Opinion of the Advocate-General

Opinion of the Advocate-General

Table of contents

I –  Introduction

II –  Legal framework

III –  Background to the dispute and the course of the procedure

A – The concentration

B – The problem of collective market dominance

C – Administrative procedure and the Commission’s first clearance decision

D – The proceedings at first instance

E – Appeal proceedings

F – New administrative proceedings and the Commission’s second clearance decision

IV –  Preliminary questions to analysis of the appeal

A – Admissibility of the individual grounds of appeal

1. Admissibility of the fifth ground of appeal

2. Admissibility of the remaining grounds of appeal

3. Interim conclusion

B – Whether the appeal is capable of achieving the result sought

1. The alleged failure of the appellants to challenge a decisive passage of the judgment under appeal

2. The grounds of the judgment under appeal which do not support its operative part

3. Interim conclusion

C – Continuing existence of interest in pursuing the appeal

V –  Whether the appeal is well founded

A – Extent of investigation and justification required of the Commission in relation to clearance decisions (first, second, third and sixth grounds of appeal)

1. Standard of reasoning for merger clearance decisions (first and third part of the sixth ground of appeal)

a) Whether clearance decisions may be challenged on the ground of failure to state reasons

b) Extent of the duty to state reasons

– The judgment under appeal

– Analysis

c) Miscellaneous matters

d) Interim conclusion

2. The Court of First Instance’s references to the statement of objections (first ground of appeal and second part of the sixth ground of appeal)

a) The Court of First Instance’s approach in general to the relationship between the clearance decision and the statement of objections

– The judgment under appeal

– Analysis

b) The specific references by the Court of First Instance to the statement of objections

c) Interim conclusion

3. The evidential value of factual assertions in the response to the objections (second ground of appeal; first part of the third ground of appeal)

a) The judgment under appeal

b) Analysis

4. Standard of proof for clearance of concentrations (second part of the third ground of appeal)

a) The alleged asymmetry between the requirements for clearance decisions and prohibition decisions in the Merger Regulation

– Merger control decisions are based on prognoses

– Symmetry of the requirements for clearances and prohibitions

– No general presumption of compatibility with the common market

– Interim conclusion

b) The standard of proof applied by the Court of First Instance in the present case

c) Interim conclusion

B – Limitations on the Court of First Instance’s power to analyse the facts and the evidence (fourth ground of appeal)

1. The Commission’s discretion and judicial review thereof

a) The judgment under appeal

b) Analysis

2. The alleged distortion of evidence

a) Paragraph 425 of the judgment under appeal

b) Paragraph 427 of the judgment under appeal

c) Paragraph 434 of the judgment under appeal

3. Interim conclusion

C – The use of undisclosed evidence in the judgment under appeal (seventh ground of appeal)

D – Interim conclusion

VI –  The supposed cross appeal

VII –  Costs

VIII –  Conclusion

I – Introduction

1. The present case presents an opportunity to develop the Court’s case-law in the area of EC merger control. In essence it concerns the question – a most important one in practice – of the extent of the investigation and reasoning which may be required of the Commission where it authorises a concentration between undertakings.

2. The background to this case is a merger control procedure concerning the markets for recorded music, on-line music and music publishing. At the end of 2003, Bertelsmann and Sony agreed that they would integrate their global recorded music businesses. Following initial doubts, by Decision dated 19 July 2004 (2) (‘the first clearance decision’) the Commission approved this proposed concentration.

3. The Independent Music Publishers and Labels Association (‘Impala’), an international association governed by Belgian law to which 2 500 independent music production undertakings belong, complained about the concentration. On its application, by judgment dated 13 July 2006 (3) (‘the judgment under appeal’) the Court of First Instance annulled the first clearance decision.

4. An appeal by Bertelsmann and Sony against the judgment under appeal has been brought before the Court of Justice. In essence, the appellants argue that the Court of First Instance applied excessive legal requirements for a Commission clearance decision and for the judicial review of that decision.

5. However, there is a preliminary question as to whether the appellants still have any legal interest in pursuing their appeal given that in the meantime, on 3 October 2007, their concentration was cleared once more by the Commission.

II – Legal framework

6. The legal framework in the present case consists of Council Regulation (EEC) No 4064/89 of 21 December 1989 on the control of concentrations between undertakings (‘the Merger Regulation’) (4) as amended by Council Regulation (EC) No 1310/97. (5)

7. Concentrations with a Community dimension within the meaning of the Merger Regulation are subject to a prohibition on implementation and must be notified to the Commission (Articles 4 and 7 of the Merger Regulation). The Commission appraises them as to their compatibility with the common market (Article 2(1) of the Merger Regulation).

8. Whether a concentration is cleared or prohibited depends on whether it creates or strengthens a dominant position as a result of which effective competition would be significantly impeded in the common market or in a substantial part of it. On that point Article 2(2) and (3) of the Merger Regulation provide as follows:

‘(2) A concentration which does not create or strengthen a dominant position as a result of which effective competition would be significantly impeded in the common market or in a substantial part of it shall be declared compatible with the common market.

(3) A concentration which creates or strengthens a dominant position as a result of which effective competition would be significantly impeded in the common market or in a substantial part of it shall be declared incompatible with the common market.’

9. The merger control procedure consists of two phases. The first phase is simply a preliminary examination of the proposed concentration. If this raises serious doubts as to the concentration’s compatibility with the common market, there follows in a second phase a formal procedure which the Commission is required to initiate in terms of Article 6(1)(c) of the Merger Regulation. (6)

10. The Commission’s powers of decision in the formal procedure are laid down in Article 8 of the Merger Regulation as follows:

‘(1) … all proceedings initiated pursuant to Article 6(1)(c) shall be closed by means of a decision as provided for in paragraphs 2 to 5.

(2) Where the Commission finds that, following modification by the undertakings concerned if necessary, a notified concentration fulfils the criterion laid down in Article 2(2) …, it shall issue a decision declaring the concentration compatible with the common market. …

(3) Where the Commission finds that a concentration fulfils the criterion laid down in Article 2(3) …, it shall issue a decision declaring that the concentration is incompatible with the common market.

…’

11. Before a prohibition decision may be taken under Article 8(3) of the Merger Regulation, the parties to the concentration are given a hearing. For this purpose the Commission sends them a written statement of objections , to which they may respond in writing and, as the case may be, orally. On this matter Article 18 of the Merger Regulation provides as follows: (7)

‘(1) Before taking any decision provided for in Article … 8 … (3) to (5) …, the Commission shall give the persons, undertakings and associations of undertakings concerned the opportunity, at every stage of the procedure up to the consultation of the Advisory Committee, of making known their views on the objections against them.

...

(3) The Commission shall base its decision only on objections on which the parties have been able to submit their observations. The rights of the defence shall be fully respected in the proceedings. …’

12. The whole merger control procedure is characterised by a requirement for speed , which is realised by a finely balanced, comparatively strict system of time-limits and which seeks to limit the length of merger control proceedings. (8) If the Commission does not make a decision as to compatibility with the common market within the specified time-limit, there is a deemed clearance in respect of which Article 10(6) of the Merger Regulation provides as follows:

‘Where the Commission has not taken a decision in accordance with Article 6 (1) (b) or (c) or Article 8 (2) or (3) within the deadlines set …, the concentration shall be deemed to have been declared compatible with the common market …’

13. As regards the consequences where a Decision by the Commission is annulled by a Community Court, Article 10(5) of the Merger Regulation provides as follows:

‘Where the Court of Justice gives a judgment which annuls the whole or part of a Commission decision taken under this Regulation, the periods laid down in this Regulation shall start again from the date of the judgment.’

14. In most merger control proceedings the Commission acts also as merger control authority for the European Economic Area (9) and decides also as to whether concentrations are compatible with the EEA Agreement.

15. In 2004 the Merger Regulation was substantially amended. However, under the new Article 26(1), the recast EC Merger Regulation (10) applies only from 1 May 2004 and is therefore not relevant to the present case; Article 26(2) provides that the previous law remains applicable to cases such as the present.

III – Background to the dispute and the course of the procedure

A – The concentration

16. Bertelsmann AG (11) is an international media company, having worldwide activities in music recording and publishing, television, radio, book and magazine publishing, printing and media services, and book and music clubs. In the area of music recording Bertelsmann operates through its wholly-owned subsidiary Bertelsmann Music Group (‘BMG’). (12)

17. Sony Corporation of America (13) is part of the Sony Group and operates worldwide in the areas of music recording, music publishing, industrial and consumer electronics and entertainment. In the recorded music sector Sony acts through Sony Music Entertainment.

18. Under a Business Contribution Agreement dated 11 December 2003 Bertelsmann and Sony agreed to combine their global recorded music businesses (with the sole exception of Sony’s business in Japan) and to transfer it into at least three new undertakings. It was intended that these joint venture undertakings should operate under the name ‘Sony BMG’.

19. Under the agreement, Sony BMG was to discover and develop artists, (14) and deal with the marketing and selling of recorded music. On the other hand, its business activities were not to include related activities such as publishing, manufacturing and distribution of recorded music.

20. As Bertelsmann and Sony confirmed in the proceedings before the Court, the concentration was brought fully into effect in 2004.

B – The problem of collective market dominance

21. As with all concentrations having a Community dimension, under Article 2 of the Merger Regulation the proposed concentration of Bertelsmann and Sony required to be appraised as to whether it was likely to create or strengthen a dominant position as a result of which effective competition would be significantly impeded in the common market or in a substantial part of it.

22. In particular, the issue was not whether Sony BMG would achieve an individual dominant position in a given market, but the risk of the creation or strengthening of a collective dominant position between five global recorded music publishers, (15) the so-called ‘majors’ of the worldwide music industry, the number of which would, by virtue of the concentration, be reduced from five to four.

23. In the case of Kali & Salz , (16) the Court has already made it clear that the concept of a dominant position for the purposes of the Merger Regulation refers not only to individual but also to collective market dominance. A collective dominant position may be constituted by two undertakings (duopoly) or several undertakings (oligopoly). (17)

24. Collective market dominance does not necessarily require that the members of an oligopoly cooperate collusively , for example by entering into anti-competitive agreements within the meaning of Article 81 EC. (18) Instead, collective market dominance may arise from the tacit coordination of market behaviour (19) of all members of the oligopoly. The members of the oligopoly are then contented with the market shares they have achieved, and there is no longer any effective competition between them. (20)

25. However, a necessary condition for a finding of collective market dominance in those circumstances is that the market in question actually admits in the long term of tacit coordination of market behaviour by the members of an oligopoly. The three criteria developed by the Court of First Instance in Airtours (21) for the purpose of determining whether collective market dominance exists, and summarised in the judgment under appeal as follows, are to be seen against this background:

‘First, the market must be sufficiently transparent for the undertakings which coordinate their conduct to be able to monitor sufficiently whether the rules of coordination are being observed. Second, the discipline requires that there be a form of deterrent mechanism in the event of deviant conduct. Third, the reactions of undertakings which do not participate in the coordination, such as current or future competitors, and also the reactions of customers, should not be able to jeopardise the results expected from the coordination.’ (22)

26. It is the first of the so-called ‘ Airtours criteria’ which is at the heart of the present case, and thus the question as to whether the markets for recorded music are sufficiently transparent for the majors to be able to coordinate their conduct tacitly. In essence the parties to the proceedings are in dispute as to the efforts the Court of First Instance may require from the Commission in terms of investigation and justification in that respect.

C – Administrative procedure and the Commission’s first clearance decision

27. On 9 January 2004 the notification of the concentration required by Article 4 of the Merger Regulation was lodged with the Commission. (23)

28. In the preliminary phase (‘Phase I’ of the merger control procedure) the Commission issued a questionnaire to a number of market participants on 20 January 2004.

29. Impala answered the questionnaire, and on 28 January 2004 it also lodged a separate submission in which it set out the reasons why, in its view, the Commission should declare the operation incompatible with the common market. In this submission Impala set out its concerns about further concentration in the market and the impact that this would have on market access, including in the retail sector, the media, the internet and consumer choice.

30. Thereafter, on 12 February 2004 the Commission initiated formal proceedings pursuant to Article 6(1)(c) of the Merger Regulation (‘Phase II’ of the merger control procedure), because the project notified raised serious doubts as to its compatibility with the common market and the EEA Agreement.

31. Under Article 11(5) of the Merger Regulation the proceedings were stayed from 7 April to 5 May 2004, because the parties to the concentration provided incomplete information to the Commission’s request for information. (24)

32.  On 24 May 2004 the Commission sent a statement of objections to the parties to the concentration, in which it provisionally concluded that the notified operation was incompatible with the common market and the functioning of the EEA Agreement, since it would strengthen a collective dominant position in the recorded music market and in the wholesale market for licences for online music and would coordinate the parent companies’ behaviour in a way incompatible with Article 81 EC.

33. The parties to the concentration replied to the statement of objections and a hearing took place before the Commission’s Hearing Officer on 14 and 15 June 2004 in the presence of, among others, Impala.

34. By its first clearance decision of 19 July 2004, the Commission finally declared the concentration compatible with the common market and the EEA Agreement pursuant to Article 8(2) of the Merger Regulation. This clearance was not subject to any conditions or obligations.

D – The proceedings at first instance

35. On 3 December 2004 Impala brought an action against the first clearance decision before the Court of First Instance and applied for the Decision to be annulled (25) and for the Commission to be ordered to bear the costs of the proceedings. The Commission applied for the action to be dismissed and for Impala to be ordered to bear the costs. It was supported in that regard by Bertelsmann and Sony, as well as by Sony BMG Music Entertainment, to whom the Court had given leave to intervene by Order of the President of the Third Chamber of 4 February 2005.

36. By the judgment under appeal the Court of First Instance annulled the first clearance decision and ordered the Commission to bear its own costs and three quarters of Impala’s costs. Impala itself was ordered to bear one quarter of its own costs. The interveners were ordered to bear their own costs.

E – Appeal proceedings

37. By their joint appeal, which was lodged at the Registry of the Court on 13 July 2006, Bertelsmann and Sony (‘the appellants’) claim that the Court should:

– set aside the judgment under appeal;

– dismiss Impala’s application for annulment of the Commission’s Decision, or, alternatively, refer the case back to the Court of First Instance for reconsideration; and

– order Impala to pay the costs of the present proceedings.

38. Sony BMG Music Entertainment adopts in full both the appeal and the form of order sought by Bertelsmann and Sony.

39. The Commission contends that the Court should:

– set aside the judgment under appeal;

– dismiss the application for annulment of the first clearance decision or, in the alternative, refer the case back to the Court of First Instance for reconsideration; and

– order Impala to pay the costs incurred by the Commission in the present proceedings.

40. Impala contends that the Court should:

– dismiss the appeal as unfounded and/or inadmissible in part or inadmissible in its entirety;

– uphold the judgment under appeal; and

– order the appellants to pay the costs of the present proceedings.

41. Before the Court the appeal procedure was first written and then, on 6 November 2007, oral.

F – New administrative proceedings and the Commission’s second clearance decision

42. In consequence of the annulment of the first clearance decision by the judgment under appeal, the Commission carried out a new merger control procedure in this case (26) (in this regard see Article 10(5) of the Merger Regulation), which necessarily ran in parallel to the present appeal proceedings , because the appeal did not have any suspensory effect (Article 60(1) of the Statute of the Court).

43. Thus, for the purpose of its new competitive assessment the concentration was notified again to the Commission on 31 January 2007, and on 1 March 2007 the Commission initiated the formal procedure under Article 6(1)(c) of the Merger Regulation (‘Phase II’).

44. The merger control procedure was concluded by decision dated 3 October 2007, in which the Commission again declared the concentration to be compatible with the common market and the EEA Agreement under Article 8(2) of the Merger Regulation (‘the second clearance decision’). The clearance was not subject to any conditions or obligations.

IV – Preliminary questions to analysis of the appeal

45. Before considering the appeal as to substance, it is necessary to consider whether the individual grounds of appeal it raises are admissible (see below, section A), whether they can at all achieve the result sought by the appellants (see below, section B), and whether the appellants’ interest in continuing to pursue the appeal has not been superseded by the second clearance decision which has since been issued (see below, section C).

A – Admissibility of the individual grounds of appeal

46. Impala resists the appeal on the ground that it is inadmissible in its entirety, because it merely seeks reconsideration of the Court of First Instance’s appraisal of the facts.

47. Given this objection, it is appropriate to recall the standard of reference laid down by Article 225(1) EC and Article 58(1) of the Statute of the Court of Justice, and which the Court has confirmed in its consistent case-law in appeal proceedings: (27) appeals are limited to points of law. The Court of First Instance thus has exclusive jurisdiction to find and appraise the relevant facts and to assess the evidence, and the appraisal of those facts and the assessment of that evidence thus do not, save where the facts or evidence are distorted, constitute points of law subject, as such, to review by the Court on appeal.

48. By contrast, whether the Court of First Instance has imposed too high a standard as regards the reasons given for a Commission decision, whether it has applied the correct criteria in its appraisal of the facts and of the evidence, and whether it has generally applied the law correctly in its judgment, are all questions of law which may be the subject of an appeal. (28) Likewise, the Court has jurisdiction to review the legal characterisation of the facts and the evidence by the Court of First Instance and to review the legal conclusions it has drawn from them, (29) as well as whether it has properly applied the provisions relating to the burden of proof and the taking of evidence. (30)

49. Applying these criteria, contrary to Impala’s submissions the present appeal cannot be regarded as wholly inadmissible. Instead, in this connection a distinction must be drawn between the fifth ground of appeal and the other grounds of appeal.

1. Admissibility of the fifth ground of appeal

50. The fifth ground of appeal, which concerns the requirements for a finding of a collective dominant position, consists in an introductory part (31) and a number of detailed points of criticism. (32)

51. The introductory part raises essentially the question whether the criteria the Court of First Instance itself developed in Airtours (33) for a finding of collective market dominance are to be applied with different degrees of strictness depending on whether what must be proved is a collective dominant position which already exists or a prognosis for the future as regards the risk that a collective dominant position will arise in consequence of a concentration. Thus, the introductory part of the fifth ground of appeal concerns the interpretation of the criteria in Airtours . This is a question of law which may indeed be the subject of an appeal.

52. The position is different as regards the remaining detailed points of criticism which are made in the context of the fifth ground of appeal and which may be summarised as follows. The Court of First Instance failed to take into account net wholesale prices and the importance of price discounts, which are decisive on the issue of market transparency, and instead satisfied itself with considering only list prices and retail prices. Moreover, the Court of First Instance erroneously inferred transparency of discounts from the impact of such discounts on average net prices. In addition, in assessing transparency the Court of First Instance wrongly dismissed the relevance of complex pricing structures. Finally, the Court of First Instance erroneously dismissed price variance and volatility of prices as irrelevant to the question of transparency.

