Société Coopérative De Production Seafrance S.A. v Competition and Markets Authority & Anor [2015] EWCA Civ 487

Case No C3/2015/0226

[2015] EWCA Civ 487

IN THE COURT OF APPEAL (CIVIL DIVISION)

ON APPEAL FROM

THE COMPETITION APPEAL TRIBUNAL

Roth J., President, Professor John Beath, Joanne Stuart OBE)

123541214

Venue Royal Courts of Justice, Strand, London, WC2A 2LL

Date Friday 15 th May 2015

Before

LADY JUSTICE ARDEN

LORD JUSTICE TOMLINSON

SIR COLIN RIMER

Between

Société Coopérative de Production SeaFrance S.A.Appellant

Competition and Markets Authority1 st Respondent

DFDS A/S2 nd Respondent

(Transcript of the Handed Down Judgment of

WordWave International Limited

A Merrill Communications Company

165 Fleet Street, London EC4A 2DY

Tel No: 020 7404 1400, Fax No: 020 7831 8838

Official Shorthand Writers to the Court)

Mr Daniel Beard QC and Mr Rob Williams (instructed by Reynolds Porter Chamberlain LLP) for Appellant

Mr Paul Harris QC and Mr Ben Rayment (instructed by The Competition and Markets Authority) for First Respondent

Mr Meredith Pickford QC and Ms Ligia Osepciu (instructed by Hogan Lovells International LLP) for Second Respondent

Hearing Hearing dates: 11-12 March 2015

Judgment

[1]Société Coopérative de Production SeaFrance S.A. (which I shall call “the SCOP”), a company incorporated in France as a workers’ co-operative, appeals against the dismissal by the Competition Appeal Tribunal (Mr Justice Roth, Professor John Beath and Joanne Stuart OBE) (“CAT 2”) of its application under section 120 of the Enterprise Act 2002 (“EA02”) for judicial review of the decision of the CMA that its merger control powers extended to the acquisition of certain assets of the former cross-Channel ferry operator, SeaFrance SA (“SeaFrance”). In a nutshell, the question was whether part of the activities of the business of SeaFrance, or just a collection of its assets, was acquired. The SCOP contends that the CMA erred in law in concluding that the statutory conditions for its intervention had arisen. The challenge is not, therefore, to the recommendations of the CMA but to the technical question of statutory jurisdiction. In addition, no-one contests the jurisdiction of the CMA on territorial grounds.

Summary of the appeal and my conclusion

[2]I shall have to consider the matters in detail, but this summary is intended to provide an overview of the rest of the judgment.

Summary of the background and this appeal

[3]The control of mergers in the UK is the responsibility by the Competition and Markets Authority (“CMA”), the first respondent to these proceedings. There are three stages. The responsibility for these stages was formerly divided between the Office of Fair Trading (“OFT”) and the Competition Commission, but the CMA is now responsible for all of them by virtue of changes made by the Enterprise and Regulatory Reform Act 2013. Those three stages may be described as follows. First, the CMA satisfies itself that certain conditions are fulfilled. One of these conditions is that the CMA believes that there is or may be a “relevant merger situation” (section 22(1) of the Enterprise Act 2002 (“EA02”)) (The expression “relevant merger situation" is summarised in the next paragraph and defined in paragraph 14, below.) Second, the CMA must make a reference (formerly to the Competition Commission but now to itself) where it considers that the merger situation may lead to a substantial lessening of competition in the UK (section 22(1) EA02). Third, the Competition Commission must carry out an inquiry: it must decide if a relevant merger situation has occurred and whether it may result in a substantial lessening of competition (section 35 EA02). In the present case, all these stages have been completed. This appeal is concerned only with one of the questions which arises at the third stage, namely whether a relevant merger situation has been created.

[4]Under the EA02, a relevant merger situation arises where two or more enterprises cease to be distinct or there is a proposal that they should do so. That means that they must be brought under common control or ownership. The definition of “enterprise” is important: it is defined by section 129(1) EA02 as "the activities, or part of the activities, of a business". A “business” is widely defined by the same provision as including “a professional practice and ... any other undertaking which is carried on for gain or reward or which is an undertaking in the course of which goods or services are supplied otherwise than free of charge”. The importance for this appeal is that part of a business’s activities must be acquired. It is not enough that just assets (“bare assets”) are acquired. There can be, however, a narrow dividing line between these two situations. The transfer of goodwill or intellectual property or human resources (formally or informally) can tip the balance and lead the competition authority to the conclusion that, in reality, part of a business was acquired. The Competition Appeal Tribunal (“the CAT”) may be asked to intervene but it can only do so applying the principles of judicial review.

[5]In this case, the relevant business was that of SeaFrance which operated passenger and freight ferry services between Calais and Dover. It had modern, specially designed vessels for this purpose, including the SeaFrance Rodin and the SeaFrance Berlioz. It went into administration in June 2010, and then into liquidation in November 2011. It ceased to trade in January 2012. Its remaining employees were then made redundant. Eventually Groupe Eurotunnel SA (“GET”) acquired its ferries. It also acquired customer lists and so on. GET and the SCOP acted as associates for the purpose of this bid. Many of SeaFrance’s former employees applied for positions with the SCOP, which entered into a contract with GET to operate the ferries and provide the employees.

[6]On 6 June 2013, the Competition Commission issued a report (“the Competition Commission Report”) in which it concluded that there was a merger situation but this conclusion was challenged before the Competition Appeal Tribunal (Marcus Smith QC, Heriot Currie QC and Dermont Glyn)

[2013] CAT 30

(“CAT 1”). On 4 December 2013, CAT 1 directed that the matter was remitted back to the Competition Commission for further consideration of whether it had jurisdiction. The Competition Commission’s functions were taken over by the CMA on 1 April 2014.

[7]In its final report, entitled Eurotunnel/SeaFrance merger inquiry remittal – Final decision on the question remitted to the Competition and Markets Authority by the Competition Appeal Tribunal on 4 December 2013 and consideration of possible material change of circumstances under section 41(3) (“the Remittal Report”), dated 27 June 2014, the CMA decided that it had jurisdiction.

[8]The SCOP made a further application to the CAT challenging this conclusion.

[9]By order dated 9 January 2015, the Competition Appeal Tribunal rejected the SCOP’s application to set aside the report of the CMA. The SCOP appeals against that decision to this court.

Summary of conclusion

[10]In this judgment, I consider in detail the arguments of the SCOP, but I conclude for the reasons given below that it has not shown that the CMA came to a decision which was irrational or wrong in law. The CMA could rationally take the view that, even though SeaFrance has been placed in liquidation, and even though its employees have been declared redundant, GET/SCOP acquired its business. The CMA made some errors in the way it described the events but the conclusion which it reached was inevitable. This summary must of course be read with my detailed reasons below.

Structure of this judgment

[11]This judgment is organised as follows: 1. Framework for UK competition authorities’ merger control (paras. 12 to 22) 2. Facts said to give rise to a merger situation (paras 23 to 29) 3. Position of the L’Autorité de la concurrence (French regulator) (para. 30) 4. Competition Commission Report, Eurotunnel 1 and remittal (paras. 31 to 44) 5. CMA’s Remittal Report (paras. 45 to 59) 6. Decision of CAT 2 (under appeal) (paras. 60 to 71) 7. Arguments on this appeal (paras. 72 to 109) 8. Discussion and Conclusions (paras. 110 to 128) Annex: MMC report in AAH Holdings/Medicopharma

[1]Framework for UK competition authorities’ merger control

[12]The CMA (as successor to the Office of Fair Trading which was named in the legislation as originally enacted) has to make a decision as to whether a relevant merger situation has been created. Thus (using the legislation as now in force), section 35(1) EA02 provides: (1) Subject to subsections (6) and (7) and section 127(3), the [CMA] shall, on a reference under section 22, decide the following questions— (a) whether a relevant merger situation has been created; and (b) if so, whether the creation of that situation has resulted, or may be expected to result, in a substantial lessening of competition within any market or markets in the United Kingdom for goods or services. (a) whether a relevant merger situation has been created; and (b) if so, whether the creation of that situation has resulted, or may be expected to result, in a substantial lessening of competition within any market or markets in the United Kingdom for goods or services.

[13]Section 35(1) therefore requires a division whether a relevant merger situation has in fact been created. The material wording appeared in the section as originally enacted.

[14]The relevant part of the definition of “relevant merger situation” reads as follows: (2) For the purposes of this Part, a relevant merger situation has also been created if— (a) two or more enterprises have ceased to be distinct enterprises at a time or in circumstances falling within section 24; and… (EA02, section 23(2)(a)) (a) two or more enterprises have ceased to be distinct enterprises at a time or in circumstances falling within section 24; and…

[15]Note that there must be two or more enterprises and so the mere acquisition of bare assets is not enough. Section 24 EA02 (referred to in section 23) imposes time limits on the making of a reference, and there is no issue on those time limits.

[16]The term “enterprise” is defined in section 129(1) EA02: “enterprise” means the activities, or part of the activities, of a business

[17]The same subsection also provides a definition of “business”: “business” includes a professional practice and includes any other undertaking which is carried on for gain or reward or which is an undertaking in the course of which goods or services are supplied otherwise than free of charge;

[18]Challenges to decisions of the CMA (in this case) that a particular situation constitutes a relevant merger situation are dealt with in section 120 EA02, which enables a person to apply to the CAT for judicial review of the CMA’s decision: (1) Any person aggrieved by a decision of the [CMA]… under this Part in connection with a reference or possible reference in relation to a relevant merger situation or a special merger situation may apply to the Competition Appeal Tribunal for a review of that decision…. (4) In determining such an application the Competition Appeal Tribunal shall apply the same principles as would be applied by a court on an application for judicial review. (5) The Competition Appeal Tribunal may— (a) dismiss the application or quash the whole or part of the decision to which it relates; and (b) where it quashes the whole or part of that decision, refer the matter back to the original decision maker with a direction to reconsider and make a new decision in accordance with the ruling of the Competition Appeal Tribunal. (a) dismiss the application or quash the whole or part of the decision to which it relates; and (b) where it quashes the whole or part of that decision, refer the matter back to the original decision maker with a direction to reconsider and make a new decision in accordance with the ruling of the Competition Appeal Tribunal.

[19]So if the CAT remits the case to the CMA, the CMA must “make a new decision in accordance with the ruling of the” CAT.

[20]Section 120(6) deals with appeals from the decision of the CAT: (6) An appeal lies on any point of law arising from a decision of the Competition Appeal Tribunal under this section to the appropriate court.

[21]The appropriate court for the purposes of this appeal under subsection (6) is this court. As this court is hearing an appeal from CAT 2, it must also apply the principles of judicial review.

[22]I shall refer to the decision of CAT 1 as “Eurotunnel 1”. It resulted in the remittal of an issue to the CMA and the production by the CMA of the Remittal Report. Eurotunnel 1 was not appealed and I shall have to consider the effect of the remittal by CAT 1 below.

[2]Facts said to give rise to a merger situation i). SeaFrance’s ferry business: cessation and liquidation

[23]Prior to November 2011, SeaFrance was a wholly-owned subsidiary of SNCF, a French company. It had a workforce of over 1,220 persons. Its profitability declined and it encountered severe financial difficulties. The key dates for the realisation of its assets were as follows: 30 June 2010 SeaFrance enters administration in France. End 2010 354 SeaFrance employees made redundant, leaving 872 employees. February to October 2011 The French government sought, but failed to secure, EU Commission approval for an injection of some €223m. July 2011 Offers for the SeaFrance business were sought but none was successful. This followed earlier attempts to find a buyer in 2010, which had also been unsuccessful. 15/16 November 2011 SeaFrance ceased to trade overnight. December 2011 Further bids invited. The SCOP makes further offer January 2012 SeaFrance formally liquidated and required to cease trading. Remaining employees made redundant. SeaFrance’s vessels were put into a condition called “hot lay-up”. This meant that they were moored outside Calais but maintained in a condition which would make it possible for them to be made ready for service in the minimum of time. The SCOP’s further offer was rejected. French court orders that the ferry business be terminated and all staff were given redundancy notices (save for 190 staff employed on the vessels in hot lay-up). Invitation of bids for liquidated assets. 23 January 2012 Court approves PSE3 containing indemnity. SeaFrance’s berthing slots in Dover and Calais were surrendered. 11 June 2012 GET’s offer to acquire vessels and certain other assets (including brand name and goodwill) is accepted. 29 June 2012 GET grants the SCOP bare boat charters of each of the three acquired ex-SeaFrance vessels. August 2012 GET recommences Dover-Calais ferry services using the SeaFrance vessels and certain ex-SeaFrance employees. This terminates a seven-month period in which the business was not operational. ii). Formation of the SCOP and employee protection plan for SeaFrance’s employees

[24]The SCOP was formed on 7 October 2011 by a group of fourteen former SeaFrance employees. Shortly thereafter, it made an offer to acquire the SeaFrance vessels, but its offer was inadequate.

[25]Under the French Labour Code, there had to be a “job saving plan” (known as PSE3) to help protect employees. This was adopted on 23 January 2012.

[26]The PSE3 was complex. Among other matters, it significantly provided for subventions to help redundant employees of €10,000 per employee if employees took a 50% stake in a company and is self-employed. This was ratcheted up to €15,000 per employee if the project was based in the Calais region. The largest subvention (€25,000 per employee) was payable if the vessels were assigned “allowing similar operation” by the SCOP or a company in which SeaFrance employees took up shares and employment. SNCF was to provide these subventions. I refer below to SNCF’s promise in favour of the SCOP as “the indemnity”: the CMA later found that the indemnity was an incentive to GET to employ ex-SeaFrance employees when it took over the vessels. The relevant provision of the PSE3 was as follows: 3.3 Aid in creating or taking over an enterprise3.3.1 The purpose of this aid is to give financial aid to persons made redundant for economic reasons who create or take a minimum 50% stake in a company in a public limited company, a limited liability company, a general partnership, an incorporated sole proprietorship, a simplified limited company with sole shareholder, a simplified joint stock company or a business in France or in a country in the European Union, and carry on actual work there. This aid of an amount of €10,000 gross will be allocated by SEAFRANCE only for the plans approved by the Employment Office, independently of aid received by employees from the State or other organisations. …

3.3.2 Employees creating or taking over an enterprise locally with the aim of contributing to economic development either in that area or within a radius of 50km of the job location in the Calais region may receive special aid relating to the scheme for developing the employment catchment area defined in paragraph 4 below. In addition to this scheme, the allowance of €10,000 stipulated in article 3.3.1 will be increased to €15,000 if the conditions stipulated in that article was fulfilled.

3.3.3 Where the bankruptcy judge in the liquidation of SEAFRANCE has to rule upon an assignment in a final ruling allowing similar operation of the vessels belonging to SEAFRANCE in favour of the SeaFrance Cooperative Enterprise or any other company (of any form) in which the employees have a direct interest (share of the equity capital) and indirect interest (employment contract), the company will be paid €25,000 per employee, on receipt of the official documents (a list of employees concerned, proof of holding of share of the capital and employment contracts). The aid of €25,000 gross may not be combined with the aid of €10,000 gross referred to above in paragraph 3.3.1. This aid of an amount of €10,000 gross will be allocated by SEAFRANCE only for the plans approved by the Employment Office, independently of aid received by employees from the State or other organisations. …

[27]As recorded in the minutes of the French court (Le Tribunal de Commerce de Paris) dated 11 June 2012, Groupe Eurotunnel (“GET”) made a bid for SeaFrance’s assets, described as a “takeover”, involving the participation of the SCOP and re-employment of ex-SeaFrance employees: The third bid submitted by SA [public limited company under French law] Groupe Eurotunnel… The bidder presents a comprehensive, integral bid bearing simultaneously on the ships and other tangible assets and intangible assets whose acquisition is proposed, as part of an individual project integrating the participation, via a SCOP [workers’ productive cooperative under French law] composed of SeaFrance’s former employees. The relevant local authorities, the Regional Council and the Calais Mayor’s Office, have clearly demonstrated their desire to be associated with the proposed recovery through a financial contribution to the acquisition of the ships on terms currently being finalized. The bidding company continues in this industrial rationale by proposing to take SeaFrance’s industrial assets and operate them through special purpose companies, in accordance with the interests of the Group and its shareholders. A partnership for the long-term between Eurotunnel and the SCOP [workers’ productive cooperative under French law] including SeaFrance’s former employees is considered; this partnership would provide for an immediate return to employment for SeaFrance’s former employees, as well a perspective for progressive hiring. The takeover of SeaFrance’s activities may be schematically summarized as follows…A special purpose company…,controlled by GET SA, shall own the ships; it shall employ no or very few staff and shall operate as lessor/charterer… … the project in which Groupe Eurotunnel is participating is aimed at providing for a partnership with SeaFrance’s former employees who shall form a SCOP in order to revive the activities previously conducted by SeaFrance…

The third bid submitted by SA [public limited company under French law] Groupe Eurotunnel…

[28]The acquisition agreement between GET and the liquidator of SeaFrance provided for GET to acquire a number of assets in addition to the vessels, namely(a) brand and domain names and customer lists;

(b) ferry management software SeaPax and SeaFret; and

[29]In August 2012, having brought the vessels back into service, GET recommenced ferry operations under the name MyFerryLink (“MFL”). GET entered into a contract with the SCOP under which the SCOP operates the services and provides the crew and other employees required for the business.

[3]Position of L’Autorité de la concurrence (French regulator)

[30]As stated above, this court is not concerned with the question of any remedies ordered by the CMA. As we understand it, the CMA has recommended remedies which have resulted in GET putting the vessels it acquired from SeaFrance on the market for sale. As this is, however, a case which concerns the competition authorities in France, it is right that I record what we understand to be the different approach taken there. By its decision of 7 November 2012, l’Autorité de la concurrence, the French regulator, found that the acquisition was likely adversely to affect competition, through the conglomerate effects on the freight transport market and the vertical effects on the cross-channel transport markets. However, l’Autorité concluded that these risks could be remedied by a series of undertakings, i.e. they were not such that the acquisition should be prohibited.

