BAILII [Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]

United Kingdom Competition Appeals Tribunal


You are here: BAILII >> Databases >> United Kingdom Competition Appeals Tribunal >> Akzo Nobel N.V. v Competition Commission [2013] CAT 13 (29 April 2013)
URL: http://www.bailii.org/uk/cases/CAT/2013/13.html
Cite as: [2013] CAT 13

[New search] [View without highlighting] [Printable RTF version] [Help]


    Case No.: 1204/4/8/13

    Neutral citation [2013] CAT 13

    IN THE COMPETITION APPEAL TRIBUNAL

    Victoria House

    Bloomsbury Place

    London WC1A 2EB

    21 June 2013

    Before:

    THE HON. MR JUSTICE NORRIS
    (Chairman)
    WILLIAM ALLAN
    PROF. GAVIN REID
    Sitting as a Tribunal in England and Wales
    _________________________
    BETWEEN:
    AKZO NOBEL N.V.

    Applicant

    - v -
    COMPETITION COMMISSION

    Respondent

    - supported by -
    METLAC HOLDING S.r.L.
    METLAC S.p.A.

    Interveners

    _________________________
    APPEARANCES
    Mr Tim Ward Q.C. and Mr Alistair Lindsay (instructed by Slaughter and May) appeared for the Applicant.
    Mr Daniel Beard Q.C. and Mr Rob Williams (instructed by the Treasury Solicitor) appeared for the Respondent.
    Mr Mario Siragusa and Mr Paul Gilbert (of Cleary Gottlieb Steen & Hamilton LLP) appeared for the Interveners.
    Note: Excisions in this judgment (marked "[…][(]") relate to commercially confidential information: Schedule 4, paragraph 1 to the Enterprise Act 2002.
    Heard at Victoria House on 18 and 19 April 2013
    _________________________
    JUDGMENT
     
    Contents
    I. INTRODUCTION 1
    A. The Transaction 4
    B. The metal packaging coatings industry 7
    II. THE COMMISSION'S INVESTIGATION AND THE REPORT 14
    A. The Commission's jurisdiction to investigate the Transaction 14
    B. The geographic and product markets 22
    C. The SLC finding 24
    D. Prohibition of the Transaction 28
    E. AkzoNobel's grounds of review 31
    III. PRINCIPLES APPLICABLE ON AN APPLICATION FOR JUDICIAL REVIEW 34
    IV. GROUND 1: CARRYING ON BUSINESS IN THE UNITED KINGDOM 42
    A. The Commission's Report 46
        (i) Corporate structure 48
        (ii) Internal governance 51
        (iii) Strategy and planning 55
        (iv) Contracting 57
        (v) The Commission's conclusions on the functioning of the AN Group 61
    B. The parties' submissions 63
    C. The Tribunal's conclusions 69
        (i) Scope of the Tribunal's inquiry 69
        (ii) The first issue: interpretation of section 86(1)(c) by reference to the specific purpose of the merger control regime 74
        (iii) The second issue: interpretation and application of section 86(1)(c) in accordance with general company law principles 90
    V. GROUND 2: METLAC COMPETES MORE AGGRESSIVELY ON PRICE 117
    A. Introduction 117
    B. The BKartA customer survey 124
        (i) The Commission's alleged reliance on a partial set of pricing data from the BKartA's survey 128
        (ii) The alleged divergence between the BKartA and the Commission's findings on price competition 133
        (iii) The overall approach of the BKartA and the Commission 137
    C. The Commission's customer survey 141
        (i) The alleged sampling bias 142
        (ii) The alleged use of leading questions 146
        (iii) The alleged conflation of evidence from FCG to B2E 152
    D. The Commission's pricing analysis 157
    E. The Commission's investigation of the reasons for any price differences 164
    F. Conclusion on Ground 2 167
    VI. GROUND 3: A LOSS OF COMPETITION IN INNOVATION 168
    VII. OVERALL CONCLUSION 181
    ANNEX: Glossary of Defined Terms Annex

