OPINION OF ADVOCATE GENERAL
WAHL
delivered on 18 January 2018(1)
Case C‑633/16
Ernst & Young P/S
v
Konkurrencerådet
(Request for a preliminary ruling from the Sø- og Handelsretten (Maritime and Commercial Court, Denmark))
(Competition — Control of concentrations between undertakings — Article 7(1) of Regulation (EC) No 139/2004 — Duty to suspend the implementation of a concentration until it has been declared compatible with the common market — Termination of a cooperation agreement by one of the merging undertakings)
1. It has been said that being too early is indistinguishable from being wrong. That is especially true for early implementation of a concentration between undertakings.
2. In order to determine when implementation of a concentration is too early and therefore wrongful, it is important to have a clear idea of the scope of the obligation to suspend the implementation of a concentration and of the corresponding powers of the European Commission to establish premature implementation under Article 7(1) of Regulation (EC) No 139/2004 (2) before the concentration has been approved (‘the standstill obligation’). More importantly, how is the exact scope of the standstill obligation to be determined?
3. Those succinct yet important issues, which are raised in the present proceedings, have hitherto gone unanswered in the case-law of the Court.
4. In ruling on those issues, the Court will have to consider whether the aim of ensuring the effective enforcement of ex ante merger control makes it necessary to extend the standstill obligation beyond the substantive scope of the merger rules as laid down in Regulation No 139/2004.
5. In what follows, I shall explain why I believe the scope of the standstill obligation must be clearly demarcated and why that is best done by recourse to a negative definition, that is to say by defining what lies outwith the scope of Article 7(1) of Regulation 139/2004.
I. Legal framework
A. Regulation No 139/2004
6. Under Article 1(1) of Regulation No 139/2004 (‘Scope’), ‘this Regulation shall apply to all concentrations with a [Union] dimension’.
7. Article 3 of Regulation No 139/2004 (‘Definition of concentration’) provides:
‘1. A concentration shall be deemed to arise where a change of control on a lasting basis results from:
(a) the merger of two or more previously independent undertakings or parts of undertakings, or
(b) the acquisition, by one or more persons already controlling at least one undertaking, or by one or more undertakings, whether by purchase of securities or assets, by contract or by any other means, of direct or indirect control of the whole or parts of one or more other undertakings.
2. Control shall be constituted by rights, contracts or any other means which, either separately or in combination and having regard to the considerations of fact or law involved, confer the possibility of exercising decisive influence on an undertaking, in particular by:
(a) ownership or the right to use all or part of the assets of an undertaking;
(b) rights or contracts which confer decisive influence on the composition, voting or decisions of the organs of an undertaking.
3. Control is acquired by persons or undertakings which:
(a) are holders of the rights or entitled to rights under the contracts concerned; or
(b) while not being holders of such rights or entitled to rights under such contracts, have the power to exercise the rights deriving therefrom.
…’
8. Article 4(1) of Regulation No 139/2004 (‘Prior notification of concentrations and pre-notification referral at the request of the notifying parties’) states:
‘Concentrations with a [Union] dimension defined in this Regulation shall be notified to the Commission prior to their implementation and following the conclusion of the agreement, the announcement of the public bid, or the acquisition of a controlling interest.
Notification may also be made where the undertakings concerned demonstrate to the Commission a good faith intention to conclude an agreement or, in the case of a public bid, where they have publicly announced an intention to make such a bid, provided that the intended agreement or bid would result in a concentration with a [Union] dimension.
For the purposes of this Regulation, the term “notified concentration” shall also cover intended concentrations notified pursuant to the second subparagraph. For the purposes of paragraphs 4 and 5 of this Article, the term “concentration” includes intended concentrations within the meaning of the second subparagraph.’
9. Article 7 of Regulation No 139/2004 (‘Suspension of concentrations’) provides:
‘1. A concentration with a [Union] dimension as defined in Article 1 … shall not be implemented either before its notification or until it has been declared compatible with the common market …
…
3. The Commission may, on request, grant a derogation from the obligations imposed in paragraphs 1 or 2. The request to grant a derogation must be reasoned. In deciding on the request, the Commission shall take into account inter alia the effects of the suspension on one or more undertakings concerned by the concentration or on a third party and the threat to competition posed by the concentration. Such a derogation may be made subject to conditions and obligations in order to ensure conditions of effective competition. A derogation may be applied for and granted at any time, be it before notification or after the transaction.
…’
10. Under Article 8 of Regulation No 139/2004 (‘Powers of decision of the Commission’):
‘...