[4]Competition Commission report, Eurotunnel 1 (CAT 1) and remittal

[31]In June 2013, the Competition Commission issued a report in which it concluded that the acquisition by GET was a relevant merger situation for the purposes of section 35(1) EA02. GET and the SCOP brought judicial review proceedings challenging this conclusion. Those proceedings led to the judgment of CAT 1 in Eurotunnel 1. This judgment is particularly important because the judgment sets out the basis on which CAT 1 remitted the jurisdiction to what became the CMA. It gave rulings on (among other matters) the question whether GET had acquired the SeaFrance business or bare assets.

[32]CAT 1 noted that it was central to the determination of what is a merger that two or more enterprises should cease to be distinct (see section 23(1)(a) EA02). It set out the statutory definitions of “enterprise” and “business” and concluded that: “Essentially an enterprise is the activities, or part of the activities, of a business.” (Eurotunnel 1, paragraph 101, underlining added in original)

[33]CAT 1 noted that there was no definition of “activities”. No authority was cited which threw light on this term. The nearest thing was the report of the Monopolies and Mergers Commission (“the MMC”) in the matter of AAH Holdings plc and Medicopharma NV (“AAH Holdings/Medicopharma”). CAT 1 drew on that report to formulate helpful guidance as to when the acquisition of assets was no more than that and when it led to a merger situation. For that reason and because the parties made submissions on it, I have set out a fuller description of the relevant part of the report in the annex to this judgment.

[34]The factors in the report to which CAT 1 drew attention were:(1) how much time passed between the transferor ceasing to carry on business and the transfer to the merged enterprise;

(2) whether the acquirer obtained benefits that it would have obtained by acquiring the business. From that CAT 1 deduced that the inquiry which the Competition Commission had to carry out was (1) to identify the assets which the acquirer obtained over and above the bare assets and (2) to ask whether those matters placed the acquirer in a different position from that in which it would have been if it had simply purchased the assets in the market.

[35]CAT 1 held: 104. Before the MMC, it was contended that no merger situation arose because Medicopharma NV's United Kingdom operation "had ceased to trade prior to the acquisition and that AAH had acquired only stock, certain assets and three depots" (see paragraph 6.62 of the MMC Report). However, the periodduring which the United Kingdom operation had not traded was extremely short – essentially comprising the period between 3 November 1991 (paragraph 6.78 of the MMC Report) and 7-8 November 1991 (paragraph 6.87 of the MMC Report). The MMC rejected the argument that no merger situation arose (paragraph 6.102 of the MMC Report): “In our view, however, although AAH did not in terms acquire the depots as going concerns, in reality it obtained much of the benefit of so acquiring them and it clearly acquired more than bare assets, as described in greater detail above.” 105. We find this approach a helpful one. Essentially, the MMC was drawing a distinction between the acquisition of "bare assets" – which would not constitute the activities of a business – and the acquisition of something more than bare assets. The key to distinguishing between "bare assets" and an "enterprise" lies in:(a) Defining or describing exactly what, over-and-above "bare assets", the acquiring entity obtained; and

(b) Asking whether – and if so how – this placed the acquiring entity in a differentposition than if it had simply gone out into the market and acquired the assets. (underlining in the original) “In our view, however, although AAH did not in terms acquire the depots as going concerns, in reality it obtained much of the benefit of so acquiring them and it clearly acquired more than bare assets, as described in greater detail above.” (a) Defining or describing exactly what, over-and-above "bare assets", the acquiring entity obtained; and (b) Asking whether – and if so how – this placed the acquiring entity in a differentposition than if it had simply gone out into the market and acquired the assets. (underlining in the original)

[36]So it was necessary to find out what had been acquired apart from bare assets and why this acquisition was different from simply going into the market and acquiring the assets. That was not all. The Competition Commission would have to go on to consider whether the difference was capable of making the acquisition one of a business and not bare assets, remembering that the answer to that question would inevitably be a question of fact and degree. CAT 1 considered that the guiding principle lay in understanding that an enterprise takes in inputs and generates not just outputs but also valuable assets, such as know-how or goodwill. That was the essence of a business. If there is an acquisition of a business not bare assets, this will continue to happen even after the acquisition: The question, then, is whether this difference is capable of constituting what would otherwise be bare assets into something that may properly be described as the activities of a business. Inevitably, this is a question of fact and degree, and there will be no single criterion giving a clear answer. However, if a guiding principle is sought, then we consider that it lies in an understanding of what an enterprise – the activities or part of the activities of a business – does. An enterprise takes inputs (assets of all forms) and by combining them transforms those inputs into outputs that are provided for gain or reward. It thereby also may generate intangible but valuable assets such as know-how or goodwill. It is in this combination of assets that the essence of an enterprise lies. In those cases where the acquiring entity takes over the business of the acquired entity, the answer will be self-evident: the same enterprise is simply continuing, albeit under different ownership or control.

[37]I would add this. It would follow that if a purchaser acquires goodwill or knowhow, that would be an indication that those assets continued in existence and that the acquirer was acquiring a business.

[38]CAT 1 recognised that the difficult case is where the previous enterprise ceases to operate. That did not necessarily mean that there could be no acquisition of a business because even where a business had been wound down to a considerable extent there could still be “the embers” of an enterprise: it could, therefore, be held in suspense. CAT 1 warned the Competition Commission against being led into thinking that there was a merger situation simply because the acquirer carried on the same business as the old business which had ceased to exist: The difficult case arises where the combination of assets is fractured, such that the assets are no longer, or no longer to the same extent, being used in combination. This case is a particularly good one, where what was clearly once an enterprise was wound down: the difficult question is whether, even though the business of SeaFrance had been wound down to a very considerable extent, there still remained the embers of an enterprise.

[39]CAT 1 then made two rather telling points: first, there could still be a merger even if the acquired business was (my word) paused for example because it was “low season”, and, second, it was not enough that the business of the merged entity was (my word) the mirror image of the acquired business: 106. In this context, it is necessary to make two points:(a) First, it is perfectly possible for an enterprise to wind down, and to wind down to such an extent that it ceases to be an enterprise.The mere fact that in the past the activities of a business were being carried on by an entity does not necessarily mean that, as at the time of the merger, that entity was an enterprise. Of course, it is also important to recognise that some businesses (e.g. those involved in tourism) trade for some periods and not for others (e.g. during the "low season"). Such a hiatus does not preclude the existence of an enterprise. Continuous trading is not essential.

(b) Secondly, the fact that the acquiring entity emulates the business of the acquired entity, and even uses that entity's assets, does not necessarily mean that the acquiring entity has acquired an enterprise. .. (Italics added) (a) First, it is perfectly possible for an enterprise to wind down, and to wind down to such an extent that it ceases to be an enterprise.The mere fact that in the past the activities of a business were being carried on by an entity does not necessarily mean that, as at the time of the merger, that entity was an enterprise. Of course, it is also important to recognise that some businesses (e.g. those involved in tourism) trade for some periods and not for others (e.g. during the "low season"). Such a hiatus does not preclude the existence of an enterprise. Continuous trading is not essential. (b) Secondly, the fact that the acquiring entity emulates the business of the acquired entity, and even uses that entity's assets, does not necessarily mean that the acquiring entity has acquired an enterprise. .. (Italics added)

[40]CAT 1 then underlined these conclusions in its comments on the arguments put forward by Mr Pickford, appearing then, as he does before us, for the second respondent, DFDS, another cross-Channel ferry operator. It dismissed his arguments that the matter could be determined by looking at the situation after the acquisition. It is not enough that there is a substantial lessening of competition or that the activity after the acquisition is that of the old company. Close attention to the wording of the EA02 was necessary: the real question was always whether the acquirer bought a business or bare assets. It had to be shown that at the time of the merger there were two or more businesses: Mr. Pickford, counsel for DFDS, contended that Eurotunnel had acquired an enterprise for precisely this reason (Day 2/page 97):

“The second overarching point…is…not disputed by the SCOP that [Eurotunnel] now operates a freight and passenger ferry service across the short sea, using the same vessels as SeaFrance on the same route, with a large proportion of ex-SeaFrance staff, targeting amongst others ex-SeaFrance customers and which [Eurotunnel] took to be a partnership [with the SCOP]. It is not in dispute that SCOP was formed for the very purpose of continuing the SeaFrance operations in so far as possible, and that it worked towards that objective for the entirety of the seven month pause in trading which took place during mainly the low season of 2011-2012. Nor is it disputed by the SCOP in its application that the transaction could be expected to lead to a substantial lessening of competition. If we take all those points together, we say it is very difficult to see how the SCOP can sensibly claim that none of the activities of the SeaFrance business came under the control of [Eurotunnel], and that the [Commission] was not therefore empowered to act to prevent what it saw as a lessening of competition.”

Mr. Pickford put the point powerfully but we find it to be misconceived: (i) Mr. Pickford's references to the creation of a situation in which a substantial lessening of competition may result is nihil ad rem. It refers to the question that the OFT and the Commission must answer after there has been a finding that a relevant merger situation has been created (see section 22(1) of the Act in the case of the OFT and section 35(1) of the Act in the case of the Commission). The SCOP's contentions were directed to this, anterior, point. If, as the SCOP contended, no relevant merger situation has been created, then the question of whether there is or may be a substantial lessening of competition simply does not arise. (ii) As regards the question of whether a relevant merger situation exists, the statutory test is not whether the acquiring entity is carrying out the same activity that was once carried out by the acquired entity, even with the same assets. The statutory test is not satisfied if the acquiring entity reconstructs a business that was once conducted by a different entity, even if the assets of that entity were used to do so. The statutory test in section 26(1) turns on two enterprises ceasing to be distinct because they are brought under common ownership or common control. It is critical that there are two enterprises, not one enterprise (the acquiring enterprise) and a collection of assets. Mr. Pickford's contentions thus address the wrong test. “The second overarching point…is…not disputed by the SCOP that [Eurotunnel] now operates a freight and passenger ferry service across the short sea, using the same vessels as SeaFrance on the same route, with a large proportion of ex-SeaFrance staff, targeting amongst others ex-SeaFrance customers and which [Eurotunnel] took to be a partnership [with the SCOP]. It is not in dispute that SCOP was formed for the very purpose of continuing the SeaFrance operations in so far as possible, and that it worked towards that objective for the entirety of the seven month pause in trading which took place during mainly the low season of 2011-2012. Nor is it disputed by the SCOP in its application that the transaction could be expected to lead to a substantial lessening of competition. If we take all those points together, we say it is very difficult to see how the SCOP can sensibly claim that none of the activities of the SeaFrance business came under the control of [Eurotunnel], and that the [Commission] was not therefore empowered to act to prevent what it saw as a lessening of competition.” (i) Mr. Pickford's references to the creation of a situation in which a substantial lessening of competition may result is nihil ad rem. It refers to the question that the OFT and the Commission must answer after there has been a finding that a relevant merger situation has been created (see section 22(1) of the Act in the case of the OFT and section 35(1) of the Act in the case of the Commission). The SCOP's contentions were directed to this, anterior, point. If, as the SCOP contended, no relevant merger situation has been created, then the question of whether there is or may be a substantial lessening of competition simply does not arise. (ii) As regards the question of whether a relevant merger situation exists, the statutory test is not whether the acquiring entity is carrying out the same activity that was once carried out by the acquired entity, even with the same assets. The statutory test is not satisfied if the acquiring entity reconstructs a business that was once conducted by a different entity, even if the assets of that entity were used to do so. The statutory test in section 26(1) turns on two enterprises ceasing to be distinct because they are brought under common ownership or common control. It is critical that there are two enterprises, not one enterprise (the acquiring enterprise) and a collection of assets. Mr. Pickford's contentions thus address the wrong test.

[41]Finally, CAT 1 contrasted the situation where GET merely bought identical vessels from a shipbuilder, which would clearly not be the acquisition of a business, with the events in this case: 107. The short, but difficult distinction that we have to draw is that between an asset purchase and the acquisition of an enterprise. Had Eurotunnel simply gone to a shipbuilder and commissioned the building of three vessels identical to the Rodin, the Berlioz and the Nord Pas-de-Calais or with similar capabilities and used these vessels to establish a Dover-Calais ferry service using a crew or crews comprising anyone other than ex-SeaFrance employees, then this would not involve the acquisition of an "enterprise". Rather, Eurotunnel would be using assets that it had acquired to create an enterprise. The question we must answer is whether the fact that the vessels were acquired from SeaFrance and the fact that the crews were largely drawn from ex-SeaFrance employees changes this outcome.

[42]The key message in these concluding passages is a warning not to use post hoc propter hoc reasoning, that is, not to assume that because one thing (the carrying on of a ferry business using ex-SeaFrance vessels and ex-SeaFrance employees) happens after another (the acquisition by GET of those vessels and other assets), that acquisition was the cause of the resurrection of SeaFrance’s old business. That business might equally have been recreated by GET.Vessels plus employees did not necessarily mean a business and not bare assets had been acquired.

[43]CAT 1 considered it possible that what had happened was indeed an acquisition of bare assets (Eurotunnel 1, para 114). For instance, it found it difficult to see how the employees were acquired but it recognised that that could happen if the workforce had migrated to the new employer so that the reality was that the workforce was transferred (Eurotunnel 1, para 116). It also recognised that the indemnity made available to the SCOP when employees took up employment with GET and the fact that the vessels were in “hot layup” might mean that there was: “a momentum and continuity in the combination between the vessels and the workforce that takes this case over the line from an asset acquisition to the acquisition of an enterprise.” (Eurotunnel 1, para 120)

[44]In the light of its doubts, CAT 1 remitted the matter back to the Competition Commission to determine in accordance with its ruling whether the merger condition was satisfied. That led to the CMA’s Remittal Report. By the date of this Report, the CMA had assumed the responsibilities of the Competition Commission.

[5]CMA’s Remittal Report

[45]There was a 45-paragraph summary at the start of the Remittal Report, notified 27 June 2014, which included the following highlights on the question whether GET acquired an enterprise: – GET’s acquisition of the SeaFrance’s vessels is likely to have reduced the commercial risk for GET/SCOP compared with buying or chartering new vessels (para 15). – The hot lay-up enabled them to be put into service within a shorter time frame than if they had been laid up cold (para 16). – The €25,000 provided a strong incentive for ex-SeaFrance employees to be employed on the ex-SeaFrance vessels. “Our conclusion was that in effect many employees transferred from SeaFrance to the SCOP.” It was easier for crew to be found from the ex-SeaFrance employees than from crewing companies (para 20). – It was easier to obtain berthing slots with the ex-SeaFrance pilots than it would otherwise have been. Some of the pilots had valid pilotage exemption certificates (para. 22). – The miscellaneous assets conferred a material benefit on GET/SCOP to start the ferry service more quickly (para. 24). – The fact that customer and supply contracts were not transferred was not material (para 25). – In conclusion, the combination of assets meant that more than bare assets were acquired (para. 26). – GET/SCOP obtained much of the benefit that they would have obtained if the assets had been acquired as a going concern (para. 26). – The assets acquired (including ex-SeaFrance employees) constituted an enterprise as they constituted the activities or part of the activities of a business (para. 27).

[46]The Remittal Report makes it clear that the CMA have looked to the substance of the arrangements and not merely their legal form (para. 2.12). The aim of the Report was to deal with the issue remitted to the Competition Commission by the CAT.

[47]The CMA concluded that the payments made to the SCOP under the indemnity enabled it to make its contribution to GET’s successful bid (para. 3.31).

[48]The Remittal Report concluded that the workforce was effectively held together by the formation of the SCOP, and that this was not akin to a situation where assets are put on the market and then purchased: 3.51 In our view, the collaboration between GET and the SCOP presented a solution that addressed two main concerns flowing from the liquidation of SeaFrance:(a) payment of creditors; and

(b) ensuring employment for ex-SeaFrance workers. Although the various schemes previously considered by the French Court had at their core the continuation of a ferry service and employment for SeaFrance employees, it had not been possible to find a viable solution producing value for creditors and the continuation of the SeaFrance operation involving all employees under their existing terms and conditions. 3.52 The liquidation process and subsequent termination of employment contracts meant that the TUPE Regulations did not apply and allowed the business and workforce to be restructured. Continuity of employment was effectively safe-guarded by the formation of the SCOP, which held the workforce together, and, to a lesser extent, due to the fact that a significant number of employees were involved in the lay-up of the vessels. 3.53 At the point the decision was taken that SeaFrance activities should cease, the French Court recognised that the aim of achieving some form of business continuity remained unchanged. This is clear from statements made by the French Court such as:

‘the end of the temporary continuance of business is not the end of the road’ and ‘there must be a trade-off between the value of the assets, which are mainly the vessels, and the continuance of employment contracts’

. PSE3 was designed to support such a business continuity solution (given the fact that the SCOP had been unable to secure finance in the market), and GET’s acquisition of the vessels provided funds to pay creditors. 3.54 Overall, we consider that a review of the background to the transaction shows that there is considerable continuity and momentum between the time of SeaFrance’s operation of the Dover–Calais ferry and the commencement of MFL’s operation of the same ferries on that route involving ex-SeaFrance employees. This is not a situation where a collection of assets (used at some point in the past to carry on a business activity) comes to the market, and a buyer is successful in acquiring them, and then uses them to set up a business similar to the one for which the assets were originally used. For reasons set out above, the circumstances of this case are fundamentally different. 3.55 We appreciate that there are material differences between the transactions involving the SCOP that were contemplated in respect of the sale of SeaFrance as a going concern and the acquisition of liquidated assets by GET. Significantly, once SeaFrance was put into liquidation, many of its employees were made redundant within two weeks. This also meant that any buyer of the liquidated assets would not have to assume any employee obligations – a fact that might well be an advantage in circumstances where the new owner envisaged a leaner operation….