    I. INTRODUCTION

  1. This is an application for review of a decision of the Competition Commission (the "Commission") of 21 December 2012, entitled "A report on the anticipated acquisition by Akzo Nobel N.V. of Metlac Holding S.r.l." (the "Report"). The application is brought by Akzo Nobel N.V. ("AkzoNobel") pursuant to section 120(1) of the Enterprise Act 2002 (the "Act"). By an order made on 30 January 2013, Metlac Holding S.r.L. ("Metlac Holding") and Metlac S.p.A. ("Metlac") were granted permission to intervene in support of the Commission.
  2. The Report related to the proposed acquisition (the "Transaction") by AkzoNobel, a company registered in the Netherlands, of the remaining share capital in Metlac Holding, a company incorporated in Italy and in which AkzoNobel already holds indirectly a 49% stake. Both companies, and in the case of AkzoNobel a large number of its subsidiaries, are active in the market(s) for metal packaging coatings. The Office of Fair Trading (the "OFT") referred the Transaction to the Commission, pursuant to section 33 of the Act, on 23 May 2012. In the Report, the Commission found that the Transaction would lead to a substantial lessening of competition (an "SLC") in the UK for one particular segment of the metal packaging coatings market. Pursuant to sections 84 and 86 of the Act, the Commission prohibited the Transaction.
  3. This judgment uses a number of terms and abbreviations, which are defined when first used. The annex to the judgment sets out a composite glossary of those defined terms.
  4. A. The Transaction

  5. As noted above, AkzoNobel presently holds, through its Dutch subsidiary Akzo Nobel Coatings International B.V. ("ANCI"), a 49% stake in Metlac Holding. The remaining 51% of the shares in Metlac Holding are owned by members of the Bocchio family. Through one of its UK subsidiaries, Mortar Investments International Ltd ("Mortar"), AkzoNobel also holds 44.44% of the shares in Metlac, with the remainder being held by Metlac Holding. The combination of these shareholdings gives AkzoNobel a 71.7% economic interest in the Metlac business but, as matters stand, the Commission found that it does not have sole control of either Metlac Holding or Metlac.
  6. In 2008, AkzoNobel acquired Imperial Chemical Industries ("ICI"). By virtue of its acquisition of ICI, AkzoNobel acquired its present shareholding in Metlac Holding, as well as Mortar, which holds the shares in Metlac. AkzoNobel also inherited a call option over the remaining 51% stake in Metlac Holding held by the Bocchio family. That option was held by ANCI and was exercised in December 2011. The terms of the call option expressly made its exercise subject to any necessary antitrust clearances.
  7. The Transaction has been considered by a number of competition authorities other than the OFT and Commission, some within the European Union and some without. The Commission is, we were told, the only one to raise concerns about the Transaction. It will be necessary to consider the decision of the Bundeskartellamt (the German Federal Cartel Office, the "BKartA"), which cleared the Transaction, in some detail in the context of AkzoNobel's second ground of review.
  8. B. The metal packaging coatings industry

  9. The Transaction relates to the metal packaging coatings (hereafter, "coatings") industry, which is concerned with the lacquer coatings applied to metal packaging including, for example, beverage cans, food tins, drums and metal tubes. Demand for coatings is driven by demand for the various forms of metal packaging to which they are applied. These tend to be divided into four main segments according to their end-use: beer and beverages ("B&B"); food coatings ("Food"); caps and closures coatings ("C&C"); and general line ("GL", together with Food and C&C, "FCG"). Those segments are then further sub-divided. The following breakdown of the various segments is taken from paragraph 12 of the Report:
  10. Metal packaging coatings End-use Sub-categories Further distinctions
    B&B B&B Beverage externals (B2E) Aluminium or steel
    Rim coatings
    Aluminium or steel
    Rim coatings
    Aluminium or steel
    Rim coatings
    Beverage internals (B2I) Aluminium or steel
    Aluminium or steel Aluminium or steel Beverage ends (BE) Internal or external