53. In reality, these detailed points of criticism do not seek a review of questions of law but instead call in question the actual appraisal of the facts and the evidence by the Court of First Instance in the present case. This is because the relevance of list prices, retail prices, net wholesale prices, average net prices, price discounts, the complexity of price structures, as well as price variance and volatility of prices to the assessment of transparency of a particular market cannot be authoritatively determined in general terms, but depends on the actual analysis of the circumstances of the individual case, and specifically on the particular circumstances of the market concerned.

54. In other words, the Court is being asked to substitute its own appraisal of the facts and of the evidence in this particular case for that of the Court of First Instance. However, this is not permissible in an appeal. At most, the plea that may be raised before the Court is that the Court of First Instance has distorted facts or evidence, or has offended against rules of logic in its assessment of the facts or the evidence. Here, however, neither of those is alleged, and apart from these exceptions it is not the Court’s task, when sitting as an appellate court, to decide for itself what relevance factors such as list prices, retail prices, net wholesale prices, average net prices, price discounts, the complexity of price structures, as well as price variance and volatility of prices have to the assessment of market transparency in a case such as the present.

55. On that basis, it is only the introductory part of the fifth ground of appeal which is admissible, and not the detailed points of criticism in it.

2. Admissibility of the remaining grounds of appeal

56. By contrast, I regard the remaining grounds of appeal as admissible in their entirety, because their subject-matter is not the appraisal of facts and evidence as such by the Court of First Instance. Instead, these grounds of appeal relate to the criteria which the Court of First Instance has applied in assessing the legality of the first clearance decision. Moreover, they concern the evidential requirements to which the Commission is subject in clearing concentrations. All of these are questions of law which are capable of being the subject-matter of an appeal.

3. Interim conclusion

57. Whereas only the introductory part of the fifth ground of appeal is admissible, the remaining grounds of appeal are admissible in their entirety.

B – Whether the appeal is capable of achieving the result sought

58. Apart from the question discussed above as to the admissibility of the individual grounds of appeal, the present case raises the additional question of whether the appeal is in any event capable of achieving the result the appellants seek, namely that the judgment under appeal is set aside. First, the appellants may have failed to challenge a decisive passage in the judgment under appeal (see below, section 1) and, second, it may be that their appeal is directed against parts of the reasons for the judgment which are not the grounds on which the operative part of the judgment under appeal is founded (see below, section 2).

1. The alleged failure of the appellants to challenge a decisive passage of the judgment under appeal

59. First, Impala submits that the appeal as a whole cannot result in the judgment under appeal being set aside, because in their appeal the appellants have failed to challenge a decisive passage in the judgment.

60. Specifically, Impala is of the view that the appeal is restricted to challenges to the reasoning of the Court of First Instance as regards the question of strengthening an existing collective dominant position, and by contrast is not directed against the reasoning of the Court of First Instance in paragraph 528 of the judgment under appeal as regards the possible creation of a collective dominant position. Thus, even if the appeal were wholly successful, according to Impala it could not result in the judgment under appeal being set aside, because the findings of the Court of First Instance contained in paragraph 528 of the judgment as regards further errors of law in the first clearance decision would still stand.

61. I do not find this submission convincing.

62. Admittedly, the appellants must indicate precisely the contested elements of the judgment which they seek to have set aside and also the legal arguments specifically advanced in support of the appeal. (34) However, this does not mean that the drafting of their appeal must correspond in every detail to the structure of the judgment under appeal, and must challenge each section of this judgment by a specific ground of appeal.

63. In the present case, the appellants have by no means restricted their appeal to only the first of the sections in the judgment called into question, that is to the reasoning of the Court of First Instance as regards the strengthening of an existing collective dominant position on the recorded music market. (35) Moreover, such a restriction would not correspond to the result they have stated they wish to achieve by the appeal, namely to have the judgment under appeal set aside in whole. This is because this result can be achieved only if the appeal is to be understood as challenging both sections of the judgment in question, (36) that is both the section concerning the strengthening of an existing collective dominant position and the section concerning the creation of such a dominant position.

64. The appellants’ introductory remarks at the start of their appeal also suggest that the appeal is not restricted to only one section of the judgment. A list of passages in the judgment under appeal which are specifically criticised expressly refers to paragraph 528, which is the one in dispute at this point, and also to paragraphs 533, 539 and 541, which likewise concern the question of the creation of a collective dominant position. (37) In addition, at a number of points in their appeal the appellants expressly incorporate into their reasoning the overall conclusion of the judgment under appeal as formulated by the Court o f First Instance in paragraphs 542 and 543. (38)

65. In these circumstances it cannot be asserted that the appeal is restricted only to the reasoning of the Court of First Instance as regards the strengthening of an existing collective dominant position and must therefore be dismissed in whole as ineffective (in French, ‘inopérant’).

2. The grounds of the judgment under appeal which do not support its operative part

66. The position is different as regards the introductory part of the fifth ground of appeal which, as already mentioned, (39) is the only admissible part of that ground of appeal.

67. In essence it raises the question whether the criteria developed by the Court of First Instance itself in Airtours (40) for establishing the existence of a collective dominant position are to be applied with different degrees of strictness depending on whether what must be established is a collective dominant position which already exists or a prognosis for the future as to the risk that such a collective dominant position might arise in consequence of the concentration. In the former case the Court of First Instance favours a less strict application of the Airtours criteria than in the latter. (41)

68. In this regard the appellants’ main point of challenge is paragraph 251 of the judgment under appeal. There the Court of First Instance states that an (already existing) collective dominant position ‘may, however, in the appropriate circumstances, be established indirectly on the basis of what may be a very mixed series of indicia and items of evidence relating to the signs, manifestations and phenomena inherent in the presence of a collective dominant position’.

69. However, as the appellants themselves correctly pointed out, (42) the disputed passage, and also the legal reasoning of the Court of First Instance connected to it, are obiter dicta . In the whole of the section of the judgment under appeal, from paragraph 245 to 253, the Court of First Instance is not considering one of the grounds of invalidity raised by Impala at first instance, but is merely stating obiter dicta on the application of the criteria for establishing whether a collective dominant position exists.

70. This becomes particularly clear if one reads the disputed passage of the judgment with paragraph 254 of the judgment under appeal, in which the Court of First Instance expressly restricts its consideration of the first clearance decision to the question whether the Airtours criteria are fulfilled: ‘[T]he Court will confine itself, in its examination of the pleas in law put forward, to ascertaining that the Decision properly applied the conditions defined in Airtours v Commission . …’. By contrast, in its judgment the Court of First Instance expressly rules out considering the question whether the existence of a dominant position can be found to be established, ‘without its being necessary positively to establish market transparency’. In the proceedings at first instance this question had, ‘not been discussed’.

71. Thus, because the introductory part of the fifth ground of appeal does not refer to the ratio decidendi of the judgment but merely to additional considerations mentioned by the Court of First Instance, it cannot result in the judgment under appeal being set aside. In accordance with the consistent case-law of the Court this part of the appeal is to be dismissed as ineffective (inopérant). (43)

3. Interim conclusion

72. Thus, only the fifth ground of appeal, so far as it is at all admissible, is to be dismissed as ineffective. By contrast, all the other grounds of appeal are capable of achieving the result sought by the appellants.

C – Continuing existence of interest in pursuing the appeal

73. As regards the grounds of appeal which are admissible and not ineffective it remains to be considered whether the appellants have in the meantime ceased to have an interest in pursuing the appeal.

74. The requirement for such an interest ensures at a procedural level that the courts are not asked to give expert opinions on purely hypothetical questions of law. Accordingly, the existence of an interest is a mandatory requirement of admissibility, which must be considered ex proprio motu and which can be relevant at various stages of the proceedings. Thus, there is no doubt that an interest must exist at the time an action is brought or an appeal lodged; however, it must also continue to exist beyond that time and up to the Court’s decision in the case. (44)

75. An interest exists so long as the appeal is likely, if successful, to procure an advantage to the party bringing it. (45)

76. It is not disputed that at the time they lodged their appeal the appellants had an interest in it. This is because, from the time the first clearance decision was annulled by the Court of First Instance, Bertelsmann and Sony did not have the necessary approval retrospectively for their concentration under the Merger Regulation. If the judgment under appeal were set aside, as the appeal seeks, this approval could have been revived. The appeal could thus have procured them this advantage.

77. However, since the second clearance decision was issued, the participating undertakings once more have clearance for their concentration under the Merger Regulation. The operative part of this second clearance does not fall short of the first in any way, both having declared the concentration to be compatible with the common market and the EEA Agreement, and neither being subject to any conditions or obligations. In addition, the state of uncertainty which has existed since the first clearance decision was annulled, and in which the participating undertakings and the markets could not be certain whether the concentration had been lawfully carried out, was removed by the renewed clearance of the concentration. Specifically, the legal approach taken in Article 7(5) of the Merger Regulation (46) has the effect that the second clearance decision has retrospective effect to the time at which the transactions by which the concentration was brought about were carried out (‘ ex tunc effect’).

78. All this does not necessarily mean that the present appeal is devoid of purpose . This is because the appeal is not aimed directly at the Commission’s first clearance decision, but against the judgment of the Court of First Instance under appeal, which continues to have effect. Accordingly, the possibility of declaring that there is no need to adjudicate on the action (Article 92(2) of the Rules of Procedure) is not applicable in appeal proceedings (see Article 118 of the Rules of Procedure). (47)

79. Nevertheless, the existence of the second clearance decision raises the question as to whether the appeal can procure any advantage for the appellants, that is whether they continue to have an interest in maintaining the appeal.

80. The sole fact that in the judgment under appeal the appellants were ordered to bear their own costs from the proceedings at first instance does not in any event establish that they have an interest. This is because Article 58(2) of the Statute of the Court provides that an appeal directed solely against a decision on costs is inadmissible. For the same reason, the Court dismisses an appeal against a decision at first instance on costs if this is the only one remaining of a number of grounds of appeal and it has already been established that none of the other grounds succeeds. (48) Having regard to the wording and purpose of Article 58(2) of the Statute, nor can an interest in setting aside the decision at first instance on costs justify continuing an appeal. Accordingly, in order to pursue appeal proceedings, it is necessary for the appellant to have an interest beyond that relating to the costs at first instance.

81. In the present case, however, the appellants have an obvious interest in obtaining as quickly as possible not only clearance but also a definitive clearance of their concentration. Only then will the participating undertakings, and indeed the market generally, have legal certainty as to whether the concentration was lawfully effected.

82. The second clearance decision cannot provide this certainty at this time, because it may not become definitive for the foreseeable future. This is because it must, in addition to being notified to its addressees (Article 254(3) EC), be published in the Official Journal of the European Union (Article 20(1) of the Merger Regulation). It is only after such publication, (49) which has so far not occurred, that the period for third parties to challenge it begins to run. (50) The beginning of the period for third parties to bring a challenge could be brought forward generally only if the complete second clearance decision had otherwise become known to the public at an earlier time, for example via the internet, and an appropriate reference had been made in the Official Journal of the European Union . (51) By contrast, the mere fact that the wording of the decision may have been brought to the knowledge of individual third parties, specifically Impala, in advance in the form of a non-confidential version, does not affect the beginning of the period for bringing a challenge (52) and thus the time at which the decision can become final. (53)

83. Should the Court of First Instance annul the second clearance decision too, following the application of a third party, a state of uncertainty would again arise, in which the participating undertakings would not have any clearance for their concentration under the Merger Regulation. Such a situation of uncertainty lasting a number of months or even years could have negative effects on the participating undertakings and on the markets generally.

84. In order, so far as possible, to avoid another such state of uncertainty and its negative effects, it is of particular importance to the appellants to continue with the present appeal and to obtain a final decision from the Community Courts on the legality of the first clearance decision as quickly as possible. Even if, should the Court allow the appeal, the state of the proceedings does not permit it to give a final decision itself on the present dispute, and it must instead refer the case back to the Court of First Instance (Article 61(1) of the Statute of the Court), it would be highly probable that the final judicial decision on the lawfulness of the first clearance decision would be given before that on the lawfulness of the second .

85. In these circumstances the requirement for speed which characterises merger control at the Community level (54) equally favours continuing the present appeal. By this means it is also possible to prevent annulment actions raised by third parties from delaying certainty for the undertakings participating in the concentration for any longer than is necessary.

86. In all, I am therefore of the opinion that at the moment the decision of the Court can certainly procure the appellants an advantage and it follows that they have a sufficient interest in maintaining the present appeal.

87. By contrast, if the second clearance decision becomes final after the close of the oral proceedings but before the decision of the Court of Justice in the present case, on the information available to me at the moment this would result in the appellants ceasing to have an interest. However, I would then consider it necessary to hear the parties again on the question of interest.

88. I n any event, I do not find convincing the idea cursorily mentioned at the oral hearing that the present proceedings could be pursued solely on the ground that the Commission supports some of the individual grounds of appeal relied upon by Bertelsmann and Sony. Community organs have a privileged status as regards appeals they have raised themselves (Article 56(2) and (3) of the Statute of the Court), and in particular need not prove an interest in such appeals. (55) However, where such an organ is not itself the appellant but only one of the other parties to the proceedings, its mere interest in the outcome of the appeal and in the clarification of certain questions of law by the Court cannot compensate for the appellants’ lack of interest.

V – Whether the appeal is well founded

89. The appellants challenge the judgment under appeal with a total of seven grounds of appeal. They raise legal questions of fundamental importance for the system of merger control at Community level. Admittedly, these questions of law arise in relation to the ‘old’ Merger Regulation, but the answers to them are still decisive for the recast EC Merger Regulation (Regulation No 139/2004), because there is no significant difference between the two regulations as regards the points in dispute in the present case.

90. Given that the individual grounds of appeal overlap in part, it is appropriate to group them according to the substantive points to which they relate, and accordingly to consider them in a different order. Only the fifth ground of appeal needs no further consideration, because, as already mentioned, it is in part inadmissible and in part ineffective (‘inopérant’). (56)

A – Extent of investigation and justification required of the Commission in relation to clearance decisions (first, second, third and sixth grounds of appeal)

91. The first, second, third and sixth grounds of appeal concern in particular the pains the Commission must take by way of investigation and giving reasons where it grants clearance for a concentration of undertakings.

92. In essence, the appellants are of the view that the Court of First Instance applied excessive legal requirements relating to a clearance decision by the Commission and to judicial review thereof. In that regard they are supported in part by the Commission. (57) Impala, on the other hand, defends the judgment under appeal in its entirety.

1. Standard of reasoning for merger clearance decisions (first and third part of the sixth ground of appeal)

93. I start my consideration with the first and third parts of the sixth ground of appeal. (58) There the appellants claim that the Court of First Instance applied erroneous and excessive standards of reasoning for Commission merger clearance decisions.

94. It must first be clarified whether Commission clearance decisions can at all be annulled on account of insufficiency of reasons. If so, we must consider the extent of the duty to state reasons and whether the Court of First Instance made any errors of law in this regard in the present case.

a) Whether clearance decisions may be challenged on the ground of failure to state reasons

95. The appellants are of the view that a merger clearance decision by the Commission cannot in any event be annulled for insufficient reasoning.

96. I am not convinced by this view.

97. Article 253 EC provides that decisions by the Commission shall state the reasons on which they are based. This duty to state reasons is a consequence of the rule of law and, in conjunction with the right to proper administration, finds expression also in Article 41(2) of the Charter of fundamental rights of the European Union. (59) It is intended not only to enable acts of the institutions to be subject to independent review by the Community Courts, but also to encourage the institutions to exercise self-control and to guard against ill-considered or ill-thought through measures. The statement of reasons for decisions also contributes to ensuring transparency of administrative action. (60)

98. The obligation to state reasons is by no means restricted to decisions which negatively affect their addressees. The principles of the rule of law and of proper administration instead require reasons to be given also for decisions which benefit their addressees. This applies a fortiori where such decisions may negatively affect the rights and interests of third parties, not least in the field of competition law. Accordingly, neither Article 253 EC nor the third indent of Article 41(2) of the Charter of fundamental rights distinguishes between decisions which benefit their respective addressees and those which are disadvantageous for them. In relation to merger control in particular, this means that reasons must be given for Commission clearance decisions just as much as for prohibition decisions.

99. As it infringes an essential procedural requirement, a breach of the duty to state reasons may be complained of inter alia by means of an annulment action before the Community Courts subject to the conditions in Article 230 EC.

100. Article 10(6) of the Merger Regulation, on which the appellants rely in the present case, does not provide for any exception to the liability of decisions to challenge on the ground of breach of the duty to state reasons. It follows simply from the hierarchy of norms that provisions of secondary law cannot restrict the scope of primary law, including Article 230 EC and Article 253 EC. Instead, Article 10(6) of the Merger Regulation is to be interpreted and applied in light of higher-ranking provisions, in particular Article 230 EC and Article 253 EC. (61)

101. In any event, to introduce an exception to the duty to state reasons, or even to confer on Commission clearance decisions an ‘immunity’ from challenge on the ground of failure to state reasons, would not correspond to the wording or to the purpose and legislative context of Article 10(6) of the Merger Regulation.

102. Article 10(6) of the Merger Regulation provides for deemed clearance only where the Commission does not give a decision in good time as to the compatibility with the common market of a concentration which has been notified to it. (62) By contrast, the provision does not relieve the Commission of its statutory duty (63) to make an express finding by way of a reasoned decision in relation to every concentration notified to it. (64) On the contrary, the provision contains a remedy should the Commission in a given case not comply with this obligation within the applicable time-limit.

103. In addition, Article 10(6) of the Merger Regulation is certainly an expression of the requirement for speed which characterises the whole of the merger control procedure. Together with the strict procedural time-limits laid down by that same article, this provision contributes to creating legal certainty as quickly as possible, and this is to the benefit not only of the undertakings participating in the concentration but also of the markets in general.

104. Nevertheless, the legitimate need for legal certainty cannot go so far as to exclude the decision on a concentration in whole or in part from review by the courts. It is only once the time-limit under Article 230(5) EC for bringing a challenge has expired, or for example an annulment action which has been raised is dismissed, that the clearance of the concentration becomes final and creates definitive legal certainty for all the participants.

105. Contrary to the submission of the appellants, Article 10(6) of the Merger Regulation is by no means deprived of its practical effectiveness (‘ effet utile ’) solely by the fact that the – explicit or deemed – clearance remains subject to review by the courts. Instead, the impending grant of a deemed clearance in case of the expiry of the time-limit (the ‘guillotine effect’) promotes self-discipline in every merger control procedure dealt with by the Commission as the merger control authority, and this to an extent which is not to be underestimated. (65) In addition, the participating undertakings are free to implement their concentration as soon as it is cleared by the Commission or deemed to have been cleared; (66) according to the information they themselves provided, the appellants indeed made use of this possibility in the present case.

