OPINION OF ADVOCATE GENERAL

EMILIOU

delivered on 21 March 2024 (1)

Joined Cases C‑611/22 P and C‑625/22 P

Illumina, Inc.

v

European Commission (C‑611/22 P)

and

Grail LLC

v

Illumina, Inc.,

European Commission (C‑625/22 P)

(Appeal – Competition – Concentrations between undertakings – Article 22 of Regulation (EC) No 139/2004 – Concentrations that do not have a Community dimension – Referral request from a competition authority not having jurisdiction under national law – Commission decision to examine the concentration – Competence of the Commission – Time limit for submitting the referral request – Obligation to act within a reasonable time – Principle of good administration – Right of defence – Legitimate expectations)

I.      Introduction

1.        Most modern anti-trust laws, both within the European Union and elsewhere, are built on a trifecta of provisions: rules on agreements and concerned practices, rules on unilateral conduct (or abuses of dominance), and rules on merger control.

2.        The peculiarity of rules of merger control lies in the fact that, unlike the other two sets of rules, they generally require the competent (administrative and/or judicial) authorities to engage in an ex ante, as opposed to an ex post, form of review: whether a proposed concentration could, if implemented, result in significant harm to effective competition. It is a particularly complex and laborious technical evaluation, ‘based not on the application of precise scientific rules but on criteria and principles which are open to question’, aimed at ‘predicting the effects of the concentration on the structure and competitive dynamics of the markets concerned, taking into consideration the many constantly evolving factors which may impinge on the future development of supply and demand on those markets’. (2)

3.        Nevertheless, that assessment must be done in the shortest possible timeframe. Indeed, in order to preserve the effectiveness of the system, most legal regimes – including that of the European Union – require the undertakings concerned to notify the transaction to the competent authorities and to suspend its implementation until they receive clearance from those authorities. Notification and suspension entail significant costs and engender some risks for the undertakings involved.

4.        Against that backdrop, the legislature’s choice of the type of thresholds and the setting of the relative amounts which, when met, trigger the notification and suspension obligations for the merging parties is of crucial importance for the proper functioning of the system. Those thresholds pursue a dual function: to ensure a ‘local nexus’ which justifies the intervention of the authorities in question, and to filter the transactions which are potentially of interest. Ideally, thresholds should be easy to calculate (to avoid uncertainties as to whether a given transaction has to be notified) and set at a level that minimises, at the same time, the number of transactions unlikely to raise competitive concerns that are caught by the system, and those that are capable of raising such concerns that fall outside it. (3)

5.        The EU system of merger control – governed by Council Regulation (EC) No 139/2004 of 20 January 2004 on the control of concentrations between undertakings (the [EU] Merger Regulation) (‘the EUMR’) – (4) is primarily based on the turnover of the merging companies. However, there are some provisions in that regulation that, by way of exception, empower the European Commission to review mergers not meeting the turnover thresholds in question, when cases are referred to that institution by the Member States’ authorities and, as the case may be, after being invited to do so by the Commission. The present case turns mainly on defining the meaning and scope of one of those provisions: Article 22 EUMR. In a nutshell, the key issue in the present proceedings is the following: does that provision enable the Commission to review a merger referred to it by a Member State’s authorities, where the latter lack any competence to review it, since the merger in question falls below the thresholds set out in their national legislation on merger control?

6.        Despite the apparent simplicity of the question, finding the correct answer is by no means a straightforward exercise. It requires from the interpreter a meticulous hermeneutic analysis to determine the proper construction of Article 22 EUMR. To do so, it is necessary not only to examine the wording, origin, context and purpose of that provision, but also to take into account the logic of the EU system of merger control as well as some fundamental principles of EU law (such as institutional balance, subsidiarity, legal certainty, territoriality, etc.). Last but not least, it is hard to overemphasise the importance that the answer to that question may have on the correct and effective function of the EU system of merger control.

II.    European Union law

7.        Article 22 EUMR, entitled ‘Referral to the Commission’, provides:

‘1.      One or more Member States may request the Commission to examine any concentration as defined in Article 3 that does not have a Community dimension within the meaning of Article 1 but affects trade between Member States and threatens to significantly affect competition within the territory of the Member State or States making the request.

Such a request shall be made at most within 15 working days of the date on which the concentration was notified, or if no notification is required, otherwise made known to the Member State concerned.

2.      The Commission shall inform the competent authorities of the Member States and the undertakings concerned of any request received pursuant to paragraph 1 without delay.

Any other Member State shall have the right to join the initial request within a period of 15 working days of being informed by the Commission of the initial request.

…

3.      The Commission may, at the latest 10 working days after the expiry of the period set in paragraph 2, decide to examine, the concentration where it considers that it affects trade between Member States and threatens to significantly affect competition within the territory of the Member State or States making the request. If the Commission does not take a decision within this period, it shall be deemed to have adopted a decision to examine the concentration in accordance with the request.

The Commission shall inform all Member States and the undertakings concerned of its decision. It may request the submission of a notification pursuant to Article 4.

The Member State or States having made the request shall no longer apply their national legislation on competition to the concentration.

4.      Article 2, Article 4(2) to (3), Articles 5, 6, and 8 to 21 shall apply where the Commission examines a concentration pursuant to paragraph 3. Article 7 shall apply to the extent that the concentration has not been implemented on the date on which the Commission informs the undertakings concerned that a request has been made.

Where a notification pursuant to Article 4 is not required, the period set in Article 10(1) within which proceedings may be initiated shall begin on the working day following that on which the Commission informs the undertakings concerned that it has decided to examine the concentration pursuant to paragraph 3.

5.      The Commission may inform one or several Member States that it considers a concentration fulfils the criteria in paragraph 1. In such cases, the Commission may invite that Member State or those Member States to make a request pursuant to paragraph 1.’

8.        The referral mechanism now set out in Article 22 EUMR was originally established in Article 22(3) to (6) (‘Application of this Regulation’) of the 1989 EC Merger Regulation (5) (‘the ECMR’) which was then amended by Council Regulation (EC) No 1310/97. (6) The ECMR was then repealed by the EUMR with effect from 1 May 2004.

III. Factual background

9.        The most relevant facts, as set out in the judgment in Case T-227/21, Illumina v Commission (‘the judgment under appeal’),(7) can be summarised as follows.

10.      On 20 September 2020, Illumina Inc. – a US-based company marketing sequencing- and array-based solutions for genetic and genomic analysis – entered into an agreement and plan of merger to acquire sole control of Grail LLC (formerly Grail, Inc.), which develops blood tests for the early detection of cancer, in which it already held a 14.5% stake (‘the concentration at issue’). On 21 September 2020, Illumina and Grail (‘the appellants’) issued a press release announcing that concentration.

11.      Since the turnover of the appellants did not exceed the relevant thresholds, in particular given the fact that Grail did not generate any revenue in any EU Member State or elsewhere in the world, the concentration at issue did not have a European dimension for the purpose of Article 1 EUMR and was not therefore notified to the Commission. Nor was the concentration at issue notified in the EU Member States or in States party to the Agreement on the European Economic Area, (8) since it did not fall within the scope of their national merger control rules.

12.      After receiving a complaint relating to the concentration at issue in December 2020, the Commission had some exchanges with the complainant, with a number of Member States’ national competition authorities (‘NCAs’) and with the United Kingdom’s Competition and Markets Authority.

13.      On 19 February 2021, the Commission informed the Member States of the concentration at issue by sending them a letter in accordance with Article 22(5) EUMR (‘the invitation letter’). In that letter, the Commission explained the reasons why it found, prima facie, that the concentration appeared to satisfy the conditions laid down in Article 22(1) EUMR, and invited the Member States to submit a referral request.

14.      In the course of a telephone conversation on 4 March 2021, the Commission informed the legal representative of each of the appellants of the invitation letter, and of the possibility of a referral request under Article 22(1) EUMR.

15.      On 9 March 2021, the Autorité de la concurrence française (French Competition Authority; ‘the ACF’) asked the Commission, pursuant to Article 22(1) EUMR, to examine the concentration at issue (‘the referral request’). On 10 March 2021, the Commission, in accordance with Article 22(2) EUMR, informed the NCAs and the EFTA Surveillance Authority (‘ESA’) of the referral request. On 11 March 2021, the Commission also informed the appellants of the referral request, stating that the concentration at issue could not be implemented provided that, and in so far as, the standstill obligation laid down in Article 7 EUMR, read in conjunction with the second sentence of the first subparagraph of Article 22(4) of that regulation, was applicable (‘the information letter’).

16.      On 16 and 29 March 2021, the appellants submitted observations to the Commission opposing the referral request. On 2, 7 and 12 April 2021, Illumina responded to the requests for information that the Commission had sent to it on 26 March and 8 April 2021.

17.      By letters dated 24, 26 and 31 March 2021, the Belgian, Greek, Icelandic, Dutch and Norwegian competition authorities requested to join the referral request, pursuant to Article 22(2) EUMR (‘the requests to join’).

18.      On 31 March 2021, the Commission published a communication entitled ‘Guidance on the application of the referral mechanism set out in Article 22 [EUMR] to certain categories of cases’. (9)

19.      By decisions of 19 April 2021, the Commission accepted the referral request and the requests to join. By those decisions, the Commission (i) found that the referral request had been submitted within the time limit of 15 working days laid down in Article 22(1) EUMR; (ii) found that the requests to join complied with the time limit laid down in Article 22(2) EUMR; (iii) found that the concentration at issue satisfied the criteria for referral under Article 22(1) EUMR; and (iv) rejected as unfounded the arguments of the appellants concerning an alleged breach of their rights of defence and other general principles of EU law.

IV.    The proceedings before the General Court, the judgment under appeal and the proceedings before the Court of Justice

20.      By application lodged on 28 April 2021, Illumina applied to the General Court, under Article 263 TFEU, seeking the annulment of the information letter, the decision to accept the referral from the ACF and the decisions accepting the request to join (‘the contested decisions’).

21.      By orders and decisions of the President of the Third Chamber (Extended Composition) of the General Court (i) Grail was granted leave to intervene in support of the form of order sought by Illumina, (ii) the Hellenic Republic, the French Republic, the Kingdom of the Netherlands and ESA were granted leave to intervene in support of the form of order sought by the Commission, and (iii) the request lodged by the Computer & Communications Industry Association to intervene in support of the form of order sought by Illumina was dismissed.

22.      Illumina, supported by Grail, asked the General Court to annul the contested decisions and the information letter and to order the Commission to pay the costs. For its part, the Commission, supported by the Hellenic Republic, the French Republic, the Kingdom of the Netherlands and ESA, asked the General Court to dismiss the action as inadmissible or, in the alternative, as in part inadmissible and in part unfounded, and to order Illumina to pay the costs.

23.      On 13 July 2022, by the judgment under appeal, the General Court dismissed the action, ordered Illumina to bear its own costs and to pay those incurred by the Commission, and ordered the Hellenic Republic, the French Republic, the Kingdom of the Netherlands, ESA and Grail to bear their own costs.

24.      In their appeals before the Court of Justice, lodged on 22 and 30 September 2022, respectively, Illumina (Case C‑611/22 P) and Grail (Case C‑625/22 P) asked the Court to set aside the judgment under appeal, annul the contested decisions and order the Commission to pay the costs of the proceedings. Grail also asked the Court to annul the ACF’s request and the Commission’s information letter.

25.      On 21 December 2022, the President of the Court decided, after hearing the Judge-Rapporteur, the Advocate General and the parties, to join the two cases for the purposes of the oral part of the procedure and the judgment, in conformity with Article 54(2) of the Rules of Procedure of the Court of Justice (‘the RoP’). By decision of 10 January 2023, the President of the Court also decided, after hearing the Judge-Rapporteur and the Advocate General, to dismiss the Commission’s request that Case C‑625/22 P be dealt with under the expedited procedure provided for in Articles 133 to 136 of the RoP and that the case be dealt with as a priority, pursuant to Article 53(3) of the RoP.

26.      By two orders of the President of the Court of 10 March 2023, Biocom California was granted leave to intervene in support of the form of order sought by Illumina in Case C‑611/22 P, and the applications for leave to intervene in support of the form of order sought by Grail in Case C‑625/22 P, submitted by the Association Française des Juristes d’Entreprise (AFJE) and the European Company Lawyers Association (ECLA), were dismissed.

27.      In their responses, the Commission, the French Republic, the Kingdom of the Netherlands and ESA asked the Court to dismiss the appeals and order the appellants to pay the costs. For their part, Grail submitted a response in Case C‑611/22 P, and Illumina submitted a response in Case C‑625/22 P each asking the Court to set aside the judgment under appeal, annul the contested decisions and order the Commission to pay the costs.

28.      The appellants submitted a reply and the respondents submitted a rejoinder. The appellants, the respondents and the interveners presented their views at the hearing before the Court that was held on 12 December 2023.

V.      Assessment

29.      In support of their appeals, each of the appellants relies on three grounds of appeal, which largely overlap. I shall, thus, examine those grounds jointly.

30.      Accordingly, I shall first assess whether the General Court erred in its interpretation of the meaning and scope of the first paragraph of Article 22(1) EUMR (A). Second, I will turn to the applicants’ allegations that the referral request was made out of time and that the Commission breached its obligation to act within a reasonable time (B). Third and final, I will deal with the alleged breaches of the principles of legitimate expectations and legal certainty (C).

A.      First ground: the meaning and scope of Article 22(1) EUMR

31.      Illumina’s and Grail’s first ground of appeal concern paragraphs 85 to 185 of the judgment under appeal. In those passages, the General Court dismissed Illumina’s first plea at first instance, alleging a lack of competence on the part of the Commission to review the concentration at issue. In particular, after considering the parties’ arguments, the General Court came to the following conclusion:

‘183       … taking account of the literal, historical, contextual and teleological interpretations of Article 22 [EUMR], it must be held that the Member States may, under the conditions set out in therein, make a referral request under that provision irrespective of the scope of their national merger control rules.

184      Accordingly, the Commission was right, by the contested decisions, to accept the referral request and the requests to join under Article 22 [EUMR]. …’

1.      Arguments of the parties

32.      Illumina argues that, in endorsing the application of Article 22(1) EUMR made by the Commission, the General Court erred in interpreting that provision. In particular, Illumina submits that the General Court has failed to (i) apply a number of fundamental principles of EU law (such as legal certainty, proportionality and subsidiarity); (ii) properly identify and consider the object of the EUMR; (iii) interpret strictly a provision which constitutes a derogation from a general rule; and (iv) recognise the importance of the context and object of the provision at issue. Similarly, Grail takes the view that a textual, historical, contextual and teleological interpretation of Article 22(1) EUMR did not support the General Court’s reading thereof.

33.      In essence, Biocom supports the arguments put forward by the appellants, emphasising the legal insecurity and the disproportionate burden on the merging parties that result from the judgment under appeal.

34.      The Commission argues that the appellants’ first grounds of appeal are inoperative, inadmissible in so far as they rely on certain preparatory documents and, in the alternative, unfounded. The Commission is of the view that the General Court has correctly interpreted Article 22(1) EUMR. In particular, the Commission contends that the appellants (i) fail to give proper consideration to the clear wording of that provision and (ii) err in considering that the interpretation retained by the General Court would result in the EUMR system not providing adequate legal certainty to the merging parties.

35.      The French and Netherlands Governments and ESA share the Commission’s views. In particular, the French Government argues that the General Court has correctly applied the principles of legal certainty, proportionality and subsidiarity. The Netherlands Government maintains that, under Article 22(1) EUMR, it had the right to request the Commission to examine a concentration such as that at issue or to join a request made by another NCA. For its part, ESA submits that the appellants err in relying on the one-shop-stop system established by the EUMR: that mechanism only concerns mergers with a Community dimension, whereas it is not applicable in relation to mergers that do not have such a dimension.

2.      Analysis

36.      In the following pages, I shall first assess some preliminary objections of a procedural nature raised by the Commission, before turning to the merits of the appellants’ first grounds of appeal.

(a)    Preliminary issues

37.      At the outset, it is appropriate to deal with the Commission’s arguments according to which (i) the appellants’ first grounds of appeal are ineffective, and (ii) Grail relies on certain documents which are inadmissible.

38.      I am not convinced by those arguments.

39.      First, a ground of appeal is ineffective when, even if it were held to be well founded, it would not be capable of leading to the judgment under appeal being set aside. (10) That is, quite clearly, not the case of the grounds of appeal being considered here. It is common ground that if the General Court has – as argued by the appellants – wrongly interpreted the nature and scope of Article 22 EUMR, with the consequence that the Commission could not examine the concentration at issue, the judgment under appeal would be vitiated by an error of law that would lead to the setting aside of that judgment and the annulment of the contested decisions.

40.      The Commission’s assertion that the appellants have not challenged the findings made by the General Court in certain passages of the judgment under appeal (paragraphs 90 to 94 thereof as regards Illumina, and paragraphs 183 and 184 thereof as regards Grail) is contradicted by the text of the appeals. In fact, the Commission’s criticism appears to relate rather to the strength of the arguments put forward by the appellants to contest the General Court’s findings in those passages. However, that is an issue which concerns the merits of the ground of appeal, not its allegedly ineffective character.

41.      Second, the Commission’s claim of alleged inadmissibility of certain documents relied on by Grail with regard to the historical interpretation of Article 22 EUMR (‘the contested documents’) is equally unfounded. The Commission essentially argues that such documents should have been first produced before the General Court in order to then be admissible on appeal before the Court of Justice. To that end, the Commission relies on the order of the President of the Court of 10 October 2023, Deutsche Lufthansa v Ryanair and Others. (11)

42.      However, a general requirement that documents must be first produced before the General Court, in order to then be admissible on appeal before the Court of Justice, is neither set out in the RoP, nor does it follow from the case-law of the EU Courts. It could not have been otherwise: such a rule would be entirely unreasonable and counterproductive. I hardly need to point out, in that regard, that actions for annulment and appeal proceedings have a different object (a decision the former, a judgment the latter), and the legal issues on which the two courts are called upon to rule may thus not fully coincide.

43.      More fundamentally, such a rule would be at odds with the principles governing the production of evidence before the EU Courts. The Court of Justice has consistently held that ‘the principle of equality of arms, which is a corollary of the very concept of a fair hearing, guaranteed in particular by Article 47 of the Charter of Fundamental Rights of the European Union [(“Charter”)], requires that each party must be afforded a reasonable opportunity to present his or her case, including his or her evidence, under conditions that do not place him or her at a substantial disadvantage vis-à-vis his or her opponent’. (12) With regard to the production of evidence, the basic rule is that any evidence can be submitted before the EU Courts. However, those courts may take into account the existence of any (intra- or extra-judicial) interest which may, by way of exception, justify the refusal to accept the evidence and balance those interests against those that plead for its acceptance. (13) That may be the case, for example, where a document has been obtained illegally or contains confidential information which should not be publicly disclosed in order to protect certain public or private interests.

44.      In the present case, the contested documents have been lawfully obtained by Grail following requests for access to documents lodged pursuant to Regulation (EC) No 1049/2001, (14)and criticise some specific passages of the judgment under appeal. Given that those passages concern one of the issues which is key to the present case (whether or not the General Court’s reading of Article 22 EUMR is supported by a historical interpretation thereof), I see no plausible reason as to why the appellants should not be allowed to rely on the contested documents. In fact, were those documents to be ruled inadmissible, the appellants would de facto be deprived of the opportunity to challenge the General Court’s findings in paragraphs 69 to 117 of the judgment under appeal. That would be against the right to an effective remedy and to a fair trial enshrined in Article 47 of the Charter.