    FCG
    Food Food external
    Food internal
    Food ends
    Often divided depending on whether two- or three- piece and by manufacturing process. Some beverage cans are three-piece and they may be included in this category. Other types of coatings required for two-piece food cans include side stripe coatings.
    Often divided depending on whether two- or three- piece and by manufacturing process. Some beverage cans are three-piece and they may be included in this category. Other types of coatings required for two-piece food cans include side stripe coatings. C&C Twist-off caps
    Tamper-proof caps
    Other
    As with Food and B&B, each different type of cap requires both internal and external coatings
    As with Food and B&B, each different type of cap requires both internal and external coatings GL General line
    Aerosol
    Collapsible tubes
    Aluminium monobloc
    As with other categories, each different type of packaging requires both internal and external coatings and, in the case of tins (eg for paint) and three-piece tubes, ends
    Source: [Commission] Source: [Commission] Source: [Commission] Source: [Commission]

  11. It is important to note that the SLC identified by the Commission, and which it sought to remedy by prohibiting the Transaction, related solely to the B2E market segment. We consider the SLC finding in more detail below.
  12. The Commission noted in the Report that the production of coatings is "relatively concentrated at the global level" with only AkzoNobel and two other producers: PPG Industries, Inc. ("PPG") and The Valspar Corporation ("Valspar"). Metlac is not a global player but does have a strong presence in the European Economic Area (the "EEA"). There are also a number of smaller players on the EEA coatings market (see Report, Summary, paragraph 13). Whilst AkzoNobel, PPG and Valspar are active across the full range of market segments, Metlac is active in the FCG sector and B2E only, although it provided the Commission with evidence that it was planning to enter both B2I and BE (see Report, Summary, paragraph 70; see also paragraph 9.117).
  13. The Commission noted that the downstream metal packaging industry is also concentrated, with only five major purchasers, namely, The Ardagh Group ("Ardagh"), Ball Corporation ("Ball"), Can Pack S.A. ("Can-Pack", which is principally active in the EEA only), Crown Holdings Inc ("Crown") and Rexam plc ("Rexam"), of which the latter four accounted for almost all EEA-demand in the B2E segment (see Report, Summary, paragraphs 16 and 17).
  14. Another important aspect of the coatings industry is the rigour of the qualification process that a supplier must satisfy before a purchaser will buy and make commercial use of a new coating. This is required even for coatings that only differ slightly from one already used by a particular purchaser or which a purchaser has already used but at a different time or in a different production facility. The process can last anywhere from several months to several years, generally taking longest in the B&B and Food segments. As such, the process brings with it relatively high costs for switching between coatings manufacturers.
  15. The Commission observed that:
  16. "In an industry where suppliers have the manufacturing equipment to produce most coatings, their ability to switch quickly to compete on price in particular segments is constrained by factors including: technological know-how (to innovate and formulate coatings); reputation; technical support; regulation and qualification; and the appetite to compete." (Report, Summary, paragraph 30)
  17. Although there are certain other features of the industry to which it will be necessary to refer, it is more convenient to set those out as they arise.
  18. II. THE COMMISSION'S INVESTIGATION AND THE REPORT