106. In all, I conclude that clearance of a concentration – whether explicit or deemed – can be annulled on the ground of breach of the duty to state reasons.

b) Extent of the duty to state reasons

107. The appellants are in addition of the view that in the present case the Court of First Instance erred in finding that insufficient reasons were given for the first clearance decision. In their opinion, the Court of First Instance thereby set itself against the established case-law of the Community Courts.

108. In essence the dispute revolves around the question of the level of detail in which the Commission was required to justify its finding in the first clearance decision that the market was not so transparent as to enable prices to be co-ordinated.(67)

109. This finding was of relevance to the Commission’s assessment that there was no sufficient basis for finding a collective dominant position of the five majors on various national markets for recorded music and that the concentration was not likely to create such a collective dominant position. (68) The clearance of the concentration was based not least on this assessment.

– The judgment under appeal

110. In the judgment under appeal the Court of First Instance reviewed various sections of the first clearance decision as to whether they provided sufficient justification for the finding that the market was not transparent, and in each case decided the issue in the negative.

111. The Court of First Instance considered, first, the section of the first clearance decision which concerned the issue of market transparency, (69) and found that the Commission ‘did not conclude that the market was opaque or not sufficiently transparent to allow a collective dominant position’. In addition, in that section the Commission mentioned ‘only factors capable of giving rise to great transparency in the market and of facilitating the monitoring of compliance with collusion, with the sole exception of the rather limited and unsubstantiated assertion that campaign discounts could reduce transparency and make tacit collusion more difficult’. Thus, this specific section of the first clearance decision could ‘clearly not, in itself, be considered to support to the requisite legal standard the assertion that the market was not sufficiently transparent’. (70)

112. The Court of First Instance then directed its attention to the Commission’s consideration of a potential ‘common price policy’ (71) between the five majors, (72) and likewise reviewed whether there were sufficient factors to prove the alleged lack of market transparency. (73) In this connection it considered both the information provided by the Commission as regards list prices and that as regards price discounts (file discounts and campaign discounts). In this regard the Court of First Instance held first, that ‘according to the very terms of the Decision, list prices … constitute a factor of market transparency’. (74) Second, ‘the few assertions relating to campaign discounts … in so far as they are imprecise, unsupported, and indeed contradicted by other observations in the Decision, cannot demonstrate the opacity of the market or even of campaign discounts …’. (75) Finally, ‘the section dealing with the smaller countries, too, contains no reasoning for the finding that the market is not transparent on account of the campaign discounts’.(76)

113. Specifically in regard to the campaign discounts relied upon by the Commission, the Court of First Instance felt that there was a lack of more specific findings in the first clearance decision as regards ‘the nature of campaign discounts, the circumstances in which such discounts might be applied, their degree of opacity, their size or their impact on price transparency’. (77) The Court of First Instance also bemoaned that ‘[t]hose assertions [by the Commission as regards campaign discounts in larger countries] are confined, moreover, to indicating that campaign discounts are less transparent than file discounts, but do not explain how they would be relevant for the transparency of the market and do not make it possible to understand how they in themselves might compensate for all the other factors of transparency of the market identified in the Decision and thus eliminate the transparency necessary for the existence of a collective dominant position’. (78)

– Analysis

114. In the appeal it is not the task of the Court to substitute its own assessment of the first clearance decision for that of the Court of First Instance. Accordingly, the Court need not itself review and assess the clearance decision as to whether its reasoning was inadequate or sufficient. Instead, the Court has to decide whether the Court of First Instance for its part made any errors of law in the judgment under appeal when it reviewed the reasons given for the clearance decision, and in particular whether the Court of First Instance based its review on correct or on excessively strict criteria.

115. The Court has consistently held that the statement of reasons required by Article 253 EC must be appropriate to the measure at issue and must disclose in a clear and unequivocal fashion the reasoning followed by the institution which adopted that measure in such a way as to enable the persons concerned to ascertain the reasons for it and to enable the competent Community Court to exercise its power of review. (79)

116. The requirements to be satisfied by the statement of reasons depend on the circumstances of each case, in particular the content of the measure in question, the nature of the reasons given and the interest which the addressees of the measure, or other parties to whom it is of direct and individual concern, may have in obtaining explanations. It is not necessary for the reasoning to go into all the relevant facts and points of law, since the question whether the statement of reasons meets the requirements of Article 253 EC must be assessed with regard not only to its wording but also to its context and to all the legal rules governing the matter in question. (80)

117. The particular features of a merger control procedure include the time pressure to which the Commission is subject on account of the requirement for speed which applies in that context and the strict procedural time-limits. (81) In addition, in giving reasons for its decisions in the field of competition law the Commission can orientate itself by reference to the degree of difficulty of the individual case, and in doing so assume the knowledge of informed market participants who are familiar with market circumstances; (82) this applies a fortiori to cases involving rights and interests of market participants who, like Impala in the present case, themselves participated in the proceedings. (83) The appellants correctly referred to both of these points.

118. Accordingly, it clearly cannot be required that, in a decision on a concentration notified to it, the Commission defines its position on matters which are manifestly irrelevant or insignificant or plainly of secondary importance. (84) The obvious need not be expressly mentioned in the decision. Moreover, the Commission need not provide more detailed reasons for its decision than is appropriate having regard to the degree of difficulty of the particular case, or than appears to be absolutely necessary from the point of view of an informed market participant familiar with the market circumstances.

119. At least the factual and legal considerations which are of decisive importance in the context of the decision must always be capable of being understood from the reasons given. (85) Accordingly, the reasons given must not be so laconic as to endanger their clarity and persuasiveness. (86) They must in addition be logical (87) and must not disclose any internal inconsistencies. (88)

120. In the present case the Court of First Instance considered that the reasons given were not persuasive, comprehensible and free of inconsistency.

121. In essence, what the Court of First Instance found to be inadequate was an imbalance in the reasons given for the first clearance decision: it contained a list of factors pointing to the existence of market transparency and explained these in detail, (89) but by contrast, as regards the factor of campaign discounts, which allegedly militate against the existence of sufficient market transparency, it made only vague assertions. (90) There was no explanation as to why the campaign discounts were relevant to market transparency, and it was not apparent how campaign discounts taken by themselves outweighed all the other factors relating to market transparency listed in the decision and thus could exclude the transparency necessary for there to be a collective dominant position. (91) Apart from that, the Court of First Instance referred to internal inconsistencies in the reasons given for the decision. (92)

122. I agree with the Court of First Instance that, in a decision in which the Commission has initially described in detail a list of factors which point to the existence of market transparency, the Commission cannot proceed on the footing that there is a lack of market transparency without giving more detailed reasons. If there is only a single factor worthy of mention – in the present case, campaign discounts – which militates against the existence of sufficient market transparency and thus a finding of a collective dominant position, it is necessary to explain in all the more detail how this factor affects the market and the extent to which precisely this factor can outweigh all the other factors pointing to the existence of market transparency.

123. In particular, it is not enough to explain that a particular factor leads or may lead to less market transparency; instead, it is necessary at least to explain why exactly this factor makes the market so opaque that it is not possible to make a finding of a collective dominant position. Otherwise, the reasons for the decision lack persuasiveness and are not comprehensible. The Court of First Instance correctly referred to exactly this in the judgment under appeal. (93)

124. Simply put, in a merger control decision which to a large extent reads as if it were a prohibition decision, it is essential that a sufficiently precise explanation is given of the considerations on the basis of which the matter ultimately turns, even for informed readers who are familiar with the market.

125. The deficiencies the Court of First Instance found to exist in the reasons given are all the more serious given that the question of market transparency, which was disputed by the parties, was not merely a subsidiary matter but was of major importance for the result of the merger control procedure. (94) This is because the Commission’s finding that the market was not sufficiently transparent as to allow prices to be coordinated was a major reason it relied upon for the first clearance decision. (95) In this regard the Court of First Instance rightly imposed high standards on the reasons given for the first clearance decision and subjected those reasons to an intense review.

126. High standards were also justified here because, as regards market transparency, the Commission was required to assess complex economic situations, in relation to which it enjoyed a not insignificant margin of discretion, as is usual in the field of merger control. (96) However, where the Commission has such a margin of discretion, observance of the guarantees conferred by the Community legal order in administrative proceedings acquires more importance. These guarantees include not least the obligation to give sufficient reasons for a decision. (97)

127. If, in the exercise of its discretion, the Commission takes into account factors which influence the functioning of the market, it must not only identify these factors in its decision but also state their effects. (98)

128. This consideration may be transposed to the field of competition law and merger control. If, in its assessment of the effect on competition of a concentration, the Commission attaches particular importance to certain factors which are relevant to the market, it must not only identify these factors in its decision but must also describe, sufficiently precisely, their effects on the functioning of the markets in question.

129. In a case in which all other factors point to the transparency of the market, the factor which, in the Commission’s view, points decisively against the market’s being sufficiently transparent cannot be simply mentioned without further explanation. As the Court of First Instance correctly emphasises, what is instead necessary is for the Commission to substantiate the effects of this factor in its decision with specific details; (99) it cannot simply rely on the surmise, hesitantly expressed, that ‘this factor could reduce transparency in the market and may make tacit collusion more difficult’, (100) and just as little on the mere conjecture that ‘it appears that campaign discounts are less transparent than file discounts’. (101)

130. In all, the Court of First Instance therefore rightly held that the Commission ought not to have restricted itself to ‘vague assertions’ (102) as regards the factors which were decisive for its decision, which were ‘unsubstantiated’, ‘rather limited’ (103) and in addition internally contradictory. (104)

131. Against this background I am of the view that the Court of First Instance did not make any error of law in finding that the first clearance decision gave insufficient reasons as regards the Commission’s finding in relation to (insufficient) market transparency. (105)

c) Miscellaneous matters

132. For the sake of completeness, in what follows I consider some further arguments which the appellants put forward in their sixth ground of appeal.

133. First, the appellants claim that less stringent requirements are to be imposed in relation to the reasoning of a clearance decision than in relation to a prohibition decision, because the weaker position of third parties in the procedure means that they cannot expect the same degree of precision in the reasoning as the parties to the concentration can in case of a refusal.

134. This argument is not convincing. Neither Article 253 EC nor the third indent of Article 41(2) of the Charter of fundamental rights distinguishes between decisions which benefit their respective addressees and those which are disadvantageous for them as regards the requirement to state reasons.

135. In any event, where a third party has been included in a merger control procedure and has been formally heard by the Commission – as was the case with Impala here – the duty to state reasons is intended also to protect its rights and interests. The third party may expect at least persuasive, comprehensible and internally consistent reasons as regards the essential factual and legal considerations on which the Commission bases its clearance decision. It is precisely these minimum requirements which are the subject of the present case. (106)

136. Admittedly, it is correct that the procedural position of third parties in merger control as regards their right to a hearing is weaker than that of parties to the concentration. (107) However, it cannot be concluded from this that a third party is likewise subject to restrictions when challenging a decision for insufficient reasoning . This is because a person who, because directly and individually affected, is able to clear the hurdle of an interest in bringing proceedings as a third party must be able to bring proceedings on the same conditions as all other applicants (Article 230(4) EC); he must be able to rely on the same grounds of invalidity as they can, including that of insufficient reasoning.

137. Second, the appellants claim that in the present case both Impala and the Court of First Instance had easily understood the reasons for the first clearance decision. From this they conclude that the decision included sufficient reasons.

138. This argument is likewise not to the point. The fact that the applicant has been able to make his other pleas sufficiently clear before the Court of First Instance may constitute an initial factor pointing to the conclusion that adequate reasons were given. (108) However, this cannot be regarded as more than a rebuttable presumption.

139. Specifically, whether a statement of reasons satisfies the legal requirements of Article 253 EC depends ultimately on objective criteria, including in particular whether the essential factual and legal considerations are logical, comprehensible and free from contradiction. (109) If, as in the present case, a statement of reasons is not logical, comprehensible and free from contradiction on an essential matter, the decision in question is to be annulled even if the applicant’s rights of defence have not suffered any damage as regards his other pleas. Otherwise, it would be in practice impossible for an applicant in an annulment action to succeed on the ground that the reasoning was insufficient while also raising other grounds of invalidity.

140. Contrary to the appellants’ view, it is incidentally by no means contradictory for the Court of First Instance in the present case to have found at one and the same time that the reasons given were insufficient and that the Commission made a manifest error of assessment. A decision by the Commission may be vitiated by both procedural and substantive flaws. The fact that the reasons given for a decision – for example on account of not being logical – are deficient does not preclude the possibility that the same decision is also substantively flawed.

141. Third, the appellants claim that, in laying down requirements as to the reasons to be given in a merger control decision, account must be taken of the confidentiality of sensitive commercial information. Thus, in their view, the Court of First Instance made an error in law in that in paragraph 411 of the judgment under appeal it required the Commission to publish details of the pricing and discounting policies of the other majors.

142. This submission too is not convincing. It is clearly based on a misunderstanding of paragraph 411 of the judgment under appeal. In that paragraph the Court of First Instance by no means required the Commission to disclose business secrets of individual market participants. It merely – and correctly – rejected the Commission’s argument that because of their confidential character certain figures could not have been included at all in the first clearance decision. This is because the obligation of professional secrecy (Article 287 EC) cannot be interpreted so broadly as to undermine the requirement to state reasons for a decision. (110)

143. So far as figures include business secrets, the Commission’s present practice in competition law is to replace such figures in its decision with ranges of figures, or otherwise to summarise them or describe them. The Court of First Instance rightly points out (111) that at other places in its first clearance decision the Commission indeed provided information, including figures, on the pricing policy of individual market participants. (112)

d) Interim conclusion

144. In all, it follows that the Court of First Instance did not make any errors in law in reaching the conclusion in paragraphs 325 and 542 of the judgment under appeal that insufficient reasons were given for the first clearance decision, which had, for that reason, to be set aside. Accordingly, the first and third parts of the sixth ground of appeal are unfounded.

2. The Court of First Instance’s references to the statement of objections (first ground of appeal and second part of the sixth ground of appeal)

145. In their first ground of appeal and in the second part of their sixth ground of appeal the appellants claim that the Court of First Instance wrongly used the statement of objections as a benchmark for its substantive assessment of the first clearance decision and wrongly required of the Commission that it justify any deviations from the objections in its decision.

146. It is not disputed that the judgment under appeal contains numerous references to the statement of objections. In the following I shall consider, as a first step, the Court of First Instance’s approach in general to the relationship between the clearance decision and the statement of objections, and then, as a second step, I shall consider in more detail the specific references the Court of First Instance made to individual passages in the statement of objections.

a) The Court of First Instance’s approach in general to the relationship between the clearance decision and the statement of objections

147. The appellants, supported by the Commission, object generally to the Court of First Instance’s discussion of the relationship between the clearance decision and the statement of objections. In their opinion the Court of First Instance misunderstood the nature and function of a statement of objections, and this ‘coloured’ the rest of the judgment.

– The judgment under appeal

148. In considering the complaint of a failure to state reasons, the Court of First Instance, by way of preliminary observation, recalled ‘that the Commission had concluded emphatically in the statement of objections that the concentration was incompatible with the common market on the ground, in particular, that a collective dominant position existed before the proposed concentration and that the market for recorded music was transparent and particularly conducive to coordination’. (113)

149. The ultimate clearance of the concentration was described by the Court of First Instance as a ‘fundamental U-turn in the Commission’s position’ which was ‘surprising, particularly in view of the late stage at which it was made’; (114) in addition, reference was made to the requirement to comply with mandatory time-limits in merger control procedures, which, according to the Court of First Instance, ‘does not allow [the Commission] to extend its investigation, thus reducing the likelihood that the Commission will fundamentally alter its position as the administrative procedure advances’. (115)

– Analysis

150. As in antitrust proceedings, in merger control proceedings it is essential to observe the rights of the defence before making a decision which may adversely affect the undertakings concerned. (116)

151. The rights of the defence include in particular the principle of the right to a hearing, which is a fundamental principle of Community law (117) and which is now also recognised in Article 41(2) of the Charter of fundamental rights. In addition, in relation to merger control proceedings this principle is laid down by statute in the second sentence of Article 18(3) of the Merger Regulation.

152. The fact that in antitrust proceedings and in merger control proceedings the undertakings concerned are issued with a written statement of objections (118) is the result of their right to a hearing. The statement of objections allows them to acquaint themselves with the evidence which the Commission has at its disposal and to render the rights of the defence fully effective. (119) They are given the opportunity to make submissions on the objections in writing and, on justified application, orally. (120) In addition, the first sentence of Article 18(3) of the Merger Regulation makes it clear that the Commission may base its decisions only on objections on which those concerned have had the opportunity of making known their views.

153. It follows from its function in administrative proceedings that a statement of objections is a preparatory document containing assessments of fact and of law which are purely provisional in nature. For that reason, the Commission may, and even must, take into account the factors emerging from the administrative procedure in order, inter alia , to abandon such objections as have been shown to be unfounded. (121)

154. The fact that, in contrast to where Articles 81 EC and 82 EC apply, in merger control proceedings the Commission is subject to strict procedural time-limits does not change this provisional nature of the statement of objections in any way. The requirement for speed in merger control proceedings naturally means that the affected undertakings too are subject to particularly short time-limits within which they can exercise their rights in defence. However, the comprehensive safeguarding of the rights of defence requires that submissions in defence by the persons affected in merger control proceedings may not be taken any the less into account than in antitrust proceedings. Accordingly, such a submission in the context of merger control can lead to a change in the Commission’s view just as much as it can in antitrust proceedings, even shortly before the expiry of the time-limit allowed for a clearance or a prohibition decision.

155. Against this background it was certainly unfortunate that in the judgment under appeal the Court of First Instance regarded the ‘fundamental U-turn in the Commission’s position’ which occurred just prior to the first clearance decision as ‘surprising’ in view of ‘the late stage at which it was made’, (122) and generally described the likelihood that the Commission would fundamentally alter its position as ‘reducing … as the administrative procedure advances’. (123)

156. If the Commission alters its assessment of a concentration which has been notified to it in consequence of submissions in defence made by one of the participants in response to the objections, this new analysis of the case is by no means ‘late’, but is made instead at the usual time in the context of merger control proceedings. Nor is such an occurrence ‘surprising’ or ‘unlikely’, (124) but an expression of observance of the rights of the defence and proves that the hearing given to the undertakings concerned is not a farce.

157. However, contrary to the appellants and the Commission, I am not of the view that the judgment under appeal was vitiated by an error of law on the basis that these rather unfortunate choices of expression by the Court of First Instance expressed a legally inaccurate basic understanding of the nature and function of a statement of objections.

