) 70 to 86, at p. 75.


45 – However, in my view it is clear from paragraph 44 of the Court’s judgment in Tetra Pak II that proof of actual recoupment is not required. This follows from the Court’s statement in that paragraph that ‘[i]t must be possible to penali[s]e predatory pricing whenever there is a risk that competitors will be eliminated. The Court of First Instance found, at paragraphs 151 and 191 of its judgment, that there was such a risk in [that] case. The aim pursued, which is to maintain undistorted competition, rules out waiting until such a strategy leads to the actual elimination of competitors.’


46 – Joined Cases C‑395/96 P and C‑396/96 P [2000] ECR I‑1365, point 129 and footnote 84.


47 – See point 78 of his Opinion in Tetra Pak II, case cited in footnote 4.


48 – See Case T‑83/91 Tetra Pak International v Commission [1994] ECR II‑755, paragraph 150.


49 – In particular paragraph 71 of the judgment.


50 – In particular paragraph 91 of the judgment.


51 – In my view, the analysis of the possibility of recoupment of losses requires an ex ante or forward-looking appraisal of market structure after exclusion has occurred, that is to say, the possibility of recoupment as it can be reasonably foreseen by the dominant undertaking at the time when it establishes its pricing. Indeed, it would appear that such an appraisal is similar to the analysis undertaken by the Commission in the area of merger control. Cf. Case C‑12/03 P Commission v Tetra Laval [2005] ECR I‑987, paragraphs 42 to 44. See also, for instance, Ezrachi, A., EC Competition Law, Hart, 2008, p. 140; Papandropoulos, P., ‘Article 82 EC reform’, Droit & économie, Concurrences No 1 – 2008; and de la Mano, M., and Durand, B., A Three-Step Structured Rule of Reason to Assess Predation under Article 82, DG Competition, European Commission, Office of the Chief Economist Discussion Paper, 12 December 2005, p. 3.


52 – See, to that effect, the Opinion of Advocate General Fennelly in Compagnie maritime belge transports and Others, cited in footnote 46, point 136. In addition, since the price/cost analysis required in predatory pricing cases is bound to be controversial and contentious (cf. Akzo, cited in footnote 3, see also: Howarth, D., ‘Unfair and Predatory Pricing under Article 82 EC’, in EC Competition Law, G. Amato and C.-D. Ehlermann ed., Hart, 2007, p. 258-262), not least with regard to questions such as which costs are variable, which fixed, what the relevant time period is and so forth, requiring proof of possibility of recoupment would in principle be beneficial to alleviate the cost of errors (especially false positives) in alleged predatory pricing cases. Furthermore, as regards the ex ante proof of the possibility of recoupment as a useful and administrable screen, I agree that, in principle, ‘it is much easier to determine from the structure of the market that recoupment is improbable than it is to find the cost a particular producer experiences’. See the US case A. A. Poultry Farms, Inc v Rose Acre Farms, Inc 881 F.2d 1396, (7th Cir. 1989), cert denied, 494 U.S. (1990). And such analysis of recoupment ‘enables a court to avoid getting into the messy area of cost analysis, examination of various accounting figures and competing expert evidence on the question of what are the relevant costs’. See the Australian case Boral (2003) 195 ALR 609, paragraph 292.


53 – Opinion in Case C‑7/97 [1998] ECR I‑7791, point 58.


54 – ‘In particular, the recoupment ability of the predator has to be carefully assessed’, in Report by the EAGCP which is a group of distinguished academic economists advising the Commission’s Directorate-General for Competition and the Commissioner for Competition, ‘An economic approach to Article 82’, July 2005, p. 52, available at: http://ec.europa.eu/comm/competition/publications/studies/eagcp_july_21_05.pdf.


55 – ‘The recoupment test should be routinely applied in predatory pricing cases’ in OECD, Predatory Foreclosure, Directorate for Financial and Enterprise Affairs, Competition Committee, DAF/COMP(2005)14, 15 March 2005, p. 8 et seq., available at: http://www.oecd.org/dataoecd/26/53/34646189.pdf. For a recent general overview of 35 jurisdictions, see The International Competition Network, Report on Predatory Pricing, prepared by The Unilateral Conduct Working Group, Presented at the 7th Annual Conference of the ICN, Kyoto, April 2008, available at: http://www.icn-kyoto.org/documents/materials/Unilateral_WG_3.pdf.


56 – The European Regulators Group for electronic communications networks and services, set up by the Commission, states at page 36 of a consultation document of 21 November 2003 on a ‘Draft joint ERG/EC approach on appropriate remedies in the new regulatory framework’ the following: ‘Predatory pricing thus has the following characteristics: (i) the price charged is below costs, (ii) competitors are either driven out of the market or excluded, and (iii) the undertaking is able to recoup its losses’ (emphasis added). That document is available at: http://www.erg.eu.int/doc/publications/erg0330_draft_joint_approach_on_remedies.pdf.


57 – Cf. ‘At the same time, however, it should be appreciated that proof of dominance does not necessarily imply that the dominant firm will also be able to recoup its losses. While proof of recoupment means that the dominant firm’s monopoly would persist in future, it is important to appreciate that the loss-making and recoupment phases do not coincide … The fact that a firm was dominant at the time it engaged in below-cost selling does not mean that it will in future be able to recover past losses by raising prices: conditions of competition may well be different in future … Second, predatory pricing claims have in practice only succeeded under Article 82 EC where recoupment was either actually established or probable on the facts’, in O’Donoghue, R., and Padilla, A.J., The Law and Economics of Article 82 EC, Oxford: Hart, 2006, p. 254. See also: ‘Agencies should be mindful, though, that testing for dominance is not equivalent to doing a recoupment analysis’, OECD, cited in footnote 55, p. 8 et seq.