    A. The Commission's jurisdiction to investigate the Transaction

  19. In considering how the Commission came to investigate the Transaction, it is first necessary to draw a distinction as to which authority exercises jurisdiction over a merger that has a potential effect in the United Kingdom.
  20. Pursuant to Article 21 of Council Regulation 139/2004/EC on the control of concentrations between undertakings[1] (the "Merger Regulation"), the European Commission has sole jurisdiction over concentrations with a Community (now, probably, better referred to as a 'Union') dimension. The definition of a concentration with a Community dimension is set out in Articles 1 and 3 of the Merger Regulation. It is a complex definition and it is, for present purposes, unnecessary to say more about it since it was common ground that the Transaction did not give rise to such a concentration. Notwithstanding that fact, there are two routes under the Merger Regulation[2] by which the European Commission might have assumed jurisdiction over the Transaction. In the event, neither arose but it will be necessary to say a little more about them in the context of AkzoNobel's first ground of review (see paragraph 88, below).
  21. Since the Transaction was not a concentration with a Community dimension, it is necessary to summarise how the Act operated to give the Commission jurisdiction to consider it.
  22. The United Kingdom, unlike many jurisdictions (including at EU level under the Merger Regulation), does not operate a system of mandatory notification of mergers. Where an anticipated merger comes to the attention of the OFT, however, section 33(1) of the Act places a duty on the OFT to refer that merger to the Commission if it believes that it is, or may be the case, that:
  23. (a) arrangements are in progress or in contemplation which, if carried into effect, will result in the creation of a relevant merger situation (an "RMS"); and
    (b) the creation of that situation may be expected to result in an SLC within any market or markets in the United Kingdom for goods or services.
  24. An RMS is defined in section 23 of the Act and it was common ground in this case that the Transaction gave rise to an RMS within the meaning of section 23(2) and (3).
  25. As will be clear from the above, the OFT is not obliged to reach a final view on the existence of either an RMS or an SLC. Where a reference is made to the Commission under section 33, however, the Commission must reach a final decision on those matters (section 36(1)). The Commission is obliged, if it considers that there is an RMS giving rise to an SLC, to decide whether, and if so what, action should be taken to remedy the anticompetitive effects it has identified (section 36(2)). In deciding those matters the Commission must have regard to the need to achieve as comprehensive a solution to the SLC as is reasonable and practicable (section 36(3)).
  26. The Commission is required to publish a report, containing its answers to those statutory questions, supported by reasons (section 38). That report must be published within 24 weeks of a merger being referred to it by the OFT, although that period can be extended by a further 8 weeks if the Commission considers there are special reasons for doing so (section 39). In the present case, the Commission did extend the period of its investigation to 32 weeks.
  27. The Commission's jurisdiction to consider the Transaction is not challenged by AkzoNobel. It was accepted that the Commission could properly investigate the Transaction and AkzoNobel played an active role in that process.
  28. B. The geographic and product markets

  29. After setting out (in considerably more detail than we have found necessary to do on this review) the salient features of the coatings market (see Report, paragraphs 2.1 to 2.75 and Appendix B), the Commission drew the following conclusions:
  30. (a) there are a number of differences between the B&B and FCG coatings markets, which led the Commission to define the B&B and FCG segments as separate product markets, which distinction reflected "industry practice—the set of suppliers and customers is more distinct between B&B and FCG than between the various segments within FCG and B&B" (Report, paragraph 7.24);
    (b) the geographic market it was called on to examine was EEA-wide. The Commission saw no evidence suggesting that competitive conditions differed markedly on a national basis within the EEA. It also found that it is generally unnecessary "to provide local support to UK plants from the UK. Therefore, [it] concluded that the relevant market is no narrower than EEA-wide" (Report, paragraph 7.31); and, therefore
    (c) the "relevant geographic market is EEA-wide but [it] focused [its] analysis on the possible impacts of the merger on customers with operations in the UK." (Report, paragraph 7.34)
  31. On that basis, the Commission analysed the likely effects of the Transaction, which would have seen a four-to-three change on the supply side, on the markets it had so defined.
  32. C. The SLC finding

  33. In its Provisional Findings, which it made available to the parties in October 2012, the Commission concluded that the Transaction would lead to an SLC on both the B&B and the FCG markets. Following the submissions it received on its Provisional Findings, the Commission carried out further pricing work and analysis, following which it decided that the SLC finding in relation to the FCG market should not be maintained (Report, paragraph 9.174).
  34. The SLC finding in relation to the B&B market was, however, maintained in the Report. The Commission found that:
  35. "... prices sought by suppliers for the B2E products that AkzoNobel and Metlac are currently qualified to supply (in the UK or somewhere in the EEA) are likely to increase post-merger. More specifically, we would expect to see an overall increase in prices sought by suppliers when B2E contracts contested by Metlac are rebid, as Metlac will have been removed as a potential low-price competitor for these contracts. We would also expect a weakening of rivalry in innovation, particularly when AkzoNobel and Metlac are head-to-head in the race to develop new formulations or minor changes to existing products (and this is also relevant to our views in relation to potential competition in B&B) ..." (Report, paragraph 9.57)
  36. The Commission summarised its overall conclusions in relation to the B&B market as follows:
  37. "In summary, we found that the proposed merger may be expected to create unilateral effects in the B&B market from a loss of actual competition. We also found that the proposed merger may be expected to create unilateral effects in the B&B market from a loss of potential competition. Metlac was in the process of becoming qualified to supply customers with additional products in B2E, and we considered it likely that Metlac will place a constraint on AkzoNobel on a larger number of product/customer circumstances in the future. We did not consider that Valspar, PPG and smaller suppliers would constrain the merged entity from raising prices or implementing non-price effects at least in the short to medium term. The merger would also remove a potential entrant from B2I and BE which reinforced our finding that the merger would result in unilateral effects in the B&B market. We found that new entry and expansion was unlikely to occur in a timely and sufficient manner to counteract the SLC in this market and that countervailing buyer power was unlikely to be sufficient to counteract the SLC in this market. We did not consider that efficiencies were likely to provide sufficient customer benefits to counteract any adverse merger impacts." (Report, paragraph 9.173)
  38. The Commission, therefore, found that the Transaction would give rise to an SLC in the "market for the supply of metal packaging coatings for B&B in the UK." (Report, paragraph 10.1)
  39. D. Prohibition of the Transaction