45.      By the same token, the Commission’s suggestion that the Court would be barred from examining documents lawfully submitted by a party is clearly untenable. As the Court has stated, in that regard, ‘the applicable principle in EU law is that of the unfettered evaluation of evidence’, (15) and ‘it is only the reliability of the evidence before the Court which is decisive when it comes to the assessment of its value’. (16)

46.      The order of the President relied on by the Commission is immaterial in this context. That case concerned a company’s application to the Court for confidential treatment, vis-à-vis the other parties in the proceedings, of certain pieces of information included in the body of and in an annex to its appeal. Importantly, the information for which confidential treatment was sought had been produced at first instance but was then removed from the case file because it was considered to be irrelevant by the General Court. Consequently, that information did not benefit from confidential treatment at first instance, since the General Court had removed it from the file without proceeding to weigh its confidentiality against the requirements linked to the right to effective judicial protection, as laid down in Article 103(2) of its Rules of Procedure. On that basis, the President dismissed the company’s request for confidential treatment, emphasising that, since the information at issue was not included in the file that served as the basis for the General Court’s ruling, that information could not, in principle, be relevant for the review by the Court of Justice of the lawfulness of that ruling at the appeal stage. There was, thus, no reason to grant confidential treatment, on appeal, to information which the appellant had voluntarily disclosed in its submissions.

47.      That order is a straightforward application of the basic principles according to which an appeal before the Court lies on points of law only, and the subject matter of those proceedings is confined to that at first instance and may not be changed in the appeal. (17) However, unlike that case, the present case does concern (i) a point of law (interpretation of Article 22 EUMR) and not the establishment of controversial facts, and (ii) an issue which was raised and discussed at first instance and upon which the General Court has ruled.

48.      It certainly does not follow from that order that, in order to challenge a crucial passage of a judgment under appeal, an appellant must have submitted the relevant evidence already at first instance. Nor could that order be read as implying that the correct interpretation of the law is an issue which is for the appellant to prove to the requisite standard, let alone having to do it at first instance. That would be in obvious contradiction of the well-established iura novit curia principle (18) and of numerous decisions of the Court. (19)

49.      That said, the Commission is correct in stating that, in principle, the appellants’ essential arguments in law must appear in the application itself, and that the documents annexed thereto have merely a supporting role. Thus, whilst the Court is not bound by the interpretation of the law proposed by the parties and, to that end, is at liberty to draw inspiration from any document that has been legitimately brought before it, it cannot be expected to seek and identify in the annexes to the appeals the complaints and the arguments on which those appeals may be based. (20) Accordingly, I will disregard all arguments which are not made explicit in the appeals and that cannot be properly understood without examining the annexes.

(b)    Merits

50.      I shall now turn to the substance of the appellants’ first grounds of appeal. In essence, those grounds raise the issue as to whether the General Court erred in law in its interpretation of Article 22(1) EUMR. As mentioned above, that court came to the conclusion that a ‘literal, historical, contextual and teleological’ interpretation of that provision supported the view that Member States may request the Commission to examine a concentration which does not have a Community dimension, even where they have no competence to review such a concentration under national law. Indeed, the General Court found that Article 22 EUMR pursues different objectives, one of which is ‘permitting effective control, as a “corrective mechanism”, of all concentrations which are capable of significantly impeding effective competition in the internal market and falling outside the scope of the merger control rules between the European Union and the Member States because the turnover thresholds have not been exceeded’. (21)

51.      In the next pages, I shall explain why I believe that the General Court erred in its interpretation of the first subparagraph of Article 22(1) EUMR. Although the arguments based on the wording of the provision, put forward by the Commission and embraced by the General Court, have some force, a number of other interpretative elements – which concern the history, context and objective of the provision as well as of a broader systemic significance – make it quite clear that the meaning and scope of the first subparagraph of Article 22(1) EUMR is not that retained in the judgment under appeal.

(1)    A textual interpretation of the first subparagraph of Article 22(1) EUMR

52.      The analysis must begin with the wording of the first subparagraph of Article 22(1) EUMR which is worth recalling: ‘one or more Member States may request the Commission to examine any concentration … that does not have a Community dimension … but affects trade between Member States and threatens to significantly affect competition within the territory of the Member State or States making the request’.

53.      As the General Court found, that provision (i) lays down certain conditions which must be satisfied for it to apply, among which it is not a requirement that the merger falls within the scope of the national merger control rules; (22) (ii) uses the broad expression ‘any concentration’; (23) and (iii) does not distinguish between Member States that have enacted a national merger control system and those not. (24) In the light of that, the General Court came to the conclusion that, in principle, a literal interpretation of the first subparagraph of Article 22(1) EUMR supported the interpretation proposed by the Commission. However, in so far as the wording of the provision did not allow a definitive conclusion on the point, the General Court considered it appropriate to complete the analysis by making use of other interpretative methods. (25)

54.      I agree with both of those points.

55.      On a prima facie reading of the text of the provision, the General Court’s broad interpretation of the first subparagraph of Article 22(1) EUMR is defendable. The elements listed above may indeed be taken to indicate that all Member States can refer any merger to the Commission, regardless of whether they have a national merger control system and, if so, whether that merger is caught by such a system.

56.      At the same time, it is also true that – as the General Court found – the concise and general wording of that provision does not provide a clear-cut answer to the interpretative issue in dispute here.

57.      The Commission disagrees in that respect. It emphasises, in particular, the broad scope of the provision which, in its view, clearly implies (or does not expressly exclude) that Member States with a national merger control system can also refer cases which are not caught by their systems. Yet implying (or not excluding) something cannot be equated, for the purposes of a textual interpretation of a provision, to expressly stating it. The issue of whether the minor premise of the Commission’s reasoning (that the scope of the provision also covers referrals such as that at stake) is the logical continuation of the Commission’s major premise (that the text of the provision is broad) cannot be resolved – as the Commission would want the Court to do – by looking at a single subparagraph of the EUMR, in ‘clinical isolation’ from the rest of the provision and, more generally, the rest of the regulation.

58.      As a matter of principle, the Commission’s contention that, where the wording of a provision appears clear enough, the Court should not make use of any other means of interpretation is perplexing. The Court is obviously free to resort to all the methods of interpretation that it sees fit in each specific situation. I think it is worth insisting on this point, which is of constitutional significance: where the issues in dispute concern the interpretation of the law, principles such as those of party disposition, burden of proof or standard of proof do not apply. Once again, the crucial principle in this context is iura novit curia.

59.      The Commission’s argument also overlooks the consistent case-law of the Court. As the Court stated very clearly in its judgment in Cilfit, ‘every provision of [EU] law must be placed in its context and interpreted in the light of the provisions of [EU] law as a whole’. (26) Indeed, it is settled case-law that ‘in interpreting a provision of EU law, it is necessary to consider not only its wording, but also the context in which it occurs and the objectives pursued by the rules of which it is part’. (27) Accordingly, the Court has never hesitated to engage in a contextual and/or teleological interpretation of a provision, even where the wording thereof was allegedly clear, in order to either confirm the literal interpretation (28) or, where appropriate, to depart from it. (29)

60.      There is, after all, nothing unusual about the importance that the Court has consistently attached, in particular, to contextual and teleological interpretation. In fact, even the Vienna Convention on the Law of Treaties, which famously distinguishes between the ‘general rule of interpretation’ and the ‘supplementary means of interpretation’, (30) includes all those elements in the former group and establishes an indissoluble connection between them. Article 31(1) thereof states: ‘A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose’. (31)

61.      That is also why the Commission’s emphasis on the expression ‘any concentration’, employed in the first subparagraph of Article 22(1) EUMR, is misplaced. One has to look at the type of concentrations Article 22 EUMR is concerned with to establish what the term ‘any’ precisely refers to. At the risk of stating the obvious, the expression ‘any concentration’ cannot but refer to any concentration which is not only caught by Article 22 EUMR but also, and a fortiori, one that falls within the scope of the EUMR. A contextual examination of that provision is thus unavoidable.

62.      By the same token, it would be absurd to suggest that the Court should stop at the examination of the text of a provision when some specific elements which call into question the allegedly clear wording of that provision have been brought to its attention. (32) That is precisely the case of the present proceedings: as it will be shown later, numerous elements point to a different reading of the provision at stake.

63.      Similarly, I find ESA’s argument emphasising the lack, in the first subparagraph of Article 22(1) EUMR, of terms indicating that the referral mechanism would only be applicable to concentrations that are capable of being reviewed under the national competition laws of Member States to be irrelevant. ESA points, in that regard, to the difference between the text of Article 4(5) EUMR (which also concerns a referral mechanism and contains those terms) and the text of Article 22(1) EUMR. However, that argument overlooks the fact that, unlike the former provision, the latter provision was originally introduced to catch concentrations that could be problematic at national level, where the Member State(s) in question lacked a national merger control system. Accordingly, Article 22(1) EUMR could not contain any expression such as that in Article 4(5) EUMR, since it would then exclude the very Member States for which that provision had been established. In fact, the General Court itself, in paragraph 126 of the judgment under appeal, refused to draw any parallel between the two provisions.

64.      In any event, the Commission’s objections of principle are not only unfounded but also moot in the present case, since there are at least two textual elements that are sufficient to cast some doubt over the literal interpretation that, according to the Commission, is so clear that any other method of interpretation of Article 22 EUMR should be outright excluded.

65.      First, the title of the provision is one such element. Article 22 EUMR is entitled ‘Referral to the Commission’. The term corresponding to ‘referral’, in the vast majority of the language versions, (33) has a specific connotation. Indeed, it suggests that the provision concerns, in principle, cases that are actually or potentially before the national authorities and are then referred (that is, passed on, transferred, handed over, assigned, etc.) to the Commission. That interpretation would be in line with the legal maxim nemo dat quod non habet (no one can give what they do not have).

66.      Second, under the first subparagraph of Article 22(1) EUMR, one of the conditions that has to be satisfied for the Commission to be able to review mergers that fall below the thresholds set out in Article 1 thereof is that the merger in question ‘threatens to significantly affect competition within the territory of the Member State or States making the request’. (34) That wording is entirely reasonable if one thinks that the provision at issue is meant, since its introduction in the original ECMR, to permit review of mergers that could distort competition in a Member State that does not have a national merger control system. Moreover, that wording is consistent with the purpose of a provision which, after its amendments in 1997 and in 2004, is – as it will be outlined later – also meant to strengthen the one-stop-shop nature of the EU merger control system by avoiding, as far as possible, multiple national filings.

67.      By contrast, the wording of the provision becomes less obvious if that provision is interpreted, as the General Court stated, as constituting a ‘“corrective mechanism” … to permit the control of concentrations likely significantly to impede effective competition in the internal market’. (35) If that is so, why did the EU legislature refer only to restrictions of competitions occurring at Member State level? Should not the provision refer, more generally or in addition, to restrictions of competition within the internal market?  More fundamentally, why would the Commission need a referral from a Member State’s authority altogether, if the competition problem is at EU level?

68.      The above textual elements appear capable of casting some doubt on the allegedly straightforward interpretation of the provision proposed by the Commission.

69.      Hence, as it is usually the case with legal provisions that are to some extent unclear, or at least not self-contained (which, I believe, is also the case of the provision at stake here: a single subparagraph of an article in a regulation), the old English adage ‘bare reading is bare feeding’ appears quite pertinent. Accordingly, in order to determine the exact meaning and scope of the first subparagraph of Article 22(1) EUMR, it is indeed necessary, as the General Court correctly held, to also resort to the other methods of interpretation used by the Court of Justice.

(2)    A historical interpretation of the first subparagraph of Article 22(1) EUMR

70.      In paragraphs 96 to 117 of the judgment under appeal, after reviewing a number of documents concerning the EUMR history the General Court came to the conclusion that ‘the historical interpretation confirms that the first subparagraph of Article 22(1) [EUMR] enables a Member State, irrespective of the scope of its national merger control rules, to refer to the Commission concentrations which do not meet the turnover thresholds in Article 1 of that regulation, but which may have significant cross-border effects’.

71.      I do not agree with such an assessment. In particular, I have four major reservations in that respect: (i) the documents referred to in the judgment under appeal have certain inherent limitations in clarifying the EU legislature’s intention; (ii) the passages of those documents cited do not support the General Court’s findings; (iii) when those documents are read in their entirety, they actually contradict those findings; and (iv) the General Court failed to take into account numerous other documents, including the relevant travaux préparatoires, that support the interpretation put forward by the appellants.

(i)    The limits of the General Court’s historical assessment (I)

72.      First, as Grail correctly points out, there are two important limitations which are inherent to the type of documents that are referred to in the judgment under appeal as supporting the conclusion drawn from them. All of those documents (the 1996 Green Paper, (36) the 2001 Green Paper, (37) the 2003 Commission Proposal, (38) and the 2009 Staff Working Paper) (39) were authored by the Commission and post-date the adoption of the ECMR. The General Court’s approach is, in my view, particularly perplexing in the present case.

73.      The Commission was asked at the hearing whether the alleged broad scope of (what is now) the first subparagraph of Article 22(1) EUMR was (i) already there in the original ECMR adopted in 1989; (ii) added when that provision was amended in 1997; or (iii) introduced when the new EUMR was adopted in 2004. The Commission answered, without hesitation, that such a broad scope was there from the very beginning: that is, in Article 22(4) ECMR, as adopted in 1989. ESA took the same view. (40)

74.      If that is so, it seems to me that, in order to carry out a historical assessment of the meaning and scope of the first subparagraph of Article 22(1) EUMR, documents which post-date the adoption of the ECMR in 1989 are of less significance than those pre-dating the adoption of that regulation. I do not think that I need to explain why preparatory documents (meant as documents used during the elaboration of a given provision) are usually more meaningful than ex post facto documents to prove the legislature’s intention.

75.      In that connection, I also find the judgment under appeal to be contradictory. In paragraph 115 of that judgment, the General Court refused, as a matter of principle, to examine five documents, authored by the Commission and referred to in the appellants’ submissions, which allegedly showed that, until recently, the Commission itself had not interpreted the first subparagraph of Article 22(1) EUMR as proposed in the present proceedings.

76.      If the broad scope of that provision was there since the ECMR’s adoption in 1989, why did the General Court take into account several documents which were authored after 1989, but not those indicated by the appellants? If, on the other hand, the scope of the provision was broadened with the adoption of the EUMR in 2004, why did the General Court not cite any document whatsoever of the legislative process leading to the adoption of that regulation, and especially those from the institution that acted as sole legislature, that is, the Council? This leads me to my next point.

77.      In fact, it is rather surprising that, to confirm the interpretation of the first subparagraph of Article 22(1) EUMR put forward by the Commission, the General Court only relied on documents authored by the Commission itself, citing none from the Council.

78.      I can certainly agree that an official document setting out the Commission’s view concerning the meaning and scope of a given provision of a regulation or directive carries a certain weight, especially where that provision was included in the original proposal and was not the object of significant discussions or amendments during the legislative process. However, the Commission’s view cannot be considered a decisive factor for the Court’s interpretation of the provision. That is a fortiori true where the provision was added by the Council at a relatively late stage of the legislative process, following lengthy discussions, as it is the case of (what is now) the first subparagraph of Article 22(1) EUMR.

79.      Against that background, I find it problematic that none of the documents referred to in paragraphs 96 to 117 of the judgment under appeal was authored by the Council and/or pre-dates the adoption of the ECMR in 1989.

(ii) The limits of the General Court’s historical assessment (II)

80.      Second, the historical documents the General Court relied on do not actually support the conclusion drawn from them, for two reasons: (i) the passages referred to in the judgment under appeal are immaterial to the issue in dispute, and (ii) other more pertinent passages in the same documents were either overlooked or their importance was wrongly downplayed.

81.      The General Court started its historical assessment of the provision by stating that ‘that referral mechanism followed the wish of the Kingdom of the Netherlands, which did not have [a merger control] system at the time, to have the Commission examine concentrations having adverse effects in its territory, provided that those concentrations also affected trade between Member States, which is why that mechanism was referred to as the “Dutch clause”’. (41) It then went on to refer to a number of relevant documents, from which it would follow that (i) the referral mechanism is generally regarded as a useful tool, especially for those Member States that did not currently have a merger control system, but its use was by no means reserved to them; (42) (ii) that mechanism is intended to enable the Member States to ask the Commission to examine a concentration with cross-border effects in a situation where the thresholds laid down in Article 1 of the regulation are not met; (43) (iii) the objectives of that mechanism were extended over time in order to permit joint referrals that would avoid multiple national filings, without calling into question its original objective; (44) and (iv) the amendments to the provision showed that the Commission favoured greater recourse to the referral mechanism. (45)

82.      All those statements by the General Court are, in my view, factually correct. It is a ‘lapalissian’ truth that Article 22(1) EUMR applies to concentrations with cross-border effects that do not meet the thresholds laid down in Article 1 thereof. Moreover, it is not even in dispute that the referral mechanism provided for in Article 22(1) EUMR can be used both by Member States which do not have a merger control system and by Member States which have a control system. Finally, there is also no doubt that the referral mechanism was amended over time with a view to enlarge its objectives and make a more frequent use thereof possible.

83.       However, nothing in those findings illuminates, directly or indirectly, the question at the heart of the present ground of appeal: whether or not the first subparagraph of Article 22(1) EUMR permits Member States which have a national merger control system to refer cases that do not fall within that system.

84.      Therefore, not only are the documents referred to in the judgment under appeal of relative persuasive value but, upon closer scrutiny, the parts of those documents which are cited by no means support the final conclusion drawn from them in paragraph 116 of that judgment. The General Court’s findings are, therefore, plainly irrelevant.

(iii) The limits of the General Court’s historical assessment (III)

85.      Third, when read in their entirety, the very documents referred to in the judgment under appeal appear to contradict the General Court’s findings and thus corroborate the interpretation put forward by the appellants. The importance of this point should be stressed. The Court of Justice has consistently stated that the documents relied on by the General Court must be read in their entirety to correctly assess their evidentiary value. Extrapolating one or more specific passages from a document, and then drawing inferences from those that are inconsistent with the real content of the document read as a whole, is an error of law. (46)

86.      Those principles are, in my view, relevant in this context.

87.      To begin with, I find it surprising that paragraph 99 of the judgment under appeal downplays the importance of the passage in the 2001 Green Paper, stating that – in view of the fact that, at the date of adoption of that paper, only the Grand Duchy of Luxembourg did not have a merger control system – ‘in practice … the potential scope for use of Article 22(3) in its original form [was] very limited’. (47) True, that passage does imply, as the General Court rightly stated, that Member States other than Luxembourg were not precluded from making use of Article 22(3) ECMR. (48) Yet again, that is not the issue in dispute. In fact, that passage suggests that, because of the limits to the use of the referral mechanism by the Member States with a merger control system, the practical use of the referral mechanism had been reduced over time. Most Member States had, in the meantime, enacted a domestic system of merger control and, therefore, had a more limited interest, and fewer opportunities, to refer a case to the Commission.