158. Indeed, in the same breath as its statements as to the ‘surprising’ and ‘late’ character of the ‘U-turn in the Commission’s position’, the Court of First Instance expressly emphasised that preparatory character of the statement of objections in the context of merger control too; in addition it recognised that according to the case-law on Articles 81 and 82 EC the Commission was not obliged to explain any deviations from the statement of objections. (125)

159. Accordingly, I am of the opinion that in its general remarks on the relationship between the first clearance decision and the statement of objections the Court of First Instance ultimately did not misunderstand the nature and function of the statement.

160. Even if one were of the view that such an error of law existed, this would not justify setting aside the judgment under appeal, because, as I will show, (126) it did not result in an assessment of the first clearance decision which was erroneous in law, and thus could not in any event have had any effect on the operative part of the judgment under appeal. (127)

b) The specific references by the Court of First Instance to the statement of objections

161. There remains to be considered the criticism by the appellants and the Commission of a series of specific references by the Court of First Instance to the statement of objections in the context of its examination of the lawfulness of the first clearance decision. It is alleged that the Court of First Instance wrongly supported its finding that the first clearance decision was unlawful on a comparison of the decision and the statement of objections, instead of deciding solely by reference to the decision.

162. This submission too is not convincing.

163. Admittedly, the judgment under appeal refers a number of times to the statement of objections. In particular, the Court of First Instance repeatedly emphasises that in the first clearance decision the Commission subdued (128) or changed (129) its description and assessment of certain facts by comparison with the statement of objections.

164. However, it is only on a superficial reading that the Court of First Instance appears to have based its findings of errors in law in the first clearance decision on such mere divergences from the statement of objections. On closer reading it becomes clear that the Court of First Instance derived its findings both of a lack of reasoning and of manifest errors of assessment from the first clearance decision itself.

165. As regards the failure to state reasons , the Court of First Instance’s discussion of the separate section in the first clearance decision, which concerns market transparency, (130) does not refer at all to the statement of objections. (131) In the rest of the Court of First Instance’s discussion on failure to state reasons, (132) references to the statement of objections are infrequent; they occur only in paragraphs 300, 302 and 308 of the judgment under appeal, and none of them is decisive for the Court of First Instance’s finding that the first clearance decision did not contain sufficient reasons as regards the lack of market transparency it claimed to exist:

– in paragraph 308 of the judgment under appeal, which concerned movements in gross and net list prices, the Court of First Instance relies upon the content of the first clearance decision. It does not make any comparison between this decision and the statement of objections, and thus does not find there to have been any divergence between them. Instead, the Court of First Instance appears to seek to draw the same conclusion from both documents; in that way the statement of objections is used by the Court of First Instance merely as an additional illustration of the information it derives from the decision itself.

– In paragraph 302 of the judgment under appeal too, the reference to the statement of objections is purely by way of illustration, the Court of First Instance being of the view that even the subdued assessment of list prices in the first clearance decision is ‘a further element that favours the transparency of the market’.

– Finally, the fact that the reference to the statement of objections in paragraph 300 of the judgment under appeal was not decisive for the line of argument taken by the Court of First Instance is expressly made clear by the Court of First Instance itself in the immediately following paragraph: ‘In any event, even considering only the observations set out in the Decision , the Commission concluded that list prices were rather aligned.’ (133)

166. It follows that any divergences between the first clearance decision and the statement of objections did not play any part in the finding by the Court of First Instance that insufficient reasons had been given for the decision, this being in itself enough to result in the decision’s being annulled. (134) Contrary to the submissions of the appellants and the Commission, the statement of objections was not used by the Court of First Instance as a benchmark for its assessment of the first clearance decision.

167. As regards the Court of First Instance’s findings of manifest errors of assessment , the references in the judgment under appeal to the statement of objections are admittedly more numerous: they are to be found in paragraphs 335, 338, 339, 341, 362, 378, 379, 398, 402, 409, 419, 424, 446, 447, 451, 456, 467, 528, 532 and 538 of the judgment under appeal. (135)

168. The references in paragraphs 338, 339, 341, 362, 402, 456, 467, 532 and 538 of the judgment under appeal to the statement of objections serve without doubt as illustrations of, and supplements to, what the Court of First Instance has already taken directly from the first clearance decision. The purely supplementary character of these references to the statement of objections is also made clear in the paragraphs of the judgment listed by additional words such as ‘moreover’ and ‘furthermore’. At no time is any kind of contradiction between the first clearance decision and the statement of objections criticised. The statement of objections was not used as a benchmark for the judicial examination of the first clearance decision.

169. The same applies for paragraphs 378 and 379 of the judgment under appeal. If one reads them in conjunction with the immediately preceding paragraph 377 of the judgment under appeal, it is clear that they too serve merely to illustrate and supplement what the Court of First Instance has already taken solely from the first clearance decision concerning the transparency of price discounts and has summarised as follows: ‘Accordingly, it must be held that the evidence, as mentioned in the Decision, does not support the conclusions drawn from it.’ (136)

170. By contrast, the references to the statement of objections in paragraphs 335, 398, 408 to 410, 419, 424, 446, 447, 451 and 528 of the judgment under appeal seem to be more problematic. At first glance it appears that the Court of First Instance in fact intended to allege that the Commission had departed from the statement of objections in the first clearance decision without sufficient explanation. (137)

171. According to case-law to which the appellants and the Commission referred in the present proceedings, the Commission is not obliged to explain in its decision potential divergences from its statement of objections, because the statement is a preparatory document the analysis in which is merely provisional in nature. (138) It follows that a decision by the Commission in either antitrust proceedings or merger control proceedings cannot be regarded as vitiated by an error of law and annulled solely on the ground that without further explanation it departs in substance from the statement of objections.

172. However, this does not preclude the possibility that a decision which diverges from the statement of objections can, for other reasons relating to the decision itself, be erroneous in law and accordingly be annulled if challenged.

173. In this connection it is to be recalled that in the context of merger control the Commission is required to analyse complex economic situations. As already mentioned, in doing so it has a not insignificant discretion, (139) but it is none the less always subject to review by the Community Courts. The latter have the task not only of establishing whether the evidence relied on is factually accurate, reliable and consistent but also whether that evidence contains all the information which must be taken into account in order to assess a complex situation and whether it is capable of substantiating the conclusions drawn from it. (140)

174. Thus, even if, in its decision as to the compatibility of a concentration with the common market, the Commission need not explain why it has departed in substance from its statement of objections, three things must be capable of being understood from its decision:

– first, that the facts on which the decision is based were accurate, and in particular that they are based on reliable and consistent evidence ( accuracy of the factual basis );

– second, that the decision does not fail to take into account any relevant information which ought to have been taken into account in assessing the concentration ( completeness of the factual basis ); and

– third, that the facts on which the decision is based are capable of substantiating the conclusions drawn from them ( strength of the factual basis ). (141)

175. Reference points for whether the Commission has, in an individual case, fully taken into account all the relevant information may appear from all the circumstances of the individual case, and in particular from all the documents which were part of the administrative file in the administrative proceedings. Those documents include not least the statement of objections. This is because this statement provides a summary of all the facts and evidence which the Commission considers at that stage in the proceedings to be relevant to its decision.

176. On this basis, in the judgment under appeal it was possible for the Court of First Instance to find that despite its provisional nature the statement of objections was not ‘wholly without merit or wholly irrelevant’. (142)

177. There is no doubt that as the procedure runs its course it may appear – in particular because of the submissions made by the participating undertakings in their defence – that the facts and evidence on which the statement of objections was based were incomplete or inaccurate, or that they were not capable of substantiating the conclusions drawn from them. The Court of First Instance correctly recognises this too, where it says in the judgment under appeal that the Commission ‘is obliged to take account of the evidence obtained during the administrative procedure and also of the arguments put forward by the undertakings concerned, and must drop any objections which might ultimately prove to be unfounded’. (143)

178. However, the Community Courts can and must ascertain whether certain facts which the Commission had established and still relied upon in its statement of objections were, in the further course of the procedure, correctly categorised as inaccurate or unreliable by the Commission and dropped. Equally, the Community Courts can and must consider whether any new facts on which the Commission now relies are in fact accurate, whether the facts thus established are complete, and whether they are capable of substantiating the conclusions the Commission has drawn from them.

179. The correctness, completeness and strength of the factual material which underpins a decision must be liable to judicial review. (144) Without such a review of the factual basis for a decision it would not be possible to assess, in a meaningful way, whether the Commission had stayed within the limits of the discretion allowed to it or had committed manifest errors of assessment.

180. Accordingly, in the judgment under appeal the Court of First Instance correctly undertook this review and in doing so explained that the Commission ‘must be in a position to explain … at least in the context of the proceedings before the Court, its reasons for considering that its provisional findings were incorrect; but above all, the findings set out in the decision must be compatible with the findings of fact made in the statement of objections, in so far as it is not established that the latter findings were incorrect’. (145)

181. Contrary to what the appellants and the Commission submit, this does not mean that the merger control authority is subject to a double burden before the courts. It is not required to defend its decision and in addition refute the statement of objections in so far as it diverges from it. It must simply – and in response to a substantiated submission by the applicant – be able to explain that the factual basis of its decision was correct and complete and that it was capable of substantiating the conclusions drawn in its decision. Any explanations as to why certain factual elements have been supplemented or abandoned or reassessed in the course of the administrative procedure are inevitably bound up with the question of the correctness, completeness and strength o f the factual basis of the decision.

182. Thus, all in all, the Court of First Instance relied on the statement of objections in an acceptable manner as a basis for its review of whether the first clearance decision was adopted on a correct and complete factual basis that was capable of substantiating the conclusions drawn by the Commission. However, the Court of First Instance did not annul the first clearance decision on account of apparent divergences from the statement of objections, but on account of its insufficient reasoning and on account of manifest errors of assessment. The Court of First Instance found there to be such errors of assessment because the factual basis for the decision had been incomplete and incapable of substantiating the conclusions drawn by the Commission from it as regards the absence of market transparency. (146)

c) Interim conclusion

183. Thus, to summarise, the first ground of appeal and the second part of the sixth ground of appeal are unfounded.

3. The evidential value of factual assertions in the response to the objections (second ground of appeal; first part of the third ground of appeal)

184. The second ground of appeal is closely connected to the first part of the third ground of appeal; for that reason I propose to consider both together. In essence, the appellants submit that the Court of First Instance set too high a standard of proof as regards their assertions in the response to the statement of objections. In addition, they submit that the Court of First Instance wrongly held that the Commission was under a duty to conduct a new market investigation following the statement of objections.

a) The judgment under appeal

185. In the course of reviewing the first clearance decision for the existence of manifest errors of assessment in relation to market transparency, the Court of First Instance stated inter alia that ‘the parties to the concentration cannot wait until the last minute before submitting evidence to the Commission with a view to refuting objections raised at the proper time by the Commission, since the Commission would then no longer be in a position to carry out the necessary investigations. In such a hypothetical situation, that evidence must at the very least be particularly reliable, objective, relevant and cogent if it is to be capable of validly refuting the objections raised by the Commission’. (147)

186. The Court of First Instance was further of the view that the Commission could not ‘go so far as to delegate, without supervision, responsibility for conducting certain parts of the investigation to the parties to the concentration, in particular where, as in the present case, those aspects constitute the crucial element on which the decision is based and where the data and assessments submitted by the parties to the concentration are diametrically opposite to the information gathered by the Commission during its investigation and also to the conclusions which it drew from that information’. (148)

187. Moreover, in various places in the judgment under appeal the Court of First Instance criticises the Commission for failing, following the response by the parties to the concentration to the statement of objections, to carry out any new market investigations in order to test the validity of the new direction of its assessment of the intended concentration. (149)

b) Analysis

188. As has already been mentioned in connection with the first ground of appeal, in merger control proceedings it is essential to observe the rights of defence before making a decision which may adversely affect the undertakings concerned (150) (see also the second sentence of Article 18(3) of the Merger Regulation).

189. For that reason, the undertakings concerned cannot be faulted for not having put forward certain – possibly decisive – arguments, facts or evidence when notifying the concentration or during the Commission’s market investigation, and instead doing so only in their submissions in defence when responding to the statement of objections. (151) The statement of objections is the first document which tells the parties to the concentration in detail what the Commission’s objections are to their intended concentration and the individual arguments and evidence on which it bases those objections. (152)

190. Thus, the sole fact that the parties to the concentration put forward certain arguments, facts or evidence for the first time in their response to the statement of objections does not by any means justify a finding that they withheld this information ‘until the last minute’. (153) Instead, the rights of defence conferred upon the undertakings concerned give them the right to put forward in the course of their written and oral hearing, that is after receipt of the statement of objections, everything which they consider appropriate to refute the objections and to persuade the Commission to clear their concentration. Such a submission is not too late, but is made at the point in time specifically envisaged for it in the merger control procedure.

191. Nor can a higher standard of proof or of cogency be imposed on the submissions put forward by the undertakings concerned in reply to the statement of objections than is imposed on the submissions by competitors, customers, and other third parties questioned by the Commission in the course of the merger control procedure. Admittedly, the Commission is obliged to review the submissions of the parties to the concentration carefully as to their accuracy, completeness and cogency, and in case of reasonable doubt to leave them out of account, but in doing so it must apply the same standards as it does in considering third parties’ submissions.

192. The rights of defence of the undertakings concerned would be devalued if they were not applied in relation to their defence submissions on individual objections raised by the Commission, or if their defence submissions were from the outset viewed as less credible and less cogent than, for example, information provided by third parties in the course of the market investigation.

193. Moreover, if in its decision the Commission considers the defence submissions made by the undertakings concerned and is persuaded by them to reconsider and even depart from the provisional conclusions in its statement of objections, this does not in any way mean that the investigation has been ‘delegated’ to them.

194. Finally, the Commission cannot be obliged in every individual case to undertake further market investigations following the statement of objections and after hearing the undertakings concerned. The time pressures arising from the comparatively strict procedural time-limits are already enough to make it impossible for the Commission to send further comprehensive requests for information about complex economic questions to numerous market participants shortly before issuing its draft decision to the Advisory Committee on concentrations. (154) Realistically, in the short time remaining it is only seldom that such an investigation could be expected to give useful results. Moreover, the undertakings concerned would have to be given a new hearing on the results of the investigation, if, for example, it was intended to rely on them for making a prohibition decision. For that reason, the appellants correctly observed that the consequence of an unclear factual situation after hearing the undertakings concerned cannot be the initiation of new market investigations. Instead, the decision must be made on the basis of the existing information. This approach is supported by Article 10(6) of the Merger Regulation, which provides for a deemed clearance if the Commission does not make a decision within the time-limits allowed to it.

195. Against this background I am of the view that the Court of First Instance misunderstood the legal position in stating that the parties to the concentration could not ‘wait until the last minute before submitting evidence to the Commission with a view to refuting objections raised at the proper time by the Commission’ and that such evidence had to be ‘ particularly reliable, objective, relevant and cogent’ if it were to be capable of validly refuting the objections raised by the Commission; in addition, the Court of First Instance was wrong to complain about the absence of further market investigations following the statement of objections, and to equate the acceptance by the Commission of the appellants’ defence submissions with an unlawful delegation of the investigation to the parties to the concentration. (155)

196. Thus, the second ground of appeal and the first part of the third ground of appeal are in substance well founded.

197. However, it does not follow that the judgment under appeal is to be set aside. (156) Specifically, the Court of First Instance did not just object that in assessing market transparency the Commission relied on the defence submissions by Bertelsmann and Sony and failed to carry out new market investigations. Instead, it also reviewed the substance of the Commission’s analysis of market transparency for manifest errors of assessment.

198. In this connection it is to be emphasised that the passages in the Court of First Instance’s judgment which are disputed here were not the first place where the Court of First Instance found there to have been manifest errors of assessment, it having already found one much earlier, at paragraph 377 of the judgment under appeal: ‘Accordingly, it must be held that the evidence, as mentioned in the Decision, does not support the conclusions drawn from it.’ That finding is by no means related to the defence submissions by the parties to the concentration in dispute here concerning campaign discounts and how the Commission took them into account.

199. In paragraphs 384 to 387 of the judgment under appeal too the Court of First Instance found the Commission to have made a manifest error of assessment which was not based on the Commission’s taking into account of the defence submissions by the parties to the concentration as regards campaign discounts, but was based on the incorrect substantive assessment of the results of the market investigation, and in particular of answers given by retailers.

200. Each of these two manifest errors of assessment was in itself sufficient to justify the Court of First Instance in annulling the first clearance decision. Apart from that, as already mentioned, the finding of a failure to give reasons, which the Court of First Instance made without any error in law, was likewise enough to justify the annulment of the first clearance decision. (157)

4. Standard of proof for clearance of concentrations (second part of the third ground of appeal)

201. By the second part of their third ground of appeal the appellants claim that the Court of First Instance erred in law by applying an erroneous and excessively high standard of proof for Commission merger clearance decisions. (158)

202. The appellants’ reasons for their criticism of the judgment under appeal are as follows. In general, the Court of First Instance failed to understand that the standards of proof the Commission was required to apply in relation to clearance decisions and prohibition decisions are not the same; they claim that there is an asymmetry as regards the standard of proof and a general presumption that concentrations are compatible with the common market. Specifically, in the present case the Court of First Instance erroneously required the Commission to obtain positive evidence of a lack of market transparency ; according to the appellants, the Court of First Instance should have reviewed only whether, at the time of the first clearance decision, the Commission had sufficient evidence for the existence of market transparency .

a) The alleged asymmetry between the requirements for clearance decisions and prohibition decisions in the Merger Regulation

203. I consider first the appellants’ argument that the standards of proof in the Merger Regulation are asymmetrical, and that there is a general presumption that concentrations are compatible with the common market.

– Merger control decisions are based on prognoses

204. By way of introduction it is to be observed that merger control decisions by the Commission are different in a material respect from decisions in cartel proceedings under Article 81 EC and in proceedings for abuses of a dominant position under Article 82 EC. In the merger control context, the Commission is not required to assess and, as the case may be, punish – supposedly unlawful – conduct by undertakings which has already occurred, but instead to give a prognosis as to the market’s future development. It must assess whether a concentration is capable of creating or strengthening a dominant position as a result of which effective competition would be significantly impeded in the common market or in a substantial part of it (Article 2(2) and (3) and Article 8(2) and (3) of the Merger Regulation).