  40. The Commission recalled that section 36(2) of the Act placed upon it a duty to decide whether, and if so what, action should be taken to remedy an identified SLC. It also noted that section 36(3) required it, in particular, to have regard to the need to achieve as comprehensive a solution as is reasonable and practicable to the SLC and any adverse effects resulting from it (see Report, paragraphs 11.2 and 11.3).
  41. In its Remedies Notice, issued on 20 September 2012, the Commission invited views on prohibition[3] of the Transaction. It did not outline other possible remedies for the SLC as it had concerns as to their effectiveness but indicated that it would consider any alternatives proposed (see Report, paragraph 11.9). In Chapter 11 of the Report, the Commission gave detailed consideration to the type of remedy that ought to be imposed, including a formal remedy proposal for undertakings offered by AkzoNobel (see Report, paragraphs 11.19-11.42) and concluded that the only effective remedy would be prohibition (see Report, paragraphs 11.43, 11.44 and 11.72). That is not a conclusion that AkzoNobel has challenged on its application to the Tribunal.
  42. What AkzoNobel has taken issue with, however, is the Commission's conclusion that it had jurisdiction to prohibit the Transaction. It was common ground that that would only be the case if it was established that AkzoNobel fell within section 86(1)(c) of the Act as "a person carrying on business in the United Kingdom". The Commission concluded that, having regard to the functional and operational structure of the AN Group (which we describe in more detail below), AkzoNobel satisfied that condition.
  43. E. AkzoNobel's grounds of review

  44. Section 120(1) of the Act provides that any person aggrieved by a decision of the Commission under Part 3 of the Act (Mergers) (into which the Report falls) may apply to the Tribunal for a review of that decision. Pursuant to section 120(4), the Tribunal is required to decide such applications by applying the "same principles as would be applied by a court on an application for judicial review."
  45. In its Notice of Application,[4] AkzoNobel challenged the Report on three grounds, averring that the Commission erred:
  46. (a) in law in its interpretation of section 86(1)(c) of the Act and/or misdirected itself in the application of that section, in concluding that AkzoNobel carries on business in the UK and could, therefore, be the subject of a prohibition order. AkzoNobel submits that the Commission had no power to impose such a remedy;
    (b) in law in finding that Metlac competes more aggressively on price than other competitors (PPG and Valspar), which finding was the basis for the Commission's theory of harm and SLC finding. In so doing, the Commission took a decision that was not supported by the evidence, failed to carry out sufficient enquiries and failed to have regard to material considerations; and
    (c) in maintaining in the Report a finding that the Transaction would lead to a loss of competition in innovation when there was no evidence to support that conclusion. The analysis in the Commission's Provisional Findings, supporting an essential aspect of that conclusion, was dropped from the Report. The Commission, therefore, made findings in the Report that were not supported by the evidence and failed to carry out sufficient enquiries.
  47. In the event that one or more of the above grounds of review is upheld by the Tribunal, AkzoNobel seeks an order under section 120(5) of the Act quashing the Report and remitting the matter to the Commission with a direction to reconsider the matter and take a new decision in accordance with the ruling of the Tribunal. In the event its application is successful, AkzoNobel also seeks its costs.
  48. III. PRINCIPLES APPLICABLE ON AN APPLICATION FOR JUDICIAL REVIEW