[49]The CMA noted that some 70% to 80% of the SCOP employees as at 29 October 2012 were ex-SeaFrance employees. The CMA drew the following conclusion on the effect of the indemnity: 3.107 In our view, the indemnity demonstrates that it is not the case that SeaFrance’s employee contracts of employment were terminated ‘with no thought as to how they might be reemployed in future’. The indemnity that SNCF–SeaFrance’s parent company at the time–agreed to pay created a strong incentive for ex-SeaFrance employees to be employed on the SeaFrance Berlioz, SeaFrance Rodin and SeaFrance Nord Pas de Calais in similar operations to those of SeaFrance. It creates a link between the vessels and the employees and it was aimed at ensuring, and ultimately did ensure, to the extent possible given the points that we highlighted in paragraph 3.77 above, that a significant number of employees transferred from SeaFrance to the operator of the vessels. We consider that this shows that a large proportion of the SeaFrance workforce effectively transferred from SeaFrance to the SCOP… 3.110 Although a TUPE transfer may be an indicator of the transfer of an enterprise, the converse is not necessarily true. There may well be circumstances, of which this, in our view, is one, where had TUPE (or its French equivalent) applied, this would have been damaging to the transfer of a viable business. The evidence indicates to us that the SeaFrance business required restructuring in part because it was overmanned and suffered from bad labour relations. The liquidation avoided a TUPE transfer of employees, and as a result GET and the SCOP were in a better position to carry on a viable ferry business (albeit on a reduced scale compared with SeaFrance) and the SCOP was able to offer employment to a number of appropriately skilled persons, drawn substantially from ex-SeaFrance employees. That, in turn, enabled GET to table an acceptable offer for the vessels and other assets, and assisted GET and the SCOP in developing a viable business plan for the Dover–Calais route. In our view, this is consistent with the situation described by the CAT and referred to in paragraph 3.64 above.

[50]The Remittal Report then considered the other assets which were acquired.

[51]It noted that GET had acquired three of the four vessels previously operated by SeaFrance. The prefix “SeaFrance” was changed but not the names of the vessels acquired which the CMA considered would be likely to generate some goodwill. The CMA’s view was that GET/SCOP were likely to have benefitted by acquiring the sister ships (that is, the SeaFrance Rodin and the SeaFrance Berlioz), which were specially designed for the route. The CMA considered the acquisition of other ships, but concluded that there were few suitable ships available.

[52]On berthing slots, the CMA noted that GET/SCOP and MFL encountered no material obstacles in obtaining berthing slots. It concluded that the process was facilitated by the fact that vessels were known in both ports and that the ex-SeaFrance officers were familiar with the ports (para. 3.179).

[53]On brand names and domain names, the CMA took into account the substantial value that GET and others placed on these assets: 3.195 Whilst we acknowledge that some of the goodwill associated with the brand and domain names is likely to have dissipated in the period of inactivity, nevertheless, GET’s offer to the French liquidator included €1 million attributable to the trade marks and domain names of SeaFrance. We find it significant that P&O bid separately for the domain names, indicating that it attached value to them despite the period of inactivity. We note also that GET did not withdraw the SeaFrance web page immediately and gained some business as a result of redirected traffic (see further Appendix D). 3.196 Ordinarily, the acquisition of intangible assets such as brand and domain names, together with tangible assets and employees, would point in the direction of the acquisition of an enterprise. That would be the case regardless of whether or not the acquirer actually decided to use the acquired brands and domain names. We consider that despite the period of inactivity, there remained some value in these intangible assets that would be of benefit to GET and the SCOP in the context of their use of the other acquired assets (noting, however, that for some freight customers the brand may have had negative associations: see paragraph 3.224 below).

[54]The CMA also attached some weight to the IT systems which GET took over and to the fact that the ex-SeaFrance employees were available to operate them: 3.206 We recognise that since the systems were ‘blank’, work would have been required to repopulate them with parameters and data. We note, however, that all of the IT staff ([%] in total) employed by the SCOP are ex-SeaFrance employees and this is likely to have been useful in overcoming any difficulties associated with use of the system and the fact that it was ‘blank’. 3.207 In our view, IT systems suitable for use on the short sea are likely to have special requirements over and above IT systems suitable for operating ferry services more generally, given the high frequency of services and multiple daily departures that are a feature of the short sea, as well as the requirement for accurate manifests. We contacted the third parties that we were told would be in a position to supply an off-the-shelf system that would be suitable. The responses we received indicated that one provider was able to offer a web-based reservation system. It appeared to us that this lacked much of the functionality of SeaFret and SeaPax. We consider that GET’s acquisition of the SeaFrance IT systems gave it access to systems that were proven in practice to be effective in managing passenger and freight operations on the short sea, reducing the risk (and cost) associated with having to introduce new unproven IT systems which may not have all the required functionality. Together with MFL’s employment of ex-SeaFrance IT staff, this places MFL at a material advantage compared with the situation where GET did not purchase the SeaFrance IT systems.

[55]Similarly, the CMA attached weight to GET’s decision to acquire the customer databases. Even though there had been a hiatus in trading, there would remain some value in these databases (para. 3.215).

[56]The CMA considered the impact of GET not acquiring supplier or customer contracts but on the facts it concluded that the absence of those assets was immaterial.

[57]In its conclusions, the CMA explained that it had followed the methodology laid down by CAT 1. Its review of the background led it to conclude that there was considerable and deliberate continuity and momentum between the time SeaFrance operated the ferry service and MFL’s resumption of operations. A significant part of the period in between was taken up with the liquidator’s sale process (para 4.5). GET benefitted by acquiring sister ships and by their having been in hot lay-up. The indemnity “forged a link between the vessels and the employees”: 4.11 The €25,000 indemnity that SNCF agreed to pay created a strong incentive for ex-SeaFrance employees to be employed on the SeaFrance Berlioz, SeaFrance Rodin and SeaFrance Nord Pas-de-Calais in similar operations to those of SeaFrance. It forged a link between the vessels and the employees and it ensured that – to the greatest extent possible – ex-SeaFrance employees transferred from SeaFrance to GET/SCOP. The indemnity reinforces our view that contracts of employment were not terminated ‘with no thought as to how they might be reemployed in future’. Our conclusion is that in effect these employees transferred from SeaFrance to GET/SCOP. As a result, around [70–80]% of the SCOP workforce comprises ex-SeaFrance employees who were made redundant as a consequence of SeaFrance’s liquidation. The steps taken in relation to the ex-SeaFrance employees were designed to ensure continuity of SeaFrance’s activities to the maximum possible, and those steps substantially achieved their aim (para 4.14). A variety of steps were taken to secure to the maximum extent possible the preservation of key assets of SeaFrance’s business. The combination of assets + employees meant that more than bare assets were acquired; the business was substantially the same as that of SeaFrance (para.4.19). In all the circumstances the collection of assets which GET/SCOP acquired met the legal definition of an enterprise.

[58]In its conclusions, after the passage headed “Assessment of jurisdictional issue applying approach in the judgment” (paras. 4.1 to 4.22), the Remittal Report set out a passage headed “Broader observations on the jurisdictional test”. This lead the CMA to conclusions which were consistent with the approach in the judgment (para. 4.23). The CMA observed that the statutory provisions should be interpreted widely and purposively. The CMA considered therefore that it should take a purposive approach to the concept of an “enterprise” (para.4.26).

[59]GET and the SCOP made an application to the CAT under section 120(1) EA02 to challenge the CMA report. (GET has not, however, taken part in this appeal.) The court only has extracts from that application, which appears to have been based on error of law and irrationality. The points made in these extracts show that a major plank in the SCOP’s application was that the SeaFrance freight business had been defunct since November 2011 and therefore it was not open to the CMA to conclude that what GET/SCOP acquired was “a business”. In any event the CMA misdirected itself into thinking that the fact that the SeaFrance employees were not made redundant without any thought as to how they might be re-employed in the future (see paragraph 49 above and paragraph 3.107 of the Remittal Report) meant that there was the acquisition of a business. In addition, the SCOP contended that there had been no transfer of the employees.

[6]Decision of CAT 2 (under appeal)

[60]The first ground of challenge before CAT 2 was that GET/SCOP did not acquire “an enterprise” and that CMA erred in law in so concluding. CAT 2 noted that there had been no appeal from Eurotunnel1. Given the terms of section 120(5)(b), it therefore declined to reconsider the approach in CAT 1. CAT 2 went on to apply that approach. CAT 2 concluded that the CMA had addressed the question of what GET/SCOP had acquired over and above “bare assets”. The expression “bare assets” was not in CAT 2’s judgment a precise term. It considered that bare assets would not include goodwill, trademarks, trade names, domain names, customer databases or lists. It noted that the successful bid included €1,000,000 specifically for trademarks, trade names, domain names and internet sites. CAT 2 did not consider that the employees were included within “bare assets”.

[61]CAT 2 noted that the main thrust of the SCOP’s submissions was that there had been no continuity between SeaFrance and the operations of MFL, and that it was irrational for the CMA to find otherwise. All the attempts by the administrators to sell SeaFrance as a going concern had failed. SeaFrance itself had gone into liquidation and ceased all trading activity. Most employees had been made redundant and were thereafter unemployed. They had to apply to the SCOP in order to be employed in the new business. MFL only commenced cross-channel ferry services nine months after SeaFrance had closed down. Over a quarter of the staff engaged in running MFL by the date of the reference were not SeaFrance employees. Accordingly, unlike AAH/Medicopharma, this was not an exceptional case where the statutory definition of enterprise should apply where a business had ceased trading.

[62]CAT 2 acknowledged that these were powerful points. However, the CMA had applied the test laid down in CAT 1. Therefore this ground of challenge was rejected. CAT 2 was satisfied that the proceedings were for judicial review and not an appeal (applying R (Thames Water Utilities Limited) v Water Services Regulation Authority [2012], EWCA Civ 218, [2012] PTFR 1147).

[63]CAT 2 accepted that the question whether there was an enterprise admitted of more than one answer. However it was a question of fact. The matter called for an exercise of judgement by the decision-maker for which there was not necessarily a clear cut answer.

[64]CAT 2 noted that the CMA had found that two passenger vessels were specially adapted for the route and were in the words of the French court “hyper-specialised vessels”. The CMA also found that there was a benefit in their being sister ships of achieving a consistency of service. The CMA also found that the vessels were in hot lay-up which enabled those vessels to be brought into operation more quickly than otherwise.

[65]The thrust of the SCOP’s attack was on the CMA’s approach to employees. CAT 1 expressly noted that the employees were not acquired from SeaFrance. However it had stated that the critical issue was whether the terms of the indemnity meant that there was “a cogent reason on the part of Eurotunnel/the SCOP to employ ex-SeaFrance employees”; and correspondingly, whether this was a benefit that “would not be gained were an employee from elsewhere to be retained” (Eurotunnel 1, Judgment, para. 119).

[66]CAT 1 had noted that, while the SCOP’s employees were recruited through open recruitment, it found that the SCOP would have been motivated by the indemnity. That finding could not be challenged. It was apparent also that a substantial percentage of the 820 SeaFrance employees at the date of liquidation in January 2012 had found employment with the SCOP when MFL started operations on 20 August 2012. Moreover some 80% to 90% (at August 2012) and 70% to 80% (at October 2012) of the SCOP employees were formerly part of the SeaFrance workforce. Given this data, CAT 2 did not think that the CMA could be criticised for using terms such as “significant number” and “large proportion”.

[67]Taking all the circumstances into account, CAT 2 was satisfied that it was open to CMA to find that the indemnity created a link between the vessels and the employees and that both the purpose and result was that a significant number of ex-SeaFrance employees were employed by the SCOP. It did not consider that this conclusion was irrational or that the CMA failed properly to have regard to the facts.

[68]The CMA acknowledged that there was no transfer for the purposes of the Transfer of Undertakings (Protection of Employment) Regulations 2006 (SI 2006/246) (“TUPE”) and that that might be an indicator of the transfer of an enterprise. CAT 2 considered that some of the wording in the remittal report could be criticised. However it took the view that its report was not to be subjected to fine analysis as if it were a statute: see R v Monopolies and Mergers Commission,ex parteThe National House Building Council [1993] ECC 388 to 23. In the circumstances, it did not consider that the CMA’s failure to consider the absence of a TUPE transfer as an indicator against the acquisition of a business was irrational.

[69]CAT 2 did not accept that there was any scope for an expansive interpretation of such matters on the term “enterprise”. However the CMA’s observations on this in the Remittal Report were supplementary to its conclusion, and that its conclusion was not dependent on it.

[70]CAT 2 dismissed the argument that there were significant factual errors: – Factual error 1: CAT 2 referred to the CMA’s finding in paragraph 3.107 of the Remittal Report (see paragraph 49 above) and accepted that the opening sentence of 3.107 could be criticised and contradicted the finding of CAT 1 (Eurotunnel 1, paragraph 115). This sentence failed to take account of the fact that at the time the SeaFrance employees were made redundant by the liquidator it was not clear what bids would be received or accepted. However, CAT 2 did not regard this as material because the CMA recognised that the employees had to be re-engaged through an open recruitment process (CAT 2, Judgment, paragraph. 78). Paragraph 3.107 referred to the indemnity creating a link between the employees and the vessels aimed at ensuring the transfer of a significant number of employees. CAT 2 did not consider that in the light of all the facts for the purpose stated could be regarded as irrational (CAT 2, Judgment, paragraph 79). – Factual error 2: CAT 2 considered it an error for the Remittal Report to say that a considerable proportion of the period between November 2011 and August 2012, was due to the “requirements of the liquidator’s sale process” (paragraphs 3.47 and 4.5). However, it did not consider that these passages undermined its overall assessment of the circumstances. – Factual error 3: CAT 2 considered it an error for the Remittal Report to say that “continuity of employment was effectively safeguarded by the formation of the SCOP” (paragraph 3.52). Again, it did not consider that these passages undermined its overall assessment of the circumstances.

[71]CAT 2 noted the additional observations which the CMA had made on a purposive interpretation of the transactions but held that these did not form part of its conclusions. In any event, the CAT did not consider that the statutory expressions with which this appeal is concerned needed to be given any expansive interpretation.

[7]arguments on this appeal (1). The SCOP’s submissions

[72]Mr Daniel Beard QC, for the appellants, submits that that there were errors of law in the way CAT 2 dealt with the SCOP’s application to it to set aside the CMA’s decision on jurisdiction because(1) the Remittal report was based on errors of law,

(2) the CMA had erred in failing to follow its own guidance and

(3) the Remittal Report contained errors and it could not be shown that these errors played no significant part in the making by the CMA of its decision.

[73]Mr Beard identifies three steps in the CMA's reasoning. First, it assessed the period of inactivity: I will call this the “hiatus point”. Second, it concluded that the indemnity linked the employees to the vessels. I will call this “the indemnity point”. Third, it prayed in aid the acquisition of other assets. The first two steps are challenged, but not the third. The CMA’s decision depends on a series of interrelated findings about the employees and what the CMA regarded as their transfer to the SCOP. Hiatus point Bare assets not enough

[74]Mr Beard’s starting point is that CAT 1 had correctly distinguished the case where the acquirer acquired assets and reconstructed a business from the case where it acquired a business. CAT 2, on the other hand, did not, on Mr Beard’s submission, carefully distinguish between the new ferry operations of MFL, in which the ex-SeaFrance employees were engaged, and the former ferry operations of SeaFrance.

AAH Holdings/Medicopharma

[75]Mr Beard has to distinguish AAH/Medicopharma. He submits that for there to be an acquisition of a business which had been terminated, the circumstances must have been exceptional, as in AAH Holdings/Medicopharma, and that follows from the statutory reference to “activities”, which he submits bears its ordinary meaning, namely the state of being active. In AAH Holdings/Medicopharma, the circumstances were exceptional because the merger resulted in minimal interruption to the business, there was a prior arrangement and the parties’ specific aim was to evade merger control. A beach cafe or company with a seasonal business is obviously distinguishable because the seasonality is known in advance. On Mr Beard’s submission, CAT 2 failed to identify exceptional reasons or to explain what they were. In refusing permission to appeal CAT 2 stated that it accepted that there had to be exceptional circumstances. There was a liquidation and no going concern. The liquidation removed the possibility of any continuity.

No appeal from Eurotunnel 1

[76]Mr Beard has to deal with the effect of the remittal by CAT 1, whose decision was not appealed. Mr Beard contends that the SCOP could not sensibly have appealed from the remittal by CAT 1. He sought to submit that there would be obvious difficulties in trying to persuade this court to hear an appeal which sought to predict what findings might be made on remittal. I should say immediately that I do not accept that point. The appeal would not have been academic if only because its outcome would or might save any further reconsideration as directed by CAT 1. In my judgment, the CMA had no option but to follow Eurotunnel 1: see section 120(5)(b) EA 02, paragraph 18 above. It is a decision which on general principle stands and is binding on all the parties unless and until it is set aside on an application made under section 120 EA 02. Moreover, although this point was not made by counsel, I am provisionally of the view that, no permission having been sought or given to appeal from the order of CAT 1, its order cannot be varied by this court. It therefore cannot be challenged in an appeal from the decision of CAT 2. I need say no more about this point. Indemnity point

[77]CAT 2 accepted that the ferry operations had not merely been suspended but unquestionably terminated. Therefore, on Mr Beard’s submission, the CMA’s decision can only stand if the employees were transferred so that there can be said to be a transfer of activities. He accepts that the transfer of employees might occur informally. However, to show this, there had to be some link between the employees’ involvement with SeaFrance and the re-employment by the SCOP. To reach the conclusion that there was such a link, the CMA relied on the indemnity, but it was wrong so to do.

[78]First, the SCOP could not ensure re-employment. The CMA understated the uncertainties attaching to the future employment prospects of the ex-SeaFrance employees. The employees had the benefit of the indemnity, but there was no guarantee that a bid involving the SCOP would succeed.

[79]Second, the fact that the indemnity created an incentive to re-employment is not enough to create a transfer of activities. The indemnity did not link the employees to the vessels and make the acquisition one of a business. The ferry business was not a going concern. In addition, in point of fact, there was a delay in the payment of the amounts due to the indemnity but this was resolved in January 2013.

[80]Moreover, on Mr Beard’s submission, the absence of a transfer of employment under TUPE or its French equivalent was a real indicator that the employees were not transferred and the CMA did not give this factor sufficient weight.

[81]In effect, SNCF was subsidising employee share ownership for the vessels. The €10 million was paid in instalments. In the minds of those putting in the bid it was cash flow for the business but the business was a new one. The indemnity did not change that. Everything was dependent on these vessels. It is not therefore enough for the CMA to point to the transfer of assets.