205. The Commission’s decision to clear or to prohibit any given concentration thus necessarily depends on an assessment of probability which is made ex ante . The Court recognised this too in Tetra Laval , where it held that merger control involved ‘a prediction of events which are more or less likely to occur in future’. (159)

206. Judicial review of such an assessment of probability is not so much about whether the Commission’s prognosis on a concentration’s positive or negative effects on competition is capable of being proved but rather about its plausibility . The Commission need provide evidence only for the facts on which it has based its prognosis, for example the elements of the market structure it has established (in the present case, for example, the diverse factors which speak in favour of or against a finding of market transparency). In this regard the Court held in Tetra Laval that the ‘evidence must support the Commission’s conclusion that … the economic development envisaged by it would be plausible’. (160)

207. The standard by reference to which the Commission’s prognosis as to the market development envisaged is plausible or implausible should be defined having regard to the specific features of merger control proceedings. In merger control decisions under Article 8 of the Merger Regulation, the Commission neither imposes any sanctions nor infringes the personal freedom or physical integrity of any natural person. Instead, the Commission grants or refuses clearances of an administrative law nature for an economic activity, namely the concentration of undertakings. Moreover, this takes place in a procedure which is characterised by a requirement for speed and a finely balanced, comparatively strict system of time-limits. (161)

208. Given these features of merger control proceedings, the appropriate standard would be, in my view, that the Commission should base its decision on the market development which, on the balance of probabilities, it considers most likely at the end of its months-long, intense investigation of a concentration. The Court of First Instance gave a very accurate summary of this not long ago in General Electric : ‘A prospective analysis … makes it necessary to envisage various chains of cause and effect with a view to ascertaining which of them are the most likely .’ (162)

209. Thus, if it is more likely than not that a dominant position will be created or strengthened, the concentration is to be prohibited; on the other hand, if it is less likely than not that a dominant position will be created or strengthened, the concentration should be cleared. It is the task of the Community Courts to review the Commission’s prognosis for any manifest errors of assessment, in other words as to whether the Commission has based it on evidence which is factually accurate, reliable and consistent and is capable of substantiating the prognosis. (163)

210. I would consider it inappropriate to set the bar higher in the context of merger control and, for example, to require that the market development envisaged by the Commission should be ‘very probable’ or ‘particularly likely’ in order to be accepted by the Court. (164) Such a higher standard of probability would seriously weaken the Commission in carrying out its competition policy functions. The Commission would have to clear concentrations with open eyes even though they would probably lead to the creation or strengthening of a dominant position, and thus would have detrimental effects on competition. The Commission would be able to intervene only in cases in which such detrimental effects of a concentration would be ‘very probable’ or ‘particularly likely’. In addition, it would be difficult to reconcile a higher standard of probability with the margin of discretion allowed to the Commission in its assessment of complex economic situations, (165) the essence of which includes the Commission’s prognosis as to how the market is expected to develop as a result of a concentration between undertakings.

211. A fortiori , it cannot be decisive whether, at the end of its extensive examination of a concentration over many months, the Commission is in a position to decide beyond a reasonable doubt that a dominant position will or will not be created or strengthened. (166) This particularly high standard is known principally in the field of criminal and quasi-criminal proceedings. In merger control proceedings it is applicable only in the preliminary phase (‘Phase I’), to compensate for the fact that at that stage the investigation of a concentration is merely a summary one . At that stage, ‘serious doubts’ as to the compatibility of the concentration with the common market will only prevent its being cleared too quickly and force the Commission to make a more extensive investigation in a formal procedure (‘Phase II’) (Article 6(1)(b) and (c) of the Merger Regulation). (167) However, after that extensive investigation the concentration must be cleared, notwithstanding any remaining doubts, provided that on the Commission’s prognosis the creation or strengthening of a dominant position is less likely than not. On the other hand, the concentration is to be prohibited if, following its more extensive investigation, the Commission regards the creation or strengthening of a dominant position as more likely than not, despite any remaining doubts.

– Symmetry of the requirements for clearances and prohibitions

212. I cannot discern any difference between the legal requirements as to clearance decisions on the one hand and prohibition decisions on the other. Contrary to the submissions of the appellants, there is no such difference either as regards the degree of plausibility of the prognosis to be made by the Commission or as regards the solidity of the factual basis supporting it.

213. Both Article 2 of the Merger Regulation, which lays down the general review programme for the control of a notified concentration by the Commission, and Article 8 of the Merger Regulation, which lists the Commission’s powers of decision, are structured completely symmetrically in their respective second and third paragraphs.

214. Ultimately, this symmetry reflects the fact that in each individual case the Commission must strike a fair balance between interests of equal value which are guaranteed by primary law, (168) being on the one hand the rights and interests of the parties to the concentration, and on the other the public interest in protecting competition against distortion (Article 3(1)(g) EC). (169) Thus, while the commercial freedom of the participating undertakings and the property rights of their shareholders (Articles 16 and 17 of the Charter of fundamental rights) undoubtedly include the right to realise concentrations of undertakings, this applies only in so far as certain conditions or obligations on, or even the prohibition of, the concentration in question are not necessary for protecting competition from distortion.

215. The case of Tetra Laval , on which the appellants rely, does not gainsay this symmetry of the requirements for clearance and for prohibition decisions. The requirement laid down there, namely ‘convincing evidence’, (170) is mentioned merely by way of reminder that the Commission must have a reliable factual basis for its assessment of probability. This appears moreover elsewhere in the judgment in Tetra Laval , where the Court says that ‘the quality of the evidence produced by the Commission … is particularly important, since that evidence must support the Commission’s conclusion that … the economic development envisaged by it would be plausible’. (171)

216. The requirement for ‘convincing’ evidence must not be misunderstood as placing the bar higher in relation to prohibition decisions than in relation to clearance decisions. As the Court made clear in Tetra Laval , the requirement for ‘convincing’ evidence ‘by no means added a condition relating to the requisite standard of proof but merely drew attention to the essential function of evidence, which is to establish convincingly the merits of an argument or … of a decision’. (172)

217. Stricter requirements in relation to prohibiting an intended concentration than in relation to clearing it can be inferred from the passage in Tetra Laval quoted above just as little as they can be inferred from the Court’s warning that the Commission must carry out its prospective analysis ‘with great care’. (173) The Commission may by no means grant clearance to a concentration ‘without care’, and in doing so support its analysis with less ‘convincing’ evidence than in the case of a prohibition. Otherwise it would not be properly performing its core function of protecting competition within the internal market against distortion.

218. Thus, if the creation or strengthening of a dominant position is more likely than not, the concentration is to be prohibited; by contrast, if the creation or strengthening of such a position is less likely than not, the concentration is to be cleared. These two assessments of probability are two sides of the same coin. Each is to be made with great care and is to be supported by a factual basis which is not only correct and complete – for that, ‘convincing’ evidence is necessary – but is also capable of substantiating the conclusions drawn from it. (174)

– No general presumption of compatibility with the common market

219. It also follows from the symmetry described above and the equal importance given to the legal interests protected (175) that the Merger Regulation is not based on a general presumption in favour of compatibility of concentrations with the common market. In each individual case the Commission must make an express finding as to whether the concentration in question is compatible or incompatible with the common market, (176) and the participating undertakings are expressly prohibited from putting their concentration into effect before the decision is made (Article 7(1) and (5) of the Merger Regulation).

220. The case of EDP v Commission , (177) on which the appellants rely, does not lead to any different conclusion. In that case the Court made it clear that mere doubts on the Commission’s part are not enough to justify prohibiting a concentration. However, it cannot be taken from that, a contrario , that there is a general presumption that concentrations are compatible with the common market. This is also shown by a glance at the 15th recital to the Merger Regulation, from which it follows that, at best, concentrations having a limited market share may be presumed to be compatible with the common market.

221. In my opinion, it is only in the following two categories of case that, very exceptionally, a concentration may be presumed to be compatible with the common market.

222. The first category of case concerns notified concentrations on which the Commission has not given a decision in time, in breach of its statutory obligation. Article 10(6) of the Merger Regulation provides that such concentrations are by operation of law deemed to be compatible with the common market (see also Article 7(1) and (5) of the Merger Regulation). However, given that it is an exception, and given its place in the legislative scheme in conjunction with the provisions as to time-limits, it is not possible to derive from Article 10(6) of the Merger Regulation a more far-reaching general presumption that concentrations are compatible with the common market. (178)

223. The second category of case is concentrations whose investigation by the Commission produces such an unclear state of evidence that it is not possible to make any reliable prognosis as to whether or not they will ultimately create or strengthen a dominant position. Advocate General Tizzano used the term ‘grey area’ to characterise these cases. (179) However, in my view this term should not be misunderstood as involving a considerable number of cases. I think that there can only be a few small and infrequent borderline cases in which, even after extensive market investigations, it is not clear on which side of the line the case falls. Only such instances of non liquet ought to be declared compatible with the common market and cleared on the basis of the principle in dubio pro libertate . However, it is not possible to derive from this category of case a more far-reaching general presumption that concentrations are compatible with the common market.

224. Regardless of the exact scope of these two categories of case, it is clear that the present case does not fall into either of them. The Commission did not, in breach of its obligations, allow the time-limit for its decision on the concentration to expire; nor, according to the Court of First Instance, was the evidence unclear. (180) Although the Commission itself referred to a lack of evidence for certain facts at a number of places in its first clearance decision, (181) so far as appears it never classified the present case as a borderline one of non liquet . Instead, at the oral hearing before the Court of Justice the Commission expressly emphasised that the discussion of borderline cases in which the evidence was unclear was purely hypothetical. (182)

– Interim conclusion

225. Altogether, therefore, the appellants’ submissions that in merger control proceedings the requirements in relation to clearance and prohibition decisions are asymmetrical and that there is a general presumption that concentrations are compatible with the common market are to be rejected. Nor does the present case fall into one of the two categories of case in which, by way of exception, there is a presumption that the concentration is compatible with the common market.

b) The standard of proof applied by the Court of First Instance in the present case

226. It remains to be considered whether in the present case the Court of First Instance applied the correct standard of proof in relation to the Commission’s findings on market transparency.

227. The appellants argue that the Court of First Instance was not entitled to require the Commission to prove positively an absence of market transparency : instead, the Court of First Instance ought to have been satisfied with reviewing whether, at the time of the first clearance decision, the Commission had sufficient evidence of the existence of market transparency . (183) It is apparent that this line of argument is based on the idea that the standard of proof for clearing a concentration must be lower than that for prohibiting it.

228. I do not find this line of argument convincing, for two reasons.

229. First, it follows from the symmetry of the standard of proof applicable in relation to clearance and prohibition decisions (184) that it can make no difference whether the Commission is considering market transparency with a view to clearing or with a view to prohibiting the concentration in question. The negative finding that a market is not so transparent as to allow a collective dominant position leads to the same result and accordingly requires the same degree of evidence as the positive finding that the market in question is so opaque as to preclude a finding of a collective dominant position. The two findings are two sides of the same coin. And both allow it to be ruled out that there is a risk that a collective dominant position will be created or strengthened, provided that the facts on which each finding is based are correct, complete, and capable of substantiating the conclusions drawn from them.

230. Second, the appellants’ criticism of the standard of proof applied by the Court of First Instance appears to me to be based on a misinterpretation of the judgment under appeal. On closer consideration, the Court of First Instance does not accuse the Commission of having failed to satisfy the standard of proof in its assessment of the concentration’s effect on competition. Instead, in reviewing the first clearance decision the Court of First Instance orientates itself on the Commission’s finding that the market ‘is not sufficiently transparent to permit a collective dominant position.’ (185) Thus, as the appellants demanded, the Court of First Instance even went so far as to review whether there was in fact the alleged insufficiency of evidence that the market was transparent.

231. The manifest error of assessment found by the Court of First Instance consisted not in a failure positively to prove a lack of market transparency but instead in a failure to take all the relevant facts into account in the first clearance decision and in the fact that the facts taken into account were not capable of substantiating the conclusion that the Commission drew. (186)

c) Interim conclusion

232. Thus, the second part of the third ground of appeal is unfounded. The same applies for the submissions in paragraphs 98 to 100 and 102 of the appeal, the substance of which overlaps with the second part of the third ground of appeal.

B – Limitations on the Court of First Instance’s power to analyse the facts and the evidence (fourth ground of appeal)

233. By their fourth ground of appeal the appellants claim that the Court of First Instance exceeded the scope of judicial review by failing to respect the Commission’s margin of discretion and, on a number of points, by substituting its own assessment of the facts and of the evidence for that of the Commission; in doing so, the Court itself committed manifest errors and fundamentally misconstrued the evidence. (187)

234. Whereas the Commission supports this ground of appeal, Impala defends the judgment under appeal.

1. The Commission’s discretion and judicial review thereof

a) The judgment under appeal

235. As support for its complaint that the Court of First Instance did not respect the Commission’s margin of discretion and instead substituted its own assessment of the facts and evidence on a number of points, the appellants give the following examples from the judgment under appeal:

– The Court of First Instance described the alignment of list prices as ‘very marked’, whereas the Commission referred simply to list prices as being ‘rather aligned’. (188)

– The Court of First Instance described the variation in the general levels of invoice discounts applied by the parties to the concentration as being ‘very low’. (189)

– The Court of First Instance regarded campaign discounts as having ‘only a limited impact on prices’. (190)

– The Court of First Instance referred to a ‘high level of transparency of prices’ and a ‘high degree of transparency on the market’, (191) and regarded the weekly reports by parties to the concentration as ‘an additional factor of the transparency of the market’; (192) the Court of First Instance regarded campaign discounts as ‘destined to become public knowledge’ (193) and ‘rather public and transparent’. (194)

– According to the Court of First Instance, the differences in the ranges of discounts could be ‘the result of differences in performance’ and did not preclude ‘the discounts being based on a known set of rules’. (195)

– The Court of First Instance wrongly regarded the economic data produced as irrelevant to the question whether campaign discounts also played a significant role for the best-selling albums. (196)

b) Analysis

236. As already mentioned elsewhere, (197) in the context of merger control the Commission has a not insignificant margin of discretion in assessing complex economic situations. According to consistent case-law, review by the Community Courts in this regard must be limited to verifying whether the rules governing procedure and the statement of reasons have been complied with, whether the facts have been accurately stated and whether there has been any manifest error of assessment or misuse of powers. (198)

237. By their criticism of the passages in the judgment under appeal quoted at paragraph 235 above, the appellants ultimately raise the question of the limits which the Commission’s discretion imposes on the judicial review of its decisions, in the context of merger control, with regard to manifest errors of assessment . (199)

238. On this point the Court expressed itself most recently in Tetra Laval as follows:

‘Whilst the Court recognises that the Commission has a margin of discretion with regard to economic matters, that does not mean that the Community Courts must refrain from reviewing the Commission’s interpretation of information of an economic nature. Not only must the Community Courts, inter alia , establish whether the evidence relied on is factually accurate, reliable and consistent but also whether that evidence contains all the information which must be taken into account in order to assess a complex situation and whether it is capable of substantiating the conclusions drawn from it.’ (200)

239. Having regard to this standard of review, it would be an error to assume that the Commission’s margin of discretion precludes the Community Courts in any event from giving their own analysis of the facts and the evidence. On the contrary, it is essential for the Community Courts to undertake such an assessment of their own where they are assessing whether the factual material on which the Commission’s decision was based was accurate, reliable, consistent and complete, and whether this factual material was capable of substantiating the conclusions the Commission drew from it. Otherwise, the Community Courts could not sensibly assess whether the Commission had stayed within the limits of the margin of discretion allowed to it or had committed a manifest error of assessment. (201)

240. The Court of First Instance exceeds the limits of judicial review of a Commission decision in the context of merger control only where the factual and evidential position reasonably allows different assessments, the Commission adopts one of them, and the Court of First Instance none the less substitutes its own different assessment for that of the Commission.

241. If one considers the examples from the judgment under appeal relied upon by the appellants from this point of view, it is clear that although the Court of First Instance in each case undertook its own assessment of the factual and evidential position it clearly remained within the proper limits of judicial review of a Commission decision in the context of merger control.

242. First, one cannot complain of an error of law in so far as the Court of First Instance reviewed the facts established by the Commission as to whether and to what extent it was evidence for or against market transparency. Thus, findings such as those in paragraphs 299, 307, 317, 347, 354 and 361 of the judgment under appeal which are criticised by the appellants were lawful, for example the statement ‘that three factors referred to in the Decision … are capable of giving rise to a high level of transparency of prices’ (paragraph 347).

243. Second, the Court of First Instance was entitled also to form its own view as to whether and to what extent the factual material established by the Commission as regards price discounts, and in particular campaign discounts, was sufficient to exclude there being enough transparency in the market to allow a finding of a collective dominant position. Thus, the findings of the Court of First Instance such as those in paragraphs 402, 403, 405, 406, 419, 420, 421, 424, 436, 444, 456 and 457 of the judgment under appeal which are criticised by the appellants were likewise lawful, for example the statement that campaign discounts represent only a very small part of the gross selling price of albums (paragraph 457).

244. In essence, in the disputed paragraphs of the judgment under appeal the Court of First Instance applied the criteria developed by the Court of Justice itself (202) and reviewed in particular whether the factual basis underlying the first clearance decision was capable of substantiating the Commission’s conclusion that the market was not sufficiently transparent as to allow a collective dominant position.

245. The Court of First Instance by no means substituted its own assessment of market transparency for an equally reasonable assessment of the Commission and it by no means took upon itself to decide whether the disputed concentration was compatible or incompatible with the common market. (203) It merely concluded that the conclusions drawn by the Commission in the present case were not supported by the factual basis of its first clearance decision. (204) This assessment by the Court of First Instance is part of its analysis of the facts and the evidence in first instance proceedings and as such is not susceptible to review in appeal proceedings, subject, however, to the question of distortion of facts and evidence, to which I will turn in a moment.

246. In all, I conclude that in its substantive review of the first clearance decision the Court of First Instance did not fail to respect the Commission’s margin of discretion.

2. The alleged distortion of evidence

247. The appellants further complain that in paragraphs 425, 427 and 434 of the judgment under appeal the Court of First Instance distorted the evidence. In the disputed passages of the judgment under appeal the Court of First Instance considered the Commission’s view that campaign discounts were not transparent.

248. Before I turn to consider those paragraphs of the judgment under appeal, I recall the strict criteria which the Court has laid down in its consistent case-law for the review of an allegation that evidence has been distorted. According to it, evidence may be found to have been distorted where, without recourse to new evidence , the assessment of the existing evidence appears to be clearly incorrect . (205) Thus, the issue is not to review whether the Court of First Instance’s analysis of the evidence is convincing from the point of view of the Court of Justice. Otherwise, the Court would simply be substituting its own analysis of the evidence for that of the Court of First Instance, and this is not its task in appeal proceedings. So long as the Court of First Instance’s analysis of the evidence is at least justifiable , there is no distortion of the evidence.