  49. As noted above, section 120(4) of the Act requires the Tribunal to decide AkzoNobel's application by applying the principles that would be applied by a court on an application for judicial review. This formulation appears in (materially) the same form in a number of sections conferring jurisdiction on the Tribunal. In addition to section 120(4), it appears also in section 179(4) of the Act, as well as section 193(7) of the Communications Act 2003 and section 57(5) of the Postal Services Act 2011.
  50. The Tribunal is called on to apply the ordinary principles of judicial review that would be applied by the Administrative Court (see Office of Fair Trading v IBA Healthcare Limited [2004] EWCA Civ 142, per Carnwath LJ, as he then was, at [88]). The 'intensity of review' to be applied in judicial review cases is on a spectrum so that, at one end, a "low intensity" review is applied in cases depending essentially on political judgment or economic policy. At the other end of the spectrum, where fundamental rights arise, unreasonableness is not equated with absurdity or perversity, such that a lower threshold must be reached to show a decision to be unreasonable (see IBA Healthcare at [91]).
  51. The parties were in broad agreement as to the principles to be applied but there was some difference of emphasis between Mr Tim Ward Q.C. (appearing for AkzoNobel) and Mr Daniel Beard Q.C. (who appeared for the Commission). We set out below the principles which we propose to apply in reviewing the Report.
  52. Mr Ward relied particularly on the following passages of Carnwath LJ's judgment in IBA Healthcare to make good his submission that the Tribunal should, in this instance, apply a high intensity review to the Commission's Report:
  53. "93. The present case ... is not concerned with questions of policy or discretion, which are the normal subject-matter of the Wednesbury test. Under the present regime (unlike the 1973 Act) the issue for the OFT is one of factual judgment. Although the question is expressed as depending on the subjective belief of the OFT, there is no doubt that the court is entitled to enquire whether there was adequate material to support that conclusion ...
    ...
    100. ... [T]he essential question ... was whether the material relied on by the OFT could reasonably be regarded as dispelling the uncertainties highlighted by the issues letter. That question was wholly suitable for evaluation by a court. It involved no policy or political judgment, such as would be regarded as inappropriate for review by the Administrative Court."
  54. Mr Ward submitted that AkzoNobel's first ground was a question of statutory construction and that its second and third grounds were concerned simply with the quality of the evidence on which the Commission relied. As such, he argued that there were no issues of economic or policy judgment in relation to which the Commission could claim a wide margin of appreciation and, therefore, the ordinary judicial review principles would be flexible enough to allow for a higher intensity review. We accept that, as Mr Ward submitted, we are not here dealing with matters of policy. We are, however, of the view that the task carried out by the Commission entailed, and its Report contains, an element of economic prediction, which the Tribunal should show restraint not to 'second guess' (see R v Director General of Telecommunications, ex p. Cellcom Ltd [1999] ECC 314; [1999] COD 105, at [26]).
  55. Both Mr Ward and Mr Beard drew our attention to paragraph 20 of the Tribunal's judgment in BAA Limited v Competition Commission [2012] CAT 3, where the Tribunal set out the principles it considered applicable to an application under section 179(4) of the Act, which, as we have said, applies the same standard as section 120(4) with which we are concerned. For present purposes, we consider that the pertinent principles to take from BAA are as follows:
  56. (a) The Commission must take reasonable steps to acquaint itself with the relevant information to enable it to answer each statutory question posed for it and the "extent to which it is necessary to carry out investigations to achieve this objective will require evaluative assessments to be made by the [Commission], as to which it has a wide margin of appreciation as it does in relation to other assessments to be made by it". The Tribunal in BAA also accepted – as we do – that in judging the steps taken by the Commission to put itself in a position to answer the statutory questions, it is a rationality test that must be applied (see [20(3)]);
    (b) To decide whether the Commission had a sufficient basis for its conclusions, that rationality test must be applied in light of the totality of evidence available (see [20(4)]). Mr Ward accepted that "totality" is very important in this context;[5]
    (c) The intrusiveness of the remedy imposed by the Commission may necessitate an adjustment to the ordinary rationality standard but does not wholly transform the approach that the Tribunal should adopt (see [20(7)]); and
    (d) The Commission's reasons for its decision must be intelligible and adequate. It is not, however, "the function of the Tribunal to trawl through the long and detailed reports of the [Commission] with a fine-tooth comb to identify arguable errors. Such reports are to be read in a generous, not a restrictive way ..." (see [20(8)]).
  57. Mr Beard also drew our attention to paragraphs 78 to 80 of the Tribunal's judgment in Tesco plc v Competition Commission [2009] CAT 6, which note that Commission reports are to be read and construed as a whole, and not analysed as one might analyse a statute (see also to that effect, Barclays Bank plc v Competition Commission [2009] CAT 27 at paragraph 76). To focus only on certain parts of a decision, "fails to do justice to the overall appraisal and assessment made by the Commission" ([80]). Furthermore, Mr Beard drew our attention to the Tribunal's judgments in Somerfield plc v Competition Commission [2006] CAT 4, paragraphs 175 and 176, and British Sky Broadcasting Group plc v Competition Commission & Anr [2008] CAT 25, paragraph 119, which emphasise that what weight is to be accorded to different pieces of evidence is primarily a matter for the Commission.
  58. We consider that those are the principles which should be applied in considering AkzoNobel's challenge to the Report in the present case.
  59. IV. GROUND 1: CARRYING ON BUSINESS IN THE UNITED KINGDOM