[82]Mr Beard submits that it is not enough to say that the indemnity held the business together. The bids had failed in the preceding 18 months. This was a case of a fracturing of vessels and staff. He accepts that it is sufficient that there is a de facto or effective transfer of employment contracts, as Eurotunnel 1 held. However, that was not the reality here. Assets were being sold to competitive bidders. There was no acquisition of a business in any meaningful sense. There was no “migration” of the workforce. “Migrates” is a synonym for transfer by continuity or novation, as under TUPE. The existence of the indemnity arrangements does not mean transfer or migration of the workforce. It was not open to CMA to conclude otherwise because that undermines the distinction in Eurotunnel 1 and EA02 between the acquisition of assets and the acquisition of a business. Remittal Report’s errors which could not be treated as insignificant Three errors relied on

Remittal Report’s errors which could not be treated as insignificant

[83]Mr Beard relies on three errors in the Remittal report: – Factual error 1: The CMA dismissed the argument that the employees’ contracts were terminated “with no thought as to how they might be reemployed in the future” (Remittal Report, para 3.107, paragraph 49 above). This was plainly wrong for the reasons explained above in the context of the indemnity point. – Factual error 2: it was not correct to say that the period of inactivity was in substantial part due to “the requirements of the liquidator’s sale process.” (paragraphs 3.47 and 4.5) – Factual error 3: the CMA was wrong to say that the continuity of employment was effectively safeguarded by the formation of the SCOP (Remittal Report, para 3.52). This was wrong for the reasons given on the indemnity point. This was also plainly wrong for the reasons explained above in the context of the indemnity point.

Legal consequence of errors

[84]Mr Beard submits that once CAT 2 found there were errors in the Remittal Report it should have concluded that the matter should be remitted to the CMA unless it was satisfied that it was inevitable that the CMA would reach the same conclusion: see R(o/a FDAv Secretary of State for Work and Pensions

[2012] EWCA Civ 332

at 67 to 68, which was cited to CAT 2:

“67. Where a decision-maker has taken a legally irrelevant factor into account when making his decision, the normal principle is that the decision is liable to be held to be invalid unless the factor played no significant part in the decision-making exercise…. 68. Even where the irrelevant factor played a significant or substantial part in the decision-maker's thinking, the decision may, exceptionally, still be upheld, provided that the court is satisfied that it is clear that, even without the irrelevant factor, the decision-maker would have reached the same conclusion. Thus, in Simplex GE (1989) 57 P&CR 306, 326, Purchas LJ approved the following passage in the judgment of May LJ in R v Broadcasting Complaints Commission ex p Owen [1985] 1 QB 1153, 1177: ‘Where the reasons given by a statutory body for taking ... a particular course of action are not mixed and can clearly be disentangled, but where the court is quite satisfied that even though one reason may be bad in law, nevertheless the statutory body would have reached precisely the same decision on the other valid reasons, then this court will not interfere by way of judicial review.’ In Smith v North East Derbyshire PCT [2006] 1 WLR 3315, para 10, (a different) May LJ said this: ‘Probability is not enough. The defendants would have to show that the decision would inevitably have been the same and the court must not unconsciously stray from its proper province of reviewing the propriety of the decision making process into the forbidden territory of evaluating the substantial merits of the decision.’” (See also per Keene LJ at [2006] 1 WLR 3315, para 16, as well as Simplex 57 P&CR 306, 327 and 329, and R v Secretary of State for the Environment ex p Brent LBC [1982] QB 593, 646.) ‘Where the reasons given by a statutory body for taking ... a particular course of action are not mixed and can clearly be disentangled, but where the court is quite satisfied that even though one reason may be bad in law, nevertheless the statutory body would have reached precisely the same decision on the other valid reasons, then this court will not interfere by way of judicial review.’ ‘Probability is not enough. The defendants would have to show that the decision would inevitably have been the same and the court must not unconsciously stray from its proper province of reviewing the propriety of the decision making process into the forbidden territory of evaluating the substantial merits of the decision.’”

[85]Mr Beard fastens on the second citation, which was from the judgment of (a different) May LJ. Keene LJ formulated a test of whether there might not have been a different decision. The context was a failure to consult. Certainly, it is a high test. The court must not fall into the trap of making the decision which the legislature has left to the decision-maker.

[86]It follows that it is not enough, submits Mr Beard, to say that the report should be read as a whole. CAT 2 could not have been satisfied that the errors could in no circumstances have affected the outcome of the CMA's analysis. I accept that this principle is directed to a different point, and that, having correctly read the finding in the context of the whole report, the court must be satisfied that the outcome was inevitable.

Application of test of inevitability

[87]Mr Beard submits that CAT 2 could not have been satisfied that the outcome would have been the same even if the erroneous statements were disregarded. The errors were not insignificant because the CMA relied on what it saw as continuity and momentum between the SeaFrance and MFL operations. The CMA concluded that there was: considerable continuity and momentum between the time of the SeaFrance’s operation of the Dover-Calais ferry and the commencement of MFL's operation of the same ferries on that route involving ex-SeaFrance employees… (see Remittal Report, paragraphs 3.54 and 4.5).

[88]So, argues Mr Beard, the CMA regarded the continuity of employment as important. The fallacious assumption of continuity meant that the reasoning in para. 4.11 of the Remittal Report (paragraph 57 above) was based on error. Moreover, the CMA’s conclusion at 4.14 could not sensibly be severed from its erroneous finding about continuity of employment. Likewise the CMA were in error about the advantages of continuity in para 4.19 of the Remittal Report (paragraph 57 above).

Failure to follow guidance

[89]Mr Beard submits that CAT 2 was wrong to say these matters were merely background. The guidance issued by the CMA states that the length of the period of inactivity is a matter which has to be considered. This provided a further ground for judicial review.

Interpreting transactions purposively and widely

[90]Finally the CMA were also in error in saying that the jurisdictional test had to be applied in a broad and purposive way. (2). CMA’s submissions

[91]Mr Paul Harris QC, for CMA, submits that CMA had to apply the directions of CAT: section 120(5) EA02. The question whether a business still exists is a multi-factorial question of fact and degree.

Hiatus point

[92]Mr Harris submits that there is no need for the business to be actively trading on the day of acquisition. He cites the decision of the OFT in Completed acquisition by Hargreaves Services plc of certain assets of Scottish Coal Company Ltd (ME/6153/13, 30 October 2013) where the OFT took the view that there could be a business of operating three coal mines which had ceased to operate because of a liquidation but which were still viable. In the present case, there was also the acquisition of goodwill and other benefits, which the SCOP ignores. The CMA came to a rational conclusion. Therefore the SCOP’s first ground of appeal should be rejected.

[93]In any event, there is no need for any exceptional circumstances to be shown before the CMA can conclude that a business has been transferred despite its earlier termination. In any event, the applicable principles are those set out in Eurotunnel 1. CAT 1 clearly held that:

“Continuous trading is not essential”

. This was not appealed.

[94]Mr Harris submits that there can be a transfer of employees even if TUPE does not apply. These ferries ply exactly the same route with exactly the same people and exactly the same vessels. In reality there had been the acquisition of a business. Importantly the ex-SeaFrance pilots had pilot exemption certificates (PECs). The customer lists were important in the freight business.

[95]Mr Harris submits that there is no basis on which it could be said that the decision of the CMA was irrational. CAT 1 made it clear that it was a question of fact and degree. It does not matter if this Court does not agree with the weight they gave to these matters: Runa Begum v Tower Hamlets LBC

[2003] 2 AC 430

, 462 to 463 per Lord Bingham.

[96]On the rulings made by CAT 1, the SCOP has to show that it was starting up a new business. So it is necessary to ask whether it was easier to start up from scratch. That was clearly not the case. A new business simply turned the key to turn on the old business. Intention is a factor: a buyer would be keener to buy if he could turn the key for the old business. What was intended during 2011 was to acquire the business as a going concern. It morphs into a situation where that has not succeeded but the interested parties try to keep that going.

Indemnity point

[97]Mr Harris submits that the indemnity holds things together and provides the missing money which is for the necessary “similar operation” of SeaFrance vessels. There is unemployment in this region. SNCF had to help regenerate employment. The CMA had to look at it in reality and it would take a long time to find crew and to train them. There were no sister ships available on the market. According to the court minutes, the court receiver thought that there had to be compromise between the value of the assets and maintaining employment contracts.

[98]As to the link between the vessels and the employees, the CMA assessed a number of issues and found in particular that the vessels were specialised, and that SeaFrance employees were known to the port authorities and were familiar with the operation of the ships. The vessels could be relatively easily reactivated for further service.

[99]In Eurotunnel 1, CAT 1 held as a matter of law that there was no need for formal continuity of employment contracts: it is sufficient that in reality there is a transfer of the workforce. CAT 1 contemplated that the indemnity might be a cogent reason for GET/SCOP to reemploy ex-SeaFrance employees. Moreover, the amounts contributed by way of indemnity enabled the SCOP to make its financial contribution to the GET/SCOP bid. The SCOP was set up to promote the employment of the SeaFrance employees and the €25,000 indemnity was only payable to the SCOP. Therefore there was material on which the CMA could find that the indemnity had “forged a link" between the vessels and other relevant employees.

Read in context, the errors made by CMA were not significant

[100]Mr Harris relies on the R (o/a National House Building Council) v Monopolies and Mergers Commission (1993) ECC 388 at 398, affirmed [1995] ECC 89 for the proposition that CAT 2 had to have regard to the CMA remittal report as a whole. Therefore this court must consider it in its entirety. He submits: – Factual error 1: The error in the opening sentence of 3.107 was not substantial. – Factual error 2: the reference to the process of the liquidator was likewise not substantial. The reference should have been to the sale process generally. The liquidation did complicate matters because of the necessity for court hearings and formal valuations etc, and in that sense there was no error. The process before the French court took six months which was two-thirds of the period of activity. – Factual error 3: This was likewise insubstantial. The underlying facts were not challenged.

[101]He submits that, as none of the errors was material, it must inevitably follow that the CMA would have reached the same conclusion.

[102]As respects momentum and continuity, Mr Harris submits that the CMA assessed the efforts and activity undertaken during the period of inactivity(a) to maintain the operational capability of the vessels,

(b) to ensure that SeaFrance’s employees were reemployed in the Dover-Calais region, preferably on the SeaFrance vessels, including the structure and amount of the indemnity payment and

(c) to make bids to the liquidator, making clear the intentions and motivation of GET and the SCOP to preserve and resume the former Sea France business. Mr Harris refers to a number of matters, including the statement by the bankruptcy judge that the liquidation of SeaFrance was "not the end of the road".

[103]Mr Harris also relies on GET’s acquisition of other assets, in particular trademarks, domain names, business information systems and software, goodwill and customer lists. These also led to conclusion that there was a business. (3). DFDS’s submissions

[104]Mr Meredith Pickford QC for DFDS, makes a number of submissions which sought to uphold the decision of CAT 2 and which are covered by the description of the submissions that I have already given. I mean no discourtesy therefore by not summarising Mr Pickford’s submissions in full. He makes additional points as well.

[105]Mr Pickford submits that the fact that the SCOP did not go so far as to safeguard continuity of employment does not mean that its role was not a relevant factor in determining whether activities were acquired. Apart from receiving the indemnity, it also supported GET’s bid for the vessels and other assets and so, as matters turned out, re-employment was effectively ensured by the formation of the SCOP. The fact that that outcome was not certain at the start did not change that. The existence of the indemnity showed that some thought had been given to how ex-SeaFrance might be employed in future. The formation of the SCOP was only 3 months before the indemnity was given.

[106]Mr Pickford takes issue with the test of inevitability advanced by Mr Beard and submits, on the basis of the National House Building case, that the court should be slow to set a decision of the CMA aside “unless any perceived errors of law are both material and substantial.” In my judgment, the correct test today when granting remedies in judicial review is that put forward by Mr Beard and I proceed on that basis for the purpose of section 120 EA02. Mr Pickford also submits that the court should be more robust in declining to set aside a decision of the CMA in the light of the practice in employment cases (see Burrell v Micheldever Tyre Services Ltd

[2014] EWCA Civ 716

). However, I do not consider that it would be helpful to elaborate on the issues there raised (imposing limits on the process of reconsideration).

[8]Discussion and Conclusions

[107]It is apparent from this judgment already that there is a wealth of fact in this case. It is, therefore, self-evident that the question whether an acquirer has acquired a business or bare assets will depend on an assessment by the decision-maker, the CMA, of the facts. Those facts will need to be weighed. Each case will turn on its facts. Thus the question will be one of fact and degree in each case.

[108]Parliament has entrusted the decision to the CMA and thus the court will not retake that decision but consider whether the CMA made an error of law. There might be error of law because the CMA has misinterpreted the law or reached an irrational conclusion on the facts. It is not open to the court to take the decision itself, or to set aside the decision, simply because it would not itself have weighed up the facts in the same way.

[109]With that introduction, I turn to the grounds of the SCOP’s appeal.

Hiatus point

[110]CAT 1 held that it is not necessary, for the purpose of establishing that a business rather than bare assets has been acquired, that the activities of the acquired business continue until the date of completion of the transaction. Otherwise merger control law could be easily evaded. CAT 1 held at [106 (1)]: Continuous trading is not essential.

[111]CAT 1 made this comment in the context of a seasonal business but there cannot be any difference between a seasonal business and a business which has been closed down provided that it is one which is capable of operating and that its acquisition is consistent with its future operation. The Hargreaves report is an example of coal mining sites which had ceased production but which were viable and could be operated again (and which were acquired for that purpose). In argument I gave the example, which Mr Harris endorsed, of a drug manufacturer which has had to cease production of a drug because its authorisation has been withdrawn. Suppose it has no technical expertise to overcome that problem. The business is at an end so far as it is concerned. But another drug manufacturer might have the knowhow to recommence production, and regain regulatory approval, and may acquire the rights to manufacture the drug and the laboratories at which it is produced. There would clearly be the acquisition of a business in that case. As already stated, it is all a question of fact and degree, and the AHH Holdings/Medicopharma case set out in the Annex to this judgment shows the level of detail of the investigation that may be required of the competition authority. It also shows that the business need not actually be operational at the date of the acquisition.

[112]I do not accept Mr Beard’s argument that a defunct business can only exceptionally be the subject of a merger situation. There is no indication in the statute that this is intended to be the position, and the court is not in a position to say whether in fact this situation occurs infrequently or not. Moreover, CAT 1 did not decide that there had to be exceptional circumstances and so it would have been inconsistent with the terms of the remittal for the CMA to introduce that pre-condition. There was therefore no error of law in the approach of the CMA to the hiatus.

[113]The next question is whether the CMA’s decision to hold that there was an acquisition of a business notwithstanding the hiatus was irrational so that it could be set aside under section 120 EA02. On the one side of the equation, there is no doubt that the services of the ex-SeaFrance employees had been terminated. In addition there was a long period – seven months or more - between the termination of their services in January 2012 and the recruitment of new staff by the SCOP from July 2012. However, a significant number of employees (190 or about 15% of the workforce) were retained to operate the vessels in hot lay-up. Furthermore, while it was a long time, it may be relevant that this was a region with high unemployment: we do not know. The structure of the bid was important – this is not a case where the SCOP comes along after the event to offer to provide crews etc. The SCOP was closely involved in GET’s bid so that it was clear that there was a plan to offer employment to ex-SeaFrance employees.

[114]On the other hand, the impact of the hiatus has to be seen in the context of the facts of the case. Undoubtedly, the CMA regarded it as most important from the point of view of operating the vessels that GET/SCOP should obtain the services of the ex-SeaFrance employees. Now it could have happened that by a pure coincidence those employees were available to be recruited by the SCOP after the vessels were acquired by GET, in which case the decision-maker would fall into the trap of holding that there was the acquisition of a business post hoc (which CAT 1 warned against), if it found that, because of the recruitment of the same workforce, the acquirer acquired the old business. But the fact is, as the CMA explained and as I shall in part explain in the next part of this judgment, that there was a quite different sequence of events. I do not, therefore, consider that the decision of the CMA could on this ground be set aside as irrational.

[115]I can conveniently take here Mr Beard’s point that the CMA did not follow its own guidance in which it states that the length of the period of inactivity is a matter which has to be considered. The Remittal Report shows that the CMA did give careful thought to the events in that period and thus inevitably it considered the length of the period. I do not therefore consider that this point takes the SCOP any further.

[116]Likewise I can dispose of the point that the CMA were in error in proposing that a jurisdictional test should be applied in a broad and purposive way. It is not entirely clear to me what the CMA had in mind in its observations but I agree with CAT 2 that the CMA had by this stage come to its conclusions and set them out, so this passage cannot form part of its reasoning or the basis for setting aside its decision. THE

Indemnity Point

[117]Mr Beard also submits that the indemnity could not turn the acquisition of the vessels into the acquisition of a business because the business was not a going concern, but I would reject that point for the same reasons as I reject the argument that a defunct business can only exceptionally be the subject of a merger situation.

[118]Mr Beard’s point is that there were many uncertainties and the SCOP could not ensure re-employment of ex-SeaFrance employees. There was no guarantee that the SCOP’s bid would succeed. There was no TUPE scheme and the CMA failed to take into account the impact of the lack of a TUPE scheme. This was not, on his submission, a case of “migration” of a workforce.

[119]This is again raises questions of fact and degree which are primarily for the decision-maker. The Remittal Report shows that, from October 2011, there was a conscious effort to promote re-employment of the ex-SeaFrance employees. The SCOP was formed and three months later the French court approved the indemnity. This, the CMA found, provided a strong incentive for the SCOP to be involved in the operation of the vessels. The plan for employment of the ex-SeaFrance employees was also in fact very successful given the proportion of employees of the SCOP who were ex-SeaFrance employees was at October 2012 some 70% to 80%. The proof of the pudding, so far as the incentive and the promotion of reemployment was concerned, was thus in the eating, and that was not dependent on post hoc reasoning. Furthermore, the delay in payment under the indemnity is not stated to have had any impact on the acquisition.

[120]However, I accept the point that the SCOP could not guarantee that it would be able to ensure re-employment of the ex-SeaFrance employees. It all depended on a suitable bid for the assets being accepted. I also accept that events could have turned out quite differently. Another ferry operator could have tried to take over the Dover-Calais route with its own vessels and offered employment to all the ex-SeaFrance employees before any bid involving the SCOP could be accepted. (In fact we understand that DFDS did start operating on the Dover-Calais route using its own vessels and some ex-SeaFrance employees in February 2012). But the fact that that could have happened is beside the point. As Mr Pickford pointed out in his submissions, the safeguarding of employees’ interests through the SCOP and by the indemnity did not have to be continuous. What mattered was that at the date of the acquisition by GET/SCOP, the participation of the employees could be assured.