88.      Read in that manner, the passage at issue fits perfectly with the excerpts of the documents cited in the preceding paragraphs of the judgment under appeal and lends support to the appellants’ position: the referral mechanism had been conceived, and considered ‘especially’ useful, for the Member States without a merger control system. Had the Member States with a merger control system been able to refer any concentration whatsoever, irrespective of whether or not they were caught by their systems, the use and expediency of the mechanism for those Member States would not have been affected much by their adoption of a national regime, and the mechanism certainly would not have been ‘limited’.

89.      In addition, other very clear and significant passages of the documents referred to by the General Court were not mentioned in the judgment under appeal.

90.      For example, when discussing the limits of the regulatory framework then in force, and the options available to amend it in order to catch more mergers with cross-border effects, the 1996 Green Paper makes no mention of an alleged possibility to refer to the Commission review of mergers that escape the national systems of merger control under Article 22 ECMR. In fact, that provision is referred to as being merely concerned with ‘the allocation of cases between the Commission and the Member States’. The 1996 Green Paper goes as far as stating that ‘below the [ECMR] thresholds, concentrations are subject to national merger control if that exists’. (49)

91.      Next, the 2001 Green Paper is even clearer in contradicting the General Court’s interpretation of the first subparagraph of Article 22 EUMR. First, the document indicates that its objectives (‘to strengthen the application of Community competition law in cases with cross-border effects, to strengthen the “one-stop shop” principle and to alleviate the problem of multiple filings’) were to be achieved by ensuring that cases leading to multiple notifications at national level could be handled by the Commission. (50) It goes without saying that cases leading to multiple notifications are not those falling below the national thresholds. In fact, that document referred extensively to referrals for cases that are subject to obligatory and/or voluntary notifications at national level, (51) but there is no hint whatsoever that the referral mechanism could also be used for concentrations that are not notifiable at national level. (52)

92.      Second, the 2001 Green Paper indicated that one of the reasons for which the referral mechanism set out in Article 22 ECMR was underused consisted in the ‘technical differences in national merger control procedures, notably concerning the event triggering the notification and the rules concerning timing of notifications’. (53) Obviously, no such consideration would have been relevant if Article 22 ECMR permitted Member States to refer concentrations to the Commission regardless of whether a notification at national level was triggered. (54) Similarly, were the General Court to be correct, the statements, in the 2001 Green Paper, according to which the possibility to make joint referrals under Article 22(3) ECMR more operational would be difficult to implement since it would be conditional upon carrying out ‘a sufficient degree of harmonisation of national laws’ would be inexplicable. (55)

93.      As regards the 2003 Commission Proposal, paragraph 21 thereof reads: ‘One of the initial purposes of Article 22 [ECMR] was to provide a possibility for Member States which do not have national merger control legislation to refer cases with an impact on trade between Member States to the Commission; today, only Luxembourg falls into this category. Nevertheless, the possibility for a single Member State to refer cases to the Commission should not be completely excluded’. (56) That suggests that the Member States’ unilateral use of the referral mechanism, although possible, was considered unlikely. Arguably, had Article 22 ECMR allowed the Member States with a merger control system to also refer cases that they could not review, the use of the referral mechanism could not have been considered unlikely.

94.      In addition, paragraphs 22 to 25 of that proposal also indicated that the main weakness of the referral provisions (Articles 9 and 22 ECMR) is the fact that they could only be used after a merger had been notified to either the Commission or the NCAs, as the case may be. Furthermore, paragraph 28 of that document makes it very clear that the possibility for the Commission to invite Member States to make a referral request was limited to cases which had already been notified.

95.      Finally, paragraph 133 of the 2009 Staff Working Paper makes clear that (i) far from being an issue which is as clear-cut as the Commission contends, whether Member States with a merger control system should be allowed to make use of Article 22 EUMR with regard to concentrations not caught by those systems was – although the language did not appear to exclude it – controversial, and most Member States that took a position on that issue leaned towards a negative answer; (57) (ii) some of the stakeholders consulted (which included the NCAs) even questioned whether a provision such as Article 22 EUMR should continue to exist at all, since allowing ‘a Member State without jurisdiction [to] refer or to join a referral under Article 22’ created issues of predictability, legal uncertainty and excessive length of the procedures; and (iii) the original reason for the existence of Article 22 having become almost obsolete, that provision still served a purpose ‘where a Member State, after a period of assessment of a transaction, forms the opinion that a case would be better assessed by the Commission’. (58)

96.      I thus conclude that the documents relied on in paragraphs 96 to 117 of the judgment under appeal not only do not support the conclusion drawn from them by the General Court but, when read in their entirety, actually contradict that conclusion.

(iv) The limits of the General Court’s historical assessment (IV)

97.      Fourth, the error made by the General Court in finding that the historical interpretation of Article 22 EUMR supported a broad scope thereof becomes even more evident if other relevant documents – among which, most notably, some travaux préparatoires, including those authored by the Council – are examined.

98.      The preparatory works show quite clearly that, during the discussions and negotiations which led to the adoption of the ECMR by the Council in 1989, some of the most controversial topics concerned the definition of the material scope of the regulation and its articulation with other (EC and national) rules which could be equally applicable to the transactions notified under that regulation. Two questions arose in particular: was the application of ECMR to be exclusive, or could the Member States also review the notified concentrations in parallel? Would the application of the ECMR exclude a priori the application of then Articles 85 and 86 EEC to the same transaction? (59)

99.      It that respect, an agreement was eventually reached, in the Council, that the Commission’s jurisdiction under the ECMR had to be exclusive and, by contrast, concentrations not meeting the thresholds set out in the ECMR were to be reviewed by the national authorities only. (60) In addition, whereas it was not possible to exclude the application of Articles 85 and 86 EEC (primary law) to the transactions caught by the regulation, it was instead possible to limit the application of the legislation implementing those provisions to them. (61) This led to the addition of two paragraphs to Article 22 of the Commission Proposal. (62)

100. That agreement within the Council raised the problem of the several Member States which, at that time, did not have a national merger control system (including Belgium, Italy, Luxembourg and the Netherlands): who would review the concentrations falling below the ECMR thresholds but having an impact on their national market? Hence the introduction of the ‘Dutch clause’, which permitted the Commission to ‘step into the shoes’ of the national authorities and act on their behalf, on an exceptional basis, when there was no merger review legislation or when those authorities, because of their relative inexperience or limited resources, considered the Commission to be a ‘better-placed’ authority to review a merger notified to them.

101. Indeed, both the Council and the Commission considered that it could be ‘reasonably assumed’ that concentrations below the ECMR thresholds had, generally, an insufficient impact on trade to justify review at EU level. (63) The Council and the Commission were aware that the ECMR thresholds could be based on a variety of values and those values set at different amounts (any amount would necessarily be a proxy). (64) Therefore, it was abundantly clear to all the parties involved in the legislative process, including the Commissioner for Competition at the time, (65) that, regardless of the type and amount of the thresholds chosen, certain concentrations which could affect the common market would, in any event, escape an ex ante review by the Commission under the ECMR. (66) However, that was considered inevitable for a number of reasons, inter alia, to keep the Commission’s workload at reasonable levels, (67) provide legal certainty to the merging parties, (68) and put into place a balanced and clear-cut division of competences between the Commission and the national authorities. (69) In any event, it was clear that then Articles 85 and 86 EEC permitted an ex post intervention for all mergers not meeting the thresholds. (70)

102. As a matter of fact, not a single document, among the vast amount of travaux préparatoires relating to the original version of the ECMR submitted by the parties, points to the referral mechanism set out in Article 22(3) to (5) ECMR as having the ‘corrective’ objective mentioned by the General Court. In order to confirm that, the Commission was asked at the hearing whether it could point to any such document, and the Commission was unable to do so. That is not surprising, to my mind, since many of the discussions which took place within the Council concerning the precise wording of that provision, as reflected in the travaux préparatoires, would become incomprehensible if concentrations below the national thresholds could still be reviewed by virtue of the referral mechanism.

103. The same holds true, mutatis mutandis, with regard to the amendments of the ECMR in 1997. As mentioned in point 82 above, it is true that the EU legislature intended to enlarge the scope of the referral mechanism set out in Article 22 ECMR. However, there is no trace in the travaux préparatoires relating to the revision of the regulation of the fact that the changes pursued the gap-filling objective suggested by the General Court. On the contrary, the very purpose of strengthening the one-stop-shop system by avoiding multiple filings sits at odds with the interpretation of Article 22 EUMR embraced by the General Court.

104. Indeed, I find something inherently paradoxical about the fact that the General Court refers to a document which explains that the rationale of the 1997 amendment to Article 22 ECMR was to avoid multiple filings, in order to endorse an interpretation of that provision which – as it will be explained later – (71) will de facto encourage undertakings that, under the EU and national merger control laws are not required to do any filing at all, to still make filings (potentially up to 30 of them) (72) just as a matter of precaution.

105. Furthermore, the historical documents relating to the adoption in 2004 of the EUMR do not support the General Court’s findings concerning the EU legislature’s intention to use the referral mechanism set out in Article 22 to remedy the alleged deficiencies stemming from the rigidity of the thresholds laid down in Article 1 of the regulation. (73) The idea behind the changes made to the provisions of Article 22 EUMR pursued the objective of reinforcing the one-stop-shop function of the referral mechanism which avoids the need for the merging parties to make multiple filings. The very wording of the amendments demonstrates that quite clearly. (74)

106. Lastly, certain documents authored by the Commission after the adoption of the EUMR also provide some useful indications. As mentioned above, their interpretative value can only be relative. However, in so far as the General Court itself relied only on Commission documents which post-date the adoption of the ECMR, those additional documents allow a more complete picture by providing interesting insights into the Commission’s historical reading of Article 22 EUMR.

107. In particular, in the 2005 ‘Commission Notice on Case Referral in respect of concentrations’, (75) published in the aftermath of the EUMR adoption, referrals under Article 22 EUMR are consistently referred to as ‘post-notification’ referrals. (76) The use of that expression is difficult to reconcile with the assertion, repeatedly made by the Commission, that it has always interpreted that provision as enabling Member States to refer cases falling below the thresholds set out in national law. If one were to follow the Commission’s arguments, it would also be odd that the same document makes no reference, when listing ‘the categories of cases normally most appropriate for referral to the Commission pursuant to Article 22’, to concentrations which give rise to serious competition concerns whilst not being caught by any merger control system within the European Union. (77) Arguably, that situation should have been on the top of the list.

108. Similarly, in its 2014 White Paper entitled ‘Towards more effective EU merger control’, the Commission proposed, inter alia, to ‘mak[e] the case referral system more efficient and effective by … amending Article 22 so that it enhances adherence to the [one-stop-shop] principle’. (78) Interestingly, the proposed amendments to Article 22 EUMR indicated expressly that only Member States that are ‘competent to review a transaction under their national law’ could request a referral to the Commission or oppose it. (79) One can legitimately doubt that, with such proposals, the Commission intended to restrict the scope of Article 22 EUMR since it would go against both the overarching objective of making the merger review system more effective and efficient, and the more specific objective of improving the referral mechanisms, ‘both before and after notification’. (80) In passing, I observe that, also in that document, the Commission referred again to the Article 22 EUMR mechanism as a ‘post-notification’ referral. (81)

109. Finally, the Commission’s 2016 Evaluation Roadmap of procedural and jurisdictional aspects of EU merger control also appears to be of interest. In that document, the Commission discusses the possibility of completing the existing turnover-based jurisdiction thresholds with others based on alternative criteria, and the need to streamline the referral system. In my view, there can hardly be two topics which are more closely related to the issue here in dispute. It is therefore nothing less than striking that in such a document no mention at all is made of the alleged broad scope of Article 22 EUMR. Incidentally, the document also refers to the referral system as being concerned with the ‘correct allocation of cases’ and to the referral from Member States to the Commission as a ‘post-notification’ mechanism. (82)

110. My interim conclusion is that a historical interpretation of the first subparagraph of Article 22(1) EUMR unequivocally supports the conclusion that the General Court erred in law with regard to the meaning and scope of the referral mechanism in question.

(3)    A contextual interpretation of the first subparagraph of Article 22(1) EUMR

111. I shall now turn to paragraphs 118 to 139 of the judgment under appeal, in which the General Court engaged in a contextual interpretation of the first subparagraph of Article 22(1) EUMR. To that end, the General Court considered 12 elements of context, included in 5 provisions (or sets of provisions) of the EUMR. After reviewing those elements, the General Court came to the conclusion that ‘it follow[ed] from the contextual interpretation that a referral request under Article 22 [EUMR] may be submitted irrespective of the scope of national merger control rules’. (83)

112. I do not agree with that conclusion, for four distinct reasons: (i) the provisions of the EUMR other than Article 22 do not confirm the interpretation endorsed by the General Court, (ii) the other paragraphs and subparagraphs of Article 22 do not either; (iii) the General Court wrongly downplayed the significance of certain elements of context which – although by no means determinative – appear to have a certain weight when properly considered; and (iv) the General Court also overlooked some other elements of context which appear to contradict its own conclusions.

(i)    The limits of the General Court’s contextual assessment (I)

113. The General Court started its contextual assessment by examining whether the wording of the provisions of the EUMR other than Article 22 thereof could shed some light on the meaning and scope of the first subparagraph of Article 22(1) EUMR. To that end, it first examined four provisions (or sets of provisions) of the regulation.

114. In the first place, the General Court found that the legal bases chosen by the EU legislature (current Articles 103 and 352 TFEU) (84) for the adoption of the ECMR first, and for the EUMR subsequently, provided no indication on the proper meaning and scope of the first subparagraph of Article 22(1) EUMR. It thus dismissed Illumina’s contention that the legal bases supported its proposed interpretation of that provision. (85)

115. That finding is, in my view, correct. It is apparent from the preamble to the ECMR and that of the EUMR, (86) and from the travaux préparatoires, (87) that the EU legislature took the view that Article 103 TFEU – which permits the adoption of legislation intended ‘to give effect to the principles set out in Articles 101 and 102 [TFEU]’ – was, taken alone, insufficient to establish a merger control system which sought to prevent the mere creation of dominant positions (as opposed to the abuse thereof, which is prohibited by Article 102 TFEU), and also caught concentrations on the market for agricultural products that, under Article 38(3) TFEU and Annex I to the FEU Treaty, (88) could be subject to a specific legal regime which included exceptions from the full application of EU competition rules. Accordingly, the EU legislature considered it necessary to base the regulation on Article 352 TFEU as well. (89)

116. Whether the EUMR’s legal bases could be relevant in the present issue was also discussed at length at the oral hearing. The Commission, for its part, argued that the legislature’s choice would indirectly confirm its position since Article 352 TFEU is a provision capable of creating a new competence for the Member States to request the Commission to review a given merger, even if there is no power to do so under national law. However, irrespective of whether Article 352 TFEU could be read in that manner, I found no trace of any such consideration by the legislature in any historical document. As stated, both the preamble and the travaux préparatoires make it abundantly clear that the choice of the legal basis by the legislature was not influenced by the scope of Article 22 EUMR. (90)

117. In the second place, the General Court referred to Article 1(1) and (2) EUMR which lays down the thresholds above which a concentration is deemed to have ‘a Community dimension’ (and thus becomes subject to the mandatory notification regime), and makes clear that such thresholds are ‘without prejudice to Article 4(5) and Article 22’. The General Court inferred from Article 1(1) and (2) EUMR that ‘the scope of [the EUMR] and, consequently, the Commission’s competence to examine concentrations depend primarily on the exceeding of the turnover thresholds defining the European dimension and, in the alternative, on the referral mechanisms provided for in Article 4(5) and Article 22 of that regulation, which supplement those thresholds by authorising the examination, by the Commission, of certain concentrations that do not have a European dimension’. (91)

118. Again, the General Court’s finding in that regard is entirely correct and none of the parties disputes that the first subparagraph of Article 22(1) EUMR enables the Commission to review certain mergers which fall below the thresholds set out in Article 1 EUMR. Nevertheless, the General Court’s finding does not shed any light on the real issue in dispute: which mergers below the EUMR thresholds can be reviewed by the Commission pursuant to Article 22 thereof.

119. In the third place, the General Court took into consideration the text of Article 4(5) EUMR. That provision includes another referral mechanism, enabling the parties to a merger that does not have a Community dimension and that is capable of being reviewed under the national competition laws of at least three Member States to request that that merger be reviewed by the Commission. As the General Court noted, those two provisions differ significantly as regards their conditions for application and their purpose. The General Court thus refused to interpret the first subparagraph of Article 22(1) EUMR in the light of Article 4(5) thereof. (92)

120. For the reasons explained in point 63 above, I find that approach justified. In my view, the text of Article 4(5) EUMR is simply inconclusive with regard to the interpretation of the first subparagraph of Article 22(1) EUMR.

121. In the fourth place, the General Court held that Article 22 EUMR ‘cannot be interpreted in the light of the referral mechanisms provided for in Article 4(4) and Article 9 of that regulation’. (93) The differences in the wording of those provisions showed, for the General Court, that those mechanisms are ‘not aligned’ and, consequently, no inference with regard to the meaning and scope of the first subparagraph of Article 22(1) EUMR could be drawn. (94)

122. Once more, the General Court’s finding is correct: the appellants’ arguments in that regard were unconvincing. At the same time, it may be worth adding that those provisions also do not support the Commission’s arguments; in fact, they say nothing about the issue here at dispute.

(ii) The limits of the General Court’s contextual assessment (II)

123. Finally, in paragraphs 130 to 138 of the judgment under appeal, the General Court examined the meaning and scope of the first subparagraph of Article 22(1) EUMR in the light of the other paragraphs and subparagraphs of that provision. For that purpose, the General Court considered eight elements in Article 22 EUMR.

124. First, contrary to what the General Court stated, (95) the wording of the second subparagraph of Article 22(1) EUMR – providing that the referral request should be made ‘at most within 15 working days of the date on which the concentration was notified, or if no notification is required, otherwise made known to the Member State concerned’ – (96) does not mean that the first subparagraph thereof ‘govern[s] … situations in which concentrations are not notified but merely made known to the Member State concerned, either because they do not fall within the scope of that system, or because no such system exists’. (97)

125. The General Court overlooked the obvious fact that the terms ‘made known’ were necessary for the provision to fulfil its essential function of the ‘Dutch clause’: permitting Member States without a national merger control system to ask the Commission to review mergers that may be problematic at national level.

126. In addition, the General Court ignored the amendments made over time to the second subparagraph of Article 22(1) EUMR. In the original ECMR, that provision merely referred to a request to be made ‘within one month … of the date on which the concentration was made known to the Member State or effected’. When the ECMR was amended in 1997, that provision read: ‘[a] request shall be made within one month … of the date on which the concentration was made known to the Member State or to all Member States making a joint request or effected’. Finally, only with the adoption of the EUMR was that provision amended so as to also include a reference to the concentration being ‘notified’. (98)

127. What do those amendments tell us? In my view, they clearly confirm what was made evident by the analysis of the travaux préparatoires: (i) Article 22 of the original ECMR was conceived to govern referrals by Member States without a merger control system (hence no mention of any notification); (ii) Article 22 ECMR was amended in 1997 to permit the referral by several Member States in order to avoid multiple filings, when the Commission was considered to be the best-placed authority (hence the introduction of the reference to joint requests); and (iii) Article 22 EUMR consolidated the Article 22 acquis and strengthened the one-stop-shop function of that provision (hence the introduction of the reference to notifications). (99) Accordingly, the General Court’s finding based on the text of the second subparagraph of Article 22(1) EUMR is, in my view, incorrect.