  60. In Ground 1, AkzoNobel claims that the Commission erred in law and/or misdirected itself as to its power to prohibit the Transaction.
  61. AkzoNobel's challenge is based upon the interpretation and application of section 86(1) of the Act which provides:
  62. "An enforcement order may extend to a person's conduct outside the United Kingdom if (and only if) he is –
    (a) a United Kingdom national;
    (b) a body incorporated under the law of the United Kingdom or any part of the United Kingdom;
    (c) a person carrying on business [including (by virtue of section 129(3)) in partnership with one or more other persons] in the United Kingdom."
  63. It is common ground that the Commission's prohibition order extends to AkzoNobel's conduct outside the United Kingdom, because it prohibits AkzoNobel's acquisition of the shares that it does not already control in Metlac Holding, an Italian company. It is also common ground that AkzoNobel does not satisfy section 86(1)(a) or (b) of the Act, so that the legality of the prohibition depends upon whether section 86(1)(c) is satisfied.
  64. There is no dispute that, in section 86, the term "person" refers to a natural or legal person so that, in the present case, the requirements of section 86(1)(c) must be satisfied in respect of AkzoNobel itself, and not the AN Group. The issue in dispute is whether that company carries on business in the UK.
  65. A. The Commission's Report

  66. The Commission's analysis of that question is shortly stated in paragraphs 11.88 – 11.99 of its Report, which have to be read in conjunction with the Commission's factual description of AkzoNobel in paragraphs 3.1 – 3.11 of its Report. As recorded in paragraph 11.89 of its Report, the Commission states that it has "analysed the evidence provided by AkzoNobel NV that described the Group corporate structure, the governance arrangements as well as the operational arrangements" including, according to paragraph 11.92, information relating to AkzoNobel's contractual arrangements with customers and suppliers entered into by UK registered companies, which AkzoNobel said were carrying on business in the United Kingdom.
  67. We set out the relevant primary facts identified by the Commission in paragraphs 48 to 60, dealing in turn with AkzoNobel and the AN Group's corporate structures, internal governance, strategy and planning, and contracting. We then set out, in paragraphs 61 and 62, the conclusions that the Commission drew on the basis of the primary facts. In both cases, paragraph references are to the Commission's Report.
  68. (i) Corporate structure
  69. AkzoNobel is a pure holding company and, as such, is the ultimate parent company for the AN Group (paragraph 3.7). It is publicly listed on the NYSE Euronext Amsterdam stock exchange (paragraph 3.1). The AN Group consists of a number of wholly owned subsidiaries incorporated in different countries (numbering, according to Mr Ward, approximately 450 in total) (see Report, paragraph 11.90).
  70. The AN Group's business activities are organised by Business Areas ("BAs"), Business Units ("BUs") and Sub Units ("SBUs") (paragraph 11.90). SBUs and BUs are organised by market or geography, depending on the specific activities and customers served (paragraph 11.91). As explained earlier in the Report (at paragraph 3.3), the business activities affected by the Transaction are located in the SBU named Akzo Nobel Packaging Coatings ("ANPG") , which forms part of the Industrial Coatings BU; that, in turn, is one of six BUs in the Performance Coatings BA which is one of three BAs in the AN Group. Neither the ANPG SBU nor the Industrial Coatings BU have separate legal personality (paragraph 11.90).