[121]However, continuity is, as I see it, logically relevant to the question whether the workforce “transferred” with the vessels. There has to be some link between the employees of the old business and the employees of the business on acquisition for it to be concluded that that the transfer was one of a business. In the absence of a TUPE scheme, or some other statutory route providing for the transfer of contracts of employment, the transfer of the workforce could only occur on an informal basis, i.e. if the workforce “migrated”, and voluntarily moved to new posts with the SCOP without any legal mechanism for transfer. The SCOP accepts that an informal transfer of this nature would be sufficient to give rise to a merger situation. The CMA took the view that, in all the circumstances, there had been a migration of the workforce. The SCOP challenges that conclusion.

[122]What constitutes a migration of employees is inescapably a question of evaluation of all the material facts. The SCOP undoubtedly recruited a substantial percentage of ex-SeaFrance employees to work in the ferry business. It is difficult to see what factor was missing which prevented a rational conclusion that there was a migration of the workforce. Mr Beard did not assist on this point.

[123]The indemnity was an important part of the jigsaw. Given that the conclusion of the CMA about migration of the workforce turned on its assessment of the facts, which I have summarised above which are not themselves challenged, I do not consider that the SCOP has shown that the CMA’s conclusion that there was a migration of employees can be said to be irrational.

The three errors made by the CMA

[124]The three errors share a common characteristic. They occur in sentences in which the CMA tries to sum up the facts or draw inferences. Mr Harris submits that all the errors were insubstantial. Moreover, the underlying facts had been set out by the CMA in great detail earlier in the Remittal Report. Mr Beard contends that it was wholly incorrect to say that the employees’ contracts had not been terminated with no thought as to the future or that the SCOP safeguarded their position throughout. I see the literal force of Mr Beard’s point but the fact is that the CMA set out the facts in detail earlier in its report. In my judgment it is inconceivable that the CMA intended these passages to do other than represent the facts which they had found and set out in the Remittal Report. In those circumstances I consider that the errors were peripheral to the CMA’s reasoning process. In my judgment, the conclusion of the CMA would inevitably have been the same if the errors were corrected since, as I have said, it had identified the correct facts earlier in the Remittal Report and those facts are not challenged. Moreover, I accept Mr Harris’s submission that, in relation to what I have called the Second Factual Error, there was in fact no error in stating that a considerable part of the period of inactivity was due to the liquidator’s sale process (see paragraph 102 above).

[125]I have considered whether the Factual Errors 1 and 3, if viewed cumulatively rather than separately, would lead to any different result. In my judgment, they do not become more serious if so viewed because it remains the fact that the CMA in both cases reviewed the detailed facts.

[126]There remains the conclusion of the CMA at para. 3.54 of the Remittal Report that: Overall, we consider that a review of the background to the transaction shows that there is considerable continuity and momentum between the time of SeaFrance’s operation of the Dover–Calais ferry and the commencement of MFL’s operation of the same ferries on that route involving ex-SeaFrance employees.

[127]Mr Beard has challenged this conclusion. It is undoubtedly correct that there was no straight line between these two times, and there were times when the eventual outcome was not clear. But I do not consider that the CMA was saying that there had to be consistent and regular progress between these two points in time. This statement in my judgment has sufficient basis in the fact that events, such as putting the vessels into hot lay-up, occurred throughout the period from the formation of the SCOP in October 2011 to the commencement of the MFL’s operations which, together with such matters as the special skill sets of the ex-SeaFrance employees and the special features of the vessels acquired, contributed to the end result. Clearly the consistent aim, which was a laudable and important aim, of many participants in these events was that, if all possible, the livelihood of the ex-SeaFrance employees should be preserved.

Overall conclusion

[128]For the reasons given above, I would dismiss this appeal. Annex

MMC REPORT IN AAH HOLDINGS PLC /MEDICOPHARMA NV

[1]The issue in AAH was the impact of the cessation of trading by the UK activities of Medicopharma on 3 November. The issue therefore was what AAH acquired. There was an asset transfer agreement and share purchase agreement. AAH accepted that in anticipation of these agreements it took steps designed to secure commercial advantages to itself but it contended that those steps had to be left out of account in ascertaining whether the enterprises of Medicopharma and AAH had ceased to be distinct. Medicopharma submitted that no merger occurred and that all that happened was that AAH acquired stock, assets and three depots. AAH argued that the burden of proof was on MMC. MMC rejected this argument. This task was to determine the facts on the basis of a balance of probabilities.

[2]The MMC rejected the argument that it could only have regard to the asset transfer and share purchase agreement. It considered that the phrase “brought under common ownership or common control” were to be interpreted in accordance with its context in section 65(1) of the Fair Trading Act 1973 (“the 1973 Act”) (enterprises ceasing to be distinct enterprises) and having regard to the purpose of the 1973 Act. Accordingly, given that one of the intentions and purposes of the Act was to enable the MMC to consider commercial realities and results, the MMC was not limited to legally enforceable agreements. The MMC did not consider that the relevant agreements were shams but it was clear that the shared purchase agreement was part of a set of arrangements which also included a side letter of 3 November from AAH to Medicopharma undertaking to use reasonable endeavours to assume the obligations and liabilities of Medicopharma in respect of pharmacy loan schemes and the issuing of redundancy notices to employees at Medicopharma. Evidence was given to the MMC that AAH would sign the agreement to purchase the various assets as soon as the redundancy letters were sent and would not do so unless this happened. The MMC found that the transaction as explained to AAH’s board was that Medicopharma would close its business and that AAH would purchase specified assets once closure had been affected. The MMC found that it was the common understanding of the parties that the redundancy letters would be sent out and a share purchase agreement would be completed before any public announcement of the decision to cease trading. It was also part of the arrangements that there would not be any period after the public announcement that Medicopharma had ceased to trade during which Medicopharma owned and controlled the three depots from which it conducted its business.

[3]The MMC accepted that, if a company decides to cease to trade, its decision, and whether and to what extent it had been given effect, is a relevant factor in considering whether the activities of a business which a company has previously carried on has been brought under common ownership or common control with enterprises of another. The mere fact of ceasing to trade did not mean that its business could not be transferred as a going concern or that its activities could not be brought under common ownership or common control with enterprises of another.

[4]The MMC considered that the position could be tested by asking what the position would be in a case where a decision to cease trading had been taken and in which existing contracts with customers and suppliers were transferred. The MMC considered that in such a case “part of the activities of a business” might well be transferred notwithstanding the decision by the transferor to cease trading. A decision to cease trading could not in itself determine that no part of activities of a business had been transferred.

[5]The MMC considered the situation lay between an acquisition of assets and an acquisition of a business as a going concern. The MMC considered that it was a question of fact and degree whether activities were brought under common ownership. So it had to consider whether part of the business carried on by AAH after the arrangements should as a matter of commercial reality be regarded and treated as only its own business or as in whole or part a continuation of the activities of the business of Medicopharma at the three depots.

[6]The MMC noted that most business with retail pharmacies and wholesalers was not on the basis of long-term contracts and the fact that there were no transfers of contracts was therefore not determinative. What was more important was the preservation of the customer base of a pharmaceutical wholesaling business such as that of Medicopharma. Thus contact between a retail pharmacy and a particular depot from which it was supplied was important. Goodwill was therefore attached to the depot and in addition AAH took over the telephone and fax numbers within the depots. There were arrangements to ensure that customers telephoned the depot that morning and were informed that AAH had acquired the depot which would be operating fully as soon as possible and that meanwhile orders could be placed at another AAH depot. Thus AAH had prior knowledge of the timing of the closure of Medicopharma. Furthermore the three depots began operating fully again on 7 and 8 November. There was no sale of goodwill but employees had knowledge of customers and a relationship with them. Particular members of staff were asked to report for duty on 4 November. They were taken on by AAH. AAH compensated Medicopharma for the costs of the redundancy payments. It took on a large number of former members of staff of Medicopharma. Former Medicopharma employees accessed the computers of Medicopharma to obtain telephone numbers. Medicopharma did not ask its liquidators to sell the customer lists at the depots. AAH would have wished to acquire Medicopharma as a going concern had it not been for the wish to avoid investigation by the competition authorities. Stock was also acquired.

[7]The MMC concluded that although AAH did not in terms acquire the depots as going concerns, in reality it obtained much of the benefit of so acquiring them and it clearly acquired more than the bare assets. It obtained three depots complete with stock and fixtures and fittings which would have carried with them a certain degree of goodwill. It also acquired the computers in those depots to which computers or terminals of Medicopharma’s customers would have had access and the telephone and fax numbers of those depots. In the particular industry orders are placed and deliveries made twice daily and retail pharmacies need to find an immediate source of supply. The arrangements involved exclusive prior knowledge for AAH of the facts and timing of the disclosure. Other wholesalers could not recruit a large number of new customers from Medicopharma when it announced its closure. The arrangements were structured to ensure that AAH could take on employees of the depots almost as surely if it had acquired those depots as going concerns. Accordingly it gained the benefit of those employees’ knowledge of Medicopharma’s customers as well as the benefit of their relationship with those customers.

[8]The MMC concluded that the overall effect of the arrangements, agreements and understandings was that AAH acquired control of part of the activities of the business of Medicopharma at the three depots and continued to carry them on. Therefore the MMC concluded that the enterprises of Medicopharma and AAH had ceased to be distinct in the manner described in Section 65(1)(a) of the Fair Trading Act 1973 (Enterprises ceasing to be distinct enterprises).

Lord Justice Tomlinson:

[129]I have had the advantage of reading in draft before preparing my own the judgments prepared by Arden LJ and Sir Colin Rimer. I agree with Sir Colin Rimer that the appeal should be allowed. As the court is divided and as we are differing from the CAT I add some observations of my own, but they should be regarded as supplementary to the judgment of Sir Colin Rimer, with which I agree.

[130]The CMA’s Guidance on its mergers jurisdiction and procedure, CMA2, January 2014, prescribes that in assessing whether a relevant merger situation has arisen, “the CMA will assess all relevant circumstances (including whether there is evidence that the closure of the business was designed to avoid merger control), with a view to determining whether the target business constitutes an enterprise under the Act” – paragraph 4.11. The necessary corollary of this is, in my view, that the CMA should also take into account as relevant the circumstance that the closure of a business has not been designed to avoid merger control.

[131]I make this observation at the outset because in my view the CMA has given little or no weight to its acceptance, at [4.26] See also [3.56]. of its remittal report, “that there was no intention to avoid merger control in this case.” That was said in the context of the CMA’s observation, [4.24], that a failure by it to interpret the relevant statutory provisions widely and purposively “would invite gaming of the system and the structuring of transactions and arrangements in such a way as to avoid merger control.” The CAT at [86] of its judgment in Eurotunnel 2 has noted that the premise of [4.24] of the remittal report is flawed, and that the relevant statutory provisions neither require nor should be given an expansive interpretation. It seems to me that the situation here under consideration is very far removed indeed from that considered in AAH Holdings Plc and Medicopharma NV where, as Sir Colin Rimer points out, the acquisition had been deliberately structured and the steps in it pre-ordained, in an attempt to circumvent the application of the UK merger regime. Here, the cessation of the SeaFrance business, and the timing thereof, was mandated by the French court. Likewise, the PSE 3 job-saving plan on which the CMA placed so much reliance was, as the CMA pointed out at [2.34] of the remittal report, a statutory arrangement, negotiated between the liquidator and the SeaFrance works council, made up of employee representatives, and was itself subject to approval by the French court. The relevant steps here were not within the control of SeaFrance, GET or the SCOP. It is therefore to my mind an inadequate characterisation of the situation simply to accept that there was no intention to avoid merger control. Not only was there no such intention, even had there been such an intention the parties would not have been in a position to ensure that it was achieved. I recognise of course that this consideration is not decisive of the objective question which the CMA had to resolve, whether the relevant activities had been brought under the ownership or control of GET/SCOP. However, against that background the conclusion of the CMA is I think counter-intuitive.

[132]The CAT in Eurotunnel 1 observed at [106]:- “In this context, it is necessary to make two points: (a) first, it is perfectly possible for an enterprise to wind down, and to wind down to such an extent that it ceases to be an enterprise. The mere fact that in the past the activities of a business were being carried on by an entity does not necessarily mean that, as at the time of the merger, that entity was an enterprise. Of course, it is also important to recognise that some businesses (eg. those involved in tourism) trade for some periods and not for others (eg. during the “low season”). Such a hiatus does not preclude the existence of an enterprise. Continuous trading is not essential.” (a) first, it is perfectly possible for an enterprise to wind down, and to wind down to such an extent that it ceases to be an enterprise. The mere fact that in the past the activities of a business were being carried on by an entity does not necessarily mean that, as at the time of the merger, that entity was an enterprise. Of course, it is also important to recognise that some businesses (eg. those involved in tourism) trade for some periods and not for others (eg. during the “low season”). Such a hiatus does not preclude the existence of an enterprise. Continuous trading is not essential.”

[133]The period of inactivity here was seven and a half months. I note that in this regard the President of the CAT in the course of argument in Eurotunnel 2 observed:- “There has certainly been no case in the whole history of UK merger control where you have such a long gap.” The President also observed that for a company to stop trading for seven and a half months is not the normal activity of an all year round trading company, rather it is quite abnormal and extraordinary and such as to prompt the question, in this context, is that business still going on. Mr Harris, for the CMA, acknowledged the force of this point and explained that this was why “so much time and attention is devoted to it by the CMA in its report,” which in another context he characterised as “pages and pages.” This is another respect in which I regard the decision of the CMA as counter-intuitive. With respect it was dealing with something very far removed from a hiatus in trading of the sort which the CAT in Eurotunnel 1 had in mind. For my part I find the analogy with a seasonal ice cream seller wholly unhelpful. In the many pages devoted to explaining why its initial conclusion is sustainable despite the criticism of it by the CAT in Eurotunnel 1, the CMA has had, as Sir Colin Rimer observes, to resort to language indicative that it has not given to the statutory definition of enterprise its straightforward meaning.

[134]In these circumstances I would require cogent and powerful reasons to justify the counter-intuitive conclusion. In agreement with Sir Colin Rimer, I do not find the reasons given by the CMA either cogent or compelling, and like him I conclude that the decision of the CMA falls outside the ambit of reasonable decision-making.

[135]At [105] the CAT in Eurotunnel 1 posed the question as whether, even though the business of SeaFrance had been wound down to a very considerable extent, there still remained the embers of an enterprise. I do not entirely accept this way of putting the question, since its formulation begs the question of the extent to which the business had been wound down. However, on any view by 9 January 2012, if not before, SeaFrance had ceased trading, and it had by 9 January become unlawful for it to continue trading. Wherein then lay the embers? Two matters are, it seems, relied upon. First, the circumstance that the vessels were maintained in hot lay-up. Second, the provision of the PSE 3 job-saving plan which was also approved by the French court on 9 January 2012.

[136]I do not understand how the circumstance that the vessels were maintained in hot lay-up assists in the evaluation of the question whether after 9 January 2012 the relevant activity of SeaFrance subsisted. The vessels were no longer in service. They had been ordered to be sold. Maintaining them in hot lay-up would render them more attractive to a prospective purchaser who wished to employ them in the forthcoming high season on any number of potential European sea routes, not to mention those further afield. The price payable would reflect the need or absence of need to take the vessel out of conventional lay-up. No shipowner likes to lay up a vessel, but the decision whether to do so and in what manner is essentially a financial one. The liquidator would no doubt not have been entitled to maintain the vessels in hot lay-up had that not been demonstrably conducive to a more beneficial sale in the liquidation. The CMA perhaps recognised this at [3.48]:- “Third, considerable efforts were made to maintain the value of the assets during the period of inactivity. One of the liquidator’s aims was to realise the assets at the best price in order to ensure that payments to the company’s creditors would be maximised. An expert shipbroker, Parimar, was engaged to advise the liquidator in this regard. The vessels were put into hot lay-up and 190 SeaFrance staff were involved directly or indirectly, in their maintenance.” I note incidentally from footnote 71 to this paragraph that less than half of these staff were subsequently employed by the SCOP. I doubt if the remainder would agree with the proposition that continuity of employment was effectively safeguarded by the SCOP, and that the SeaFrance workforce effectively transferred from SeaFrance to GET/SCOP.

[137]Importantly, the CMA does not suggest that the liquidator adopted this course so that the ships could be more quickly redeployed in the short sea trade, as opposed to other employment, although even if it had that would not in my view have assisted resolution of the essential question whether what ultimately occurred involved the acquiring entity carrying out the same activity as that once carried out by the acquired entity, or rather reconstructing a business that was once conducted by a different entity – see the formulation by the CAT in Eurotunnel 1 at [106 (b)(ii)].

[138]The critical finding of the CMA has therefore to be that the provisions of PSE 3 “forged a link between the vessels and the employees” which “had both the purpose and the result that a significant number of ex-SeaFrance employees were employed by the SCOP” – CMA [4.11] and CAT in Eurotunnel 2 [79]. Allied to this finding is that of the CMA that there was “a considerable continuity and momentum between the time of SeaFrance’s operation of the Dover-Calais ferry and the commencement of MFL’s operation of the same ferries on the route involving ex-SeaFrance employees – [3.54].

[139]Dealing with the latter point first, the CAT in Eurotunnel 1 had encouraged investigation of any “momentum or continuity in the combination between the vessels and workforce that takes this case over the line from an asset acquisition to the acquisition of an enterprise.” I am not sure that I entirely understand what is meant by the expression “momentum or continuity in the combination between the vessels and workforce” but in any event there was plainly, in my judgment, no such momentum or continuity as the greater part of the workforce was dismissed consequent upon the cessation of trading by SeaFrance. They had no guarantee of re-engagement. Further, whilst the provisions of PSE 3 might have created an incentive to use the ships in a “similar operation” to that of SeaFrance, whatever precisely that means, the vessels could equally have proved attractive to an operator on another route with a pressing need, or an opportunity for profitable employment, which rendered the passing up of the bounty offered by PSE 3 of no consequence.