128. Second, the General Court found that the appellants could not rely on the wording of the first subparagraph of Article 22(2) EUMR which requires the Commission to inform ‘the competent authorities of the Member States’ of a referral request. That reference is generic and does not imply that a notification at national level has been made or is at any rate possible. (100)

129. I agree, in part, with the General Court. If taken in and by itself, that element appears inconclusive to determine the meaning and scope of the first subparagraph of Article 22(1) EUMR. Nevertheless, as I shall explain in points 152 to 162 of this Opinion, the provision in question is not deprived of any significance, when assessed in combination with other relevant provisions.

130. Third, the General Court found that the second subparagraph of Article 22(2) EUMR – providing that ‘any other Member State shall have the right to join the initial [referral] request’ – was ‘consistent with Article 22(1) [EUMR] and confirms that any Member State may submit a request for referral or joinder under that article, irrespective of the scope of its national merger control rules’. (101)

131. That is, admittedly, an element which, as the General Court stated, appears to support the Commission’s position. However, the persuasive value of such an element is rather limited, for the following four reasons.

–        At the outset, as the General Court itself noted, the wording of the second subparagraph of Article 22(2) EUMR is similar to that of the first subparagraph of Article 22(1) EUMR. Given the strict and inherent connection between the two provisions (governing who can make a request and who can make a joint request, respectively), that is quite logical. It is, thus, not surprising that both provisions contain terms which are similarly unqualified. However, to the extent that the former provision is allegedly unclear, the corresponding wording in the latter can hardly be considered as giving reliable guidance on the meaning of the former.

–        The wording of the second subparagraph of Article 22(2) EUMR is also unclear for another reason. Indeed, recital 15 of the EUMR – which concerns Article 22 EUMR – states that ‘other Member States which are also competent to review the concentration should be able to join the [referral] request’. (102) At the very minimum, that recital casts some doubt on the General Court’s interpretation of the second subparagraph of Article 22(2) EUMR, since that hints at the fact that the Member State which makes the referral must be competent.

–        In addition, even if one were to agree with the interpretation of the second subparagraph of Article 22(2) EUMR retained by the General Court, it would not create any incongruence with the interpretation of the first subparagraph of Article 22(1) EUMR proposed by the applicants. The Commission acquires a potential competence to review a merger that falls below the thresholds set out in Article 1 EUMR when a request for referral is made by a Member State that is competent under Article 22 EUMR. Hence, when one or more Member States join a (validly made) referral request from another Member State, the merger has already entered the scope of the EUMR. It is thus neither problematic nor anomalous that any Member State can join such a request.

–        Finally, the fact that one or more Member States join (ab initio or successively) a referral request (validly) made or being made by another Member State has no adverse consequence for the undertakings concerned in terms of legal certainty and predictability of the procedures. (103) That is in stark contrast with the consequences that would follow, in that respect, from interpreting the first subparagraph of Article 22(1) EUMR in the manner suggested by the Commission. (104)

132. Fourth, the General Court found that the fact that, according to the third subparagraph of Article 22(2) EUMR, ‘all national time limits relating to the concentration shall be suspended’, did not corroborate the appellant’s interpretation of the first subparagraph of Article 22(2) EUMR. (105)

133. On that point too I share the General Court’s immediate conclusion: the third subparagraph of Article 22(2) EUMR, taken by itself, does not shed any light on the scope of the first subparagraph of Article 22(1) EUMR. (106)

134. Fifth, the General Court turned to the wording of the third subparagraph of Article 22(3) EUMR, according to which ‘the Member State or States having made the request shall no longer apply their national legislation on competition to the concentration’. In that respect, the General Court found that such a provision did not support the appellants’ arguments: the national legislation in question also refers to the national provisions on anticompetitive agreements and abuses of dominance. (107)

135. In that regard, I fully share the General Court’s assessment. Indeed, the third subparagraph of Article 22(2) EUMR does not support the interpretation of the first subparagraph of Article 22(1) EUMR put forward by the appellants (and, by the same token, that proposed by the Commission).

136. Sixth, the General Court examined the first subparagraph of Article 22(4) EUMR, according to which Article 2, Article 4(2) and (3) and Articles 5, 6 and 8 to 21 thereof apply where the Commission accepts to examine a concentration referred, and Article 7 EUMR applies ‘to the extent that the concentration has not been implemented on the date on which the Commission informs the undertakings concerned that a request has been made’. From the wording of that provision, the Commission infers that the standstill obligation contained in Article 7 EUMR is applicable to ‘both situations in which the concentration that is the subject of the referral request falls … outside the scope of any national legislation, and those in which such legislation is applicable but does not provide for suspension of that concentration’. (108)

137. The General Court’s inference is puzzling. Taken literally, it is correct. (109) However, it would also be of no relevance to the issue here in dispute. Therefore, I understand the General Court’s inference as meaning that the standstill obligation laid down in Article 7 EUMR is also applicable with respect to mergers that fall outside the scope of the national merger control system of the Member State that submits the request.

138. Nevertheless, there seems to be a gap in the reasoning of the General Court: it is, indeed, not immediately apparent how such an inference follows from the wording of the first subparagraph of Article 22(4) EUMR. At any rate, I believe that such an inference is erroneous.

139. The first subparagraph of Article 22(4) EUMR makes the standstill obligation applicable to all mergers for which a referral request has been made in order to ensure the effectiveness of the system of review and prevent distortions of competition from taking place before it is decided whether the Commission will review the case.

140. The fact that the standstill obligation applies only to the extent that ‘the concentration has not been implemented on the date on which the Commission informs the undertakings concerned that a request has been made’ is the inevitable result of the fact that a merger in respect of which a referral request has been submitted may have (lawfully) been implemented prior to that submission. There are various reasons for which that is possible. In particular, a referral request may come from a Member State (or EEA/EFTA State) (110): (i) that has no merger control system; (ii) that has a merger control system that does not provide for a standstill obligation; (111) and (iii) in which a standstill obligation, albeit existing, was not applicable in the specific case. As regards this last point, it is indeed important to mention that the scope of the suspension obligations, including the exemptions and possible derogations therefrom, as well as the length of the waiting periods applicable, vary from Member State to Member State. (112)

141. The General Court’s finding with regard to Article 7 EUMR constitutes, therefore, a non sequitur. In my view, the first subparagraph of Article 22(4) EUMR sheds no light on the proper construction of the first subparagraph of Article 22(1) EUMR.

142. Seventh, the General Court pointed out that, in accordance with Article 22(5) EUMR, ‘the Commission may inform one or several Member States that it considers a concentration fulfils the criteria in paragraph 1 [of that article]’. Since that wording refers only to those criteria, which appear to be exhaustive, the General Court took the view that that provision does not require the concentration to fall within the scope of national merger control rules. (113)

143. In my view, the General Court is reading too much into that provision. Article 22(5) EUMR complements Article 22(1) EUMR: the referral mechanism in question can be initiated by one or more Member States, but also by the Commission – in both cases the two substantive conditions required by Article 22 EUMR have to be fulfilled, which explains the very similar language used in both. It would actually have been odd if Article 22(5) EUMR were to be more detailed than, or include any substantial difference with, Article 22(1) EUMR. Thus, as I stated in point 131 above, such a provision can hardly be used as a reliable source of contextual interpretation for the provision whose wording it reflects.

144. In addition, even if one were to consider the wording of Article 22(5) EUMR to be relevant, I see at least two other explanations for such a wording: not only they do not support the Commission’s position, but they may even be considered favourable to the applicants’ position.

145. One such explanation becomes apparent if we turn our attention to paragraph 110 of the judgment under appeal. In that passage, the General Court noted that, in a past case (‘Kesko’), it had already ruled that ‘it was not for the Commission to rule on the competence of a national competition authority to submit a referral request under Article 22 [EUMR], but that it was required only to verify whether that request was, prima facie, that of a Member State’. (114) That ruling is correct in that whether a given merger is notifiable under national law is not an issue of EU law but one of national law. Thus, it cannot be for the Commission to inform a Member State under Article 22(5) EUMR that, in its view, not only the substantive conditions for the referral are satisfied, but also that the national thresholds are met.

146. Another explanation stems from the lack of any indication, in Article 22(5) EUMR, regarding the criteria that the Commission should use in order to identify the ‘one or several Member States’ that, under that provision, it may first contact and then invite to make a request. Are those the Member States on whose territory competition may be affected? If so, can the Commission freely choose only some of them (and on the basis of what criteria?) or is it required to treat them equally? The wording of the provision could, at first sight, appear somewhat ambiguous in that regard. Or perhaps not. It could be argued that the Commission is given wide discretion on that matter because, inter alia, it may need to consider, in each specific case, which Member States are prima facie competent to refer a merger and which are not.

147. Accordingly, I am of the view that Article 22(5) EUMR is not helpful to determine the nature and scope of the first subparagraph of Article 22(1) EUMR either.

148. Finally, the General Court held that the other provisions of Article 22 EUMR ‘contain[ed] no relevant information capable of casting further light on the content of the first subparagraph of Article 22(1) of that regulation’. (115) As I shall explain in the next sections of this Opinion, I disagree with that last finding.

149. On the basis of the various elements of context illustrated above, the General Court came to the conclusion that a contextual interpretation of the first subparagraph of Article 22(1) EUMR confirmed that a referral request under Article 22 EUMR may be submitted irrespective of the scope of national merger control rules. However, as I have explained, such a conclusion does not follow from the contextual analysis carried out by the General Court. In that analysis, the General Court relied, overall, on 12 elements of context. Of those elements examined:

–        7 are, in the view of the General Court itself, irrelevant for the interpretation of the first subparagraph of Article 22(1) EUMR; they were, in fact, mainly examined in order to dismiss some arguments of the appellants. The General Court did not state (or even suggest) that those elements may support the Commission’s position; and

–        1 was referred to by the General Court in order to corroborate a point that, however, is not in dispute and, what is more, offers no guidance with regard to the disputed interpretation.

150. Accordingly, even if one were to follow entirely the General Court’s reasoning, quod non, its conclusion would only be based on four contextual elements. Yet, three of those elements have, as explained, been assessed erroneously and one, although admittedly favouring the Commission’s position, does not appear to be particularly persuasive.

151. Moreover, I find the contextual analysis made in the judgment under appeal problematic for two additional reasons: (i) the General Court wrongly excluded the significance of some elements of context which – although by no means determinative – do have an indicative value when properly considered; and (ii) the General Court ignored other elements of context which appear to contradict its conclusions.

(iii) The limits of the General Court’s contextual assessment (III)

152. To begin with, some elements of context whose significance the General Court ruled out (116) acquire, in my view, a certain hermeneutic value when due consideration is given to two aspects which that court overlooked: their connection and the time factor.

153. Allow me to explain. The elements that I am referring to are provisions and recitals of the EUMR that, when examined in isolation, may appear to be of no particular significance to the interpretation of the first subparagraph of Article 22(1) EUMR. However, in reality, when one takes a step back and looks at those provisions and recitals together, taking into account when and why they were introduced in the regulation, some useful indications may actually be gleaned.

154. At first instance, the appellants relied on a number of provisions and recitals of the EUMR which appear to be based on the premiss that (i) the merger subject to a referral under Article 22 EUMR is either notified or notifiable at the national level; (117) (ii) that merger must in any event be reviewed somewhere, even if the Commission decides not to do so; (118) or (iii) the national authorities making a reference must be competent to review the merger. This last point deserves a brief explanation.

155. As mentioned above, the first subparagraph of Article 22(2) EUMR requires the Commission to ‘inform the competent authorities of the Member States and the undertakings concerned of any request [of referral] received’. Like the General Court, I too would tend to read the terms ‘competent authorities’ as referring to the national authorities that are generally in charge of mergers – as opposed to authorities competent to review the specific merger under national law.

156. Yet, that reading is called into question, as mentioned in point 131 above, by recital 15 of the EUMR – a recital dealing precisely with the referral mechanism in question, and more specifically with the conditions that must be satisfied for its use under the first subparagraph of Article 22(1) EUMR. That recital reads as follows: ‘a Member State should be able to refer to the Commission a concentration which does not have a Community dimension but which affects trade between Member States and threatens to significantly affect competition within its territory. Other Member States which are also competent to review the concentration should be able to join the request’. (119) Does the wording of that recital not suggest – as argued by the appellants – that the Member State that makes a referral must be competent, under national law, to review the merger in question?

157. The General Court gave short shrift to the appellants’ arguments: those provisions and recitals cannot be read as implying that, to be referred, a given merger must be notified or notifiable in the Member State that triggers the mechanism. (120) That much is clearly correct. I hardly need to point out again that the Article 22 EUMR referral mechanism can be used by – and was in fact primarily conceived for – Member States that do not have a national merger control system.

158. However, it is somewhat too simplistic to stop the legal analysis at that, as the General Court did. Similarly, I find it surprising that the Commission also did not linger any longer over the wording of those provisions in its observations, given the utmost importance that it attaches to textual interpretation in the context of the present case.

159. One cannot help but wonder, in this context, whether there is not a certain inconsistency in the arguments put forward by the Commission as well as in the statements of reasons in the judgment under appeal. They both rely heavily on the (allegedly clear) wording of certain provisions and then discard what appears to follow from the (allegedly clear) wording of other provisions because of the mere fact that the latter cannot be reconciled with the interpretation given to the former. In my view, discarding the indications given by certain provisions because those indications do not fit with the provisional conclusion reached before is not a thorough contextual interpretation. It comes close to circular reasoning.

160. A more prudent interpreter should have, I believe, wondered why some provisions and recitals of the EUMR might not mean what their wording suggests. In my view, the reason behind the idiosyncrasy of those recitals and provisions lies in the fact that none of them were included in the original ECMR in 1989. They were all introduced later, when the ECMR, after being amended on that point in 1997, was eventually repealed by the EUMR.

161. Since this point has already been dealt with extensively, I need not dwell on it again. The EUMR intended to develop the ‘one-stop-shop’ objective of the referral mechanism. Therefore, since that objective concerns only notified or notifiable mergers, it is quite obvious that the wording of those provisions and recitals was drafted with those transactions in mind.

162. When seen in that light, the wording of those provisions and recitals makes perfect sense and is consistent with the rest of the EUMR. Consequently, those elements of context too suggest that Article 22 EUMR was never intended to allow Member States to refer to the Commission mergers falling below the national thresholds. Otherwise, they would have probably been drafted differently. Using another English idiom, I would thus say that, with regard to those provisions and recitals, the General Court ‘could not see the wood for the trees’.

(iv) The limits of the General Court’s contextual assessment (IV)

163. Furthermore, the General Court has overlooked other aspects of the legal context that, in my view, also appear to support the interpretation of the first subparagraph of Article 22(1) EUMR put forward by the appellants.

164. On this point I can again be brief. Indeed, I have already referred to some of those elements in previous passages of this Opinion.

165. To begin with, recital 15 in fine states that, under Article 22 EUMR, the Commission acquires the ‘power to examine and deal with a concentration on behalf of a requesting Member State or requesting Member States’. (121) The language of that recital is hard to reconcile with a provision that – according to the Commission and the General Court – gives competence to the Commission to review certain mergers that affect competition in the internal market. If the problem is in the internal market, why should the Commission act in the interest of, in lieu of or in the name of (122) a national authority, a fortiori one that is not competent to review the merger in question?

166. My doubts on this point are compounded by the wording of Article 22(5) of the original ECMR, which reads: ‘pursuant to paragraph 3 the Commission shall take only the measures strictly necessary to maintain or restore effective competition within the territory of the Member State at the request of which it intervenes’. (123) The express limit placed on the powers granted to the Commission in those circumstances (124) shows, in my view unequivocally, that Article 22 EUMR was not intended to have the broad corrective function attributed to it by the General Court.

167. It also interesting to note that, in the judgment under appeal, no mention is made of Article 1(4) and (5) EUMR, which provides for a simplified procedure (125) to enable the Council, on a proposal from the Commission, to ‘revise the thresholds and criteria’ which, under that provision, define the scope of the EUMR. (126) It is important to note that this provision refers not only to ‘thresholds’ (meant as turnover thresholds) but also to ‘criteria’. This means that the EU legislature may, should it deem it necessary, decide to replace or integrate the turnover thresholds with criteria based on other types of values (for example, price paid by the buyer, value of the transaction, market shares, share of supply, value of the local assets to be transferred, potential impact on the relevant markets, etc.). There is, therefore, a systemic corrective mechanism built in the EUMR which permits a rapid adjustment of the scope of that regulation if the jurisdictional criteria in use become, because of market developments, no longer apt to capture potentially harmful concentrations.

168. I agree with the Commission that, taken by itself, the hermeneutic value of this element should not be overemphasised. However, it does raise questions about the need to have, in the regulation, an ad hoc corrective mechanism such as that envisaged by the General Court. In addition, this element of context becomes much more relevant for the interpreter when examined from a different angle.

169. It must be borne in mind that a provision similar to Article 1(4) and (5) EUMR was already present in the original ECMR and, in the latter regulation, the connection between the thresholds adjustment mechanism and the referral mechanism was direct and explicit. Interestingly, the referral mechanism set out in Article 22(3) to (5) ECMR was initially conceived as a temporary one. Indeed, Article 22(6) ECMR provided that ‘paragraphs 3 to 5 shall continue to apply until the thresholds referred to in Article [1(2)] have been reviewed’. This means that the EU legislature considered, in 1989, that the referral mechanism was destined to become obsolete once the experience ‘on the ground’ permitted to do the appropriate adjustments to the turnover thresholds. (127) Obviously, such a consideration would have been utterly meaningless if the referral mechanism was, as the Commission argues, meant to also catch concentrations falling below the national thresholds: its usefulness would be wholly unaffected by any change of the ECMR thresholds. A fortiori, if the referral mechanism was intended to catch mergers below the national thresholds, why make it temporary?

170. My interim conclusion is that, on the whole, a contextual interpretation of the first paragraph of Article 22(1) EUMR also supports the conclusion that the General Court erred in law with regard to the meaning and scope of the referral mechanism in question. Indeed, although there are elements going in both directions, those pointing to a narrower scope of that provision are far more numerous and more relevant than those pointing to a larger scope thereof.

(4)    A teleological interpretation of the first paragraph of Article 22(1) EUMR

171. Next, in paragraphs 140 to 151 of the judgment under appeal, the General Court carried out a teleological interpretation of the first paragraph of Article 22(1) EUMR, focusing mainly on the wording of the preamble. In particular, it emphasised that, as follows from recitals 5, 6, 8, 24 and 25 of that regulation, its objective is ‘to permit effective control of all concentrations with significant effects on the structure of competition in the European Union’. The General Court also stressed that, in recital 11, referral mechanisms are referred to as ‘corrective mechanism[s]’, which indicates that they give rise to ‘a subsidiary power of the Commission which confers on it the flexibility necessary to achieve the objective of that regulation’. On that basis, it was concluded that ‘a teleological interpretation confirms that a referral request under Article 22 [EUMR] may be submitted irrespective of the scope of national merger control rules’.