4. Where the Commission finds that a concentration:
(a) has already been implemented and that concentration has been declared incompatible with the common market, or
(b) has been implemented in contravention of a condition attached to a decision … ,
the Commission may:
– require the undertakings concerned to dissolve the concentration, in particular through the dissolution of the merger or the disposal of all the shares or assets acquired, so as to restore the situation prevailing prior to the implementation of the concentration; in circumstances where restoration of the situation prevailing before the implementation of the concentration is not possible through dissolution of the concentration, the Commission may take any other measure appropriate to achieve such restoration as far as possible,
– order any other appropriate measure to ensure that the undertakings concerned dissolve the concentration or take other restorative measures as required in its decision.
…
5. The Commission may take interim measures appropriate to restore or maintain conditions of effective competition where a concentration:
(a) has been implemented in contravention of Article 7, and a decision as to the compatibility of the concentration with the common market has not yet been taken;
…
(c) has already been implemented and is declared incompatible with the common market.
...’
B. National law
11. Paragraph 12c of the Konkurrenceloven (3) (Law on Competition) states:
‘(1) The Konkurrence- og Forbrugerstyrelsen [Competition and Consumer Authority, “the KFST”] shall decide whether a concentration may be approved or prohibited.
(2) A concentration that will not significantly impede effective competition, in particular due to the creation or strengthening of a dominant position, shall be approved. A concentration that will significantly impede effective competition, in particular due to the creation or strengthening of a dominant position, shall be prohibited.
…
(5) A concentration that is subject to the provisions of this Law shall not be implemented either before it has been notified or until it has been approved by the [KFST] under subparagraph 1 above …
…’
12. According to the order for reference, the explanatory recitals accompanying the legislative amendment which gave rise to Paragraph 12c state that the Danish merger control provisions are based on those of Regulation No 139/2004 and are to be interpreted accordingly as regards both the definition and the scope of the concept of a ‘concentration’ and the standstill obligation.
II. Facts, procedure and the questions referred
13. By agreement concluded on 18 November 2013 between KPMG Statsautoriseret Revisionspartnerselskab, Komplementarselskabet af 1. januar 2009 Statsautoriseret Revisionsaktieselskab, and KPMG Ejendomme Flintholm K/S (referred to collectively as ‘KPMG DK’), on the one hand, and Ernst & Young P/S, Ernst & Young Europe LLP, Ernst & Young Godkendt Revisionsaktieselskab, Ernst & Young Global Limited and EYGS LLP (these legal entities are referred to collectively as ‘EY’), on the other hand, KPMG DK and EY decided to merge (‘the merger agreement’).
14. At the time of conclusion of the merger agreement, KPMG DK and EY were both auditing firms active on the Danish market in auditing and accountancy services. KPMG DK was a member of an international network of independent auditing firms, KPMG International Cooperative (‘KPMG International’) by virtue of a cooperation agreement signed on 15 February 2010 (‘the cooperation agreement’).
15. The cooperation agreement gave the participants the exclusive right to be included in the KPMG network on the national level and to use the trademarks of KPMG International for marketing purposes. The cooperation agreement also contains provisions on the allocation of customers, the obligation to service clients from other territories and the annual compensation for participation in the cooperation. The cooperation agreement presupposes that the auditing firms participating in the KPMG network at no point in time participate in a partnership/joint venture or the like.
16. Under the merger agreement, KPMG DK was to terminate its agreement with KPMG International immediately after the signing of the merger agreement, so that it could withdraw from its cooperation with the latter with a view to merging with EY instead and becoming part of the EY group.
17. It is common ground that the merger agreement was subject to notification, since EY was to acquire control of KPMG DK and the turnover of the undertakings involved exceeded the quantitative thresholds laid down in the Danish competition rules.
18. After signing the merger agreement with EY on 18 November 2013, KPMG DK terminated the cooperation agreement with effect from 30 September 2014, firstly by a telephone call informing the chairman of the board of KPMG International in advance and subsequently by letter of 18 November 2013 to KPMG International. The actual termination of the cooperation agreement was not in itself subject to approval by the competition authorities.
19. The conclusion of the merger agreement was made public on 19 November 2013. From that date until the approval of the merger, a number of events took place.
20. On 20 November 2013 KPMG International announced in an article in the online newspaper Business.dk its intention to remain on the Danish market. KPMG International chose to establish a new auditing business in Denmark. It also concluded a cooperation agreement with a tax consultancy firm despite the fact that its cooperation agreement with KPMG DK remained in force and the cooperation continued as before, and the notice of termination notwithstanding.
21. In addition, some auditing clients of KPMG DK, including two of the very largest; the Carlsberg and Mærsk groups, chose to recommend a change of auditors at their general meetings, so that KPMG International would be used as their auditor for the accounting year 2014. Similarly, other companies chose to switch from KPMG DK to other auditing firms in this period. In most cases, the ordinary general meetings take place in early spring, because the companies generally take the calendar year as their accounting year and because, under Danish company law, the auditor must be selected at the ordinary general meeting.