[140]For the reasons I have already given, there was still less momentum or continuity between “the time of SeaFrance’s operation of the Dover-Calais route and the commencement of MFL’s operation.” Whether the vessels or the workforce would be employed in that way was dependent on a number of imponderables over which SeaFrance and GET/SCOP had no control. The CMA accepted that GET’s involvement was, in its own words, opportunistic: it saw a business opportunity and it seized it – [3.56].

[141]We are left therefore with the “link” between the vessels and the workforce. With respect, I do not understand how the existence of this “link” assists in the question whether an enterprise was acquired. The statutory definition of enterprise required concentration upon the identification of an ongoing activity, not the identification of a link between the assets acquired. The existence of an incentive seems to me to tell against rather than in favour of the momentum and continuity which the CMA sought.

[142]I should add that for my part I see great force in Mr Beard’s submission that the three key errors in the CMA’s analysis identified by the CAT in Eurotunnel 2 vitiated the CMA’s conclusion. The three errors relate to:-(i) The conclusion that the terms of PSE 3 demonstrated that it is not the case that the SeaFrance employees’ contracts of employment were terminated with no thought as to how they might be re-employed in future, a conclusion which was in any event not open to the CMA in the light of [115] of Eurotunnel 1;

(ii) The relevance or significance of the time taken by the requirements of the liquidator’s sale process; and

(iii) The conclusion that continuity of employment was effectively safeguarded by the formation of the SCOP. I reject the ambitious submission of Mr Pickford, for DFDS, that these were “linguistic infelicities” rather than errors of analysis. I have already discussed points (i) and (iii). Point (ii), as the CAT in Eurotunnel 2 observed at [83], overlooks that the liquidation was itself a consequence of there being no viable purchaser for SeaFrance as a business. However, as will be apparent, I do not think it necessary to analyse this aspect in terms of whether these findings, by definition irrelevant factors to have taken into consideration, played a significant or substantial part in the decision-maker’s thinking, as discussed by this court in R (F.D.A) v Secretary of State for Work and Pensions

[2012] EWCA Civ 332

at paragraphs 67 and 68. Rather, I regard these errors of analysis as in themselves part and parcel of and indicative of a flawed process of reasoning which has lead to an irrational conclusion. They are key building blocks in an overall flawed analysis.

[143]For these reasons, as well as those given at greater length by Sir Colin Rimer, I conclude that there was no rational basis upon which the CMA could conclude “that the collection of tangible and intangible assets (including the transferred ex-SeaFrance employees) that GET/SCOP acquired meets the legal definition of an enterprise in that together they constitute the activities or part of the activities of a business” – [4.20]. I too would allow the appeal.

Sir Colin Rimer:

[144]I have had the advantage of reading Arden LJ’s comprehensive judgment in draft. I have, however, come to a different conclusion as to the disposition of the appeal. I would allow it.

[145]Following the remittal to the Competition and Markets Authority (‘the CMA’) by the Competition Appeal Tribunal (‘the CAT’) by its decision of 4 December 2013 in Eurotunnel 1

[2013] CAT 30

, the CMA had to reconsider (inter alia) the same fundamental question that the Competition Commission (‘the Commission’) had originally sought to answer by its report of 6 June 2013. That was whether, upon or following the acquisition in July 2012 by Groupe Eurotunnel SA (‘GET’) of the three vessels and other assets that it purchased from the liquidator of SeaFrance SA (‘SeaFrance’), ‘the activities, or part of the activities’ of SeaFrance were brought under the ownership or control of GET/SCOP acting together as associated persons: the SCOP had earlier come to an arrangement with GET that, if GET’s bid succeeded, it would provide the labour required to operate the vessels. That way of posing the critical question is, I consider, a sufficient summary of the combined sense of the relevant provisions of sections 23, 26, 127 and 129 of the Enterprise Act 2002. If the answer to the question was yes, the CMA was entitled to conclude that a ‘relevant merger situation’ had arisen. If no, it was not. The CMA answered the question affirmatively and the CAT, in the decision now under appeal (

[2015] CAT 1

), rejected GET/SCOP’s challenge to the rationality of that answer.

[146]The key word in the question is ‘activities’: Parliament’s intention in the Enterprise Act 2002 was that it was only if activities were brought under common ownership or control that a ‘relevant merger situation’ would be capable of arising. The ‘activities’ of SeaFrance were the provision of ferry services (passenger and freight) across the short sea, using the four vessels that it did, staffed with the crews that it employed. It must therefore have been the CMA’s conclusion that it was those ‘activities’ that were brought under the ownership or control of GET/SCOP in or following July 2012. It might, however, be said that there is an apparent logical difficulty with that conclusion, namely that SeaFrance had ceased its ferry activities in November 2011 and did not thereafter resume them. Nor, by 9 January 2012, was there any prospect that it could or would do so. It is important to consider the events leading to the demise of SeaFrance.

[147]On 30 June 2010, SeaFrance entered into the equivalent of administration. During the administration, with financial support from SNCF, SeaFrance’s parent, the administrators continued to operate SeaFrance’s business whilst also seeking, unsuccessfully, to sell its business as a going concern. Towards the end of 2010 the administrators made some 354 of SeaFrance’s employees redundant, leaving a workforce of 872 employees. SeaFrance had been heavily overmanned.

[148]The SCOP, a workers’ cooperative, was formed in October 2011 by a group of 14 ex-SeaFrance employees who had foreseen SeaFrance’s likely final demise and whose objective, via the SCOP, was to secure employment for SeaFrance’s then present and former employees, in particular by ensuring that SeaFrance’s vessels continued to trade between Dover and Calais. The SCOP had, before its formal establishment, itself made an unsuccessful bid for the SeaFrance business.

[149]On 16 November 2011, SeaFrance’s administration came to an end and, by order of the Paris commercial court, SeaFrance entered into compulsory liquidation. The court authorised the liquidator to continue to carry on SeaFrance’s business until 28 January 2012, although he did not in fact do so and SeaFrance’s operations ceased on the night of 15/16 November 2011. By the end of 2011, the SCOP had 827 subscribers, who had made a minimum contribution of €50 per head and had done so on the basis, or hope, that the SCOP would acquire the SeaFrance fleet. In December 2011, the SCOP made a further bid for the acquisition of the SeaFrance business (for €1, and not very different from its previous bid) but it came to nothing. The SCOP’s problem was that it lacked the finance to carry out its plan of operating the vessels on the short sea crossings.

[150]On 9 January 2012, the French court formally ordered the end of any continuation of SeaFrance’s business (and so, prima facie, also of its ‘activities’) and ordered the liquidator to sell its assets by way of a sealed bid private sale. That order triggered an obligation under French law to make the SeaFrance employees redundant within 15 days. There were then about 820 employees and the liquidator promptly gave redundancy notices to some 630 and dismissed them. The liquidator retained the remaining 190 employees in order to assist him with the liquidation, in particular to keep the vessels mothballed in a state of ‘hot lay-up’ which would enable any buyer to put them into serviceable operation more easily and quickly than if they were not so maintained. Those retained employees were destined to be made redundant and dismissed within the following ten months or so.

[151]Also on 9 January 2012, the court approved the PSE3 job-saving plan (‘plan de sauvegarde de l’emploi’), a plan of a type required to be entered into under the French Labour Code for the benefit of former employees of a liquidated company, although the particular terms of PSE3 were the fruit of negotiation rather than statutorily imposed. The Code prescribes that when in a company of at least 50 employees it is envisaged that at least ten employees are to be made redundant, the employer must establish such a plan in order to avoid redundancies or to limit them in number and to facilitate the redeployment of people for whom redundancy cannot be avoided. PSE3 recognised that, as the French court had decided on 9 January 2012 upon the cessation of the continuation of SeaFrance’s business, SeaFrance was itself no longer able to offer redeployment solutions. PSE3 recorded that, insofar as redeployment opportunities were available in other companies in the SeaFrance group, they would be explored with those ex-SeaFrance employees who had the requisite skills.

[152]Section B of PSE3 was devoted to measures aimed at facilitating the return to work of those employees who could not be redeployed within the SeaFrance group and were made redundant. Paragraph 3 set out the various heads of assistance that would be provided to help such former employees to obtain re-employment, with paragraph 3.3 setting out the types of benefit by way of payments that could or would be made to them. Paragraph3.3.1 provided for a payment of €10,000 to each employee who obtained a defined minimum interest in a variety of types of business entity in France or other European Union countries and actually worked there. Paragraph

3.3.2 increased that to €15,000 per employee in the circumstances there prescribed, which extended to employees ‘creating or taking over an enterprise locally with the aim of contributing to economic development either in that area or within a radius of 50km of the job location in the Calais region …’. Paragraph

3.3.3 included the provision material to the present case:

“Where the bankruptcy judge in the liquidation of SEAFRANCE has to rule upon an assignment in a final ruling allowing similar operation of the vessels belonging to SEAFRANCE in favour of [the SCOP] or any other company (of any form) in which the employees have a direct interest (share of the equity capital) and indirect interest (employment contract), the company will be paid €25,000 per employee, on receipt of the official documents (a list of employees concerned, proof of holding of share of the capital and employment contracts). The aid of €25,000 gross may not be combined with the aid of €10,000 gross referred to above in paragraph 3.3.1.”

[153]The €25,000 payment has been described as an ‘indemnity’, although I am not clear why. The payments were to be made by SeaFrance’s parent, SNCF. One of the reasons why SNCF undertook that burden was explained in paragraph 4 of section B:

“Because of its particular sensitivity about employment, the SNCF Group has made the voluntary commitment to contribute to the measures implemented by SEAFRANCE in the present Employment Safeguard Plan aimed at facilitating the external redeployment of SEAFRANCE employees through action to develop the employment catchment area in the Calais region.”

[154]It is agreed that the paragraph 3.3.3 indemnity was intended to be payable in respect of each ex-SeaFrance employee who was later re-employed on former SeaFrance vessels that were purchased from the liquidator for use for an ‘operation similar’ to that of SeaFrance. There is no question that PSE3 created a high likelihood that any buyer of SeaFrance’s vessels and other assets for the purpose of using them for such a similar operation would be approached by the former SeaFrance employees for employment and would readily take them on. This duly happened; and the ‘link’ between the PSE3 payments made to the SCOP by SNCF (in fact only paid in and following early 2013, following litigation brought by the SCOP) and the taking on by the SCOP of the former SeaFrance employees played a key part in the CMA’s conclusion that the GET/SCOP acquisition of SeaFrance’s assets resulted in the creation of a ‘relevant merger situation’ for the purposes of the Enterprise Act 2002. The CMA must, therefore, have decided that the effect of the acquisition was that SeaFrance’s ‘activities’, or part of them, were brought under the ownership or control of GET/SCOP.

[155]GET made its bid to buy the SeaFrance assets in May 2012. The bid detailed its plans for the ‘recommencement of SeaFrance operations’ and stated that ‘the project for which [GET] is signing up is intended … to allow a partnership with the former SeaFrance employees who will be involved as part of a SCOP, so as to revive [faire renaître] the operations previously undertaken by SeaFrance.’ That language does not unambiguously suggest that GET regarded its intended project as involving the continuing, or the taking over the control of, SeaFrance’s activities rather than simply starting up similar operations of its own.

[156]In July 2012, GET, in association with the SCOP, acquired from the liquidator three of SeaFrance’s four vessels and certain other assets (brand, goodwill and customer lists, but not customer and supplier contracts); and MyFerryLink SAS (operating under the ‘MyFerryLink’ brand) started to trade on 20 August 2012, some seven weeks later. MyFerry used two of the three vessels for ferry operations and the third as a freight only vessel. The vessels, owned by GET subsidiaries, were chartered to the SCOP by three separate charterparties.

[157]Crucially, many ex-SeaFrance employees who had been dismissed as redundant by the liquidator became re-engaged by the SCOP; and the crews of the three vessels bought by GET became largely, but not exclusively, made up of such employees. They were a particularly attractive workforce because not only did they know the ropes of the short sea crossings, they each came with the benefit of the €25,000 indemnity payable by SNCF to the SCOP.

[158]We had much argument about whether the ‘activities’ of a business can be regarded as continuing even though on one view they might be said to have ceased. In particular, we were given the example of a seasonal seaside business that trades from, say, May to October and then closes until the following May. If such a business is sold in, say, December, are its (on one view, then inactive) ‘activities’ brought under the ownership or control of the purchaser? The answer is likely, as usual, to turn on the particular facts, but even if it is yes (as I regard as likely in most cases) it does not provide a helpful analogy for the purposes of the present case, which is materially different. That is because when SeaFrance was ordered finally to cease its trading activities in January 2012 (they had in fact ceased by 16 November 2011), there was no prospect of their being resumed in the future. Their continuation was judicially prohibited, SeaFrance’s ferry activities were therefore at an end and it was anyway incapable of carrying them on. All that remained to be done was for the liquidator to dismiss its employees as redundant, dispose of its assets and, presumably, then to give SeaFrance its final quietus under French law.

[159]We were not shown any judicial authority on the question of when ‘activities’ are or are not said to be continuing for the purposes of the Enterprise Act 2002. We were, however, shown the May 1992 report by the Monopolies and Mergers Commission as to whether a ‘relevant merger situation’ had arisen upon the acquisition by AAH Holdings plc of the assets of Medicopharma NV, a case turning on provisions in the Fair Trading Act 1973 that are in the material respects identical to those in the Enterprise Act 2002. Medicopharma had ceased to trade and dismissed all its employees immediately before the AAH acquisition, so enabling both parties to assert, as they did, that all that AAH acquired from Medicopharma were stock, certain assets and three depots – but no ‘activities’, which had ceased. It is apparent that the acquisition had been deliberately structured, and the steps in it pre-ordained, so as to enable the making of such assertion. The MMC declined to find that any of the arrangements to that end was a sham but, for reasons summarised in [6.102] of its report, nevertheless concluded that although AAH did not in terms acquire the depots as going concerns, in reality it obtained much of the benefit of so acquiring them and it clearly obtained more than the bare assets it claimed to have acquired. The conclusion was, therefore, that Medicopharma’s ‘activities’ were brought under AAH’s ownership or control, although the report does not in my view offer a very clear statement of the principle that it applied in arriving at such conclusion.

[160]We were also referred to the relevant parts of the CAT’s decision in Eurotunnel 1

[2013] CAT 30

, which concerned the GET/SCOP challenge to the original Commission report. The Commission expressly accepted before the CAT that the ‘activities’ carried out by SeaFrance with which the case was concerned were ‘the provision of ferry services (both passenger and freight) across the short sea, using the vessels that it did staffed with its crews …’ (see [111] of the CAT’s decision).

[161]The key part of that CAT decision is in [105] and [106], which were inspired by [6.102] of the Medicopharma MMC report. The essence of the guidance that the CAT provided is that:(i) the mere acquisition of what the CAT calls ‘bare assets’ will not constitute the acquisition of ‘activities’; but

(ii) if the acquirer can be said to have acquired ‘something more’ than bare assets that can be said to have placed it in a different position than if it had simply gone out into the market and acquired the assets, then

(iii) a further question arises (one of fact and degree, with no single criterion giving a clear answer) as to whether such difference ‘is capable of constituting what would otherwise be bare assets into something that may properly be described as the activities of a business.’ The reference to ‘activities’ is necessarily to those of the business whose assets the acquirer has acquired. The CAT explained that in a case where the acquirer takes over the business of the acquired entity, the answer is self-evident: the same enterprise is simply continuing under different ownership. The CAT continued in [105]:

“… The difficult case arises where the combination of assets is fractured, such that the assets are no longer, or no longer to the same extent, being used in combination. This case is a particularly good one, where what was clearly once an enterprise was wound down: the difficult question is whether, even though the business of SeaFrance had been wound down to a very considerable extent, there still remained the embers of an enterprise.”

[162]I do not find the CAT’s test an easy one. Its application is not made easier by the qualifications added in [106], where the CAT explained that:(i) it is perfectly possible for an enterprise to wind down to such an extent that it ceases to be an enterprise; and

(ii) the fact that the acquiring entity uses the acquired assets to emulate the business of the entity from which it acquired the assets does not necessarily mean that the acquiring entity has acquired an enterprise. As the CAT explained:

“… The statutory test is not satisfied if the acquiring entity reconstructs a business that was once conducted by a different entity, even if the assets of that entity were used to do so. The statutory test in section 26(1) turns on two enterprises ceasing to be distinct because they are brought under common ownership or common control. It is critical that there are two enterprises, not one enterprise (the acquiring enterprise) and a collection of assets.”

The concept of an ‘acquiring enterprise’ may, with respect, be inaccurate. The acquiring entity will not be an ‘enterprise’: an ‘enterprise’ in this context means the ‘activities’ of a business.

[163]The CAT noted that the Commission had failed to direct itself as to the difference between the acquisition of ‘bare assets’ (which would not constitute the ‘activities’ of a business) and the acquisition of ‘something more’ than bare assets: i.e. it had not applied the CAT’s approach identified in [105]. The CAT did, however, consider whether the facts found by the Commission enabled a conclusion that ‘something more’ than bare assets had been acquired. The Commission had identified three factors that the CAT regarded as pointing to the conclusion that GET/SCOP had done ‘no more than the acquisition of assets’, namely:(i) SeaFrance’s cessation of operations, or activities, on 16 November 2011 and the court’s prohibition of their continuation on 9 January 2012;

(ii) SeaFrance’s surrender of its Dover and Calais berthing slots; and

(iii) the dismissal of SeaFrance’s remaining workforce and the placing of the vessels in hot lay-up. Whilst the Commission had taken these factors into account, it had regarded them as outweighed by the considerations that: (i) the vessels were acquired in a condition that enabled them to be put into service within two months of the acquisition; (ii) the SCOP had ‘acquired’ former SeaFrance employees, who represented some 75% of the staff engaged in the MyFerry operation; and (iii) the acquisition of the brand and goodwill of SeaFrance, which the Commission said only carried ‘some, but limited, positive value.’ The Commission had also discounted the acquisition of SeaFrance’s customer lists as carrying any separate weight.

[164]The first three points just mentioned pointed away from a conclusion that SeaFrance’s ‘activities’ had come under the ownership or control of GET/SCOP. As to the last three points, upon which the Commission had relied for arriving at a contrary conclusion, the CAT was unconvinced by point(i) in the absence of a fuller explanation of it. It was also unconvinced by point

(ii) on the basis that, on the face of it, the ex-SeaFrance employees formed no part of the GET/SCOP acquisition from SeaFrance. The CAT made no separate comment on (iii), which I infer it did not regard it as, by itself, saving the Commission’s report; and the Commission had itself indicated the limited value it attached to it.