172. I have to disagree with the General Court once again. To explain why, I will attempt to address two questions that, in this context, shed light on the meaning and scope of the first paragraph of Article 22(1) EUMR. First, what are the specific objectives of that provision? Second, is the alleged gap-filling objective pursued by that provision consistent with the overarching objectives of the EUMR?

(i)    The limits of the General Court’s teleological assessment (I)

173. The answer to the first question is, at this point of my analysis, clear in part. Indeed, both the historical and the contextual assessment of the EUMR revealed two objectives that are undoubtedly pursued by the referral mechanism set out in Article 22 thereof. The first objective, which prompted the inclusion of the referral mechanism in the original ECMR (the ‘Dutch clause’), has been to permit review of mergers that could distort competition locally, where the Member State in question does not have any national merger control system. The second objective, introduced with the reform of the ECMR in 1997 and then strengthened with the adoption of the EUMR, is the ‘one-stop-shop’ objective: to permit the review of a merger notified or notifiable in several Member States by the Commission, in order to avoid multiple national filings.

174. The first objective is not obvious in the wording of the preamble to the original ECMR. However, the fact that the referral mechanism was initially introduced to pursue that objective has been ascertained by the General Court and it is common ground between the parties. At any rate, the absence of any reference to that objective in the ECMR’s preamble is not surprising since, as explained above, its scope and significance was originally meant to be very limited. Indeed, it was initially only destined to apply temporarily, that is, until the turnover thresholds were adjusted, and exceptionally, given its narrow scope, as the Commissioner for Competition at the time expressly stated. (128)

175. By contrast, the second objective is expressly (and emphatically) referred to in the preamble to both the 1997 Regulation and the EUMR.(129) That too is unsurprising, given the importance of the changes made to the referral mechanism in question.

176. I should now turn to the question whether a third objective allegedly pursued by Article 22 EUMR – the gap-filling one, permitting control of concentrations which fall below both the EU and the national thresholds – can be identified. The General Court found confirmation of that objective in recital 11 of the EUMR, according to which ‘the rules governing the referral of concentrations from the Commission to Member States and from Member States to the Commission should operate as an effective corrective mechanism’.

177. In that regard, I take the view that the General Court misread that recital. The expression ‘corrective mechanism’ should not be read in isolation but considered in its proper context.

178. First, what is the subject matter of recital 11? Its context is important. Recital 8 clarifies the basic principles concerning the allocation of competences between the Commission and the NCAs. Recitals 9 and 10 concern the EUMR turnover thresholds for mergers to be of a ‘Community dimension’. In turn, recital 12 concerns mergers that fall below the EUMR turnover thresholds but ‘qualify for examination under a number of national merger control systems’. With regard to the latter, recital 12 notes that ‘multiple notification of the same transaction increases legal uncertainty, effort and cost for undertakings and may lead to conflicting assessments’, and for that reason concludes that ‘the system whereby concentrations may be referred to the Commission by the Member States concerned should therefore be further developed’. Recitals 13 to 16 then point to the cooperation that is to be established between the Commission and the NCAs to that end, and illustrate the functioning of the various referral mechanisms.

179. To my mind, the context above suggests that recital 11 refers to a mechanism having a corrective function in terms of allocation of competences between the Commission and the NCAs. That recital is not concerned with the establishment, as the General Court stated, of ‘a subsidiary power of the Commission which confers on it the flexibility necessary to achieve the objective of that regulation’. (130)

180. The above consideration finds additional support, first, in the fact that this recital did not figure in the original ECMR, but was only introduced in the EUMR. In fact, the usefulness of the referral mechanism with regard to the allocation of cases between different authorities, all competent to review a given merger, only came about in 1997 before acquiring more importance in 2004.

181. In fact, paragraph 94 of the 2001 Green Paper confirms that: it reads, ‘in order to make Article 22(3) operational as a generally applicable corrective mechanism to the multiple filing problem, the system would most likely need amendment of more than just the Merger Regulation’. (131) That paragraph allows two conclusions to be drawn. First, that the term ‘corrective mechanism’ in recital 11 of the EUMR refers to the singular problem of multiple filings, and not the broader issue of all of the deficiencies inherent to a merger control system based on thresholds. Further, the multiple filing issue only arises because mergers may be subject to several national merger control systems and not because they escape such systems. Second, drawing upon Article 22 as a remedy to the multiple filing problem is something that required discussion and legislative amendment and was therefore not that article’s initial purpose. It would follow that engaging Article 22 to remedy other, broader problems would also require discussion and amendments.

182. When recital 11 is read in its entirety, the above considerations are further confirmed. That recital states: ‘the rules governing the referral of concentrations … should operate as an effective corrective mechanism in the light of the principle of subsidiarity; these rules protect the competition interests of the Member States in an adequate manner and take due account of legal certainty and the “one-stop shop” principle’. I draw two inferences from this text. First, the reference to the principle of subsidiarity and to the adequate protection of the Member States’ competition confirms a narrow scope for the referral mechanism: it is only intended to remedy situations in which competition is affected locally. Second, the reference to legal certainty and the ‘one-stop-shop’ principle also suggests that the referral mechanisms are aimed at replacing several national procedures with one centralised procedure, which presupposes that the mergers in question meet the national thresholds.

183. Hence, I remain unconvinced by the General Court’s reading of recital 11 of the EUMR. Nor am I persuaded by the General Court’s reliance on recitals 6 and 24 of the EUMR, in so far as they refer to effective control of all concentrations.

184. Once again, when those recitals are read in their entirety and in their proper context, it appears rather clear that the term ‘all’ does not mean that each and very merger occurring in the world, provided it may raise some competitive concern in some Member State, should be subject to ‘effective’ control under the EUMR. Recital 6 states: ‘A specific legal instrument is therefore necessary to permit effective control of all concentrations in terms of their effect on the structure of competition in the Community and to be the only instrument applicable to such concentrations’. Similarly, recital 24 states: ‘this Regulation must permit effective control of all concentrations from the point of view of their effect on competition in the Community’.

185. Several textual elements in those recitals clearly contradict the implications that they are concerned with the referral mechanism in question. First, not ‘all concentrations’ can be controlled under the EUMR: unless the regulation’s thresholds are met, the merger is normally to be reviewed by other competition authorities (of the EU Member States and/or of third States). Second, with respect to those mergers that would – if one were to follow the Commission’s theory – enter the scope of the EUMR ‘through the backdoor’ (that is, those in respect of which, in principle, neither the Commission nor the relevant NCAs are competent), it cannot be said that the EUMR will be ‘the only instrument applicable to such concentrations’; indeed, Article 22 EUMR permits parallel procedures before the Commission (when requested by one or more NCAs) and one or more NCAs (those that do not join the referral request). Third, under Article 22 EUMR, the Commission does not review mergers ‘from the point of view of their effect on competition in the Community’, as recitals 6 and 24 state, (132) but only in the territories of the Member States making the referral (Article 22(1) and (5) EUMR). In fact, the EU Courts have consistently interpreted the term ‘all concentrations’ figuring in the EUMR preamble as referring to those ‘with a Community dimension’. (133)

186. If that is so, however, one question arises: what does the term ‘all’ refer to in the context of those recitals? The answer lies again in the wording of those recitals, and is confirmed by their history and purpose. The expressions used in those recitals can be traced back to the seventh recital of the original ECMR (134) and aims to make it absolutely clear that, under the merger regulation, all concentrations will be assessed ‘from the point of view of their effect on competition’. This clarification, which may admittedly appear obvious and thus unnecessary today was, at the time of the ECMR’s adoption, not anodyne by any means. Indeed, another reason that stalled, for numerous years, the negotiations within the Council was the very marked difference of views between various Member States concerning the criteria which the Commission was to use when deciding whether or not to clear a merger. Whereas the Commission and numerous Member States favoured a pure antitrust analysis, certain Member States opposed that idea, taking the view that mergers should also be assessed in the light of other considerations, and in particular on industrial policy grounds. Eventually, the first view prevailed and the compromise was to include in the regulation the so-called German clause (then Article 21(3) ECMR, now Article 21(4) EUMR), which granted some residual power of intervention to the Member States. (135) The EU Courts’ case-law appears to confirm my reading of the recital. (136)

187. Accordingly, the General Court’s reliance on recitals 6, 11 and 24 in this context is, to my mind, misplaced. On closer scrutiny, there is no reference in, or inference from, the preamble to any of the three merger regulations of any gap-filling function attributed to Article 22 EUMR. The silence on this point is quite meaningful, given the potentially extraordinary impact that such provision would have on the functioning of a merger control system that (i) is ‘based on the principle of a precise allocation of competences between the [Commission and the national authorities]’, (137) and (ii) whose scope is ‘limited by quantitative thresholds’. (138)

188. That said, another question in this context is whether the gap-filling objective attributed to it by the General Court would be consistent with the overarching objectives of the EUMR.

(ii) The limits of the General Court’s teleological assessment (II)

189. In paragraph 140 of the judgment under appeal, the General Court examined the preamble to the EUMR and came to the conclusion that the gap-filling objective attributed to the referral mechanism in question was consistent with ‘the objective of that regulation [which] is to permit effective control of all concentrations with significant effects on the structure of competition in the European Union’. (139)

190. I find two main problems in that analysis: the General Court ignored some key elements of the preamble and misread certain recitals.

191. First, the General Court repeatedly stressed the EUMR’s objective to ensure effective control of concentrations, going as far as referring to it as ‘the objective’, that is, the only one.

192. There can be no doubt, in my view, that the objective to ensure effective control of concentrations is the very raison d’être of the regulation and its importance is, accordingly, stressed in the preamble to the EUMR. However, that cannot be the only objective or, put differently, that objective does not exist in a vacuum. As a matter of fact, Article 2 EUMR refers to ‘concentrations within the scope of this Regulation [to] be appraised in accordance with the objectives of this Regulation’. (140)

193. Indeed, the pursuit of the objective of permitting effective control of concentrations goes hand in hand with the pursuit of other objectives, some of which are particularly relevant in the present case. The first such objective, which is the result of the lengthy and (dare I say) heated discussions which ultimately led to the adoption of the ECMR, after nearly 20 years of negotiations within the Council, is to establish a system in which jurisdiction is shared between the Commission and the NCAs. (141) The second objective is to realise, at EU level, an efficient system based on the ‘one-stop-shop’ principle: the Commission has sole jurisdiction to review the mergers notified under the EUMR, which need no additional filings at Member State level, and the national authorities can no longer apply their national competition laws to those transactions. (142) The third objective is to establish an efficient and predictable system capable of offering legal certainty to the undertakings concerned. (143) The General Court itself refers, in paragraph 226 of the judgment under appeal, to the ‘fundamental objectives of effectiveness and speed underlying [the EUMR]’, and the EU legislature’s intention ‘to make a clear allocation between the interventions to be made by the national and by the EU authorities’.

194. Whereas the two first objectives mentioned in the previous point are, for obvious reasons, specific characteristics of the EU system of merger control, the third one is not. Indeed, every system of merger control existing at global level seeks to strike a balance between effective scrutiny of competition and avoidance of unnecessary costs and delays for both the merging parties and the public administration itself. (144) To ensure that balance, merger rules are usually based on thresholds which filter the transactions to be reviewed, and impose on the authorities specific time limits to complete their assessment. It is, thus, impossible to overemphasise the importance that predictability and legal certainty have, especially for the merging parties. Undertakings that are potentially subject to notification and suspension obligations need to know, with a relatively high level of confidence, whether their proposed deal will be subject to antitrust scrutiny and by which authorities, and when a definitive answer from those authorities may be expected. (145)

195. That is true, as mentioned, at global level. However, it is even more so for mergers that could be reviewed in the European Union. Not only because, within the European Union, various enforcement authorities co-exist (the Commission and NCAs) – with everything that entails in terms of complexity – but also because, unlike the vast majority of merger control regimes in the world, the EUMR imposes on the merging parties a world-wide bar on closing. That means that the implementation of a notified transaction must, in principle, be suspended in its entirety until the Commission adopts a final decision. The merging parties cannot, as a consequence, accelerate that implementation by, for example, holding certain local assets, units or businesses separate until the pending approval is granted. The costs and risks imposed on the merging parties are, consequently, even more significant, and those undertakings must thus be in a position to take appropriate precautions in that regard.

196. To that end, as the Court held, the EUMR ‘contains provisions which, for reasons of legal certainty and in the interests of the undertakings concerned, are designed to limit the duration of the controls which the Commission must carry out’. Indeed, the EU legislature ‘wished to ensure a control of mergers within deadlines compatible with both the requirements of sound administration and the requirements of the business world’. (146)

197. In the light of the above, I agree that ensuring the effectiveness of the system (meant as capacity to catch the potentially harmful mergers) is the primary objective of the EUMR. However, that effectiveness cannot be achieved at the expense of a satisfactory pursuit of the other objectives of the regulation. Thus, the references in the preamble to ‘effectiveness’ cannot lead the interpreter to maximise the scope and purpose of the provisions of the EUMR to the point that their reach goes beyond the clear intentions of the EU legislature, upsetting the carefully devised balance it has envisaged between the various objectives.

198. Against that background, is the gap-filling objective of Article 22 EUMR advocated by the Commission and endorsed by the General Court consistent with the other objectives described above, and the balance struck between them? To my mind, the answer to such a question is a clear ‘no’. It seems to me that the interpretation of the first subparagraph of Article 22(1) EUMR retained by the General Court sits at odds with the three objectives referred to in point 193 above and is capable of upsetting the balance between them that the EU legislature has sought to attain.

199. First, the ‘competence sandwich’ that would follow from the General Court’s interpretation of the first subparagraph of Article 22(1) EUMR – Commission (large mergers) / NCAs (mergers below the EUMR thresholds but above the national thresholds) / Commission (mergers below the national thresholds) – appears hardly consistent with a system, as the Court of Justice pointed out, ‘based on the principle of a precise allocation of competences between the national and [EU] control authorities’. (147)

200. That construction also appears odd when observed in the light of the principle of subsidiarity, a principle referred to in no less than four recitals of the EUMR. (148) That is a regulation which – in the words of the Commissioner for Competition at the time – represented ‘an excellent example of how [that principle] can be put into practice’. (149) Subsidiarity is a principle which, simply put, has mainly a downward effect: in an area of shared competence, it tends to push competence for a specific action down to the Member States. (150) Naturally, in some circumstances that principle may also have an upward effect: pushing competence up to the European Union when a given action appears to be, by reason of its scale or effects, more effective if undertaken at EU level. However, I wonder whether a situation in which the competence for doing something (here, reviewing a merger) is given to an EU institution (here, the Commission), for the very reason that a Member State has considered that the scale or effects of situations such as that at issue are not significant enough to warrant any action at national level, would not be against the logic of subsidiarity.

201. Second, it is common ground between the parties that one of the consequences flowing from the interpretation of the first subparagraph of Article 22(1) EUMR retained by the General Court is that the undertakings wishing to have certainty that a proposed merger could not be challenged by the Commission after completion, despite that merger not being notifiable anywhere in the European Union and not being subject to any suspensory obligation, would have to: (i) temporarily suspend implementation; and (ii) bring the merger to the attention of (potentially) all EU and EEA/EFTA States (for an overall 30 different national authorities) in order to trigger the 15-working-day time period provided for in the second subparagraph of Article 22(1) EUMR.

202. In that context, it seems important to add that, according to the Commission, the merging parties’ communication to the national authorities in question should contain all the data and information needed by those authorities in order to determine whether the two substantive conditions set out in the first subparagraph of Article 22(1) EUMR – the merger affects trade between Member States and threatens to significantly affect competition within the territory of the Member State in question – are satisfied. Nevertheless, it is evident to me that a sound assessment of those conditions is not an easy exercise, let alone within only 15 working days. Therefore, it is likely that the informal notifications addressed to the national authorities might need, in many cases, to be rather elaborate and detailed and, thus, not too different from the submissions normally required for a formal notification.

203. That means, in practice, that undertakings entering into a transaction which, in principle, falls outside each and every system of merger control in the European Union may end up being driven to file informal notifications to all the national authorities just to avoid a future use of the referral mechanism in question which could have, from their perspective, dramatic consequences.

204. Moreover, should one NCA which does not have competence to review a merger submit a referral request, thereby triggering the referral mechanism, and one or more NCAs which, by contrast, are competent to review it decide not to join the request, the referral mechanism may have the effect of multiplying the procedures running in parallel. Indeed, the procedures before the competent NCAs would co-exist with an additional one before the Commission, which would not have existed if it were not for the referral mechanism.

205. The above shows that the General Court’s interpretation of Article 22 EUMR would result in the introduction of a far-reaching exception to the one-stop-shop principle, hardly consistent with one of the main objectives of the EUMR, and would also be at variance with the aim pursued by the EU legislature when it amended Article 22 in 1997 and 2004.

206. Third – and this is the most problematic aspect, in my view – the procedure(s) that would result from a broad interpretation of the first subparagraph of Article 22(1) EUMR would hardly be efficient, predictable and capable of ensuring legal certainty to the parties.

207. To begin with, it is clear and undisputed by the Commission that, unless the merging parties take positive action to inform the 30 national authorities of the existence of a non-notifiable merger, those parties cannot have any legal certainty as to whether the Commission will, at some point in the future, be asked to review the merger on the basis of Article 22 EUMR and, if so, within which time frame.

208. The Commission retorts that the merging parties may nonetheless be able to obtain legal certainty if, as mentioned earlier, they bring the proposed merger to the attention of those 30 authorities by means of informal notifications. That would ‘start the clock’ and, if no request for referral is made within 15 working days, those parties can be sure that the merger will escape any review in the European Union.

209. Yet, I am not sure that such a course of action provides much more, or at any rate adequate, legal certainty to those parties. The main problem is that this is an informal procedure that is nowhere provided for in the EUMR or, to my knowledge, in the legislation of the Member States. Hence, non-notifiable mergers are neither subject to the national procedural rules nor to those laid down in the EUMR itself. True, Article 22(4) EUMR renders some provisions of the EUMR applicable to the review of those mergers, but only after the Commission has accepted the referral. The period before that is a sort of legal ‘no man’s land’ in respect of which there is very little clarity and predictability.

210. For example, who is entitled to trigger the informal procedure? Should it be only the merging parties, or can third parties (for example, competitors of the merging parties) also do so? The wording of the second subparagraph of Article 22(1) EUMR suggests the latter. If so, could the NCA make a referral under Article 22 EUMR on the basis of the information provided by those third parties and, as the case may be, without hearing the merging parties? Given that the authority has only 15 working days to make a decision, a superficial treatment of the situation cannot be excluded. What if that information is inaccurate or incomplete? The consequences for the merging parties arising from an erroneous assessment, by a national authority, of the substantive conditions for a referral may not be negligible.

211. The Commission is of the view, however, that the period does not start to run until the NCAs have sufficient information to carry out the analysis required by Article 22 EUMR. However, that means that the 15-working-day time limit is likely to become illusory given that it can (and possibly will) often be prolonged by one or more authorities, by means of one or more requests for information, thereby leaving the merging parties with no predictable time frame.