22. KPMG DK and EY started the pre-notification procedure immediately after the merger agreement had been announced in order to obtain approval of the merger, with the first exploratory contacts being made with the KFST on 21 November 2013.
23. The first draft notification was sent to the KFST on 13 December 2013, the notification was completed on 7 February 2014, and the merger was approved with the consent of the Konkurrencerådet (the Competition Council) by decision of 28 May 2014. Once the merger was approved by the Competition Council, KPMG DK and KPMG International agreed to end their cooperation with effect from 30 June 2014.
24. On 17 December 2014 the Competition Council issued a decision finding that KPMG DK had infringed the standstill obligation under Paragraph 12c(5) of the Law on Competition by giving notice to terminate the cooperation agreement with KPMG International on 18 November 2013, in accordance with the terms of the merger agreement, before the merger had been approved by the Competition Council (‘the contested decision’).
25. The contested decision is based on an overall assessment of the circumstances of the case by the KFST. However, the KFST laid particular emphasis on the fact that the termination of the agreement was (i) merger-specific, (ii) irreversible, and (iii) had the potential to have market effects in the period between the notice of termination and the approval of the merger, though those three elements were stated not to be exhaustive.
26. As regards the potential for market effects, the contested decision stated that there could be several reasons for the market effects noted, including the fact that it was not possible to determine the background to the market reactions in concrete terms, but since, in the view of the Competition Council, the termination itself had an inherent potential for market effects and hence could be characterised as an implementing action, it was not necessary to demonstrate any actual effects of the termination.
27. Disagreeing with the contested decision, on 1 June 2015 Ernst & Young P/S (‘Ernst & Young’) brought an action seeking its annulment before the referring court. In that connection, the referring court states that the outcome of the main proceedings is relevant to the question of the imposition of possible fines, since on 11 June 2015 the case was referred by the KFST to the Statsanklager for Særlig Økonomisk og International Kriminalitet (the State Prosecutor for Serious Economic and International Crime) with a view to its examination under provisions of criminal law.
28. Entertaining doubts as to the correct interpretation of Article 7(1) of Regulation No 139/2004, in order to interpret Paragraph 12c(5) of the Law on Competition accordingly, the referring court decided to stay the proceedings and to refer the following questions to the Court for a preliminary ruling:
‘(1) What criteria are to be applied in assessing whether the conduct or actions of an undertaking are covered by the prohibition in Article 7(1) of Regulation No 139/2004 ([the standstill obligation]), and does implementing action within the meaning of Article 7(1) thereof presuppose that the action, wholly or in part, factually or legally, forms part of the actual change of control or merging of the continuing activities of the participating undertakings which — provided the quantitative thresholds are met — gives rise to the obligation of notification?
(2) Can the termination of a cooperation agreement, such as in the present case, which is announced under circumstances corresponding to those described in the order for reference constitute an implementing action covered by the prohibition in Article 7(1) of Regulation No 139/2004, and what criteria are then to be applied in making a decision?
(3) Does it make any difference in answering Question 2 whether the termination has actually given rise to market effects relevant to competition law?
(4) If the answer to Question 3 is in the affirmative, clarification is requested as to what criteria and what degree of probability should be applied in a given case whether the termination has given rise to such market effects, including the significance of the possibility that those effects could be attributed to other causes.’
29. Observations in the written proceedings have been lodged by Ernst & Young, the Danish Government and the Commission. Those parties also presented oral argument at the hearing held on 15 November 2017.
III. Analysis
A. Jurisdiction of the Court
30. In its written submissions, the Commission questions whether the Court has jurisdiction to interpret the questions referred even though Article 7(1) of Regulation No 139/2004 does not directly govern the situation at issue in the main proceedings. More specifically, the Commission questions whether reference to that regulation in travaux preparatoires of the national legislation is sufficient to confer jurisdiction under Article 267 TFEU.
31. In a long line of case-law stemming from the judgment in Dzodzi, (4) the Court has held that in order to forestall future differences of interpretation, provisions or concepts taken from EU law should be interpreted uniformly, irrespective of the circumstances in which they apply. That approach was ultimately confirmed by the Court in Allianz Hungária Biztosító and Others. (5)Consequently, the Court has jurisdiction to give preliminary rulings on questions concerning EU law in situations where the facts of the case considered by a national court are outside the direct scope of EU law but where the national legislature has chosen to treat internal situations and situations governed by EU law in the same way. (6)
32. Therefore I conclude that the Court has jurisdiction to answer the questions referred for a preliminary ruling by the Sø- og Handelsretten (Maritime and Commercial Court, Denmark).