58 – Cited in footnote 31.


59 – According to which alignment with prices of competitors ‘might become [abusive and objectionable] where it is aimed … at strengthening and abusing [the undertaking’s] dominant position’.


60 – See Schröder and Others v Commission, cited in footnote 42, paragraph 35. Such an interpretation of an appeal would amount to nothing more than an attempt to have the case re-tried and no provision is made for retrials by the Court. See, for instance, Case C‑499/03 P Biegi Nahrungsmittel and Commonfood v Commission [2005] ECR I‑1751, paragraphs 37 and 38.


61 – Opinion in Joined Cases C‑403/04 P and C‑405/04 P [2007] ECR I‑729.


62 – Pleas which are in essence directed against an act or failure to act by the institution(s) are inadmissible. The resubmission of those pleas would mean that it was an ‘ordinary’ appeal. See, inter alia, the judgments in Case C‑401/96 P Somaco v Commission [1998] ECR I‑2587, paragraph 49, and Case C‑161/97 P Kernkraftwerke Lippe-Ems v Commission [1999] ECR I‑2057, paragraphs 76 and 77.


63 – See, inter alia, Schröder and Others v Commission, cited in footnote 42, paragraphs 35 and 38 to 42, and Case C‑257/98 P Lucaccioni v Commission [1999] ECR I‑5251, paragraphs 61 and 62.


64 – See Case C‑51/92 P HerculesChemicals v Commission [1999] ECR I‑4235, paragraph 113.


65 – Where additional findings of fact are needed or a fresh look has to be taken at those already made, the Court will refer the case back to the Court of First Instance. See, in this connection, Case C‑68/91 P Moritz v Commission [1992] ECR I‑6849, paragraphs 41 and 42.


66 – See Tetra Pak II, cited in footnote 4, paragraph 24.


67 – See the Opinion of Advocate General Fennelly in Compagnie maritime belge transports and Others, cited in footnote 31, point 127. Cf. Tetra Pak International v Commission, cited in footnote 48, paragraph 147: ‘it may be acceptable for an undertaking in a dominant position to sell at a loss in certain circumstances’. See, inter alia, also Bellamy & Child, European Community Law of Competition, Oxford , 6th ed., 2008, p. 956-957, section 10.071: ‘To the extent that [the Akzo rule] implies a rule that pricing below average variable costs … must be predatory, it seems too rigid. Prices below [average variable costs] may be set pro-competitively …’, Motta, M., Competition Policy, Cambridge University Press, 2004, p. 453; Faull, J., and Nikpay, A., The EC law of competition, Oxford University Press, 2007, p. 379, section 4.287; Ezrachi, A., cited in footnote 51, p. 136, OECD, cited in footnote 55, p. 18: ‘Just because a firm is pricing below cost does not necessarily mean its actions are harming competition. There are some circumstances under which such pricing is not only benign, but actually pro-competitive. Agencies should thoroughly investigate any justifications that alleged predators proffer for their pricing’, Whish, R., cited in footnote 44, p. 649, Jones, A., and Sufrin, B., cited in footnote 44, p. 339, Garzaniti, L., and Liberatore, F., ‘Recent Developments in the European Commission’s Practice in the Communications Sector: Part 2’ (2004) E.C.L.R. 25(4) p. 234-240.


68 – In this connection, so as to avoid repetition, I refer back to points 85 and 86 of the present Opinion, which are also applicable here.


69 – See paragraphs 143 to 151 of the judgment under appeal for the Court of First Instance’s reasoning why that is so.


70 – An appeal cannot merely repeat the pleas in law and arguments already submitted to the Court of First Instance without putting forward arguments to establish that it erred in law. See order in Case C‑30/96 P Abello and Others v Commission [1998] ECR I‑377, paragraph 45.


71 – See paragraph 386 of the contested decision.


72 – See the 76th recital of the contested decision and paragraph 136 of the judgment under appeal.


73 – See paragraphs 154 and 155. With regard to the discussions on the net present value approach see, for instance, the UK case in NappPharmaceutical Holdings Ltd and subsidiaries v Director General of Fair Trading [2002] CAT 1, paragraph 260.


74 – Paragraph 95 of WIN’s application at first instance, in the plea concerning the existence of a dominant position.


75 – Paragraph 272 of WIN’s application at first instance, in the plea concerning the fine.


76 – See Case C‑68/05 P Koninklijke Coöperatie Cosun v Commission [2006] ECR I‑10367, paragraphs 95 to 98 and the case-law cited there.


77 – That reference is to paragraph 214 of the judgment under appeal.


78 – See Lenaerts, K., Arts, D., Maselis, I., Procedural Law of the European Union, 2nd edition, London 2006, p. 464: ‘This may be inferred from Case C‑354/92 P Eppe v Commission [1993] ECR I‑7027, paragraph 13’.


79 – See Abello and Others v Commission, cited in footnote 70, paragraph 45.


80 – Case C‑185/95 P Baustahlgewebe v Commission [1998] ECR I‑8417, paragraph 24.


81 – Joined Cases C‑24/01 P and C‑25/01 P Glencore and Compagnie Continentale v Commission [2002] ECR I‑10119, paragraphs 65 to 69. Such distortion must be obvious from the documents on the Court’s file, without there being any need to carry out a new assessment of the facts and the evidence. See Case C‑551/03 P General Motors v Commission [2006] ECR I‑3173, paragraphs 51, 52 and 54.


82 – See Case C‑274/99 P Connolly v Commission [2001] ECR I‑1611, paragraph 121, and Belgium v Commission, cited in footnote 34, paragraph 81.