212. Moreover, no indication whatsoever can be found in the EUMR of the type and level of detail of the information that the merging parties may be expected to include in their informal notifications. Surely the parties could take as a model the EU official submission forms (as recently amended: Form CO, Short Form CO, Form RS and Form RM). (151) However, even the Commission did not go so far as suggest it. That would have, obviously, implied that the EUMR notification procedure may de facto apply to non-notifiable mergers. The Commission instead suggested at the hearing that the parties might want to draw inspiration from the information that, under Article 14 of the Digital Market Act, (152) undertakings defined as gatekeepers are to provide to the Commission when they intend to carry out certain mergers. However, quite apart from the oddity of suggesting the parties find guidance in a different regulatory instrument, adopted after the EUMR, and only applicable to some specific sectors of the economy, I am not sure that the information listed in that instrument is sufficient for the purposes of Article 22 EUMR.

213. Furthermore, a small but important detail: in which language should that information be communicated? The Commission argued that any language commonly understood by the staff of the relevant national authority (for example, English) would be adequate. I struggle to see on which basis the Commission can sustain that view. In any event, I doubt the national authorities would accept to carry out a rather complex analysis, within a very tight schedule, on the basis of a submission (possibly with some annexes) drafted in a language that is not theirs.

214. In the light of the above, I take the view that the teleological interpretation of the first subparagraph of Article 22(1) EUMR carried out by the General Court is erroneous, as it is inconsistent with a number of objectives that the merger control system established by the EUMR seeks to pursue, and is capable of upsetting the balance between those objectives envisaged by the EU legislature. The importance of such a balance has not escaped the Court. In its recent judgment in CK Telecoms, for example, the Court noted that ‘the need for speed which characterises the general scheme of the [EUMR]’ is of such significance that even a harmful merger will be deemed approved, unless the Commission takes a decision within the prescribed period. (153)

(5)    Other considerations concerning the interpretation of the first subparagraph of Article 22(1) EUMR

215. Lastly, I shall briefly explain why the General Court’s interpretation of the first subparagraph of Article 22(1) EUMR raises, in my view, a number of systemic issues when account is taken of various general principles of EU law.

216. It is important to stress, at the outset, that the General Court’s reading of the provision gives rise to a very significant extension of the scope of the EUMR and of the Commission’s jurisdiction. (154) In one fell swoop, by means of an original interpretation of Article 22 EUMR, the Commission gains the power to review almost any concentration, occurring anywhere in the world, regardless of undertakings’ turnover and presence in the European Union and the value of the transaction, and at any moment in time, including well after the completion of the merger. That much is clear and undisputed. Indeed, when asked at the hearing a specific question on the point, the Commission confirmed that, in theory, that is true. Nevertheless, it added that, in practice, that will not be the case as the Commission has no interest in using that power frequently and will thus act with discipline in that respect. In the view of the Commission, the merger at issue had certain specific characteristics, unlike the vast majority of the other mergers that could be caught under Article 22 EUMR.

217. However, in the present case we are not only concerned with the application of that (possibly new) power of review in the merger at issue. The Court is indeed called upon to interpret, for the first time, the meaning and scope of Article 22 EUMR which has the potential to apply in an indefinite number of cases. The Commission’s position cannot but raise concerns in various regards.

218. First, I doubt that that position is consistent with the principle of institutional balance, a principle characteristic of the institutional structure of the European Union, deriving from Article 13(2) TEU, which requires, in essence, that each of the institutions exercise its powers with due regard for the powers of the other institutions. (155)

219. One of the most fundamental elements of the EUMR is the definition of the thresholds that, under Article 1(1) to (3) thereof, trigger the notification obligation. However, under the Commission’s interpretation of Article 22 EUMR, the value of these thresholds and, indirectly, of the thresholds and criteria set out in national laws becomes only relative. A merger may well not be notifiable anywhere in the European Union, but that would by no means exclude the possibility that the Commission could claim jurisdiction to review it under Article 22 EUMR.(156)

220. I am certainly not excluding that, in a world which is increasingly based on an ‘Economy 2.0’, it may be desirable, and perhaps even necessary, to change the current thresholds for a merger review. In that context, it may be interesting to note that, very recently, two Member States (Austria and Germany) amended their domestic legislation to include thresholds based on deal values. Other States use different jurisdictional thresholds specifically conceived to permit a review of mergers despite the target company not generating any local revenue (such as the United Kingdom, with the ‘share of supply test’). These and other options could naturally be considered with a view to amending the EUMR. But that is the task of the EU legislature, not of the Commission.

221. Second, the General Court’s broad interpretation of the first subparagraph of Article 22(1) EUMR creates significant potential for cases in which a conflict with the principle of territoriality of EU law might arise. It should be recalled that, in order to comply with international law, the application of EU law presupposes an adequate link to the EU territory. (157) More specifically, it follows from the judgments in Intel and Gencor that the application of EU competition law to the conduct of undertakings is legitimate, regardless of where it takes place, in so far as that conduct has foreseeable, immediate and substantial effects in the European Union (‘the qualified effects test’). (158)

222. I certainly agree with the Commission that the substantive conditions set out in Article 22(1) EUMR are, in principle, capable of ensuring an adequate link to the EU territory. However, it must be borne in mind that, as mentioned, the verification of those conditions is made on prima facie basis only and within a particularly tight time-frame (15 working days). Therefore, it cannot be excluded that, under Article 22 EUMR, the European Union could claim jurisdiction to review a merger (with everything that entails, including the sudden triggering of the obligation to suspend, on a worldwide basis, any act of implementation thereof) which might later turn out to have no foreseeable, immediate and substantial effect in the territory of the relevant Member State.

223. Third, that situation may create issues under the principle of international comity. I am well aware that the contours of such a principle and its legal implications are rather hazy. (159) However, it seems to me that from such a principle one may derive, at the very least, a general requirement for States to consider, before claiming jurisdiction in cases with a significant foreign element and a rather weak domestic connection, whether the application of their laws could not have the effect of undermining the effective application of the laws of third States with a stronger territorial link with those cases. Such a reading of the principle appears broadly consistent with the suggestions of other learned Advocates General, (160) the European Union’s international agreements on this matter, (161) and the findings of other courts, including in competition law matters. (162) Against that backdrop, I wonder whether the Commission’s view of its far-reaching jurisdiction to review mergers under Article 22 EUMR is fully in line with the principle of international comity.

224. Fourth, the appellants’ claim that the interpretation of the first subparagraph of Article 22(1) EUMR embraced by the General Court conflicts with the principles of equality and proportionality does not seem to me unfounded, as the Commission contends. Indeed, undertakings with limited or non-existent sales in the European Union would de facto end up in a situation which is considerably worse than that of undertakings with more significant activities in the European Union.

225. The latter can benefit from the one-stop-shop system established by the EUMR or, in the alternative, will only need to file one or more national filings in those countries where they meet the national thresholds. The number of those filings can be calculated in advance, and the merging parties are aware of the authorities which will review the merger and of the manner in which and the time-frame within which they will do so. By contrast, as explained above, the undertakings which are parties to non-notifiable mergers have no means to predict the fate of their merger unless they file, in the EEA, no less than 30 informal notifications; and even then, many aspects of the procedure, including their duration, remain uncertain.

226. That situation appears, to my mind, problematic under the principle of equality, which requires that comparable situations must not be treated differently and that different situations must not be treated in the same way, unless such treatment is objectively justified. (163) It also appears to create a disproportionate burden, in terms of costs and risks, for the undertakings which entered into transactions which, as I have said, have a rather limited activity in the European Union. (164)

227. Fifth, the principle of effectiveness cannot lead to the scope of the provision in question being extended beyond what is reasonable and necessary for the purposes of the ECMR. I have explained my view on this matter in point 197 above. I only need to add a final element in that context: I am not convinced by the Commission’s claim regarding the need to fill a gap in the scope of the EUMR.

228. As the Court has consistently held – most recently in the judgment in Towercast – (165) Articles 101 and 102 TFEU are applicable to mergers which do not meet the thresholds laid down in the EUMR (including those that do not meet the national thresholds). Those provisions permit the NCAs to intervene, ex post, with regard to mergers that turn out to be anticompetitive. True, an ex post intervention may often be ‘second best’, in comparison with an ex ante review. However, the differences between those two forms of review were, as follows unequivocally from the travaux préparatoires, an aspect which was duly taken into account by the EU legislature during the process which led to the adoption of the ECMR. The Commission’s considerations cannot, accordingly, call into question specific choices made by the EU legislature.

229. In addition, I am also not persuaded by the argument, put forward by the Commission and some of the governments which intervened in the present proceedings, that enforcement under Articles 101 and 102 TFEU would be ineffective and time-consuming.

230. As the Court has recently confirmed in European Superleague Company, an abuse of a dominant position is established where a conduct has ‘the actual or potential effect – or even the object – of impeding potentially competing undertakings at an earlier stage, through the placing of obstacles to entry or the use of other blocking measures …, from even entering that or those market(s) and, in so doing, preventing the growth of competition therein to the detriment of consumers, by limiting production, product or alternative service development or innovation’. (166) To my mind, a so-called killer acquisition falls squarely within that description, offering a paradigmatic example of a ‘by object’ abuse of dominant position. (167)

231.  As such, I do not think it would require an overly lengthy or complex investigation to establish an infringement. Especially because an ex post review of a consommated merger – an activity which is not unusual in a number of jurisdictions – (168) may entail a degree of inconvenience, but it also has a very significant advantage: the authorities need not make any prediction about the undertakings’ future behaviour. Indeed, in its assessment, the competition authority can examine both pre-merger evidence (for example, to establish the intent of the acquirer and whether that undertaking viewed the target as a viable threat to its market position), together with post-merger evidence, which shows what has actually happened in the market following the acquisition (for example, to establish whether there were appreciable effects on price, output and innovation, or whether the target’s operations were terminated or significantly reduced). (169)

232. In addition, it must be borne in mind that, when investigating possible breaches of Articles 101 and 102 TFEU, NCAs are to enjoy the powers set out in the so-called ECN+ Directive. (170) Pursuant to that directive, where an infringement is detected, the competent authority is not only empowered to impose financial penalties (Articles 13 to 16 thereof), but it can also, under Article 10(1) of that directive, require the undertakings in question ‘to bring that infringement to an end. For that purpose, [the authority] may impose any behavioural or structural remedies which are proportionate to the infringement committed and necessary to bring the infringement effectively to an end’.. This may well include, in particularly serious cases, a partial or complete dissolution of the merged entity. (171) In addition, pursuant to Article 11(1) of that directive, the NCAs can also, ‘act on their own initiative to order by decision the imposition of interim measures on undertakings …, at least in cases where there is urgency due to the risk of serious and irreparable harm to competition, on the basis of a prima facie finding of an infringement of Article 101 or Article 102 TFEU’. Such measures could, for example, take the form of suspensive injunctions. (172)

233. Sixth, the very broad scope given by the General Court to a provision that is unquestionably an exception to the provisions of Article 1 EUMR goes against the well-accepted principle of interpretation according to which exceptions to, and derogations from, the general scheme or the general rules of a legal instrument are to be interpreted strictly so that those rules are not negated. (173) In fact, the General Court has already held that principle to be relevant when interpreting the scope of the referral mechanism provided for in Article 9 EUMR. (174) It is unclear to me on which grounds it has, in the judgment under appeal, decided to discard the relevance of this interpretative principle with regard to the referral mechanism provided for in Article 22 EUMR. (175)

234. On the basis of the above considerations, I take the view that the General Court erred in law in its interpretation and application of the first subparagraph of Article 22(1) EUMR. For that reason, the judgment under appeal is to be set aside.

235. If, however, the Court of Justice disagrees with my assessment of the first ground of appeal, I consider that it should dismiss the appeals. In the following section I shall briefly explain why I consider the appellants’ second and third grounds of appeal to be unfounded.

B.      Second ground: timing of the referral request and the Commission’s obligation to act within a reasonable time

236. Illumina’s and Grail’s second ground of appeal is concerned with the General Court’s rejection of Illumina’s second plea at first instance, alleging that the referral request was made out of time and, in the alternative, that the principles of legal certainty and ‘good administration’ were breached. In particular, the appellants take issue with paragraphs 190 to 211 of the judgment under appeal in which the General Court came to the conclusion that:

‘the concept of “made known to the Member State concerned”, as set out in the second subparagraph of Article 22(1) [EUMR], must be interpreted as meaning that it requires the relevant information to be actively transmitted to that Member State, enabling it to assess, in a preliminary manner, whether the conditions for a referral request under that article have been satisfied. Consequently, according to that interpretation, the period of 15 working days laid down in that provision starts to run, where notification of the concentration is not required, from the time when that information was transmitted’.

237. The appellants also take issue with paragraphs 240 and 242 to 245 of the judgment under appeal, in which the General Court found, inter alia, that (i) ‘[Illumina] was not capable of specifying, to the requisite standard, the alleged “significant factual errors” vitiating the contested decision which formerly vitiated the invitation letter and could therefore have had a decisive influence on the content of the referral request of the ACF’; and (ii) ‘[the appellants] concerned … had several opportunities to make known their views during the administrative procedure leading to the adoption of [the contested] decisions’.

1.      Arguments of the parties

238. By their second ground of appeal, the appellants contend that the General Court erred in law by (i) not deriving any legal consequence from the correct finding that the Commission took an unreasonable period of time to send the invitation letter to the Member States concerning the concentration at issue, and (ii) finding that the Commission did not breach the parties’ right of defence during the procedure leading to the adoption of the contested decisions.

239. The Commission claims that that ground of appeal is unfounded and, in part, inadmissible.

2.      Analysis

240. I am not convinced by the appellants’ arguments.

241. First, I do not think that the General Court erred in law in interpreting the terms ‘made known to the Member State concerned’ contained in the second subparagraph of Article 22(1) EUMR. As pointed out in paragraph 192 of the judgment under appeal, a comparison of the different language versions of the regulation shows that, in order to trigger the 15-working-day period, it is not enough that the merger is publicly announced in the Member State in question – for example by means of a press release or through media coverage – (176) so that the relevant authorities may become aware of it. Instead, that provision requires an active communication of the merger to those authorities. That reading seems to me to be in line with the aim of the provision, which is to enable the authorities to carry out a preliminary examination in order to assess whether, prima facie, the substantive conditions set out in the first subparagraph of Article 22(1) EUMR are satisfied. (177)

242. The appellants did not, in my view, put forward any argument which can cast any doubt on that interpretation of the second subparagraph of Article 22(1) EUMR.

243. Second, although I am not fully convinced by the legal framework applied by the General Court to determine the consequences flowing from the Commission’s failure to send the invitation letter within a reasonable period of time, I consider the conclusion reached in the judgment under appeal to be correct.

244. In my view, the crux of the matter is not whether the appellants could establish that, because of the Commission’s delay in taking action, the appellants’ rights of defence were breached. The crucial point is rather whether the appellants could provide sufficient indications that, without the procedural irregularity in question, the outcome of the procedure might have been different.

245. As I have explained in my Opinion in HSBC, the EU Courts’ case-law appears to distinguish between two main forms of procedural errors: infringements of ‘essential procedural requirements’, which automatically trigger the invalidity of the act in question, and infringements of other rules of procedure, which are subject to a ‘harmless error test’. This means that ‘ordinary’ procedural errors lead to the setting aside of the challenged act unless the error could be deemed innocuous, in the sense that it did not have, or could not have, any impact on the outcome of the procedure. Importantly, that test has, depending on the characteristics of the rule breached, been applied in three different forms: (i) breaches of a serious and structural nature giving rise to a (rebuttable) presumption that the error has influenced the outcome of the procedure, where the burden of disproving the presumption lies with the defendant; (ii) ‘standard’ errors which may or may not have influenced the outcome of the procedure, for which the applicant is to prove that, in the absence of any error, the challenged act may have been different; and (iii) irregularities of a lesser nature, which result in the annulment of the act in question, if the applicants establish that, in that absence, the outcome of the procedure would have been different. (178)

246. In the light of the text of the relevant provision (which does not provide for any specific time limit), (179) and the purpose and logic of the system of merger control set up by the EUMR (which aims at ensuring effective control of potentially anticompetitive concentrations, by means of an efficient and predictable system capable of offering legal certainty to the undertakings concerned), (180) it seems to me that the Commission’s failure to act within a reasonable period cannot be considered an infringement of an essential procedural requirement, and that the standard test for procedural errors should apply. (181)

247. Neither in the observations submitted at first instance, nor in the context of the present proceedings, have the appellants provided any concrete element capable of indicating that, had the Commission acted within a reasonable time, its assessment concerning the possibility and suitability of the merger in question being the object of a referral under Article 22 EUMR may have been different.

248. At any rate, I also agree with the Commission that the appellants failed to (i) expressly invoke a breach of their rights of defence at first instance, with the consequence that this part of the ground of appeal is inadmissible, and (ii) prove, to the requisite standard, that their ability to exercise their rights of defence during the procedure that led to the adoption of the contested decisions has been negatively affected. As regards the last point, it is true that the appellants provided a number of elements suggesting that, vis-à-vis those undertakings, the Commission may not have acted with the degree of transparency and fairmindedness that one should normally expect from public administration. (182) That is, obviously, regrettable since such forms of conduct may have an impact on the manner in which the public perceives the functioning of a service that – because of the significant powers vested in it – should consistently act with the utmost impartiality and objectivity. However, the fact remains that the Commission’s conduct did not deprive the appellants of the possibility to put forward their arguments of fact and law during the procedure initiated under Article 22 EUMR, with a view to influencing the outcome thereof.

249. For the reasons given above, the second ground of appeal should be dismissed.

C.      Third ground: principles of legitimate expectations and legal certainty

250. By their third ground of appeal – against paragraphs 254 to 260 of the judgment under appeal – Illumina and Grail criticise the General Court for rejecting Illumina’s third plea at first instance, alleging breach of the principles of the protection of legitimate expectations and legal certainty. In those passages, the General Court only assessed the arguments concerned with legitimate expectations, since the arguments concerning legal certainty had not been adequately developed.

251. As regards the principle of the protection of legitimate expectations, the General Court found that the main elements relied on by Illumina did not substantiate ‘the existence of the Commission’s alleged policy on which [Illumina relied]’, and could not be considered to constitute ‘precise, unconditional and consistent assurances from the Commission in relation to the treatment of the concentration at issue’.

1.      Arguments of the parties

252. The appellants are of the opinion that the statement of reasons in the judgment under appeal is vitiated by a number of errors of law. In particular, they argue that the General Court (i) distorted the sense of Illumina’s argument at first instance concerning legitimate expectations; (ii) erred in considering that there could only be legitimate expectations if the assurances on which those expectations were based related specifically to the concentration at issue; (iii) erred in evaluating the significance of a speech of Ms Margrethe Vestager, the Commission’s Executive Vice-President and Commissioner for Competition, delivered only a few months before the Commission sent the invitation letter; (183) and (iv) failed to address their arguments based on a breach of the principle of legal certainty.

253. The Commission responds that the General Court did not commit any error of law on this matter.

2.      Analysis

254. Again, although I find some of the relevant passages of the judgment under appeal to be unconvincing, I am of the view that the General Court did not err in dismissing Illumina’s third plea.