B. Substance
1. Preliminary observations
33. Without wanting to prejudge the necessity of advance notification of concentrations, I would call to mind that in many jurisdictions, no (national) standstill obligation exists. For instance, in the European Union, that appears to be the case in Italy, (7) Latvia (8) and the United Kingdom (where notification is, moreover, described as ‘voluntary’). (9)
34. At EU level, the initial merger control regulation only provided for a three week standstill obligation following notification of the concentration. That obligation could be prolonged by decision of the Commission. (10)
35. That indicates that while the standstill obligation might be useful, it would seem excessive, as the Commission does, to classify it as an indispensable tool for merger control. Its function is simply to deter undertakings from implementing concentrations prematurely, pending assessment by the competition authorities, and to reduce the risk that a concentration will have to be undone in case it is not approved. Put differently, the standstill obligation essentially allocates to the undertakings concerned the financial burden associated with delaying a concentration until the competition authority’s assessment is complete, as it does the financial risk associated with its potentially necessary dissolution in case of unlawful pre-implementation.
36. Those features must be kept in mind when exploring the limits of the standstill obligation under Regulation No 139/2004. Indeed, under EU merger rules, the relationship between the ex ante and ex post regulation of concentrations, as embodied by Articles 7 and 8 of Regulation No 139/2004, has yet to be fully clarified by the Court.
37. As regards the assessment of the questions referred, they may be dealt with as two pairs. On the one hand, the first and second questions referred both concern the scope of the standstill obligation and how it ought to be determined (the second question specifically asking whether the obligation arose in the main proceedings). On the other hand, the third and fourth questions referred are more specific. They both centre on the relevance to the standstill obligation of any possible market effects, including which criteria may be used to show such an effect and the evidentiary threshold which must be satisfied in that regard by the competition authority when investigating a possible breach of that obligation.
2. The first and second questions referred
38. By its first two questions, the referring court essentially asks the Court to clarify the scope of the standstill obligation set out in Article 7(1) of Regulation No 139/2004 and in which circumstances actions by undertakings will be caught by that provision. Specifically, the referring court seeks guidance on how to determine the scope of the standstill obligation, referring to specific criteria used by the Danish competition authority in the infringement decision at issue in the main proceedings.
(a) Observations of the parties to the preliminary ruling proceedings
39. On the one hand, in Ernst & Young’s view, the merger control rules, in particular the standstill obligation, only apply to the subset of measures forming part of a notifiable concentration. The possible effect on competition of measures falling outwith the scope of the standstill obligation must, instead, be assessed under Articles 101 and 102 TFEU. Referring to the judgment of the General Court in Aer Lingus Group v Commission, (11) Ernst & Young argues that the concept of ‘concentration’ is the cornerstone of the Commission’s jurisdiction under the merger control rules, and the standstill obligation therefore cannot have a scope which differs from the scope of that concept or which indeed goes beyond the scope of that concept. Measures which are caught by the standstill obligation all relate, in Ernst & Young’s view, to a transfer of control.
40. On the other hand, referring to the judgment of the General Court in Electrabel v Commission, (12) the Danish Government and the Commission emphasise the system of ex ante control which the merger rules provide, and that in case a merger is refused, it may prove impossible to restore the situation existing prior to the implementation.
41. In particular, the Danish Government argues that the standstill obligation cannot be restricted to measures which, in themselves, entail an actual change of control or a merger of the parties’ activities, but must include any measures which, given the circumstances, are apt to restrict or render more difficult the effective ex ante merger control. However, that government considers that internal preparatory steps ought to be excluded from the scope of the standstill obligation if they are not relevant to the competition authorities’ approval of the merger and are not otherwise liable to have structural market effects.
42. Similarly, the Commission, referring to recital 34 of Regulation No 139/2004, (13) argues that it is not a prerequisite of a measure being held to constitute implementation of a concentration under Article 7(1) of that regulation that that measure forms, in whole or in part, in law or in fact, part of the process leading to the actual change of control. The Commission considers that a partial implementation of a merger may, inter alia, arise in respect of measures which (i) consist of preparatory steps in the course of a procedure leading to a change of control; or (ii) allow the party obtaining control to gain influence over the structure or market behaviour of the target undertaking; or (iii) otherwise pre-empt the effects of the merger or significantly affect the prevailing competitive situation.
(b) Introductory remarks
43. To my knowledge, the Court has to date not specifically ruled on the scope of the standstill obligation in Article 7(1) of Regulation No 139/2004 nor, by way of consequence, on the Commission’s powers to monitor compliance therewith. That is noteworthy, as the fines imposed by the Commission have, as of late, in no way been insignificant. (14) That lack of judicial review seems to have allowed the Commission to continue its regulatory activities unchecked. The Commission also proposes in the present case not to establish any criteria for delimiting its monitoring powers.