249. The Court of Justice undertakes its review exclusively by reference to the case file.(206)

a) Paragraph 425 of the judgment under appeal

250. In paragraph 425 of the judgment under appeal the Court of First Instance considers some of the tables produced by the Commission and finds:

‘In effect, the calculation of the differential between minimum and maximum discounts by customer … was carried out incorrectly, in most cases, in consideration of the discounts granted by the other party, whereas … that calculation must be made on the basis of the differential between the minimum and maximum discounts granted by one and the same party to its various customers.’

251. The view of the appellants is that this is a distortion of the evidence. Specifically, in the proceedings at first instance the Commission stated to the Court of First Instance that it had calculated the differential between minimum and maximum discounts by reference to price discounts from the invoice amount which a party to the concentration had given to one and the same customer for each of its top 20 albums. This was apparent from paragraphs 19 to 22 of the Commission’s document dated 21 September 2005, (207) and clearly followed from Annex E.2 to that document.

252. In order to review the allegation of distortion of evidence sensibly, there must first be ascertained exactly which tables the Court of First Instance was referring to in paragraph 425 of the judgment under appeal. Regrettably the authoritative English language version of the judgment is very imprecise on this point. It none the less appears even from this English language version of paragraph 425 that the Court of First Instance was referring to tables ‘which are intended to show the maximum campaign discounts granted by Sony and BMG for their best-selling albums’ (208) and in which a differential between minimum and maximum discounts was calculated by customer.

253. In this respect the French language version of the judgment under appeal is more precise. Even though it is not authoritative in the present case, this language version, in which the judgment was drafted and deliberated, contains additional indications of what the Court of First Instance was in fact intending to refer to in paragraph 425, namely ‘the tables in Annex E.4.2’. (209) And indeed in the case file the cover sheet to Annex E.4.2 includes a heading which is clearly similar to the introductory words of paragraph 425 of the judgment under appeal, (210) and the tables in Annex E.4.2 identify the differential referred to in paragraph 425 between minimum and maximum discounts by customer.

254. Accordingly, I proceed on the footing that paragraph 425 of the judgment under appeal refers to the tables in Annex E.4.2. However, this Annex does not disclose anything suggesting that the Court of First Instance’s discussion in paragraph 425 is clearly incorrect . Admittedly, the tables printed in this Annex, with the exception of the heading in the Annex’s cover sheet, do not contain any explanatory additions. None the less, even a fleeting glance at the first and second tables in Annex E.4.2. shows that the Court of First Instance’s criticism is justified. In calculating the differential between minimum and maximum discounts by customer, the values for Bertelsmann and Sony are sometimes thrown into the same pot. In colloquial terms, one might say that apples have been compared with pears. (211)

255. Nor does it follow from the Commission’s statements in paragraphs 19 to 22 of its document of 21 September 2005 that the Court of First Instance manifestly misinterpreted the tables in Annex E.4.2. According to their wording, the information provided by the Commission there refers only to Annexes B.6, B.8 and E.2. By contrast, the passages in the Commission’s document to which the appellants refer do not contain any explanation of the tables in Annex E.4.2 in question at this point.

256. Against this background I find that the complaint of distortion of evidence is unfounded as regards paragraph 425 of the judgment under appeal.

b) Paragraph 427 of the judgment under appeal

257. In paragraph 427 of the judgment under appeal the Court of First Instance states:

‘In any event, even on the assumption that the various tables drawn up by the parties to the concentration and produced by the Commission are in fact capable of establishing the more or less important variations alleged, the fact remains that … those variations are of doubtful relevance in so far as … they show only brackets without analysing the weighted averages and variations by reference to the averages …’

258. The appellants’ opinion is that, to the contrary, many of the items in the tables and charts produced by the Commission to the Court of First Instance in its reply of 11 February 2005 at first instance are indeed based on weighted average values. In this connection, they refer to Annexes B.4, B.8, B.9, B.10 and B.13, as well as to the additional explanations given by the Commission at first instance in its document of 14 March 2007.

259. On this point it is to be noted that in its judgment the Court of First Instance makes a clear linguistic distinction between tables and charts . (212) In paragraph 427 of the judgment under appeal the Court of First Instance expressly refers to the various tables produced by the Commission. I would therefore have expected that to prove their allegation of distortion of evidence the appellants would likewise refer to parts of the case files which contained tables . However, after having looked at Annexes B.4, B.8, B.9, B.10 and B.13, I have to find that although they contain a number of charts , none of these Annexes contain any tables . In these circumstances the Annexes referred to are just as little capable of establishing the alleged distortion of evidence as are any explanations the Commission has given of these Annexes.

260. It is much more probable that in paragraph 427 the Court of First Instance was referring to tables such as those in Annexes B.6 and B.7, which concern invoice discounts and which, moreover, contain the ‘brackets’ referred to by the Court of First Instance. However, the appellants do not make any submissions in relation to these.

261. Against this background I find that the complaint of distortion of evidence is unfounded as regards paragraph 427 of the judgment under appeal.

c) Paragraph 434 of the judgment under appeal

262. In paragraphs 431 to 434 of the judgment under appeal the Court of First Instance considers the question whether it is possible to determine net selling prices to retailers on the basis of retail prices using ‘reverse engineering’. The background to this is Impala’s submission at first instance that retailers’ mark-ups were generally transparent and were known with a high degree of accuracy. (213)

263. Paragraph 434 of the judgment under appeal includes the following:

‘… the study drawn up by the economic advisers to the parties to the concentration does not present data that are sufficiently reliable, relevant and comparable … While it is indeed probable that the different types of retailer (supermarkets, independents, specialist chains, etc.) apply different mark-up policies, and that there are differences within each category of operators, and even differences for each individual operator, according to the types of album or their degree of success, it is very unlikely, on the other hand, and the study contains no data in that regard, that a retailer will apply a different sales policy for the same type of album.’

264. The appellants claim that the Court of First Instance disregarded the study drawn up by their economic experts and produced by the Commission as Annex B.17, (214) and in particular section 2 thereof. This allegedly contains comprehensive economic information which indicates that the Court of First Instance was wrong to find that retailers followed a uniform mark-up policy.

265. This submission does not convince me. It is clear from merely reading the disputed paragraph 434 of the judgment under appeal that the Court of First Instance, far from disregarding this study, considered it.

266. As regards the substance of the study, it is correct that its section 2 concerns retailers’ mark-up policy and in particular whether retailers apply standard mark-ups. However, the Court of First Instance found the study to lack information as to whether ‘a retailer will apply a different sales policy for the same type of album’. (215) After reading the study in Annex B.17, I have reached the same conclusion as the Court of First Instance.

267. Against this background I find that the complaint of distortion of evidence is unfounded as regards paragraph 434 of the judgment under appeal.

3. Interim conclusion

268. Thus, the fourth ground of appeal is unfounded in its entirety, as are the points made in paragraphs 101 and 102 of the appeal, the substance of which overlaps with the fourth ground of appeal.

C – The use of undisclosed evidence in the judgment under appeal (seventh ground of appeal)

269. By their seventh ground of appeal the appellants claim that the Court of First Instance erred in law by basing its judgment on evidence that was not disclosed to the applicants, on which they had no opportunity of commenting, and which was not before the Commission at the time it adopted the first clearance decision. Given that such evidence could not have provided a basis for the Commission’s decision, the setting aside of the first clearance decision could likewise not be based on it.

270. The main point of attack of this ground of appeal is the Court of First Instance’s discussion of price transparency, and in particular of the majors’ – disputed – opportunity of overseeing the retail market with the assistance of weekly reports from their agents. (216) In this connection the Court of First Instance first emphasises one of the Commission’s findings, namely that Sony and Bertelsmann had a system of weekly reports which contained information about their competitors too. (217) By way of supplement to this, in its judgment the Court of First Instance refers also to certain documents produced by Impala and classified as confidential. (218)

271. Since the appellants complain inter alia that they were not able to comment on the latter documents before the Court of First Instance , it must first be considered whether there has been any breach of the adversarial principle before the Court of First Instance .

272. There is no doubt that a judicial decision cannot be based on facts and documents of which the parties themselves – or one of them – have no knowledge and on which they therefore have not been able to comment either. (219) However, in the present case, in the proceedings at first instance the appellants were not in the position of a party to the action, but participated merely as interveners in support of the Commission. As such they have a weaker position in the proceedings than an applicant or a defendant.

273. In particular, the second sentence of Article 116(2) of the Rules of Procedure of the Court of First Instance expressly provides that the President may, on application by one of the parties, omit secret or confidential information from being disclosed to an intervener, and this is not infrequent in competition law proceedings in particular. (220) Particularly in Community competition cases the taking of evidence is characterised by the fact that the documents considered often contain trade secrets or other information which may be published either not at all or only subject to significant restrictions. (221) Thus, it is already inherent in the scheme of the Rules of Procedure that the Court of First Instance can base its judgment, if necessary, on evidence which was not available to interveners (see also Article 67(3)(1) of the Rules of Procedure of the Court of First Instance, and moreover Article 287 EC).

274. Indeed, contrary to what might appear at first sight, the confidential treatment of documents submitted by Impala is less a problem in the adversarial proceedings before the Court of First Instance than a question of observing the rights of defence of the parties to the concentration in the administrative procedure .

275. The appellants correctly submit that the Court of First Instance was not entitled to rely on the disputed documents lodged by Impala to support the annulment of the first clearance decision, because the Commission itself ought not to have used these documents, owing to their confidential nature. If the Court of First Instance considers certain documents to be confidential to such an extent that it withholds their entire content from the undertakings participating in the concentration, (222) the Commission cannot reasonably be expected to rely on them in the administrative procedure to justify a possible prohibition decision, or even simply to rebut particular arguments of the parties to the concentration.

276. Against this background the Court of First Instance erred in law when it supported its decision to annul the first clearance decision by relying on the confidential documents lodged by Impala.

277. However, this error of law does not mean that the judgment under appeal is to be set aside, as the Court of First Instance’s findings were supported by other reasons which were unrelated to the confidential documents submitted by Impala. Thus, reliance on those documents was merely one of a number of factors from which the Court of First Instance concluded ‘that the already strong transparency … is increased even further’. (223) Accordingly, even if, having regard to the appellants’ rights of defence, the Court of First Instance had declined to take the confidential documents submitted by Impala into account, this would not have undermined the conclusions it drew in the judgment under appeal in any way. (224)

278. It follows that the seventh ground of appeal is also unfounded.

D – Interim conclusion

279. Accordingly, the appeal is to be dismissed in its entirety.

VI – The supposed cross appeal

280. At the end of the Commission’s response to the appeal there appears a separate section with ‘additional observations’ (225) on what the Court of First Instance described as the ‘essential grounds’ of the first clearance decision.

281. There the Commission submits that in paragraphs 474 and 476 of the judgment under appeal the Court of First Instance wrongly classified the Commission’s findings as regards retaliation measures (226) as an essential ground of the first clearance decision. In particular, if it were to be decided in the present proceedings that, contrary to the opinion of the Court of First Instance, the Commission’s findings as regards a lack of market transparency were not vitiated by an error of law, the first clearance decision would have to be upheld, regardless of whether it contained any error of law in relation to retaliation measures.

282. Impala considered these observations by the Commission to be a cross appeal , and under reference to Article 117(2) of the Rules of Procedure replied to them by a separate document. With permission of the President of the Court further documents were submitted on this matter.

283. However, this in itself by no means constitutes a definitive finding by the Court in the present case that there is in fact a cross appeal. Under Article 117(2) of the Rules of Procedure, for a submission to be regarded as a cross appeal it must seek to set aside, in whole or in part, the judgment under appeal on a plea in law which was not raised in the appeal. Whether this is the case here is to be determined by reference to the wording, aim and context of the passage in question in the Commission’s response to the appeal.

284. In that connection it is to be emphasised that the term ‘cross appeal’ is not used by the Commission anywhere in its document. Instead, the heading ‘Additional observations’ indicates that it contains supplementary points which are intended merely to achieve a better understanding of the Commission’s actual submission in response to Bertelsmann’s and Sony’s appeal. In particular, the ‘additional observations’ seek to clarify what the consequence for the validity of the first clearance decision would be if the judgment under appeal were set aside, in part as the case may be, at the instance of Bertelsmann and Sony. (227)

285. In the further course of the procedure the Commission itself made it clear that its ‘additional observations’ were by no means intended to constitute a cross appeal, and moreover expressly declined to assume any liability for costs in that regard; furthermore, it emphasised that its ‘additional observations’ did not have any independent effect, but were relevant only if the appeal lodged by Bertelsmann and Sony succeeded (in part as the case might be). (228)

286. Against this background it is to be held that the Commission’s ‘additional observations’ did not contain any cross appeal, and that it is not necessary for the Court to make a separate decision on it, such decision even being ultra petita .

VII – Costs

287. If, as I suggest in the present case, the appeal is dismissed, the Court decides on the costs (Article 122(1) of the Rules of Procedure) according to Article 69 in conjunction with Article 118 of the Rules of Procedure.

288. Article 69(2) of the Rules of Procedure provides that the unsuccessful party is to be ordered to bear the costs if they have been applied for; where there are several unsuccessful parties the Court is to decide how the costs are to be shared. In derogation from this, Article 69(3)(1) of the Rules of Procedure provides that where each party fails on some and succeeds on other heads, the Court may order that the costs be shared; this applies also where an appeal is dismissed notwithstanding that the appellant has succeeded in part of its submissions. (229)

289. As the appellants, Bertelsmann and Sony, have been unsuccessful in the result, but part of their grounds of appeal have in substance succeeded, the costs should be shared. I suggest that Bertelsmann and Sony should each be ordered to pay their own costs and three quarters of Impala’s costs; moreover, to the extent that they are to bear Impala’s costs it appears appropriate to order that Bertelsmann and Sony should be jointly and severally liable. (230) Impala should be ordered to bear one quarter of its own costs.

290. Other parties to the proceedings who support an appeal by applications to the Court may be ordered to bear their own costs, applying Article 69(4) by analogy. The Commission and Sony BMG Music Entertainment having supported the appeal of Bertelsmann and Sony with their own applications, and having been unsuccessful in the result in doing so, it appears appropriate to order each of them to bear its own costs. (231)

VIII – Conclusion

291. For the foregoing reasons I suggest to the Court that it should decide as follows:

(1) The appeal is dismissed.

(2) Bertelsmann AG and Sony Corporation of America are each ordered to bear their own costs, and to pay three quarters of the costs of the Independent Music Publishers and Labels Association, jointly and severally; the Independent Music Publishers and Labels Association shall bear one quarter of its own costs.

(3) The Commission of the European Communities and Sony BMG Music Entertainment BV shall each bear its own costs.

(1) .

(2)  – Commission Decision 2005/188/EC of 19 July 2004 declaring a concentration compatible with the common market and the functioning of the EEA Agreement (Case No COMP/M.3333 — SONY/BMG) (notified under document number C(2004) 2815) (OJ 2005 L 62, p. 30).

(3)  – Case T-464/04 Impala v Commission [2006] ECR II-2289.

(4)  – OJ 1989 L 395, p. 1; following correction, re-published in OJ 1990 L 257, p. 13.

(5)  – Regulation of 30 June 1997 amending Regulation (EEC) No 4064/89 on the control of concentrations between undertakings (OJ 1997 L 180, p. 1, with corrigenda in OJ 1998 L 3, p. 16, and OJ 1998 L 40, p. 17).

(6)  – If, as is in practice often the case, a concentration gives no cause for serious doubts as to its compatibility with the common market, it is cleared without the initiation of a formal merger control procedure already on the basis of the preliminary examination. Under Article 6(1)(b) of the Merger Regulation, in such cases the Commission is to decide not to oppose it and is to declare it compatible with the common market.

(7)  – In addition, see Articles 11 to 15 of Commission Regulation (EC) No 447/98 of 1 March 1998 on the notifications, time-limits and hearings provided for in Council Regulation (EEC) No 4064/89 on the control of concentrations between undertakings (OJ 1998 L 61, p. 1).

(8)  – Case C-170/02 P Schlüsselverlag J.S. Moser v Commission [2003] ECR I‑9889, paragraph 33, and Case C-42/01 Portugal v Commission [2004] ECR I-6079, paragraph 51); see also the judgments of the Court of First Instance in Case T-290/94 Kaysersberg v Commission [1997] ECR II-2137, paragraph 113, and Case T-251/00 Lagardère and Canal+ v Commission [2002] II‑4825, paragraph 108, as well as my Opinion in Case C-202/06 P Cementbouw Handel & Industrie v Commission [2007] ECR I-00000, paragraph 41.

(9)  – Article 57(1) and (2)(a) of the Agreement on the European Economic Area (‘EEA Agreement’, OJ 1994 L 1, p. 3).

(10)  – Council Regulation (EC) No 139/2004 of 20 January 2004 on the control of concentrations between undertakings (the EC Merger Regulation) (OJ 2004 L 24, p. 1).

(11)  – Hereinafter, ‘Bertelsmann’.

(12)  – On this and the following, see paragraphs 3 to 6 of the judgment under appeal.

(13)  – Hereinafter, ‘Sony’.

(14)  – The so-called ‘A & R’ (‘Artists and Repertoire’).

(15)  – In the first clearance decision and in the judgment under appeal this term was taken to mean the following publishers: Bertelsmann Music Group (BMG), Sony Music Entertainment (SMEI), Universal Music Group (UMG), Warner Music Group (WMG) and EMI Group (originally Electric and Musical Industries).

(16)  – Joined Cases C-68/94 and C-30/95 France and Others v Commission (Kali & Salz) [1998] ECR I‑1375, paragraphs 164 to 178; see also the judgment of the Court of First Instance in Case T-102/96 Gencor v Commission [1999] ECR II‑753, paragraphs 123 to 156.

(17)  – Gencor (cited above, footnote 16), paragraphs 276 and 277, and Case T-342/99 Airtours v Commission [2002] ECR II‑2585, paragraphs 59 to 61.

(18)  – Joined Cases C-395/96 P and C-396/96 P Compagnie maritime belge transports and Others v Commission [2000] ECR I‑1365, paragraph 45, in relation to Article 82 EC.

(19)  – Normally, this occurs by means of anti-competitive parallel behaviour . However, in Case COMP/M.2201 – MAN/Auwärter the Commission assessed a concentration for whether it enabled a permanent, tacit coordination of market behaviour by the members of a duopoly apart from the classic forms of anti-competitive parallel behaviour (Decision 2002/335/EC, OJ 2002 L 116, p. 35; see in particular recitals 33 to 35).

(20)  – See generally Gencor (cited above, footnote 16), paragraph 276.

(21) – Airtours (cited above, footnote 17), paragraph 62.

(22)  – Paragraph 247 of the judgment under appeal. To the same effect see Case T-347/00 Verband der freien Rohrwerke and Others v Commission [2003] ECR II‑2275, paragraph 186, and – in relation to Article 82 EC – Case T-193/02 Piau v Commission [2005] ECR II‑209, paragraph 111.