255. To begin with, the appellants’ allegation that the General Court mischaracterised the sense of the arguments based on legitimate expectations is unpersuasive. The appellants claim that Illumina’s argument was that the Commission had created an expectation that it would not encourage referral requests for mergers below national thresholds, whereas the General Court examined whether the Commission could lawfully accept such referrals. This appears to me to amount to ‘splitting hairs’. Clearly, the two aspects are complementary and can hardly be dissociated.

256. The thrust of the appellants’ argument was, in essence, that they could not foresee the Commission’s sudden change of policy with regard to the interpretation of Article 22 EUMR. What truly matters, in that respect, is whether the appellants could, because of the information received from the Commission, legitimately believe that their merger would not be subject to a referral under Article 22 EUMR. Whether, in that context, the referral in question was triggered by one or more NCAs acting of their own motion, or because they were invited to do so by the Commission, seems to me to be of no significance.

257. In addition, I am also unconvinced by the appellants’ arguments lamenting a failure, by the General Court, to deal with their allegations of a breach of the principle of legal certainty. Having reviewed their observations at first instance, I have to agree with the General Court that the appellants had not submitted any specific argument in that regard; in other words, none which could be distinguished from their arguments relating to legitimate expectations, which the General Court has expressly addressed in the judgment under appeal.

258. Furthermore, I do not believe that all the conditions under which a party can rely on the principle of legitimate expectations are satisfied in the present case.

259. Admittedly, certain passages of the judgment under appeal are, in my view, incorrect. In paragraph 254 of that judgment, the General Court referred to a line of case-law according to which precise, unconditional and consistent information given by the administration may give rise to legitimate expectations, provided that, inter alia, such information ‘complies with the applicable rules’. Then, in paragraph 265 of that judgment, referring to that case-law, the General Court added that ‘in so far as it is apparent from the first plea that the contested decisions were based on a correct interpretation of the scope of that article, the applicant cannot rely on the reorientation of the Commission’s decision-making practice’.

260. I cannot agree with the General Court in that respect. The case-law referred to by the General Court (which, as far as I can see, is essentially composed of its own judgments) cannot logically mean that individuals can only rely on legitimate expectations if the assurances provided by the administration comply with the relevant rules. Indeed, if the assurances are in line with the applicable law then there would be no need for the individuals in question to invoke the protection of legitimate expectations: their position would be duly protected by the very provisions referred to by the administration. The rationale of the principle of legitimate expectation is, clearly, to protect individuals which, through no fault of their own, are misled by the administration’s interpretation of the applicable law.

261. In my view, that line of case-law can only be accepted if understood as excluding the ability of individuals to rely on the principle of legitimate expectation where a reasonably circumspect individual would realise that the assurances provided by the administration are not in compliance with the relevant rules. Accordingly, if, in the case at hand, the appellants had actually received ‘precise, unconditional and consistent assurances’ from the Commission, the fact that that institution had subsequently applied Article 22 EUMR correctly could not have precluded those undertakings from relying on a breach of the principle of legitimate expectations.

262. That said, I agree with the General Court that, in any event, no such assurances can be derived from the speech of the Commissioner referred to by the appellants. As the General Court rightly pointed out, both the object of the speech (which ‘concerned the Commission’s general policy on concentrations and did not mention the concentration at issue’) (184) and the gist and tenor of it (which stated that, in the past, ‘the Commission has had a practice of discouraging national authorities from referring cases to [it] which they didn’t have the power to review themselves’) (185) exclude that such a speech could be regarded as giving rise to assurances which are ‘precise, unconditional and consistent’ within the meaning of the Court of Justice’s case-law. (186)

263. In the light of the above, the third ground of appeal should, in my view, also be rejected.

VI.    Consequences of the assessment: disposition of the present case

264. In accordance with the second sentence of the first subparagraph of Article 61 of the Statute of the Court of Justice of the European Union, the Court of Justice may, if the decision of the General Court is set aside, give final judgment in the matter where the state of the proceedings so permits.

265. In my view, that is clearly the case in the present proceedings. The General Court erred in its interpretation and application of Article 22 EUMR. Under a proper construction, that provision does not empower the Commission to adopt decisions such as those challenged by the appellants in the present proceedings. Those decisions should, thus, be annulled.

266. However, ACF’s request and the Commission’s information letter cannot be annulled because (i) the former act was not challenged at first instance (187) (quite apart from the fact that it is not an act of the EU institutions), and (ii) the information letter, albeit challenged at first instance, was found to be an act not open to challenge by the General Court. (188) The relevant passages of the judgment under appeal have not been appealed by the appellants either.

VII. Costs

267. According to Article 138(1) and Article 184(1) of the RoP, the unsuccessful party shall be ordered to pay the costs if they have been applied for in the successful party’s pleadings. Since the appellants have applied for the costs and their appeals have been successful, the Commission should be ordered to pay the costs relating to the proceedings.

268. In accordance with Article 140 and Article 184(1) of the RoP, the Member States which have intervened in the proceedings, ESA and Biocom should bear their own costs.

VIII. Conclusion

269. In the light of the foregoing, I suggest that the Court of Justice:

–        set aside the judgment of the General Court of 13 July 2022, Illumina v Commission (T‑227/21, EU:T:2022:447);

–        annul Commission Decision C(2021) 2847 final of 19 April 2021, accepting the request of the Autorité de la concurrence française to examine the concentration relating to the acquisition by Illumina, Inc. of sole control over Grail, Inc. (Case COMP/M.10188 – Illumina/Grail), Commission Decisions C(2021) 2848 final, C(2021) 2849 final, C(2021) 2851 final, C(2021) 2854 final and C(2021) 2855 final of 19 April 2021, accepting the requests of the Belgian, Dutch, Greek, Icelandic and Norwegian competition authorities to join that referral request, and the European Commission’s letter of 11 March 2021 informing Illumina and Grail of that referral request;

–        order the Commission to pay the costs of the proceedings; and

–        order the French Republic, the Kingdom of the Netherlands, the European Free Trade Association Surveillance Authority and Biocom California to bear their own costs.

1      Original language: English.

2      Opinion of Advocate General Tizzano in Commission v Tetra Laval (C‑12/03 P, EU:C:2004:318, point 73).

3      On those issues, see International Competition Network Merger Working Group Notification & Procedures Subgroup, ‘Setting Notification Thresholds for Merger Review’, April 2008 (available on the network’s website).

4      OJ 2004 L 24, p. 1.

5      Council Regulation (EEC) No 4064/89 of 21 December 1989 on the control of concentrations between undertakings (OJ 1989 L 395, p. 1). Article 22(3) to (6) thereof read as follows:

‘3.      If the Commission finds, at the request of a Member State, that a concentration … that has no Community dimension within the meaning of Article 1 creates or strengthens a dominant position as a result of which effective competition would be significantly impeded within the territory of the Member State concerned it may, in so far as the concentration affects trade between Member States, adopt the decisions provided for in Article [8(2)], second subparagraph, (3) and (4).

4.      Articles [2(1)(a) and (b)], 5, 6, 8 and 10 to 20 shall apply. … The request must be made within one month at most of the date on which the concentration was made known to the Member State or effected. …

5.      Pursuant to paragraph 3 the Commission shall take only the measures strictly necessary to maintain or restore effective competition within the territory of the Member State at the request of which it intervenes.

6.      Paragraphs 3 to 5 shall continue to apply until the thresholds referred to in Article [1(2)] have been reviewed.’

6      Regulation of 30 June 1997 amending Regulation (EEC) No 4064/89 on the control of concentrations between undertakings (OJ 1997 L 180, p. 1) (‘the 1997 Regulation’). It amended Article 22 ECMR by, inter alia: (i) introducing a reference to joint requests of two or more Member States in paragraph 3; (ii) introducing the phrases ‘Article 7 shall apply to the extent that the concentration has not been put into effect on the date on which the Commission informs the parties that a request has been made’ and ‘the request must be made within one month at most of the date on which the concentration was made known to the Member State or to all Member States making a joint request or effected’ in paragraph 4; and (iii) deleting paragraph 6.

7      EU:T:2022:447.

8      OJ 1994 L 1, p. 3; ‘the EEA Agreement’.

9      OJ 2021 C 113, p. 1.

10      See, for example, judgment of 22 June 2023, DI v ECB (C‑513/21 P, EU:C:2023:500, paragraph 47 and the case-law cited).

11      C‑457/23 P, EU:C:2023:760.

12      See, inter alia, judgment of 12 July 2022, Nord Stream 2 v Parliament and Council (C‑348/20 P, EU:C:2022:548, paragraph 128 and the case-law cited). Emphasis added.

13      See, to that effect, ibid., paragraphs 130 and 131, and, with further references, Opinion of Advocate General Bobek in Nord Stream 2 v Parliament and Council (C‑348/20 P, EU:C:2021:831, point 120).

14      Regulation of the European Parliament and of the Council of 30 May 2001 regarding public access to European Parliament, Council and Commission documents (OJ 2001 L 145, p. 43).

15      Judgment of 12 July 2022, Nord Stream 2 v Parliament and Council (C‑348/20 P, EU:C:2022:548, paragraph 129 and the case-law cited).

16      See, for example, order of 12 June 2019, OY v Commission (C‑816/18 P, EU:C:2019:486, point 6 of the Advocate General’s Opinion as quoted in paragraph 4 of the order, and the case-law cited).

17      See, in fact, paragraphs 9 and 10 of the abovementioned order.

18      See, again, Opinion of Advocate General Bobek in Nord Stream 2 v Parliament and Council (C‑348/20 P, EU:C:2021:831, point 177 and the case-law cited).

19      See, for example, judgments of 6 September 2017, Intel v Commission (C‑413/14 P, EU:C:2017:632, paragraph 86), and of 25 March 2021, Xellia Pharmaceuticals and Alpharma v Commission (C‑611/16 P, EU:C:2021:245, paragraph 153).

20      See, by analogy, judgment of 30 September 2021, Court of Auditors v Pinxten (C‑130/19, EU:C:2021:782, paragraphs 310 and 311 and the case-law cited).

21      Paragraph 177 of the judgment under appeal. Emphasis added.

22      Paragraphs 89, 90 and 92 of the judgment under appeal.

23      Paragraph 91 of the judgment under appeal (emphasis added). That is true with regard to the majority of the language versions of the regulation; indeed, a minority of language versions (such as the Dutch and Swedish ones) do not use the expression ‘any concentration’ but others terms which could be translated as ‘a concentration’.

24      Paragraph 93 of the judgment under appeal.

25      Paragraphs 94 and 95 of the judgment under appeal.

26      Judgment of 6 October 1982, Cilfit and Others (283/81, EU:C:1982:335, paragraph 20). Emphasis added.

27      See, among many others, judgment of 6 October 2020, Jobcenter Krefeld (C‑181/19, EU:C:2020:794, paragraph 61 and the case-law cited). Emphasis added.

28      Ibid., paragraphs 62 to 66.

29      See, for instance, judgments of 19 November 2009, Sturgeon and Others (C‑402/07 and C‑432/07, EU:C:2009:716, paragraphs 40 to 69), and of 27 October 2016, Commission v Germany (C‑220/15, EU:C:2016:815, paragraphs 38 to 47).

30      (1969) UNTS Vol. 1155, p. 331. See Article 31 and Article 32, respectively.

31      Emphasis added.

32      As Advocate General Wathelet stated, the Court may stop at the literal interpretation of the provision, where the text in question is absolutely clear and unambiguous, but it need not stop (see, in that respect, Opinion in France v Parliament, C‑73/17, EU:C:2018:386, point 25 and the case-law cited).

33      See, among others, the terms ‘postoupení’ (Czech), ‘Verweisung’ (German), ‘παραπομπή’ (Greek), ‘remisión’ (Spanish), ‘renvoi’ (French), ‘áttétel’ (Hungarian), ‘rinvio’ (Italian), ‘remessa’ (Portuguese) and ‘napotitev’ (Slovenian).

34      Emphasis added.

35      Paragraph 142 of the judgment under appeal (emphasis added). See also paragraphs 141, 165, 177 and 182 of that judgment.

36      Commission Green Paper of 31 January 1996 on the review of the Merger Regulation (COM(96) 19 final), referred to in paragraphs 97 and 98 of the judgment under appeal.

37      Commission Green Paper of 11 December 2001 on the Review of Council Regulation (EEC) No 4064/89 (COM(2001) 745 final), referred to in paragraphs 97, 99, 101 and 103 of the judgment under appeal.

38      Commission Proposal for a Council Regulation on the control of concentrations between undertakings (‘The EC Merger Regulation’) (OJ 2003 C 20, p. 4), referred to in paragraphs 97 and 106 to 113 of the judgment under appeal.

39      Commission Staff Working Paper accompanying the Communication from the Commission to the Council – Report on the functioning of Regulation No 139/2004 of 30 June 2009 (SEC(2009) 808 final/2), referred to in paragraphs 97 and 115 of the judgment under appeal.

40      Unfortunately, the General Court did not take an express position on this point, nor can that position be deduced by examining the statement of reasons in the judgment under appeal. Indeed, the General Court referred indiscriminately to elements present in the original ECMR and to elements added subsequently by the 1997 Regulation or by the EUMR.

41      Paragraph 97 of the judgment under appeal.

42      Paragraph 98 of the judgment under appeal.

43      Paragraph 102 of the judgment under appeal.

44      In particular, in 1997 when the provision was amended. See paragraph 103 of the judgment under appeal.

45      Paragraph 109 of the judgment under appeal.

46      See, inter alia, judgments of 18 July 2007, Industrias Químicas del Vallés v Commission (C‑326/05 P, EU:C:2007:443, paragraphs 60 to 68), and of 30 May 2017, Safa Nicu Sepahan v Council (C‑45/15 P, EU:C:2017:402, paragraph 76).

47      Paragraph 99 of the judgment under appeal.

48      That is still correct at the time this Opinion was being written. However, I understand that the situation may change in the near future, since the Luxembourgish Government introduced, in August 2023, a bill of law establishing a merger control system in the country.

49      Paragraphs 9 and 10 of the 1996 Green Paper (emphasis added).

50      Paragraph 86 of the 2001 Green Paper, referred to in paragraph 103 of the judgment under appeal.

51      By ‘voluntary’ notifications, it referred to those filed with the United Kingdom competition authority since the United Kingdom (which, at that time, was an EU Member State) operates a merger control regime which, unlike both the EU and the other Member States’ regimes, is not based on compulsory notifications but on voluntary ones.

52      See, in particular, page 4 (‘Executive summary’) and paragraphs 72 to 88 of the 2001 Green Paper.

53      Paragraph 53 (emphasis added).

54      This point is further stressed by the reference in the 2001 Green Paper to the lack of definition of ‘making a concentration known to a Member State’ but that ‘it appears natural to use the date of a national notification as the triggering date in Member States where a notification requirement exists’. Again, the Commission is clearly concerned with cases that come to the attention of the national authorities because they are subject to their national merger control systems.

55      See, in particular, paragraphs 93, 95 and 99 of the 2001 Green Paper.

56      Emphasis added.

57      See paragraph 138 of the 2009 Staff Working Paper: ‘on the issue of whether or not a Member State should be able to make or join a referral without having jurisdiction in the case, five thought that it should be allowed while nine thought that it should not. This does raise the question of whether or not a Member State should be able to refer a case when its jurisdiction is not triggered but where the activity of the parties does have an effect in that Member State’ (emphasis added).

58      See paragraphs 133, 140 to 142 and 144 of the 2009 Staff Working Paper (emphasis added). See, similarly, paragraph 86 of the 2001 Green Paper.

59      See, in particular, Council reports of 7 November 1988 (9114/88), 10 November 1988 (9265/88) and 8 December 1988 (10054/88).

60      Draft minutes of the 1 339th Council meeting on 18 July 1989 (8016/89 PV/ CONS 47), p. 2.

61      In primis, EEC Council Regulation No 17 of 6 February 1962: First Regulation implementing Articles 85 and 86 of the [EEC] Treaty (OJ, English Special Edition 1959-1962 (I), p. 87).

62      Commission Amended proposal for a Council Regulation (EEC) on the control of concentrations between undertakings (COM(88) 97 final) (OJ 1988 C 130, p. 4). Article 22 of that proposal, entitled ‘Exclusive application of this Regulation’, read as follows: ‘Regulations No 17, (EEC) No 1017/68, (EEC) No 4056/86 and (EEC) No 3975/87 shall not apply to concentrations falling within the scope of this Regulation’.

63      See Council, Note de la Présidence au Conseil, 7 April 1989 (5857/89 (RC 9)), annex, p. 4; reports of 12 April 1989 (6267/89, RC 12); draft minutes of the 1 339th Council meeting on 18 July 1989 (8016/89 PV/ CONS 47), p. 13; reports of 9 November 1989, (9672/89 (RC 41)), p. 3. See, also, letter of Sir Leon Brittan to the Council (SG (89) D/5429) of 24 April 1989, p. 2.

64      See Report from the Commission to the Council on the implementation of the Merger Regulation of 28 July 1993 (COM(93) final, p. 14) (‘the 1993 Report’); Commission, Note from G. Drauz to the Legal Service (COMP/HT.60), Council Working Group, 6 June 2003 (11430), paragraph 4.

65      See Sir Leon Brittan, Competition Policy and Merger Control in the Single European Market, Grotius, 1991, pp. 33 and 49. Similarly, Jones, C., ‘Procedures and Enforcement under EEC Merger Regulation, in Hawk, B. (ed.), Annual Proceedings of the Fordham Corporate Law Institute, 1990, p. 476.

66      See Commission, the 1993 Report, p. 7. See, also, Communication from the Commission to the Council and to the European Parliament regarding the revision of the Merger Regulation (COM(96) 313 final, p. 5). See, also, Levy, N., Rimsa, A. and Buzatu, B., ‘The jurisdictional reach of EC merger control: Striking the right balance’, in Kokkoris, I. and Levy, N., Research Handbook on Global Merger Control, Edward Elgar Publishing, 2023, p. 219: ‘No workable system of merger control can capture every transaction capable of affecting competition in a given jurisdiction’.

67      See Council, Résultats des travaux du Groupe des questions économiques (contrôle des concentrations), 8 March 1989 (5770/89 RC 8) p. 4. See, also, letter of Sir Leon Brittan to the Council, 30 March 1989 (SG (89) D/4008), p. 2.

68      See Council, Report to the Committee of Permanent Representatives, 9 December 1988, (10189/89 RC 36), p. 8; and avis du service juridique, 11 July 1989 (7896/89 JUR 98 RC 24), p. 10. See also Commission, the 1993 Report, p. 14.

69      See Sir Leon Brittan, footnote 65, op. cit., pp. 39, 48 and 53.

70      See Council, avis du service juridique, 11 July 1989 (7896/89 JUR 98 RC 24), p. 4.

71      See points 201 and 208 of this Opinion.

72      That number reflects all of the EU Member States (with the exception of Luxembourg) and the EEA/EFTA States (Iceland and Norway) that currently have a national merger control regime.

73      On the contrary, the adoption of the EUMR was intended to enhance the assets of the ECMR. See Commission, Note from G. Drauz to the Legal Service (COMP/HT.60), Council Working Group, 6 June 2003 (11430), p. 7; and the 2003 Commission Proposal, p. 10.