44. In general, I agree with the Commission that it would not be effective for the Court to set out in detail a general and exhaustive list of criteria with the aim of capturing all the possible measures that could potentially be caught by the standstill obligation. Characterising thus, in positive terms, the standstill obligation would run the risk of excluding certain measures, thereby potentially prejudging the outcome of future cases, to the detriment of both the Commission’s activity as regulator and the Court’s review thereof. Indeed, if the Court were to approve the use of certain criteria to positively demarcate the scope of the standstill obligation, it might even narrow the scope of that obligation if those criteria were to be applied systematically and so as to be a priori determinative.
45. Rather, a negative definition of the standstill obligation is to be preferred. That requires the Court to provide a definition of those measures that will not be caught by the obligation, thereby creating enhanced legal certainty for the undertakings concerned while at the same time retaining the flexibility necessary for effective merger control.
46. Before further elaborating on this, I will first illustrate why the criteria submitted by the referring court do not lend themselves to delimiting the scope of the standstill obligation, as all three suggested criteria contain inherent flaws. In a further analysis, I will then set out the considerations to be taken into account in delimiting the scope of the standstill obligation.
(c) The three criteria of the national decision
47. The referring court submitted three criteria, used by the Danish competition authority in the decision at issue in the main proceedings, with a view to clarifying what types of measure will be caught by the standstill obligation. Accordingly, the measure in question must (i) be merger-specific; (ii) be irreversible; and (iii) potentially create market effects.
48. Merger-specificity is a prerequisite for, not a criterion of application of the standstill obligation. It is therefore not merely lawful to have recourse to that factor. It is mandatory. (15)
49. A criterion of merger-specificity might make it easier to exclude certain measures which patently have nothing to do with the implementation of a concentration. In reality, however, it does not provide any added value as without it the Commission’s powers under Regulation No 139/2004 would not come into play. (16)
50. Second, whether the measure allegedly pre-implementing a concentration is irreversible does not strike me as relevant to the standstill obligation. Attaching importance to the irreversibility of the measure itself would, in the matter under consideration, not only involve speculation as to whether KPMG DK might be able to renew its bonds with KPMG International but would also disconnect the standstill obligation from that which it seeks temporarily to suspend, namely, the concentration.
51. More importantly, under Article 8(5)(a) and (c) of Regulation No 139/2004, the Commission may order the reversal of measures which implement a concentration prematurely. It would seem contradictory for the Commission to have that power if application of the standstill obligation depended on the irreversibility of a measure.
52. Rather, the irreversibility of a measure may be a relevant factor for the competition authority when considering a request for derogation from the standstill obligation under Article 7(3) of Regulation No 139/2004. (17)
53. Last, as regards the potential to have market effects, I consider that criterion, too, to be of no value in determining the scope of the standstill obligation.
54. Indeed, in the first place, commercial measures almost invariably have some effect on the market. If a mere potential to have market effects sufficed to trigger the standstill obligation, that criterion would almost systematically be fulfilled and, thus, be an otiose criterion. Conversely, if the criterion proposed were instead one of actual market effects, the scope of the standstill obligation might be too restricted.
55. In the second place, a market effects criterion would essentially overlap with the merits of the request to approve the concentration. If such a criterion were relevant, it would in reality make the standstill obligation reminiscent of a type of automatic injunction procedure based on whether the measure allegedly implementing the concentration might have a market effect. Such an approach would risk allowing the standstill obligation to pre-empt the assessment of the concentration’s compatibility with the common market. It would also render superfluous the power of the Commission, under Article 8(5) of Regulation No 139/2004, to take appropriate interim measures to restore or maintain conditions of effective competition where the standstill obligation has been infringed.
56. In the last place, making the standstill obligation dependent on a (potential) market effect disregards the complexity of such an economic assessment which cannot with certainty be made by undertakings. This is why Article 7(3) of Regulation No 139/2004 enables the Commission to grant derogations from the standstill obligation where the threat to competition posed by the measure in question is not grave.
57. In sum, none of the criteria suggested seem of use in determining the scope of the standstill obligation. Indeed, the three criteria submitted by the referring court are all good examples of why the Court ought not to involve itself in a positive enumeration of the relevant criteria leading to a breach of the standstill obligation.
58. In what follows, I will elaborate on why a negative definition is better suited for delimiting the scope of the obligation in Article 7(1) of Regulation No 139/2004.
(d) The scope of the standstill obligation under Article 7(1) of Regulation No 139/2004
59. Under Article 1(1) of Regulation No 139/2004, that regulation, including the obligation to suspend a concentration under Article 7(1) thereof, applies to concentrations with an EU dimension (or, in the case of national legislation which mirrors that regulation, the applicable threshold). Whereas the issue of whether a concentration has an EU dimension finds an answer in Article 1(2) of the regulation, the concept of ‘concentration’ itself is governed by Article 3.