(23)  – On this point and the following see paragraphs 2 and 7 to 11 of the judgment under appeal.

(24)  – On this point see the Final report of the hearing officer (OJ 2005 C 59, p. 2).

(25)  – On a subsidiary basis Impala applied for the clearance decision to be annulled in certain respects: see the third indent of paragraph 29 of the judgment under appeal.

(26)  – On this and the following points see the Commission’s press releases of 1 March 2007 (IP/07/272) and of 3 October 2007 (IP/07/1437).

(27) – See, by way of example, Case C-24/05 P Storck v OHIM [2006] ECR I‑5677, paragraph 36, and Case C-25/05 Storck v OHIM [2006] ECR I‑5719, paragraph 40; see also Case C-105/04 P Nederlandse Federatieve Vereniging voor de Groothandel op Elektrotechnisch Gebied v Commission [2006] ECR I‑8725, paragraphs 69 and 70, Case C-328/05 P SGL Carbon v Commission [2007] ECR I-3921, paragraph 41, and Case C-167/06 P Komninou and Others v Commission [2007] ECR I-00000, paragraph 40.

(28)  – To this effect see Case C-403/04 P Sumitomo Metal Industries v Commission [2007] ECR I‑735, paragraph 40; see also my Opinion in Case C-95/04 P British Airways v Commission [2007] ECR I-2331, paragraph 113. In relation specifically to the requirement to give reasons see Case C-188/96 P Commission v V [1997] ECR I‑6561, paragraph 24, and Case C-166/95 P Commission v Daffix [1997] ECR I‑983, paragraph 35.

(29)  – Case C-185/95 P Baustahlgewebe v Commission [1998] ECR I‑8417, paragraph 23, Nederlandse Federatieve Vereniging voor de Groothandel op Elektrotechnisch Gebied (cited above, footnote 27), paragraph 69, Sumitomo Metal Industries (cited above, footnote 28), paragraph 39, and SGL Carbon (cited above, footnote 27), paragraph 41.

(30)  – Baustahlgewebe (cited above, footnote 29), paragraph 24, Nederlandse Federatieve Vereniging voor de Groothandel op Elektrotechnisch Gebied (cited above, footnote 27), paragraph 70, Sumitomo Metal Industries (cited above, footnote 28), paragraphs 38 and 39, SGL Carbon v Commission (cited above, footnote 29), paragraph 41, and Komninou and Others (cited above, footnote 27), paragraph 40.

(31) – Paragraphs 65 to 69 of the appeal.

(32) – Paragraphs 70 to 80 of the appeal.

(33)  – See point 25 above, and the judgment in Airtours (cited above, footnote 17), paragraph 62.

(34)  – This is the consistent case-law of the Court; see, for example, Case C-266/05 P Sison v Council [2007] ECR I-1233, paragraph 23, and Case C-227/04 P Lindorfer v Council [2007] ECR I-00000, paragraph 45.

(35)  – Paragraphs 44 to 481 of the judgment under appeal.

(36)  – Both paragraphs 44 to 481 and paragraphs 482 to 541 of the judgment under appeal.

(37)  – Paragraph 17 of the appeal. Similar references may be found in paragraphs 26 and 94 and in footnote 6 of the appeal.

(38)  – Paragraphs 17, 59 and 81 of the appeal.

(39)  – See above, points 50 to 55.

(40)  – See above, point 25, and Airtours (cited above, footnote 17), paragraph 62.

(41)  – To this effect see paragraphs 249 to 253 of the judgment under appeal.

(42)  – Paragraph 67 of the appeal.

(43)  – Case C-35/92 P Parliament v Frederiksen [1993] ECR I‑991, paragraph 31; Case C-122/01 P T. Port v Commission [2003] ECR I‑4261, paragraph 17; Joined Cases C-189/02 P, C-202/02 P, C-205/02 P to C-208/02 P and C-213/02 P Dansk Rørindustri and Others v Commission [2005] ECR I-5425, paragraph 148; and Case C-443/05 P Common Market Fertilizers v Commission [2007] ECR I-00000, paragraph 137.

(44)  – As regards the requirement of an interest in the appeal, see Case C-19/93 P Rendo and Others v Commission [1995] ECR I-3319, paragraph 13; to the same effect, in relation to interest at first instance, see Case C-362/05 P Wunenburger v Commission [2007] ECR I-00000, paragraph 42, and the order in Case T-28/02 First Data and Others v Commission [2005] ECR II‑4119, paragraphs 35 to 37.

(45)  – Rendo and Others (cited above, footnote 44), paragraph 13; Case C-174/99 P Parliament v Richard [2000] ECR I-6189, paragraph 33; and Case C-277/01 P Parliament v Samper [2003] ECR I‑3019, paragraph 28.

(46)  – Given that the disputed concentration was implemented at a time when the first clearance decision was still in force, Article 7(5) of the Merger Regulation is not applicable to the facts of the present case. The legal concept underlying this provision, namely temporary invalidity and retrospective cure of temporary invalidity by a clearance decision can, however, be usefully applied in the present case.

(47)  – Notwithstanding this, the Court has from time to time declared that there is no need to adjudicate on an appeal, where it is found that the appellant has already achieved its aim (orders of 23 October 2001 in Case C-281/00 P Una Film ‘City Revue’ v Parliament and Council , not published in the ECR, paragraphs 4 and 5, and Case C-313/00 P Davidoff v Parliament and Council , not published in the ECR, paragraphs 4 and 5), or where it is clear that it can no longer achieve its aim (Case C‑13/03 P Commission v Tetra Laval [2005] ECR I‑1113, paragraphs 21 to 23; orders in Case C-477/01 P(R) Reisebank v Commission [2002] ECR I-2117, paragraphs 24 to 28, and Case C‑480/01 P(R) Commerzbank v Commission [2002] ECR I-2129, paragraphs 23 to 27). As I will show in the following, however, neither of these is the case here.

(48)  – See, for example, Case C-396/93 P Henrichs v Commission [1995] ECR I‑2611, paragraph 66, Joined Cases C-302/99 P and C-308/99 P Commission and France v TF1 [2001] ECR I‑5603, paragraph 31, Joined Cases C-57/00 P and C-61/00 P Freistaat Sachsen and Others v Commission [2003] ECR I‑9975, paragraph 124, and Case C-301/02 P Tralli v ECB [2005] ECR I‑4071, paragraph 88.

(49)  – To be precise, Article 102(1) of the Rules of Procedure of the Court of First Instance provides that the period for challenges laid down by Article 230(5) EC begins with the expiry of the 14th day after publication of the decision in the Official Journal of the European Union . Admittedly, this presupposes that the non-confidential version of the grounds for the decision is published in full in the Official Journal, or at least is made available to the public at the same time via the internet. If this is not the case, the period for bringing a challenge begins to run only when the potential claimant receives a complete version, so far as he has not received it already (to this effect see Case 236/86 Dillinger Hüttenwerke v Commission [1988] ECR 3761, paragraph 14.

(50)  – Case T-110/97 Kneissl Dachstein Sportartikel v Commission [1999] ECR II‑2881, paragraphs 41 and 42, and Case T-123/97 Salomon v Commission [1999] ECR II‑2925, paragraphs 42 and 43, as well as the order in Case T-264/03 Schmoldt and Others v Commission [2004] ECR II‑1515, paragraphs 51 to 53 and 56.

(51)  – Case T-17/02 Olsen v Commission [2005] ECR II‑2031, paragraph 80, confirmed by the order of 4 October 2007 in Case C-320/05 P Olsen v Commission , not published in the ECR, and order in Case T‑321/04 Air Bourbon v Commission [2005] ECR II‑3469, paragraphs 34 and 37.

(52)  – Case T-17/02 Olsen (cited above, footnote 51), paragraph 81. As regards the subsidiarity of the time at which the claimant acquires knowledge of a decision which requires to be published in the Official Journal of the European Union , see also Case C-122/95 Germany v Council [1998] ECR I‑973, paragraph 35, and Case T-296/97 Alitalia v Commission [2000] ECR II-3871, paragraph 61, as well as the order in Case T-392/05 MMT v Commission [2006] ECR II-00000, paragraph 25.

(53)  – For the sake of completeness only, it may be mentioned that a formal notification (notice) of the decision to a third party by virtue of which the time period for it to bring a challenge might start to run is not relevant. Article 254(3) EC provides that decisions shall be notified only ‘to those to whom they are addressed’. In merger control, those are the undertakings concerned and the competent authorities of the Member States; see, for future cases, Article 8(8) of Regulation (EC) No 139/2004.

(54)  – See above, point 12.

(55)  – Case C-49/92 P Commission v Anic Partecipazioni [1999] ECR I‑4125, paragraph 171, and Case C-141/02 P Commission v max.mobil [2005] ECR I-1283, paragraph 48.

(56)  – See above, points 50 to 55 and 66 to 71.

(57)  – According to its reply to the appeal the Commission supports the first and second grounds of appeal, and the first part of the third ground of appeal.

(58)  – So far as the sixth ground of appeal complains about the Court of First Instance’s references to the statement of objections as errors in law (the second part of the sixth ground of appeal), I consider it below in conjunction with the first ground of appeal (see points 145 to 183). So far as the sixth ground of appeal concerns the substantive requirements of a clearance decision (fourth part of the sixth ground of appeal) it overlaps with the second part of the third ground of appeal and with the fourth ground of appeal and shall be considered in conjunction with them (points 201 to 232, and points 233 to 268).

(59)  – OJ 2000 C 364, p. 1. Admittedly, the Charter of fundamental rights does not yet as such have binding legal effect comparable to that of primary law, but as a source of recognition of law it does shed light on the fundamental rights guaranteed by Community law; on this point see Case C-540/03 Parliament v Council [2006] ECR I‑5769, paragraph 38), and point 108 of my Opinion in that case; see also Case C-432/05 Unibet [2007] ECR I-2271, paragraph 37.

(60)  – As regards the principle of transparency, see also Article 1(2) of the Treaty on European Union, which provides that in the European Union ‘decisions are taken as openly as possible’.

(61)  – According to consistent case-law, a provision of secondary Community law is to be interpreted, so far as possible, so that it is compatible with the Treaty and with the fundamental principles of Community law: see Joined Cases 201/85 and 202/85 Klensch and Others [1986] ECR 3477, paragraph 21; Case C-98/91 Herbrink [1994] ECR I-223, paragraph 9; Case C-1/02 Borgmann [2004] ECR I‑3219, paragraph 30; and Case C-457/05 Schutzverband der Spirituosen-Industrie v Diageo Deutschland [2007] ECR I‑00000, paragraph 22.

(62)  – In that event, there is a presumption that the concentration is compatible with the common market (as regards the concept of the presumption, see also Article 7(1) and (5) of the Merger Regulation).

(63)  – This obligation arises from Article 6(1) of the Merger Regulation as regards the preliminary phase and from Article 8(1) of the Merger Regulation as regards the formal investigation procedure.

(64)  – The Commission is relieved of this obligation only in so far as the concentration is referred to a national competition authority under Article 9 of the Merger Regulation, except for cases in which the notification is withdrawn because the intention to form a concentration is abandoned.

(65)  – So far as appears, in the long history of European merger control the Commission has only once failed to comply with a procedural time-limit, and then on account of a mistake in calculation (Case IV/M.330 – McCormick/CPC/Rabobank/Ostmann); in relation to this see Von Koppenfels, U., in Drauz / Jones (eds), EU Competition Law , Volume II – Mergers and Acquisitions (Leuven 2006), paragraph 6.27.

(66)  – Annulment actions at the instance of third parties under Article 242(1) EC do not have suspensory effect.

(67)  – On this point, see paragraphs 278 to 325, and in particular paragraphs 287 and 325, of the judgment under appeal.

(68)  – Recitals 153, 158 and 183 of the first clearance decision.

(69)  – Recitals 111 to 113 of the first clearance decision.

(70)  – Paragraph 294 of the judgment under appeal.

(71)  – The German translation of recital 69 of the first clearance decision is inaccurate in that regard, given that it uses the term ‘Preisabsprachen’ for the term ‘common price policy’.

(72)  – The Commission’s consideration appears essentially in recitals 74 to 80 of the first clearance decision, and the Court of First Instance reviewed the information provided there as regards the situation in the United Kingdom as representative of the reasons for the decision in general.

(73)  – Paragraphs 295 to 324 of the judgment under appeal.

(74)  – Paragraph 303 of the judgment under appeal.

(75)  – Paragraph 320 of the judgment under appeal.

(76)  – Paragraph 324 of the judgment under appeal.

(77)  – The last sentence of paragraph 289 of the judgment under appeal.

(78)  – The last sentence of paragraph 320 of the judgment under appeal.

(79)  – See, for example, Case C-367/95 P Commission v Sytraval and Brink’s France [1998] ECR I‑1719, paragraph 63, Joined Cases C-138/03, C-324/03 and C-431/03 Italy v Commission [2005] ECR I‑10043, paragraph 54, and Sison (cited above, footnote 34), paragraph 80.

(80)  – Sytraval and Brink’s France (cited above, footnote 79), paragraph 63, Italy v Commission (cited above, footnote 79), paragraph 55, and Sison (cited above, footnote 34), paragraph 80.

(81)  – As regards the influence of the temporal conditions in which a decision is issued on the duty to state reasons, see Case 16/65 Schwarze [1965] ECR 877, at p. 888, Case C-350/88 Delacre and Others v Commission [1990] ECR I‑395, paragraph 16, and Verband der freien Rohrwerke (cited above, footnote 22), paragraph 186.

(82)  – To this effect see Joined Cases 275/80 and 24/81 Krupp Stahl v Commission [1981] ECR 2489, paragraph 13, and, in relation to media policy, Joined Cases T-369/94 and T-85/95 DIR International Film and Others v Commission [1998] ECR II‑357, paragraphs 119 to 121.

(83)  – Case 32/86 SISMA v Commission [1987] ECR 1645, paragraph 9, and Case T-266/94 Skibsværftsforeningen and Others v Commission [1996] ECR II‑1399, paragraph 239.

(84)  – Sytraval and Brink’s France (cited above, footnote 79), paragraph 64, last sentence; see also Joined Cases C-465/02 and C-466/02 Germany and Denmark v Commission [2005] ECR I‑9115, paragraph 106, and Verband der freien Rohrwerke (cited above, footnote 22), paragraph 186, last sentence.

(85)  – Case 6/54 Netherlands v High Authority [1955] ECR 103, 111), Case 24/62 Germany v Commission [1963] ECR 63, 69, Case 41/69 ACF Chemiefarma v Commission [1970] ECR 661, paragraph 78, and Case T-206/99 Métropole télévision v Commission [2001] ECR II‑1057, paragraph 44, last sentence.

(86)  – Germany v Commission (cited above, footnote 85), p. 69 .

(87)  – ACF Chemiefarma (cited above, footnote 85), paragraph 78.

(88)  – Germany v Commission (cited above, footnote 85), p. 69, and Case 158/80 Rewe-Markt Steffen [1981] ECR 1805, paragraph 26.

(89)  – See for example paragraphs 294, 303, 319, 320 (second sentence) and 321 (last sentence) of the judgment under appeal.

(90)  – The last sentence of paragraph 289 of the judgment under appeal.

(91)  – The second sentence of paragraph 320 of the judgment under appeal; to the same effect see the last sentence of paragraph 289 of the judgment under appeal.

(92)  – The first sentence of paragraph 320 of the judgment under appeal.

(93)  – Again, see in particular the last sentence of each of paragraphs 289, 294 and 320 of the judgment under appeal.

(94)  – This distinguishes the present case from Kaysersberg (cited above, footnote 8), paragraphs 159 and 160, in which the Court of First Instance was of the opinion that the inadequacy complained about in the reasons given did not involve a fundamental part of the Commission’s decision.

(95)  – See paragraphs 286 and 289 of the judgment under appeal, according to which – as according to the view expressed by the Commission in the proceedings at first instance – ‘transparency is the essential, and indeed the only, ground for the assertion that there is no collective dominant position on the markets for recorded music’ (paragraph 289).

(96) – Kali & Salz (cited above, footnote 16), paragraph 223, and Case C-12/03 Commission v Tetra Laval [2005] ECR I‑987, paragraph 38; see also Gencor (cited above, footnote 16), paragraph 246, and Case T-210/01 General Electric v Commission [2005] ECR II‑5575, paragraph 60, from the Court of First Instance.

(97)  – Case C-269/90 Technische Universität München [1991] ECR I‑5469, paragraph 14; Joined Cases C-258/90 and C-259/90 Pesquerias De Bermeo und Naviera Laida v Commission [1992] ECR I‑2901, paragraph 26, as well as Case C-525/04 P Spain v Lenzing [2007] ECR I-00000, paragraph 58. In addition, see Joined Cases 36/59 to 40/59 Präsident Ruhrkohlen-Verkaufsgesellschaft and Others v High Authority [1960] ECR 423, at p. 439 et seq.

(98)  – To this effect, an example from the field of agricultural policy is Case C-358/90 Compagnia italiana alcool v Commission [1992] ECR I‑2457, paragraph 42.

(99)  – The last sentence of paragraph 289 of the judgment under appeal.

(100)  – On this point see recital 111 of the first clearance decision, quoted in paragraph 289 of the judgment under appeal (emphasis added).

(101)  – On this point see recital 80 of the first clearance decision, quoted in paragraphs 315 and 316 of the judgment under appeal (emphasis added).

(102)  – The last sentence of paragraph 289 of the judgment under appeal.

(103)  – Paragraph 294 of the judgment under appeal.

(104)  – Paragraph 320 of the judgment under appeal.

(105)  – Paragraph 325 of the judgment under appeal.

(106)  – See above, point 119 et seq.

(107)  – Kaysersberg (cited above, footnote 8), paragraph 105.

(108)  – To this effect see Case C-342/03 Spain v Council [2005] ECR I‑1975, paragraph 59, and Kaysersberg (cited above, footnote 8), paragraph 160.

(109)  – See above, point 119.

(110)  – Joined Cases 296/82 and 318/82 Netherlands and Others v Commission [1985] ECR 809, paragraph 27.

(111)  – Paragraph 413 of the judgment under appeal.

(112)  – On this point see for example recitals 74, 81, 88, 95 and 102 of the first clearance decision.

(113)  – Paragraph 282 of the judgment under appeal.

(114)  – Paragraphs 282 and 283 of the judgment under appeal.

(115)  – Paragraph 285 of the judgment under appeal.

(116)  – Case C-87/96 Assicurazioni Generali and Unicredito v Commission [1999] ECR II‑203, paragraph 88.

(117)  – This is the consistent case-law of the Court since Case 85/76 Hoffmann-La Roche v Commission [1979] ECR 461, paragraphs 9 and 11, and Joined Cases 100/80 to 103/80 Musique Diffusion française and Others v Commission [1983] ECR 1825, paragraph 10.