74      See above, footnotes 5 and 6.

75      OJ 2005 C 56, p. 2.

76      See, in particular, paragraphs 33, 45, 47 and 50 of the notice.

77      See paragraph 45 of the notice.

78      Paragraphs 2 and 79 of the 2014 White Paper.

79      Paragraphs 69 and 70 of the 2014 White Paper.

80      Paragraph 61 of the 2014 White Paper.

81      Paragraphs 21, 63 and 69 of the 2014 White Paper.

82      Available on the Commission website. See, in particular, Sections A.1, B.2 and B.3.

83      Paragraph 139 of the judgment under appeal.

84      Previously Articles 87 and 235 EEC.

85      Paragraphs 119 and 120 of the judgment under appeal.

86      See seventh recital of the ECMR and recital 7 of the EUMR.

87      See, inter alia, Council, Résultats des travaux du Groupe des questions économiques (contrôle des concentrations), 29 May 1989 (7752/89 RC 20), p. 5; Résultats des travaux du Groupe des questions économiques (contrôle des concentrations), 22 June 1989 (7827/89 RC 22), p. 1, Annex II, p. 3; and avis du service juridique, 11 July 1989 (7896/89 JUR 98 RC 24), p. 4.

88      Then Article 38 EEC and Annex II to the EEC Treaty.

89      In essence, Article 352(1) TFEU permits the Council to adopt appropriate measures where an action by the Union proves necessary, within the framework of the policies defined in the Treaties, to attain one of the objectives set out in the Treaties, whilst the Treaties have not provided the necessary powers.

90      On this issue, see, also, Dashwood, A., Community Report, XIVth FIDE Congress, Madrid, 2010.

91      Paragraphs 121 to 124 of the judgment under appeal. Emphasis added.

92      Paragraphs 125 and 126 of the judgment under appeal.

93      Article 4(4) EUMR permits the parties to a merger to ask the Commission to refer the examination, in whole or in part, of a concentration that has a Community dimension to the authorities of a Member State where that concentration ‘may significantly affect competition in a market within a Member State which presents all the characteristics of a distinct market’. In turn, Article 9 EUMR enables the Commission, in certain circumstances, to refer a concentration notified to it to the competent authorities of the Member States concerned.

94      Paragraphs 127 to 129 of the judgment under appeal.

95      Paragraph 130 of the judgment under appeal.

96      Emphasis added.

97      Paragraph 130 of the judgment under appeal. Emphasis added.

98      Emphasis in those provisions added.

99      See points 100, 103 and 105 of this Opinion. As regards, more specifically, the EUMR, see also recital 12 thereof. This evolution, also due to the progressively smaller scope for the use of the referral mechanism, has been stressed in legal scholarship: see, for instance, Albors‑Llorens, A., Goyder, D.G. and Goyder, J., Goyder’s EC Competition Law, 5th edition, Oxford University Press, 2009, p. 431; and Frenz, W., Handbook of EU Competition Law, Springer, 2016, p. 1308.

100      Paragraph 131 of the judgment under appeal.

101      Paragraph 132 of the judgment under appeal. Emphasis added.

102      Emphasis added. I shall come back to this issue in points 155 and 156 of this Opinion.

103      Indeed, the procedure has been validly triggered and, if anything, a joint referral request by several Member States enhances the consistency of the system: if the request is accepted, all the Member States concerned can ‘no longer apply their national legislation on competition to the concentration [in question]’ (Article 22(3) EUMR), including their provisions on anticompetitive agreements and abuses of dominance to the merger in question. See, on the latter point, paragraph 134 of the judgment under appeal.

104      See, extensively, points 206 to 214 of this Opinion.

105      Paragraph 133 of the judgment under appeal.

106      But see points 152 to 162 of this Opinion.

107      Paragraph 134 of the judgment under appeal.

108      Paragraphs 135 and 136 of the judgment under appeal.

109      Indeed, (i) if a Member State without a national merger control system submits a referral request, the standstill obligation laid down in Article 7 EUMR applies to the merger in question regardless of whether that merger falls within the scope of one or more other national merger control systems, and (ii) if a Member State submits a referral request, the standstill obligation laid down in Article 7 EUMR applies to the merger in question by virtue of the ECMR, and thus regardless of whether that Member State’s legislation provides for an equivalent obligation.

110      As ESA rightly pointed out in its submissions, the EUMR is an act that is, under Article 57 of the EEA Agreement, also applicable in the ‘EEA EFTA States’ (Iceland, Liechtenstein and Norway); Liechtenstein does not have a national merger control regime.

111      That is notoriously the case of the United Kingdom, which was still a Member State of the European Union when the ECMR and the EUMR were adopted.

112      For a good overview on those specific aspects of the system, see the ‘Merger Notification and Procedures Templates’ submitted by many EU Member States to the International Competition Network (available on the network’s website).

113      Paragraph 137 of the judgment under appeal.

114      Judgment of 15 December 1999, Kesko v Commission (T‑22/97, EU:T:1999:327, paragraph 84).

115      Paragraph 138 of the judgment under appeal.

116      See points 129 to 133 of this Opinion.

117      See, in particular, the third subparagraph of Article 22(2) EUMR: ‘All national time limits relating to the concentration shall be suspended’. Emphasis added.

118      See, again, the third subparagraph of Article 22(2) EUMR: ‘until, in accordance with the procedure set out in this Article, it has been decided where the concentration shall be examined’. Emphasis added.

119      Emphasis added.

120      As regards the third subparagraph of Article 22(2) EUMR, see paragraphs 133 and 150 of the judgment under appeal. However, the General Court deals only with the terms ‘all national time limits’ and not with the terms ‘until … decided where the concentration shall be examined’. As regards recital 15 (‘also competent’), see paragraphs 149 to 151 of the judgment under appeal.

121      Emphasis added.

122      See also the similar expressions in, for example, the German (‘für’), Greek (‘για λογαριασμό’), Spanish (‘en nombre de’), French (‘au nom d[e]’), and Italian (‘per conto di’) versions of the regulation. Emphasising that the Commission appears to act under a sort of delegation of the powers held by the relevant national authority: Cohen-Tanugi, C., et al., La pratique communautaire du contrôle des concentrations, De Boeck Université, 1995, p. 56. Similarly, Sir Leon Brittan, footnote 65, op. cit., p. 52.

123      Emphasis added. The provision was only subject to a minor amendment in 1997 and then repealed by the EUMR, since it was no longer consistent with the new one-stop-shop function of Article 22 EUMR. See Cook, J. and Kerse, C., EC Merger Control, 5th edition, Sweet&Maxwell, 2005, p. 343.

124      The fact that the limited powers of the Commission implied a limited scope for the referral mechanism set out in then Article 22 ECMR was emphasised by, for example, Cook, J. and Kerse, C., EEC Merger Control – Regulation 4064/89, 1st edition, Sweet&Maxwell, 1991, pp. 60 and 61.

125      Indeed, amending the EUMR would normally require unanimity (because of the legal basis under Article 352 TFEU), but Article 1(5) EUMR allows the Council to amend the thresholds ‘acting by a qualified majority’.

126      See also recital 9 of the EUMR which states ‘… The Commission should report to the Council on the implementation of the applicable thresholds and criteria so that the Council, acting in accordance with Article 202 of the Treaty, is in a position to review them regularly … in the light of the experience gained; this requires statistical data to be provided by the Member States to the Commission to enable it to prepare such reports and possible proposals for amendments. The Commission's reports and proposals should be based on relevant information regularly provided by the Member States’ (emphasis added). In the light of that recital, I understand Article 1(4) and (5) of the EUMR as allowing for the use of the simplified procedure at any point following the submission of the report due by 1 July 2009. However, I acknowledge that the language of the provision leaves room for ambiguity, which could lead one to believe that the simplified procedure was only applicable for the amendments proposed following the adoption of the 2009 report. Nevertheless, even setting aside the specific wording of recital 9 of the EUMR, the idea that this provision is applicable only once appears illogical. Indeed, with the passage of time, the necessity for adjusting the thresholds becomes even more apparent.

127      Emphasising the temporary nature of the mechanism, Downes, T.A. and Ellison, J., The legal control of mergers in the EC, Blackston, 1991, pp. 63 to 65.

128      Sir Leon Brittan, footnote 65, op. cit., p. 42: ‘This provision is narrowly defined and would not permit the Commission to deal with mergers below the threshold on a general basis, even if it were inclined to evade the spirit of the threshold provision in this way’ (emphasis added). See, also, Ibid., ‘The Law and Policy of Merger Control in the EEC’, European Law Review, 1990, p. 245.

129      See, in particular, recital 10 of the 1997 Regulation and recitals 11, 12 and 14 of the EUMR.

130      As stated in paragraph 142 of the judgment under appeal.

131      Emphasis added.

132      Emphasis added.

133      See, inter alia, judgment of 4 March 2020, Marine Harvest v Commission (C‑10/18 P, EU:C:2020:149, paragraph 108 and the case-law cited). See also judgment of 12 December 2012, Electrabel v Commission (T‑332/09, EU:T:2012:672, paragraph 246).

134      That recital read: ‘Whereas a new legal instrument should therefore be created in the form of a Regulation to permit effective monitoring of all concentrations from the point of view of their effect on the structure of competition in the Community and to be the only instrument applicable to such concentrations’.

135      According to that provision, notwithstanding the Commission’s exclusive competence to review the mergers falling under the scope of the EUMR, ‘Member States may take appropriate measures to protect legitimate interests other than those taken into consideration by this Regulation and compatible with the general principles and other provisions of Community law’.

136      See, to that effect, judgment of 7 September 2017, Austria Asphalt (C‑248/16, EU:C:2017:643, paragraph 21). See, also, judgments of 25 March 1999, Gencor v Commission (T‑102/96, EU:T:1999:65, paragraph 314), and of 22 September 2021, Altice Europe v Commission (T‑425/18, EU:T:2021:607, paragraph 299).

137      See judgment of 22 June 2004, Portugal v Commission (C‑42/01, EU:C:2004:379, paragraph 50), and recital 8 of the EUMR.

138      See recital 9 of the EUMR.

139      See, in particular, paragraph 140 of the judgment under appeal.

140      Emphasis added. In fact, legal scholarship also referred to the ECMR as an instrument pursuing several objectives: see, for example, Navarro Varona et al., Merger Control in the EU: Law, Economics and Practice, 1st edition, Oxford University Press, 2001, pp. 1 to 5.

141      See the references to the principle of subsidiarity in recitals 6, 8, 11 and 14. See also recital 8 in fine: ‘concentrations not covered by this Regulation come, in principle, within the jurisdiction of the Member States’.

142      See the references to the one-stop-shop principle in recitals 8 and 11, to the Commission’s ‘exclusive competence’ in recital 17 and to the ensuing limits for Member States’ action in recitals 18 and 19.

143      See the references to efficiency in recitals 14, 15 and 16, to predictability in recital 15, and to legal certainty in recitals 11, 25 and 34. See also the 1996 Green Paper, paragraph 29. In legal scholarship, see, inter alia, Blaise, J.B., ‘Concurrence – Contrôle des opérations de concentration’, Revue trimestrielle de droit européen, 1990, p. 743; and Venit, J., ‘The “merger” control regulation: Europe comes of age … or Caliban’s dinner’, Common Market Law Review, 1990, p. 44.

144      Similarly, Whish, R. and Bailey, D., Competition Law, 8th edition, Oxford University Press, 2018, pp. 832 and 833.

145      See, generally, Irarrazabal Philippi, F., ‘Merger control procedure’, Global Dictionary of Competition Law, Concurrences, Art. N° 12342.

146      Judgment of 22 June 2004, Portugal v Commission (C‑42/01, EU:C:2004:379, paragraphs 51 and 53). See, also, Opinion of Advocate General Kokott in Cementbouw Handel & Industrie v Commission (C‑202/06 P, EU:C:2007:255, point 44).

147      See judgment of 22 June 2004, Portugal v Commission (C‑42/01, EU:C:2004:379, paragraph 50). See also recital 8 (‘concentrations not covered by this Regulation come, in principle, within the jurisdiction of the Member States’) and recital 9 (‘the scope of application of this Regulation should be … limited by quantitative thresholds in order to cover those concentrations which have a Community dimension’).

148      See above, footnote 141.

149      Sir Leon Brittan, ‘Subsidiarity in the Constitution of the EC’, Robert Schuman Lecture, European University Institute, 1992, p. 12.

150      Article 5(3), first paragraph, TEU: ‘Under the principle of subsidiarity … the Union shall act only if and in so far as the objectives of the proposed action cannot be sufficiently achieved by the Member States, either at central level or at regional and local level, but can rather, by reason of the scale or effects of the proposed action, be better achieved at Union level’.

151      See Commission Implementing Regulation (EU) 2023/914 of 20 April 2023 implementing Council Regulation (EC) No 139/2004 on the control of concentrations between undertakings and repealing Commission Regulation (EC) No 802/2004 (OJ 2023 L 119, p. 22).

152      Regulation (EU) 2022/1925 of the European Parliament and of the Council of 14 September 2022 on contestable and fair markets in the digital sector and amending Directives (EU) 2019/1937 and (EU) 2020/1828 (OJ 2022 L 265, p. 1).

153      Judgment of 13 July 2023, Commission v CK Telecoms UK Investments (C‑376/20 P, EU:C:2023:561, paragraph 72 and the case-law cited).

154      As stressed by all scholarship; see, for example, Bushell, G., ‘Chapter II’, in Jones, C. and Weinert, L. (eds), EU Competition Law, Vol. II, Book One, Edward Elgar Publishing, 2021, p. 41.

155      See, recently, judgment of 22 November 2022, Commission v Council (Accession to the Geneva Act) (C‑24/20, EU:C:2022:911, paragraph 83).

156      That is a fortiori true if one takes the view, as I do, that the simplified procedure provided for in Article 1(4) and (5) EUMR for the amendment of those thresholds is still applicable. See footnotes 125 and 126 above.

157      See, among many others, judgment of 24 November 1992, Poulsen and Diva Navigation (C‑286/90, EU:C:1992:453, paragraph 28).

158      Judgments of 6 September 2017, Intel v Commission (C‑413/14 P, EU:C:2017:632, paragraphs 40 to 47), and of 25 March 1999, Gencor v Commission (T‑102/96, EU:T:1999:65, paragraph 243).

159      See Opinion of Advocate General Darmon in Joined Cases Ahlström Osakeyhtiö and Others v Commission (89/85, 104/85, 114/85, 116/85, 117/85 and 125/85 to 129/85, EU:C:1988:258, point 57).

160      See, for example, as regards competition law, Opinions of Advocate General Wathelet in InnoLux v Commission (C‑231/14 P, EU:C:2015:292, points 39 to 42), and of Advocate General Wahl in Intel Corporation v Commission (C‑413/14 P, EU:C:2016:788, points 283 and 300); and, in another context, Opinion of Advocate General Szpunar in Nikiforidis (C‑135/15, EU:C:2016:281, point 88).

161      See, for example, Article I.2(b) and Article IV of the Agreement between the European Communities and the Government of the United States of America on the application of positive comity principles in the enforcement of their competition laws of 4 June 1998 (OJ 1998 L 173, p. 28).

162      See, in particular, Opinion of the United States Supreme Court, F. Hoffmann-La Roche, Ltd. v. Empagran S.A. (124 S. Ct. 2359 (2004)).

163      See, inter alia, judgment of 17 December 2020, Centraal Israëlitisch Consistorie van België and Others (C‑336/19, EU:C:2020:1031, paragraph 85). As regards the application of that principle in the present context, see mutatis mutandis paragraph 236 of the judgment under appeal.

164      As Korah, V. stated, it can be very expensive for businesses to deal with, and supply info to, several authorities, in various languages and in different forms within differing – but in any event short – time periods (see An Introductory Guide to EC Competition Law and Practice, 8th edition, Hart, 2004, p. 356).

165      Judgment of 16 March 2023 (C‑449/21, EU:C:2023:207).

166      Judgment of 21 December 2023 (C‑333/21, EU:C:2023:1011, paragraph 131).

167      See, mutatis mutandis, United States Department of Justice and the Federal Trade Commission, Horizontal Merger Guidelines, 2010, Section 6.4.

168      See, with references, Organisation for Economic Co-operation and Development (OECD), ‘Disentangling Consummated Mergers: Experiences and Challenges’, Competition Policy Roundtable Background Note, 2022.

169      On this topic, see, for example, Ginsburg, D.H. and Wong-Ervin, K.W., ‘Challenging Consummated Mergers Under Section 2’, Competition Policy International, May 2020.

170      Directive (EU) 2019/1 of the European Parliament and of the Council of 11 December 2018 to empower the competition authorities of the Member States to be more effective enforcers and to ensure the proper functioning of the internal market (OJ 2019 L 11, p. 3). On that directive, see generally Arsenidou, E., ‘The ECN+ Directive’, in Dekeyser, K. et al. (eds), Regulation 1/2003 and EU Antitrust Enforcement – A Systematic Guide, Wolters Kluwer, 2023, pp. 143 to 149.

171      Cf. Opinion of Advocate General Kokott in Towercast (C‑449/21, EU:C:2022:777, point 63).

172      On interim measures, see, recently, OECD, ‘Interim Measures in Antitrust Investigations’, Competition Policy Roundtable Background Note, 2022.

173      See, for example, judgment of 28 October 2022, Generalstaatsanwaltschaft München (Extradition and ne bis in idem) (C‑435/22 PPU, EU:C:2022:852, paragraph 119 and the case-law cited).

174      Judgment of 3 April 2003, Royal Philips Electronics v Commission (T‑119/02, EU:T:2003:101, paragraph 354).

175      See paragraph 182 of the judgment under appeal. The meaning of that passage remains rather obscure to me.

176      See, in that regard, paragraph 203 of the judgment under appeal.

177      See, in that regard, paragraph 199 of the judgment under appeal.

178      See, with further references, my Opinion in HSBC Holdings and Others v Commission (C‑883/19 P, EU:C:2022:384, points 38 to 59).

179      See, in that regard, paragraph 221 of the judgment under appeal.

180      See, in that regard, paragraph 226 of the judgment under appeal.

181      See, by analogy, the EU Courts’ case-law referred to in paragraph 240 of the judgment under appeal.

182      It is, in particular, difficult to understand why the appellants were contacted by the Commission, and subsequently informed of its concerns, almost three months after the Commission received a complaint about the merger, despite the Commission having – throughout that period – numerous exchanges with the complainant, several NCAs, other Member States’ authorities and the Competition and Markets Authority.

183      Speech entitled ‘The Future of EU Merger Control’, delivered at the International Bar Association’s 24th Annual Competition Conference on 11 September 2020.

184      See, in that regard, paragraph 260 of the judgment under appeal.

185      See, in that regard, paragraph 261 of the judgment under appeal.

186      See, for example, judgments of 20 May 2021, Riigi Tugiteenuste Keskus (C‑6/20, EU:C:2021:402, paragraph 49), and of 31 March 2022, Smetna palata na Republika Bulgaria (C‑195/21, EU:C:2022:239, paragraph 65).

187      See, in that regard, paragraph 62 of the judgment under appeal.

188      See, in that regard, paragraphs 79 and 80 of the judgment under appeal.