60. Specifically, a concentration presupposes, under Article 3(1) of Regulation No 139/2004, ‘a change of control on a lasting basis’ which results from a merger or an acquisition. ‘Control’ itself is constituted, under Article 3(2) thereof, ‘by rights, contracts or any other means which, either separately or in combination and having regard to the considerations of fact or law involved, confer the possibility of exercising decisive influence on an undertaking’. ‘Control’ is, moreover, under Article 3(3) thereof, something which is ‘acquired’. Lastly, specifically as regards acquisitions, under Article 3(1)(b), control may be ‘direct or indirect’.
61. Article 7(1) of Regulation No 139/2004, for its part, refers to a ‘concentration’ with an EU dimension, as defined in Article 1 thereof, which may not be ‘implemented’ before its notification or until it has been approved.
62. Hence, the concept of a ‘concentration’ is crucial to the standstill obligation. It is that concept or, in fuller terms, the acquisition of the possibility of exercising decisive influence on a (target) undertaking, which gives rise to that obligation. (18) Accordingly, measures preceding a concentration should not fall within the scope of the standstill obligation.
63. At the same time, as submitted by the Commission, the standstill obligation must cover partial as well as full implementation of a concentration. While the obligation cannot apply to merely internal preparatory measures preceding a concentration, (19) measures that are intrinsic to a concentration must be caught. The difficulty lies in drawing the line between legitimate preparatory measures and partial implementation, as exemplified by the present case.
64. In that regard, the General Court has held that the Commission may, in certain circumstances, consider a group of interrelated transactions as one single merger operation where the transactions at issue are interdependent in such a way that none of them would be carried out without the others and that the result consists in conferring on one or more undertakings direct or indirect economic control over the activities of one or more other undertakings. (20) Such a staggered or bundled approach in dealing with the concept of a ‘concentration’ finds support in the wording of Article 3(1)(a) of Regulation No 139/2004, which refers to indirect (de facto) control. (21) That approach is justified in order to prevent circumvention of the merger rules by artificially creating interrelated transactions. Whether or not the merger rules are artificially circumvented must, however, be considered on a case-by-case basis. The mere fact that several transactions might be interdependent does not necessarily mean that they constitute a single concentration. (22)
65. Conversely, simply because a measure was taken in the process leading up to a concentration cannot automatically bring it within the purview of the standstill obligation. As long as the measure precedes and is severable from the measures actually leading to the acquisition of the possibility of exercising decisive influence on a target undertaking, it should not be subject to the standstill obligation and, consequently, the Commission’s powers of compliance therewith.
66. Against such a view, it might be argued that the standstill obligation ought, for reasons of efficiency, not to be interpreted narrowly, and that its scope, as the compound product of a ‘concentration’ and an ‘implementation’, ought even, as a precaution, to be wider than the scope of the concept of ‘concentration’. That position seems inherent in the arguments submitted by the Danish Government and the Commission. The judgment in Aer Lingus Group v Commission (23) lends some support to that view. The General Court held that ‘the acquisition of a [minority shareholding] which does not, as such, confer control for the purposes of Article 3 of [Regulation No 139/2004] may fall within the scope of Article 7’, thereby seemingly detaching the standstill obligation from the concept of a ‘concentration’.
67. However, to look at the context of that finding, in paragraph 82 of that judgment the General Court analysed the derogation from the standstill obligation set out in Article 7(2) of Regulation No 139/2004, and not that set out in Article 7(1). Then, in paragraph 83, it made the general statement reproduced above, without giving any explanation other than a summary of its understanding of the Commission’s approach. The ambit of that statement is therefore somewhat ambiguous. In any event, that statement is not compatible with the subsequent conclusion reached, in paragraphs 84 and 85 of that judgment, in relation to the circumstances of that case — with which I agree — that the acquisition of a minority shareholding cannot, generally speaking, be regarded as a partial implementation of a concentration capable of giving rise to a measure adopted on the basis of Article 8(4) and (5) of that regulation. (24)
68. I would argue, moreover, that an interpretation of the scope of the standstill obligation under Article 7 of Regulation No 139/2004 which is disconnected from that of the concept of ‘concentration’ would create a grey area where certain measures which, although within the periphery of a concentration but not in and of themselves inextricably linked to the transfer of control, would be caught thereby. However, if the reach of that grey area were greater than the scope of the concept of a ‘concentration’, it would imply extending that obligation beyond the scope of the regulation itself as expressed in Article 1 thereof. That would be unsustainable.