[165]The CAT was not, however, prepared simply to quash the Commission’s decision. It considered there needed to be a fuller investigation into the background to GET/SCOP’s engagement of the former SeaFrance employees than had been done by the Commission – in particular, as to the origins and workings of PSE3. The CAT assessed that the ‘indemnity’ that came with each ex-SeaFrance employee seems to have been ‘a benefit emanating from employing an ex-SeaFrance employee that would not be gained were an employee from elsewhere to be retained’, and they regarded that consideration as possibly relevantly linked to the point about the state of the vessels that GET/SCOP had acquired. They said:

“120. The nature of that relationship, however, is not explored by the Commission in the Decision. The Decision does note that the condition of the vessels was such that they could be brought into operation extremely quickly …, but it does not consider the significance of having a crew (namely, the ex-SeaFrance employees) comprising persons fully familiar with both these particular vessels and their intended operation (across the short sea). It may very well be the case that this combination enabled MyFerry to begin operations much more quickly than it could have done had it acquired crew and vessels from other sources. In short, there may have been a momentum or continuity in the combination between the vessels and workforce that takes this case over the line from an asset acquisition to the acquisition of an enterprise.”

[166]In light of those considerations, the CAT remitted the case to the Commission to re-consider it ‘using the approach that we have described’, which in [123] it described as that set out in [105]. It must have been implicit that the Commission also had to have regard to the qualifying clarification of [105] provided in [106].

[167]I admit to some respectful doubts as to the guidance given by the CAT in [105]. The statutory language in section 129 of the Enterprise Act 2002 defining an ‘enterprise’ as meaning ‘the activities, or part of the activities’ of a business shows in my view that Parliament’s intention was focused only on the case in which the acquiring entity takes over another business as a going concern. I would expect that in most cases that would cover the acquisition in December of the type of seasonal seaside business to which I referred earlier: it would be odd if a seasonal business that is in abeyance pending the arrival of the new season would not be regarded as continuing to be relevantly ‘active’ during that period. It would also have covered this case if GET had bought the SeaFrance business from the administrators, who had sought to sell it as a going concern. The facts with which the Commission were presented in this case were, however, manifestly different. The ‘activities’ of SeaFrance (accepted by the CMA as being short sea ferry crossings for passengers and freight) had, by the time of the GET/SCOP acquisition in July 2012, ceased many months before; SeaFrance had in the meantime been judicially barred from continuing them, and the workforce required for their carrying on had been dismissed as redundant. The notion, therefore, that GET/SCOP were, following the acquisition, in some manner capable of taking over and continuing the same, long-ceased activities (as compared with starting up like activities afresh) is one I do not find easy to grasp.

[168]This appeal is not, however, an appeal against the decision in Eurotunnel 1, nor did Mr Beard QC, for the SCOP, challenge the correctness of the guidance given in it, which he accepted the CMA was obliged to apply. His criticism of the CMA’s decision both before the CAT and this court was as to the CMA’s application of the guidance to the facts. I shall, therefore, say no more about my own doubts, upon which we anyway had no argument. The heart of Mr Beard’s primary case for the SCOP was that the CMA nevertheless still failed to identify the ‘activities’ that were acquired by GET/SCOP and that its apparent conclusion that relevant ‘activities’ had been so acquired was in a material respect irrational. He submitted that the application of the CAT’s test in [105] and [106] still required the CMA on the remitted inquiry to decide whether GET/SCOP had taken over SeaFrance’s activities rather than simply commenced like activities with the SeaFrance assets.

[169]Before turning to the CMA’s report, I should note the submission advanced for the CMA by Mr Harris QC that the stance of the SCOP on this appeal suggests that it ought to have appealed against the decision in Eurotunnel 1 but did not, with the consequence that the court ought now to regard the present appeal as an abuse of the process. I regard that argument as misplaced. The SCOP achieved material success in Eurotunnel 1, albeit not as much as it had hoped for. The criticism is that it did not ask the Court of Appeal to quash the Commission’s decision altogether. Since it does not ever appear to have been the SCOP’s wish to challenge the correctness of the guidance in Eurotunnel 1, my view is that it is probable that any appeal against the CAT’s decision would have been given short shrift. The court’s attitude would likely have been that the better course would be to allow the Commission/CMA to reconsider the case on the basis directed by the CAT and for the SCOP then to consider whether it wished to mount a further challenge to the further report when it was delivered. I would reject the abuse of process argument.

[170]I turn to the CMA report. I have summarised the essence of the course of events leading to the GET/SCOP acquisition of SeaFrance’s assets, but should flesh some of it out a little further. I quote first the CMA’s summary of its explanation of PSE3:

“2.34. The SCOP is not a party to the PSE3, had no role in the negotiation of the €25,000 special indemnity and the payment it received is not a SeaFrance asset. The PSE3 is a statutory arrangement, in this case negotiated between the liquidator and the SeaFrance comité d’entreprise (the works council, made up of employee representatives). PSE3 provides that Groupe SNCF … will fund certain payments in order to incentivise third parties to employ ex-SeaFrance employees; that is the capacity in which the SCOP received the indemnity payments. …”

Later the CMA also recorded as follows:

“3.31. GET’s offer document [of May 2012] explained that in addition to the financing of the three ships, the project was based on financing by the employees with the SCOP that was limited to approximately €10 million before corporation tax, and therefore involved additional financing of between €20 million and €30 million. We understand that the indemnity payments that the SeaFrance works council negotiated with the liquidator … and which the SCOP subsequently received … enabled it to be fund its €10 million contribution. 3.32 The French Court Minutes dated 11 June 2012 describe GET’s bid as aiming to revive the activities previously conducted by SeaFrance and as ‘the takeover of SeaFrance’s activities’ [the Minutes also summarised the GET project in the terms I have referred to earlier, namely as ‘providing for a partnership with SeaFrance’s former employees who shall form a SCOP in order to revive [faire renaître] the activities previously conducted by SeaFrance’]. Based on this understanding, and given that [GET’s] bid was the best bid to compensate the creditors, the Court ordered the sale of assets to GET. The Court noted that, while job creation was not a criterion established for the sole realisation of assets in liquidation, it remained a significant factor in the subjective assessment; a quick sale would provide for resuming the ships’ operating starting next season. The sale to GET of the vessels and assets was duly authorised. The acquisition completed on 2 July 2012. Operations on the Dover-Calais route recommenced on 20 August 2012 under the MyFerryLink brand. …”

. Later the CMA also recorded as follows:

[171]In [3.45] to [3.56] the CMA summarised its views on the history of the transaction and of the period from 15/16 November 2011 down to the acquisition during which SeaFrance was not trading – what it called ‘the period of inactivity’. The CMA noted that such period was considerably longer than in the Medicopharma case (that was perhaps something of an understatement: the difference was seven and a half months as compared with a few days); and said also that it was unnecessary, for merger consideration purposes, for the activities of the acquired business to continue until the date of completion of the transaction, since ‘were it otherwise, it would be very easy for businesses to evade UK merger control law’. There is of course no basis for any suggestion that the period of inactivity in this case was referable to any evasive intent: SeaFrance ceased its activities because its insolvency prevented their continuance. The CMA also noted that a considerable portion of such period was due to the requirements of the liquidator’s sale process. With respect, I regard that observation as carrying little weight. I cannot see that it would or could have made any difference to the issues in this case if, by some improbable turn of events, the GET/SCOP acquisition had been effected in exactly similar circumstances in, say, April 2012 rather than July 2012: the SCOP’s case would still have been that SeaFrance’s activities had finally ceased at the latest on 9 January 2012 and that there was no question of GET/SCOP doing, or being able to do, anything other than start up a like business afresh with the assets bought from the liquidator and using the ex-SeaFrance employees whose services had become available for hire in the market and whom the SCOP employed.

[172]The CMA next made the point that the liquidator took efforts to maintain the value of the assets during the period of activity in order to achieve the best price for them: and the CMA referred to his keeping the vessels in ‘hot lay-up’, for which exercise he kept on 190 SeaFrance employees, all other SeaFrance employees having been dismissed. It is, however, obvious that the liquidator would want to maintain the value of the assets pending their sale and the retention of the staff to that end adds nothing material to the primary considerations with which the CMA was faced. The ‘activities’ of SeaFrance were not suggested as being the maintenance of its vessels in hot lay-up for the purposes of enabling a more beneficial sale in its liquidation.

[173]The CMA turned then to the point that was central to its ultimate conclusion on the ‘activities’ issue, namely that all the various transactions leading up to the sale to GET/SCOP ‘had the aim of continuing SeaFrance’s activities in some form and providing employment to SeaFrance employees’ (my emphasis). The CMA referred in particular to PSE3, under which the highest indemnity was payable in the event that ‘the ex-SeaFrance staff were re-employed on the SeaFrance vessels used in a similar operation and which ultimately provided the SCOP with a substantial amount of working capital.’ As to that, it appears to me, with respect, that the CMA’s use of the word ‘continuing’ begged the central question it had to decide. The court’s decision of 9 January 2012 required the cessation of SeaFrance’s activities and the sale of its assets; the prompt subsequent dismissal of the bulk of its staff (save for the 190 retained to help with the liquidation) was a consequence that followed automatically under French law. True it is that everyone had in mind that, by reason of PSE3, the dismissed staff might, and ideally would, find new employment with a purchaser of the vessels in circumstances in which the vessels were again used for Dover/Calais ferry crossings. It is, however, obscure to me how any intention to achieve such an ultimate outcome can objectively be assessed as the interim ‘continuing’ of SeaFrance’s former ‘activities’, when in reality they had come finally to an end. The intention can at best have been to achieve a situation in which a ‘similar operation’ (see clause 3.3.3 of PSE3) would be established by the purchaser of the vessels, who would also employ many of the dismissed SeaFrance employees.

[174]The CMA decision continued:

“3.52 The liquidation process and subsequent termination of employment contracts meant that the TUPE Regulations did not apply and allowed the business and the workforce to be restructured. Continuity of employment was effectively safe-guarded by the formation of the SCOP, which held the workforce together, and, to a lesser extent, due to the fact that a significant number of employees were involved in the lay-up of the vessels. 3.53 At the point the decision was taken that the SeaFrance activities should cease, the French Court recognised that the aim of achieving some form of business continuity remained unchanged. This is clear from statements made by the French Court such as: “the end of the temporary continuance of business is not the end of the road” and “there must be a trade-off between the value of the assets, which are mainly the vessels, and the continuance of employment contracts”

[175]In citing that, I record Mr Beard’s point that, when making the two quotations that it did in [3.53], the CMA may not have appreciated that the quoted observations were not part of the judgment of the French court, they were merely a recitation of the submissions to it made by the bankruptcy judge. Second, the reference in [3.54] to ‘continuity and momentum’ is taken (in reverse order) from the last sentence of [120] in Eurotunnel 1 which I have quoted; and Mr Beard made the point that the critical question under the Enterprise Act 2002 does not fall to be answered by reference to considerations so expressed, but rather by reference to whether there has been a taking over of the activities of another business. If the activities have continued down to the takeover, there will be no difficulty; but, he said, it is not legitimate to assert that there has been continuity when plainly there has not and then to draw a conclusion that ‘activities’ have been taken over. There was, he said, no relevant continuity of employment in this case. The SeaFrance employees had all been dismissed save only for those retained to assist in the disposal of its assets.

[176]The position with regard to the ex-SeaFrance employees was obviously central to the CMA’s ultimate decision that SeaFrance’s activities were brought under GET/SCOP’s ownership or control; and in a long section of its report, starting at [3.57], the CMA devoted itself to answering the question ‘whether employees transferred to (or were “acquired by”) the SCOP from SeaFrance’. This was in response to GET/SCOP’s case that they did not ‘acquire’ such employees from SeaFrance: they hired them in the market following the liquidator’s dismissal of such employees on grounds of redundancy.

[177]I shall not summarise the CMA’s extensive discussion about this. Their conclusion is expressed as follows:

“3.107 In our view, the [PSE3] indemnity demonstrates that it is not the case that SeaFrance’s employee contracts of employment were terminated “with no thought as to how they might be employed in future” [a reference to a statement in [115] of the judgment in Eurotunnel 1]. The indemnity that SNCF – SeaFrance’s parent company at the time – agreed to pay created a strong incentive for ex-SeaFrance employees to be employed on the SeaFrance Berlioz, SeaFrance Rodin and SeaFrance Nord Pas-de-Calais in similar operations to those of SeaFrance. It creates a link between the vessels and the employees and it was aimed at ensuring, and ultimately did ensure, to the extent possible given the points that we highlighted in paragraph 3.77 above, that a significant number of employees transferred from SeaFrance to the operator of the vessels. We consider that this shows that a large proportion of the SeaFrance workforce effectively transferred from SeaFrance to the SCOP.”

[178]Mr Beard said that represented the crux of the CMA’s decision that SeaFrance’s ‘activities’ were brought under the ownership or control of GET/SCOP upon or following the latter’s acquisition of the SeaFrance assets. The CMA referred also to the acquisition as including other assets (brand, goodwill, trademarks, domain names), which it also discussed at length, but I agree with Mr Beard that it was not the acquisition of these assets that was the principal driver of the conclusion of the CMA that there was a taking over of SeaFrance’s activities. That conclusion was centrally dependent upon the CMA’s view that there had been an acquisition of the vessels and a transfer – or ‘effectively’ a transfer – of a large proportion of SeaFrance’s workforce to the SCOP which then made up the crews of vessels operated on the same Dover/Calais crossings. It might be thought fairly obvious that SeaFrance’s activities could only be continued or resumed so long as it still retained the employees to carry them out, whereas the liquidator had dismissed the employees in early 2012. If GET/SCOP can fairly be regarded as having acquired from SeaFrance’s liquidator both the vessels and the crew that had manned them, I can see a possible basis for a conclusion that SeaFrance’s ‘activities’ as short sea ferry operators were brought under their ownership or control. The central question is, in my view, whether the CMA’s conclusion that the SeaFrance employees had transferred or, as the CMA said, ‘effectively’ transferred from SeaFrance to the SCOP was a sustainable one.

[179]The CMA summarised its overall conclusions in [4.1] to [4.22]. In [4.2], it reminded itself of the CAT guidance in [105] of Eurotunnel 1, although it there referred only to the first two steps of the guidance so given: it did not at that point also remind itself that, were it to decide that the acquiring entity had acquired something over and above ‘bare assets’ and so found itself in a different position than if it had simply gone out into the market and acquired the assets, it still had to answer the critical further question as to whether such difference ‘is capable of constituting what would otherwise be bare assets into something that may properly be described as the activities of a business’. In [4.5], it repeated its ‘continuity and momentum’ point and its point that a considerable portion of the period of inactivity post-9 January 2012 was attributable to the liquidator’s sale process; and it added to this latter point that such process followed the SCOP’s two failed bids to purchase the SeaFrance business as a going concern. The relevance of that last observation is unclear to me: those bids were made and rejected before the court ordered the cessation of the SeaFrance business and the sale of its assets, an order leading directly to the liquidator’s prompt dismissal of the majority of the SeaFrance employees. The point appears to me rather to underline that, contrary to the CMA’s conclusion, SeaFrance was no longer a going concern when GET/SCOP acquired its assets. In [4.10], the CMA repeated that the SCOP was established for the purpose of providing employment for ex-SeaFrance staff. In [4.11], it repeated its view that the €25,000 indemnity:

“… forged a link between the vessels and the employees and it ensured that – to the greatest extent possible – ex-SeaFrance employees transferred from SeaFrance to GET/SCOP. … Our conclusion is that in effect these employees transferred from SeaFrance to GET/SCOP. As a result, around 70-80% of the SCOP workforce comprises ex-SeaFrance employees who were made redundant as a consequence of SeaFrance’s liquidation.”

As the employees were dismissed as redundant by SeaFrance’s liquidator before being engaged as employees by the SCOP, it appears to me that there is a fundamental question as to the rationality of the CMA’s description of these employees as having either ‘transferred’ from SeaFrance to the SCOP, or as having ‘in effect’ so transferred. Many, for example, may have obtained other employment before being engaged by the SCOP.

[180]In [4.14], the CMA said this:

“Overall, we are persuaded that the steps taken in relation to staff were similar in nature to the steps taken in relation to the vessels. They were designed to ensure that there would be continuity of SeaFrance’s activities to the maximum extent possible in the circumstances of the liquidation. In the result, those steps substantially achieved their aim.”

I have respectful difficulty with that statement. For all relevant practical purposes, the liquidator’s dismissal of the staff rendered irreversible the cessation of SeaFrance’s activities ordered on 9 January 2012; and the assertion that the steps taken in relation to the staff were designed to ensure ‘continuity’ in such non-existent activities appears to me to be illogical. The dismissals in fact achieved the reverse of the potential for any sort of continuity, of which since 16 November 2011 there had anyway been none. PSE3 obviously achieved a high likelihood of a buyer of SeaFrance’s vessels being able to re-start like activities by employing many of the ex-SeaFrance staff. That, however, was all it was capable of achieving.