(118)  – Article 13(2) of Regulation No 447/98 provides that in merger control proceedings the Commission shall address its objections in writing to the notifying parties and to other involved parties, and shall set a time-limit within which they may inform the Commission of their views in writing.

(119)  – SGL Carbon v Commission (cited above, footnote 27), paragraph 55.

(120)  – See Article 13(2) and Article 14 of Regulation No 447/98.

(121)  – Joined Cases C-204/00 P, C-205/00 P, C-211/00 P, C-213/00 P, C-217/00 P and C-219/00 P Aalborg Portland and Others v Commission [2004] ECR I‑123, paragraph 67; to the same effect see also the earlier cases of Musique Diffusion française (cited above, footnote 117), paragraph 14, and Joined Cases 142/84 and 156/84 BAT and Reynolds v Commission [1987] ECR 4487, paragraph 70.

(122)  – Paragraph 283 of the judgment under appeal.

(123)  – Paragraph 285 of the judgment under appeal.

(124)  – In the history of European merger control there are numerous examples of clearance decisions which were preceded by a statement of objections. In this context the appellants rightly refer, by way of example, to the following cases: COMP/M.1940 – Framatome/Siemens/Cogema/JV, COMP/M.2499 – Norske Skog/Parenco/Walsum, COMP/M.2498 – UPM-Kymmene/Haindl, COMP/M.2314 – BASF/Pantochim/Eurodiol, COMP/M.2201 – MAN/Auwärter, COMP/M.2706 – Carnival Corporation/P & O Princess, COMP/M.3056 – Celanese/Degussa/European OXO Chemicals and COMP/M.3216 – Oracle/Peoplesoft.

(125)  – Paragraphs 284 and 285 of the judgment under appeal; to the same effect, see paragraphs 300, 335, 410 and 446 of the judgment under appeal.

(126)  – Paragraphs 161 to 182 below.

(127)  – According to consistent case-law, if the grounds of a judgment of the Court of First Instance reveal an infringement of Community law but its operative part appears well founded on other legal grounds the appeal must be dismissed (see, for example, Case C-167/04 P JCB Service v Commission [2006] ECR I‑8935, paragraph 186; similarly, see Commission v Tetra Laval, cited above, footnote 96, paragraph 89).

(128)  – See, for example, paragraphs 300, 302 and 338 of the judgment under appeal.

(129)  – See, for example, paragraphs 378, 379, 398 and 447 of the judgment under appeal.

(130)  – Recitals 111 to 113 of the first clearance decision.

(131)  – Paragraphs 289 to 294 of the judgment under appeal.

(132)  – Paragraphs 295 to 324 of the judgment under appeal.

(133)  – Paragraph 301 of the judgment under appeal (emphasis added).

(134)  – Paragraph 325 of the judgment under appeal.

(135)  – It is not disputed that the further passage criticised by the appellants and the Commission in paragraph 491 of the judgment under appeal likewise contains a reference to the statement of objections. However, it cannot in any event be regarded as containing an error of law, because it merely summarises Impala’s submissions. For that reason this paragraph of the judgment under appeal is not considered in more detail hereafter.

(136)  – Paragraph 377 of the judgment under appeal.

(137)  – Below, I have particular regard to paragraph 335 of the judgment under appeal, because it was the subject of particularly heavy criticism by the appellants. However, my analysis applies mutatis mutandis to the other paragraphs of the judgment under appeal which have been called into question.

(138)  – BAT and Reynolds (cited above, footnote 121), paragraph 70.

(139)  – On this point see the case-law cited above, footnote 96.

(140)  – Tetra Laval (cited above, footnote 96), paragraph 39, and Spain v Lenzing (cited above, footnote 97), paragraph 57.

(141)  – As regards these three points, see again Tetra Laval (cited above, footnote 96), paragraph 39, and Spain v Lenzing (cited above, footnote 97), paragraph 57; in relation specifically to the completeness of the factual basis see also Technische Universität München (cited above, footnote 97) paragraph 14, Spain v Lenzing (cited above, footnote 97), paragraph 58, and Komninou (cited above, footnote 27), paragraph 51, which emphasises the obligation of the competent organ to consider carefully and impartially all the relevant features of the individual case.

(142)  – The penultimate sentence of paragraph 335 of the judgment under appeal; see also paragraphs 410, 419 and 446 of that judgment.

(143)  – The second sentence of paragraph 335 of the judgment under appeal.

(144)  – Again, in relation to this point see Tetra Laval (cited above, footnote 96), paragraph 39, and Spain v Lenzing (cited above, footnote 97), paragraph 56; and to the same effect Technische Universität München (cited above, footnote 97), paragraph 14, Spain v Lenzing (cited above, footnote 97), paragraph 58, and Komninou (cited above, footnote 27), paragraph 51.

(145)  – The last sentence of paragraph 335 of the judgment under appeal; to the same effect see paragraphs 410, 419 and 446 of that judgment.

(146)  – Paragraphs 459 and 475 of the judgment under appeal.

(147)  – Paragraph 414 of the judgment under appeal.

(148)  – Paragraph 415 of the judgment under appeal; to similar effect see paragraph 452 of the judgment under appeal.

(149)  – See, for example, paragraphs 398, 428 and 451 of the judgment under appeal.

(150)  – See above, in particular points 150 to 152.

(151)  – Of course, this does not affect the obligation of the undertakings concerned to lodge a notification of their concentration which is materially correct and complete (on this point, see Article 3(1) of Regulation) No 447/98). In addition, the parties to the concentration are obliged to answer any requests by the Commission for information comprehensively, truthfully and within the applicable time-limit (Article 11(1) in conjunction with Article 11(4) and (5) of the Merger Regulation).

(152)  – The parties to the concentration are made aware of the progress of the procedure prior to the statement of objections (discussions take place, and the notifying parties are told the areas in which the Commission has serious doubts as regards the compatibility of the concentration with the common market when the formal investigation procedure is initiated under Article 6(1)(c) of the Merger Regulation), but this information is much less detailed than a statement of objections and depending on how the market investigation proceeds can also change as matters progress.

(153)  – If it should appear clearly that in their notification of the concentration or in their answer to the Commission’s questionnaire the undertakings concerned supplied incorrect or incomplete information, this would be an infringement of their obligation to cooperate with the Commission in the merger control procedure and could also lead to the legal consequences laid down in Article 8(5)(a) of the Merger Regulation and Article 14(1)(b) and (c) of the Merger Regulation (revocation of the clearance decision and imposition of fines).

(154)  – Article 19(3) to (7) of the Merger Regulation.

(155)  – On this point, see again in particular paragraphs 414 and 415 of the judgment under appeal (emphasis added).

(156)  – On this point, see again the case-law cited above, footnote 127.

(157)  – See above, points 93 to 144.

(158)  – The following discussion applies also to the part of the sixth ground of appeal contained in paragraphs 98 to 100 and 102 of the appeal.

(159)  – Tetra Laval (cited above, footnote 96), paragraph 42; see also General Electric (cited above, footnote 96), paragraph 64.

(160)  – Tetra Laval (cited above, footnote 96), paragraph 44 (emphasis added).

(161)  – Paragraph 12 above.

(162)  – General Electric (cited above, footnote 96), paragraph 64, last sentence (emphasis added).

(163)  – Tetra Laval (cited above, footnote 96), paragraph 39, Spain v Lenzing (cited above, footnote 97), paragraph 57, and General Electric (cited above, footnote 96), paragraph 63; see also above, paragraphs 173 and 174.

(164)  – For a different view, see the Opinion of Advocate General Tizzano in Tetra Laval (cited above, footnote 96), paragraph 74, who holds that the notified transaction must very probably lead to the creation or strengthening of a dominant position if it is to be prohibited.

(165)  – See paragraphs 126 and 173 of the judgment under appeal.

(166)  – To similar effect, see the Opinion of Advocate General Tizzano in Tetra Laval (cited above, footnote 96), paragraph 74: ‘It therefore cannot be claimed that in order to prohibit a concentration the Commission must establish with absolute certainty that the concentration would lead to the creation or strengthening of a dominant position as a result of which effective competition would be significantly impeded in the common market or in a substantial part of it’ (emphasis added).

(167)  – For an example from environmental law, see Case C-127/02 Landelijke Vereniging tot Behoud van de Waddenzee and Nederlandse Vereniging tot Bescherming van Vogels [2004] ECR I‑7405, paragraphs 44 and 55 to 59.

(168)  – On this point see, on the one hand, the third and fourth recitals to the Merger Regulation and, on the other, the fifth recital thereto.

(169)  – See also the first, second and fifth recitals of the Merger Regulation. The protection of competition against distortion also serves the interests of all market participants, including consumers (on this point, see Case 6/72 Europemballage and Continental Can v Commission [1973] ECR 215, paragraph 25, Case 322/81 Michelin v Commission [1983] ECR 3461, paragraph 125, and Case C-95/04 P British Airways v Commission [2007] ECR I-2331, paragraph 106).

(170)  – Tetra Laval (cited above, footnote 96), paragraph 41; see also Kali & Salz (cited above, footnote 16), paragraph 228, which refers to, ‘cogent and consistent evidence’. The German translation of Tetra Laval , which refers to ‘eindeutig’ evidence, does not appear to me to be a correct translation of the English phrase ‘convincing evidence’; accordingly, here and in the following I use the adjective ‘überzeugend’ to mean ‘convincing’.

(171)  – Tetra Laval (cited above, footnote 96), paragraph 44.

(172)  – Tetra Laval (cited above, footnote 96), paragraph 41 (emphasis added).

(173)  – Tetra Laval (cited above, footnote 96), paragraph 42.

(174)  – Tetra Laval (cited above, footnote 96), paragraph 39, and General Electric (cited above, footnote 96), paragraph 63; in addition, see above, points 173 and 174.

(175)  – These interests are, on the one hand, the rights and interest of the parties to the concentration and, on the other, the public interest in the protection of competition from distortion; again, on this point see above, point 214.

(176)  – See also General Electric (cited above, footnote 96), paragraph 61.

(177)  – Case T-87/05 EDP v Commission [2005] ECR II‑3745, paragraph 64.

(178)  – In relation to Article 10(6) of the Merger Regulation, see also above, point 102.

(179)  – Opinion of Advocate General Tizzano in Tetra Laval (cited above, footnote 96), paragraphs 76 to 81, in particular paragraph 76.

(180)  – See for example paragraphs 290, 294, 303, 347, 362, 407 and 435 of the judgment under appeal.

(181)  – See for example recitals 80, 87, 94, 101, 108, 111, 113, 150, 153 and 158 of the first clearance decision.

(182)  – At first instance too, the Commission said it felt it possible to defend the first clearance decision on the basis of a ‘balance of probabilities’ (see paragraph 7 of its response at first instance).

(183)  – In particular, the appellants refer to paragraphs 289, 366, 381 to 387, 389, 407, 420, 428, 429, 433, 449 to 457 and 459 of the judgment under appeal.

(184)  – See above, points 212 to 218.

(185)  – Paragraphs 289 and 459 of the judgment under appeal; see also paragraphs 287, 366 and 371 of that judgment.

(186)  – See in particular paragraphs 377, 390, 459 and 542 of the judgment under appeal.

(187)  – The following discussion applies also in relation to the part of the sixth ground of appeal contained in paragraphs 101 and 102 of the appeal.

(188)  – Paragraph 299 of the judgment under appeal.

(189)  – Paragraph 307 of the judgment under appeal; the appellants refer in addition to paragraphs 421, 419, 424, 444 and 457 of the judgment under appeal, in which they identify similar statements by the Court of First Instance.

(190)  – Paragraph 317 of the judgment under appeal.

(191)  – Paragraphs 347 and 361 of the judgment under appeal.

(192)  – Paragraph 354 of the judgment under appeal.

(193)  – Paragraph 402 of the judgment under appeal. It should be observed that this claim is based on the – authoritative – English language version of the judgment (‘destined to become public knowledge ’), whereas the French language version states that a campaign discount, ‘semble, par essence, avoir vocation à revêtir un caractère de publicité ’, which seems not to be very similar to the English term, ‘ public knowledge ’ (emphasis added).

(194)  – Paragraphs 403, 405, 406 and 436 of the judgment under appeal.

(195)  – Paragraph 420 of the judgment under appeal.

(196)  – Paragraph 456 of the judgment under appeal.

(197)  – See in particular the discussion concerning the first and sixth grounds of appeal in paragraphs 126 and 173 above.

(198)  – Case T-342/00 Petrolessence and SG2R v Commission [2003] ECR II‑1161, paragraph 101, EDP (cited above, footnote 177), paragraph 151, Case T-177/04 easyJet v Commission [2006] ECR II‑1931, paragraph 44.

(199)  – In passing it may be mentioned that the context in which the appellants’ criticism of some of the findings appears, namely those in paragraphs 299, 307 and 317 of the judgment under appeal, indicates that they belong to the formal review of the first clearance decision for failure to state reasons . That criticism is principally addressed by what has been said above as regards the third part of the sixth ground of appeal (see in particular paragraphs 114 to 131). However, given that the appellants appear to assume that paragraphs 299, 307 and 317 of the judgment under appeal contain also elements of the review of the substance of the first clearance decision for manifest errors of assessment, in the following I consider them also on this footing.

(200)  – Tetra Laval (cited above, footnote 96), paragraph 39; likewise Spain v Lenzing (cited above, footnote 97), paragraphs 56 and 57, and General Electric (cited above, footnote 96), paragraph 63.

(201)  – See already above, paragraph 179, albeit in a different context.

(202)  – Again, see Tetra Laval (cited above, footnote 96), paragraph 39, most recently confirmed in Spain v Lenzing (cited above, footnote 97), paragraph 57.

(203)  – See paragraph 479 of the judgment under appeal: ‘It is not for the Court to rule on the compatibility of the concentration with the common market, but to review the lawfulness of the findings made in the Decision.’

(204)  – Paragraph 452 of the judgment under appeal.

(205)  – Case C-229/05 P PKK and KNK v Council [2007] ECR I‑445, paragraph 37, Case C-326/05 P Industrias Químicas del Vallés v Commission [2007] ECR I-00000, paragraph 60, and Case C-260/05 P Sniace v Commission [2007] ECR I‑00000, paragraph 37.

(206)  – Case C-551/03 P General Motors v Commission [2006] ECR I‑3173, paragraph 54, JCB Service (cited above, footnote 127), paragraph 108, and Wunenburger (cited above, footnote 44), paragraph 67.

(207)  – Commission document of 21 September 2005 responding to written questions from the Court of First Instance.

(208)  – ‘As regards the tables, which are intended to show the maximum campaign discounts granted by Sony and BMG for their best-selling albums , …’ (emphasis added).

(209)  – In the French language version of the judgment under appeal paragraph 425 begins, ‘S’agissant des tableaux de l’annexe E 4.2 qui ont pour objet de montrer les remises promotionnelles maximales accordées par Sony et BMG pour leurs albums les mieux vendus, …’ (emphasis added).

(210)  – The heading on the cover sheet to Annex E.4.2 reads: ‘Invoice discounts granted to each major customer for each top album listed in Annex B.13, with an estimate of the highest campaign discount granted to each customer for such albums’.

(211)  – For example, in the third column of the first table in Annex E.4.2 the lowest discount granted by Sony (SMEI) is compared with the highest discount granted by Bertelsmann (BMG); the same occurs in the third column of the second table in Annex E.4.2.

(212)  – See, for example, paragraphs 393, 401, 415, 416, 420 to 428 and 455 to 457 of the judgment under appeal, which refer to ‘tables’ (in German, ‘Tabellen’), whereas, for example, paragraphs 129 and 419 of the judgment under appeal refer to ‘charts’ (in German, ‘Grafiken’).

(213)  – This context appears from paragraphs 431 and 433 of the judgment under appeal.

(214)  – This is Annex B.17 to the Commission’s response at first instance.

(215)  – Paragraph 434 of the judgment under appeal (emphasis added).

(216)  – Paragraphs 352 to 361 and 451 of the judgment under appeal.

(217)  – See paragraph 352 of the judgment under appeal, which quotes from recital 113 of the first clearance decision.

(218)  – Paragraphs 356 to 360, 389 and 451 of the judgment under appeal.

(219)  – Joined Cases 42/59 and 49/59 Snupat v High Authority [1961] ECR 53, at p. 84, Case C-480/99 P Plant and Others v Commission and South Wales Small Mines [2002] ECR I‑265, paragraph 24, and Case C-199/99 P Corus UK v Commission [2003] ECR I‑11177, paragraph 19). See also the judgments of the European Court of Human Rights in Feldbrugge v The Netherlands (1986) Series A, No. 99, p. 16, paragraph 44, and Aksoy (Eroğlu) v Turkey , 31 October 2006, Application No. 59741/00, paragraph 21, and the case-law referred to there.

(220)  – See for example the order in Case T-271/03 Deutsche Telekom v Commission [2006] ECR II‑1747.

(221)  – Case C-411/04 P Salzgitter Mannesmann v Commission [2007] ECR I‑959, paragraph 43.

(222)  – Even if I have considerable doubt after looking at the case file from the proceedings at first instance as to whether such an approach was justified in the present case, it is not the task of the Court in the present appeal proceedings to call in question the Court of First Instance’s assessment as to the confidentiality of these documents.

(223)  – Paragraphs 348 to 362, in particular paragraph 362, of the judgment under appeal.

(224)  – To the same effect, see Aalborg Portland (cited above, footnote 121), paragraph 72, Musique Diffusion française (cited above, footnote 117), paragraph 30, and Case T-30/91 Solvay v Commission [1995] ECR II‑1775, paragraph 58.

(225)  – Paragraphs 37 to 39 of the Commission’s response to the appeal, which is headed, ‘6. Additional observations: On the “essential grounds” of the Decision’.

(226)  – Paragraphs 114 to 118 of the first clearance decision.

(227)  – See in particular the last sentence of paragraph 39 of the Commission’s response to the appeal.

(228)  – Letter by the Commission dated 15 May 2007 to the Registrar of the Court.

(229)  – Case C-93/02 P Biret International v Council [2003] ECR I‑10497, paragraph 72, and Case C‑94/02 P Biret and Cie v Council [2003] ECR I‑10565, paragraph 75.

(230)  – Joined Cases C-122/99 P D and Sweden v Council [2001] ECR I‑4319, paragraph 65.

(231)  – To this effect see for example Case C-23/00 P Council v Boehringer [2002] ECR I‑1873, paragraph 56, and Joined Cases C-172/01 P, C-175/01 P, C-176/01 P and C-180/01 P International Power and Others v NALOO [2003] ECR I‑11421, paragraph 187.