69. Such a broad interpretation would, moreover, go beyond what is necessary in order to maintain an effective merger control. As mentioned above in points 33 to 36, it would seem excessive to consider the standstill obligation to be an indispensable tool. Indeed, the Commission has powers under Article 8(4) and (5) of Regulation No 139/2004 to order the restoration or maintenance of conditions of effective competition, in particular, to order the dissolution of a concentration. The risk alone of having to pay the costs associated with such an order, for merging undertakings, is certainly dissuasive. Requiring merging undertakings to wait until clearance is given even in respect of measures which are not in themselves inextricably linked to the transfer of control would be excessive and might cause unnecessary delays.
70. In addition, Regulation No 139/2004 expressly sets out under which circumstances it is to be applicable even prior to the actual emergence of a concentration. For instance, under Article 4(1) of Regulation No 139/2004, a concentration normally presupposes the conclusion of an agreement, the announcement of a public bid, or the acquisition of a controlling interest. However, the second subparagraph of Article 4(1) allows parties to notify a concentration even before it has arisen if the intention to merge is unequivocal, such as a public announcement of a bid which would result in a concentration (‘intended concentrations’). The third subparagraph of Article 4(1) of the regulation then sets out the specific circumstances when Regulation No 139/2004 is to apply even though no concentration has arisen yet, and distinguishes in that regard between ‘notified concentrations’ and ‘concentrations’ (the inverted commas appearing in the provision itself). Pursuant to the first sentence of the third subparagraph of Article 4(1), intended concentrations amount to a ‘notified concentration’ for the purpose of the regulation. Under the second sentence of that subparagraph, an intended concentration is also to be equivalent to a ‘concentration’, but only as regards the procedure set out in Article 4(4) and (5) of the regulation.
71. Against that backdrop, and given the fact that the term ‘notified concentration’ does not appear in Article 7 of Regulation 139/2004, (25) I am led to conclude that Article 7 — and hence the standstill obligation — does not apply as regards intended concentrations. That confirms that the EU legislature did not want the standstill obligation to apply where a concentration has not yet materialised.
72. That is not to say that the term ‘implemented’, as used in Article 7(1) of Regulation No 139/2004, ought not to carry a broad meaning, thereby ensuring that partial implementations can be monitored by the Commission. (26) Be that as it may, a concentration can necessarily only be ‘implemented’ if it exists.
73. The arguments submitted by the Danish Government and the Commission to justify the idea that application of the standstill obligation should not be linked in any way to the concept of ‘concentration’ leave me unpersuaded.
74. First, linking the standstill obligation with the concept of ‘concentration’ in no way undermines the system of ex ante review of concentrations set up under Regulation No 139/2004.
75. Second, the fact that the Commission might, in its administrative practice, have adopted a broad approach as to what falls to be classified as premature implementation of a concentration is irrelevant. The Commission’s practice does not bind the Court.
76. Third, a distinction between internal preparatory measures and external preparatory measures, as essentially advocated by the Danish Government, has no basis in Regulation No 139/2004. That regulation instead refers to ‘a change of control on a lasting basis’, regardless of whether that change is the result of internal or external measures.
77. Fourth, the findings of the General Court in Electrabel v Commission, (27) cited by the Danish Government and the Commission, concern, as essentially noted by Ernst & Young, the gravity of the infringement of the standstill obligation and the proportionate nature of the fine imposed in consequence thereof, and not the existence as such of such an infringement.
78. On the basis of the foregoing considerations, I am of the opinion that the obligation to suspend a concentration set out in Article 7(1) of Regulation No 139/2004 does not affect measures which, although taken in connection with the process leading to a concentration, precede and are severable from the measures actually leading to the acquisition of the possibility of exercising decisive influence on a target undertaking.
(e) Application to the circumstances of the main proceedings
79. By the second question referred, the referring court wishes the Court to state its position on the circumstances of the case pending before it.
80. In preliminary ruling proceedings, the application of the interpretation of EU law given by the Court to the facts is a matter for the national courts, which are in a better position than the Court to establish and weigh all the relevant circumstances. Still, the Court may provide the national court with all the elements of interpretation which may be of assistance in adjudicating on the case pending before it. (28)
81. In the main proceedings, on the one hand, the merger agreement itself was subject to approval by the Competition Council and was not to be implemented before that approval was given.
82. On the other hand, as stated in the request for a preliminary ruling, the termination of the cooperation agreement between KPMG DK and KPMG International was not subject to the same requirement of approval, and notice was given before receipt of the merger approval.
83. The contested decision held that, in doing so, KPMG DK had infringed the standstill obligation. Implicitly at least, it thus appears that the KFST considered that the termination of the cooperation agreement should be classified as a partial implementation of the merger.
84. Plainly, I agree with the Danish Government that the two operations mentioned above in points 81 and 82 were not self-standing, but interrelated. The merger agreement provided that KPMG DK was to terminate its cooperation with KPMG International. Termination of the cooperation agreement was therefore a necessary prerequisite for that merger to take effect.
85. However that does not suffice. The termination did not, in any way, contribute to a shift in control between KPMG DK and EY.