[181]The CMA’s conclusions were rounded up and explained as follows in [4.19] to [4.22], although I shall also quote what it said in [4.24]:

“4.19 We therefore conclude that: • The combination of acquired assets (in particular, but not limited to, the vessels and employees) means that what was acquired was more than a “bare asset” in that it enabled the acquirer to establish ferry operations, more quickly, more easily, more cheaply and with less risk than if the relevant assets had been acquired otherwise in the market. • Although, in light of the period of inactivity, GET/SCOP did not acquire the SeaFrance assets “as a going concern”, in reality they obtained much of the benefit of so acquiring them. That is because, in our view, the commercial operability and coherence of the assets used by SeaFrance for the Dover-Calais ferry service was actively maintained, and thus impairment was minimised, during the period of inactivity. • The result of the combination of steps taken in relation to the vessels and the staff was that substantially the same business activities as had previously been undertaken by SeaFrance were able to be, and were in fact, resumed within a very short period of time following the acquisition. The intention, for good and understandable commercial and employment reasons, was to seek to preserve the former business or something as closely approximating to it as possible. That intention was achieved. • Moreover, GET was significantly motivated to acquire the assets that it did by the advantages of continuity (and the consequent ability to resume substantially the same operations as had previously been undertaken by SeaFrance on the Dover-Calais route) that those steps had preserved. 4.20 We conclude that the collection of tangible and intangible assets (including the transferred ex-SeaFrance employees) that GET/SCOP acquired meets the legal definition of an enterprise in that together they constitute the activities or part of the activities of a business. 4.21 We are satisfied that the acquired assets (including the transferred ex-SeaFrance employees) are under the control of the associated persons: GET and the SCOP. 4.22 In its judgment, the CAT remitted to the [Commission] the question of whether GET/SCOP had acquired an “asset” or an “enterprise” and to that extent, our decision was quashed. As a result, the only matter on which we are required to make a new decision is this specific jurisdictional point. We have decided that GET/SCOP have acquired an enterprise, and therefore that a relevant merger situation has arisen. In our view the effect of this is to reinstate the Report on all other matters. … 4.24 In our view, when considering what constitutes an enterprise for the purpose of deciding whether enterprises have ceased to be distinct and establishing that there is a relevant merger situation, it is important to have regard to the purpose of the legislation. The purpose of the legislation is to enable UK competition authorities to review transactions which might substantially lessen competition in a particular market. In this context, and having regard to the overall nature of the UK merger control regime under which notification of transactions is voluntary, we consider that it is appropriate that provisions enabling authorities to review transactions are interpreted widely and purposively. To do otherwise would invite gaming of the system and the structuring of transactions and arrangements in such a way as to avoid merger control.”

[182]I make three comments. First, the first bullet point of [4.19] shows that the principal driver for the CMA’s overall conclusion was its finding that GET/SCOP had acquired from SeaFrance not just the vessels and other assets they had bought but also the ex-SeaFrance employees who then made up the majority of the workforce that operated the acquired vessels. If, however, the CMA’s finding that the SeaFrance employees had transferred, or ‘effectively’ transferred, to the SCOP was unsustainable, that would appear to me to undermine the overall conclusions.

[183]Second, whilst Mr Beard submitted that the CMA made no finding as to the ‘activities’ of SeaFrance that were said to have come under the ownership or control of GET/SCOP, I consider that a fair interpretation of the passages quoted is that the CMA was there at least making a generalised finding that there had been a takeover of SeaFrance’s former activities, although the finding is perhaps a somewhat qualified one: I have in mind the third bullet point of [4.19], in which the ‘substantially’ and the ‘or something as closely approximating to it as possible’ appear to raise questions as to precisely what former SeaFranceactivities the CMA found GET/SCOP to have taken over.

[184]Third, as to the CMA’s points in [4.24], there is no question in this case of any of the transactions having been ‘structured’ so as to avoid merger control. Further, whilst no doubt the legislation must be interpreted purposively, there is no warrant for its being interpreted ‘widely’ if by that is meant that it can be read as bearing a meaning that goes beyond one derived from its fair, purposive interpretation. Why did the CMA say what it here did say unless it was tacitly acknowledging that it had interpreted the relevant legislation ‘widely’ so as to enable it to hold that SeaFrance’s ‘activities’ were brought under the ownership or control of GET/SCOP? What extra ‘width’ was the CMA giving to the word ‘activities’?

[185]I turn to the decision of the CAT. The thrust of the SCOP’s submission to the CAT was that there was no continuity of activities between those of SeaFrance and those of MyFerryLink, that is there was no acquisition of SeaFrance’s ‘activities’ and the CMA was irrationally wrong to hold otherwise: see [69] of the CAT decision. The CAT regarded the points made as powerful but said that they were all argued in Eurotunnel 1 where they did not succeed and so they could not succeed now: see [70]. That response to the SCOP submission appears to me to overlook that the like submission in Eurotunnel 1 went a good way to achieving total success, and might well have done so but for the fact that the CAT considered that it was necessary to throw a lifeline to the CMA by way of an opportunity to reconsider its apparentlydubious conclusion by having particular regard to the PSE3 arrangements in relation to the former SeaFrance employees, which is what the CMA did. I cannot see why, when that was done, it was not open to the SCOP to submit that the revised CMA decision failed on irrationality grounds, which is what the SCOP asserts.

[186]The approach of the CAT was that Eurotunnel 1 had set out the test for the CMA to apply, that it had faithfully applied it and that there was no legitimate basis upon which its report could be challenged on judicial review grounds. Perhaps the heart of the CAT’s decision is in [79], where the CAT said:

“It should also be recalled that: (a) the SCOP was set up with the aim of promoting the employment of a significant number of SeaFrance employees on a ferry service involving the SeaFrance vessels under new ownership; (b) the indemnity under PSE3 was payable in respect of former SeaFrance employees engaged through a SCOP, or equivalent, in the operation of the SeaFrance vessels; (c) the SCOP was associated with Eurotunnel in making the acquisition at issue; (d) it was inherent in the bid by Eurotunnel that the SCOP would provide funding of about €10 million to make the operation viable, and it was evidently envisaged that this would be derived through the indemnity payments under PSE3; (e) the “assets” acquired by Eurotunnel/SCOP included a significant number of ex-SeaFrance employees who were employed by the SCOP; and (f) the amount obtained by the SCOP through the indemnity payments was very close to the €10 million figure. Taking all this into account, we consider that it was clearly open to the CMA to find that the indemnity “creates a link between the vessels and the employees” and had both the purpose and result that a significant number of ex-SeaFrance employees were employed by the SCOP. On the particular circumstances of this case, we do not think it is irrational or fails properly to have regard to the facts for the CMA to have concluded that for those ex-SeaFrance employees who were hired by the SCOP there was in effect (although of course not as a matter of the legal relationship) a transfer of that part of the ex-SeaFrance workforce from SeaFrance to Eurotunnel/SCOP.”

[187]Mr Beard’s criticism of that paragraph is the same as he levelled at the CMA’s reasoning, namely that the finding that the ex-SeaFrance employees either transferred, or ‘in effect’ transferred, to GET/SCOP is not a sustainable or rational one and that the flaw in it undermines any conclusion that SeaFrance’s ‘activities’ were brought under the ownership or control of Eurotunnel/SCOP. I also draw attention to [86] of the CAT judgment, in which it criticised as wrong the CMA’s assertion in [4.24] of its report that it was entitled to interpret the relevant legislation ‘widely’.

[188]Mr Harris, for the CMA, submitted in response that the CMA had no choice but to re-consider the case in accordance with the direction given by Eurotunnel 1 and he emphasised that it had done just that. He referred us to paragraph 4.10 of the CMA’s own guidance (Mergers: Guidance on the CMA’s jurisdiction and procedure), which states that the fact that a target business may no longer be actively trading does not in itself prevent it from being an enterprise for the purposes of the Enterprise Act 2002; and that, whilst the relevant criteria may vary according to the particular circumstances, the CMA will consider ‘for example’(i) the period of time elapsed since the business was last trading;

(ii) the extent and cost of the actions required to reactivate the business as a trading entity;

(iii) the extent to which customers would regard the acquiring business as, in substance, ‘continuing from the acquired business’; and

(iv) whether despite the fact that the business is not trading, goodwill or other benefits beyond the physical assets and/or site themselves could be said to be attached to the business and part of the sale. The Guidance adds, however, that none of these factors individually is likely to be conclusive and that the CMA will assess all relevant circumstances.

[189]Thus, said Mr Harris, it does not follow from the mere fact that the target is not actively trading at the time of its acquisition that there cannot be a takeover of its activities; and he emphasised that the Guidance shows that the assessment of whether there has or has not been such a takeover is a multi-factorial exercise. He pointed out that amongst the assets that GET acquired from the liquidator were various intangible assets, including trademarks, brand names, logos, domain names, internet sites, customer lists and goodwill (although with no separate value attached to the goodwill). It is said that these assets were more than bare assets and enabled GET to resume the same trading activities as SeaFrance had carried out.

[190]The extent to which, in particular, the goodwill was of any value to GET is disputed. Mr Beard’s position was that it amounted to ill-will rather than goodwill; and he also pointed out that MyFerryLink had never used the SeaFrance name, logos or trademarks. It is not wholly obvious to me why the acquisition of intangible assets of this nature points to the taking over the ownership or control of the target entity’s ‘activities’ as compared with the acquisition of assets that will or may enable the setting up of a new, but similar business operation. Further, as the CAT explained in Eurotunnel 1, the original Commission report plainly did not assess the GET/SCOP acquisition of the intangible assets as decisive in the Commission’s favour: had it been otherwise the CAT would not have remitted the case for re-consideration. In addition, I have said that I regard the CMA’s summary of its conclusions in [4.19] as reflecting that the principal driver to its acquisition of ‘activities’ conclusion was GET/SCOP’s combined acquisition from the liquidator of the vessels and employees.

[191]Put the other way, I do not read the CMA report as saying that the acquisition of the intangible assets by themselves would have been sufficient to decide the ‘relevant merger situation’ question adversely to GET/SCOP. Mr Harris also accepted that GET’s acquisition of the intangible assets did not, by itself, justify the CMA’s conclusions: he said that, whilst they are important, they were merely factors that, with all other factors, the CMA had to consider in the round. That is why the points about goodwill etc did not (nor were advanced) as by themselves fatal to the SCOP’s case in Eurotunnel 1. So I do not consider it necessary to focus further on them. If the SCOP is right in its challenge to CMA’s conclusion on the ‘transfer of employees’ point, that is enough to undermine the CMA report.

[192]Mr Harris emphasised repeatedly that the only task before the CMA was to carry out the guidance given to it by Eurotunnel 1, which he said it had manifestly done, with the result that the scope for challenge was limited. He said it was perfectly rational for the CMA to find that there had been a migration of the workforce from SeaFrance to GET/SCOP and he relied on [116] of Eurotunnel 1, where the CAT said that if ‘the reality is that a workforce is being transferred, then the fact that wholly new relationships are forged as part of that process should not affect the position.’ It matters not, said Mr Harris, that in the meantime the employees had suffered a period of redundancy. It was, said Mr Harris, no part of the SCOP’s submission that the guidance provided to the CMA in Eurotunnel 1 was wrong. The CMA applied that guidance and came to the decision that it did. It matters not that it may have made mistakes of fact or of law en route to its ultimate conclusion. It is also irrelevant that this court might perhaps have been disposed to come to a different conclusion. The decision was one for the CMA, which was, he said, entitled to conclude, as a matter of commercial reality, that the effect of the GET/SCOP acquisition was to achieve the revival or regeneration of the SeaFrance business. We were also reminded, as the court usually is on such appeals, that the appeal is one against the decision of an expert tribunal whose decisions must be regarded as close to sacred.

[193]All this is very familiar stuff and I of course fully recognise that the CMA was the fact finding tribunal and that it was and is neither for the CAT nor for this court to substitute any different factual finding that it might, had it been the primary decision maker, perhaps have been disposed to make. The SCOP’s challenge to the CMA’s conclusion is, however, that in one centrally material respect it was irrational, that is it dependent upon a finding that was beyond the responses of a reasonable decision. That is a challenge of a nature that the SCOP was and is entitled to advance, and it is one that required both the CAT and this court to scrutinise the rationality of the CMA finding that it is challenged. Moreover, in my view, having regard to two particular considerations, that scrutiny calls to be carried out with special care.

[194]First, I have referred to [4.24] of the CMA report, where the CMA asserted that ‘it is appropriate that provisions enabling authorities to review transactions [which might substantially lessen competition in a particular market] are interpreted widely and purposively’ (my emphasis); and the opening words of [4.24] show that the CMA regarded that approach as applying to what constitutes an ‘enterprise’ (i.e. ‘the activities, or ‘part of the activities, of a business’). The inference I draw is that it did regard itself as applying a ‘wide’ interpretation to the sense of ‘enterprise’: if not, why else did it say that it considered it was entitled to do so? The assertion that it was so entitled was wrong and was rightly criticised by the CAT. It raises in my view a concern as to the unspoken width that the CMA may have attached to the concept of an ‘enterprise’. Second, the CMA’s decision was of the greatest importance: it went to whether the CMA had jurisdiction to wield its very considerable powers in a manner that would or might have the potential to affect hundreds of jobs. In this combination of circumstances, this court should scrutinise with particular care the SCOP’s rationality challenge.

[195]I have already said enough to make clear the particular difficulty I have with the CMA’s overall conclusion. That is, that it was materially dependent upon the finding that GET/SCOP not only acquired from SeaFrance’s liquidator the three vessels and other assets to which I have referred, but also took a transfer (or ‘in effect’ did so) of the bulk of the SeaFrance employees who had formerly carried on SeaFrance’s activities when it was trading. GET/SCOP then started to carry on ferry activities similar to those that SeaFrance had carried on, and did so using the combination of ships and men that had been essential to the carrying on of such activities. If the CMA was entitled to find that the GET/SCOP acquisition of the assets it purchased also carried with it the transfer from SeaFrance of the workforce that had carried on the like activities for SeaFrance, I can of course see a basis upon which the CMA could properly come to a conclusion that SeaFrance’s ‘activities’ had come under GET/SCOP’s ownership or control. But was the CMA entitled to make the finding that it did?

[196]Implicit in my acknowledgment just made is that I have no difficulty with the proposition that, for the ‘enterprise’, or ‘activities’, of a business to come under the control of an acquirer, it is not necessary for the business actually to be trading, or carrying on any ‘activities’, at the moment of acquisition, although of course every case will turn on its facts. I have already indicated my instinctive view in relation to a sale out of season of a seasonal business. As for the present case, if, say, between 16 November 2011 and 9 January 2012 (when SeaFrance was not trading) the liquidator had sold the SeaFrance assets to GET/SCOP, I can well see that such a sale would probably have carried with it the transfer to GET/SCOP of all the employees’ contracts under the French equivalent of TUPE, so enabling a decision that GET/SCOP acquired the ownership or control of SeaFrance’s ‘enterprise’.

[197]The difficulty, however, that I have with the CMA decision is whether there was any rational basis for the finding that, in the events that actually happened, the ex-SeaFrance workforce transferred, or ‘in effect’ transferred, or ‘effectively’ transferred from SeaFrance to GET/SCOP, the words in quotation marks being the (understandably) somewhat slippery language that the CMA felt compelled to use at various points in its report. The essence of the CMA’s finding is that, at least as a matter of substance or reality, although not of form, the workforce was to be regarded as having transferred from SeaFrance to GET/SCOP as part of the latter’s acquisition of the SeaFrance assets.

[198]I respectfully regard that finding as unsustainable. The effect of the court order of 9 January 2012 was that the ‘activities’ in which the workforce had formerly been engaged were finally to cease and that the employees must be dismissed within 15 days as redundant, as they were. At the point of dismissal, the employees’ connections with SeaFrance were finally severed. It is also true that, at such point, PSE3 was already in place. It was a statutory job-saving plan directed at assisting the dismissed employees to find re-employment elsewhere than with SeaFrance; and paragraph 3.3.3, its most generous provision, of course raised a high likelihood that they, or most of them, would obtain re-employment with any purchaser of the SeaFrance vessels which proposed to use them for ferry operations similar to SeaFrance’s.

[199]PSE3 was, however, in no manner directed at preserving any connection between the employees and SeaFrance, let alone its activities (which had ceased), nor did it do so. In the event, following the successful GET bid, many of the former employees were later re-engaged by the SCOP. There is no sustainable basis for any conclusion that such engagements by SCOP resulted from, or were referable to, or were explained by any ‘transfer’, or by what was said to be ‘in effect’ a transfer, by SeaFrance to GET/SCOP as part of the GET/SCOP acquisition. That is not what happened as a matter of law or according to any rational assessment of the facts or by reference to the supposed ‘reality’ of the situation. Nor would any objective observer of the scene at the time that PSE3 was adopted have considered that if, at some future stage, there were to be a mass re-employment of the ex-employees by a purchaser of the SeaFrance vessels, such re-employment could at that point be characterised as, in reality, a transfer of the employees to the purchaser by SeaFrance together with the purchased vessels. They would foresee such re-employment as being simply that which it was, namely a true re-employment of employees whose services were available for hire in the market, albeit a re-employment incentivised by the terms of PSE3.

[200]The CMA’s different finding that upon such mass re-employment there was in reality a transfer, or a transfer ‘in effect’ by SeaFrance, is one that I therefore regard as irrationally wrong. It is one that could not properly have been made. If one were to explain the facts to the ubiquitous reasonable man and ask him whether the employees either transferred, or ‘effectively’ transferred, from SeaFrance to the SCOP, or so transferred ‘in effect’ or as a matter of reality, I would expect him to respond testily with a robust negative. He would make the obvious point that they could not have so transferred because they had been dismissed from SeaFrance before GET was relevantly on the scene. He would say that the simple reason for their re-employment by the SCOP was referable to the combination of GET’s successful bid and the various incentives provided by PSE3 for the re-employment of SeaFrance’s ex-employees. He would be right. He might wonder why he was being asked such a peculiar question.

[201]In my view, it follows that the CMA’s decision was materially flawed. The CMA’s finding about the employees was, I consider, central to its decision that SeaFrance’s ‘activities’ came under GET/SCOP’s ownership or control. Take it away and a material part of the foundation for the decision collapses. The decision of the CAT under appeal was wrong in failing so to hold.

[202]I would allow the SCOP’s appeal.

Société Coopérative De Production Seafrance S.A. v Competition and Markets Authority & Anor [2015] EWCA Civ 487

Summary of the appeal and my conclusion

Summary of the background and this appeal

Summary of conclusion

Structure of this judgment

The third bid submitted by SA [public limited company under French law] Groupe Eurotunnel…

AAH Holdings/Medicopharma

No appeal from Eurotunnel 1

Remittal Report’s errors which could not be treated as insignificant

Legal consequence of errors

Application of test of inevitability

Failure to follow guidance

Interpreting transactions purposively and widely

Hiatus point

Indemnity point

Read in context, the errors made by CMA were not significant

Hiatus point

Indemnity Point

The three errors made by the CMA

Overall conclusion

MMC REPORT IN AAH HOLDINGS PLC /MEDICOPHARMA NV

Lord Justice Tomlinson:

Sir Colin Rimer:

Cited in 2 later judgments