86. Although it can be argued, as the Danish Government does, that the termination of the cooperation agreement was part of the merger agreement, it was not inextricably linked to the transfer of control, giving EY the possibility of exercising decisive influence on KPMG DK, as defined in Article 3(2) of Regulation No 139/2004. (29) The effect of that termination was simply that KPMG DK was no longer part of the KPMG network and would revert to the status of independent trader in the market for accountancy services. Although the termination might have had some effect on the market, it would not have meant that KPMG DK would no longer have been a competitor for EY. (30)
87. Indeed, accepting such a line of argument would mean that any measure implemented by EY or KPMG DK between signing the merger agreement and the competition authority’s approval potentially could have been caught by the standstill obligation. However, for the reasons given above in point 71, the mere intention to effect a concentration does not give rise to the standstill obligation.
88. Therefore, a preparatory measure of that type does not, in my view, fall to be classified as a premature implementation of a concentration.
3. The third and fourth questions referred
89. By its third and fourth questions, the referring court essentially wishes to know whether it is of relevance to the standstill obligation that the measure claimed to be a premature implementation of a concentration had any market effects and, if so, which criteria may be used to show such an effect and the evidentiary threshold which must be satisfied in that regard by the competition authority when investigating an alleged infringement of that obligation.
90. Those questions are asked since the contested decision recognised that while it was not feasible for the KFST to determine with certainty whether the termination of the cooperation agreement caused some of KPMG DK’s clients to recommend a change of auditor at their general meetings, it held that it was only necessary to show a potential to have market effects. Ernst & Young counters that those recommendations were simply issued following the regular practice of large Danish companies when selecting or changing auditors, and that the KFST’s assessment of the merger was made at the point of time of the year where a change of auditor normally takes place.
91. From my answer to the first two questions and, in particular, the criterion relating to the potential of the measure under scrutiny to have market effects (see points 53 to 56 above) it follows that, in my opinion, the possible market effects of a measure are of no relevance for application of the standstill obligation. Rather, that obligation must be complied with at the same time as, and in so far as, a concentration which meets the applicable thresholds arises.
92. Should the Court take a view which differs from mine, I shall limit myself to the following brief observations in the alternative.
93. The Court has held that, in the field of merger control, the prospective analysis called for, which consists of an examination of how a concentration might alter the factors determining the state of competition on a given market in order to establish whether it would give rise to a significant impediment to effective competition, makes it necessary to envisage various chains of cause and effect with a view to ascertaining which of them is the most likely. That standard applies both as regards decisions authorising a concentration and decisions prohibiting it. (31)
94. Should the Court, unlike me, consider the criterion of potential market effects to be relevant, notwithstanding the overlap with the substantive assessment of the compatibility of the concentration with the common market, I see no obvious reason why the standard of review, mentioned in the previous point, ought not to apply — mutatis mutandis — to decisions finding implementation. The assessment to be performed would be reminiscent of that concerning requests for derogation from the standstill obligation under Article 7(3) of Regulation No 139/2004, which require, inter alia, taking account of the threat to competition posed by the concentration.
95. In accordance with ordinary rules of evidence, it is for the Commission (or the relevant competition authority) to prove the alleged infringement. (32) That authority must therefore show, on the basis of an assessment of different hypotheses involving various chains of cause and effect, that the hypothesis in which the measure at issue is capable of giving rise to a significant impediment to effective competition is the one which is the most likely. (33)
96. As regards the type of evidence which may be relied on by the Commission (or the relevant competition authority), the principle which prevails in EU law is that of the unfettered evaluation of evidence. Specifically, as long as the evidence has been lawfully adduced, the only relevant criterion for the purpose of assessing its probative value relates to its credibility. (34) In that regard, there is no requirement to use a particular method or test. (35)
97. It is for the referring court to determine whether the Danish competition authorities have shown, on the basis of an assessment of different hypotheses involving various chains of cause and effect, that the hypothesis in which the termination of the cooperation agreement was capable of giving rise to a significant impediment to effective competition is the one which is the most likely. In concrete terms, it must therefore be deemed more plausible than the line of argument submitted by Ernst & Young concerning the typical behaviour of large Danish companies.
IV. Conclusion
98. On the basis of the foregoing, I propose that the Court should answer the questions referred by the Sø- og Handelsretten (Maritime and Commercial Court, Denmark) to the effect that, on a proper interpretation, the obligation to suspend a concentration set out in Article 7(1) of Council Regulation (EC) No 139/2004 of 20 January 2004 on the control of concentrations between undertakings (the EC Merger Regulation) does not affect measures which, although taken in connection with the process leading to a concentration, precede and are severable from the measures actually leading to the acquisition of the possibility of exercising decisive influence on a target undertaking.