  71. The subsidiaries that comprise the AN Group sit within the BUs (paragraph 11.90). The Commission specifically considered the corporate arrangements of seven companies (selected from a list of companies identified by AkzoNobel as carrying on business in the UK) whose activities appear to include the supply of coatings and related products (paragraph 11.91, footnote 222). Several of these companies had another wholly-owned subsidiary as company secretary and a second wholly-owned subsidiary as one of their directors: the two directors of both those companies were, respectively, the head of UK Tax for AkzoNobel and UK Corporate Controller and Country Co-ordinator at AkzoNobel; the first subsidiary served as secretary to the second subsidiary.
  72. (ii) Internal governance
  73. AkzoNobel operates a two-tier board structure, as required by Dutch law, with a Board of Management that reports to an independent Supervisory Board (paragraph 3.8).
  74. Day-to-day management of AkzoNobel, including strategic direction, is the responsibility of the Executive Committee ("ExCo") which comprises the four members of the Board of Management together with four executives who have responsibilities for BAs ("BA Responsibles"), functions and specific countries or regions (paragraphs 3.8, 3.9 and 11.95). ExCo members report to the CEO (paragraph 11.95).
  75. ExCo has issued various policy documents to give general steering and direction to each of the BAs, BUs and SBUs, including the so-called AkzoNobel Authority Schedule, which sets out the level within the functional hierarchy at which approval for specified actions is required (paragraph 3.10).
  76. The AkzoNobel Authority Schedule extends to such matters as strategy, operational plan, investments and disposals, organisation, restructuring, finance, control and accounting, human resources, insurance, risk management and pensions (paragraph 11.96). In several of these areas, financial thresholds are set "as low as […][(]" (paragraph 11.96). The Report also sets out specific examples of the approval requirements (see paragraphs 3.9 and 3.10, and footnote 44). In argument, Mr Ward acknowledged that some of the thresholds for approval were quite low but described ExCo's activities as purely reactive, in the sense that it reacted to things placed before it rather than actively directing the activities of the AN Group companies.[6]
  77. (iii) Strategy and planning
  78. In accordance with the AkzoNobel Authority Schedule, ExCo decides (or, at least, approves) strategic plans for BUs and Countries and also (following consultation with the relevant BU and BA) sets targets for each BA which includes targets for BUs (paragraph 3.9, footnote 44, paragraph 3.10 and paragraph 11.95). ExCo also decides on starting up new business in a country where the AN Group is not yet present (paragraph 11.95). Proposals for the operational plan and target are developed by BUs but submitted to the BA Responsibles for review and advice (paragraph 3.10).
  79. There is no strategic plan for individual subsidiaries (paragraph 3.9, footnote 44).
  80. (iv) Contracting
  81. As neither the ANPG SBU nor the Industrial Coatings BU have separate legal personality, it appears that contracts are concluded by an appropriate subsidiary (paragraph 11.90).
  82. There was no customer contract concluded by a UK registered company carrying on business in the United Kingdom to which AkzoNobel was also a party (paragraph 11.92(a)). It is to be noted that, at paragraph 21 of its Notice of Application, AkzoNobel stated that it has "subsidiaries which carry on business in the UK (although not generally in respect of the supply of metal packaging coatings) ...". The parenthetical caveat concerns one draft contract relating to B2I.
  83. Similarly, AkzoNobel was not party to supplier contracts concluded by such UK companies. Agreements with key suppliers, and most other suppliers, are completed under a Framework Agreement entered into by AkzoNobel Sourcing BV (a wholly owned subsidiary of AkzoNobel, acting on behalf of all AkzoNobel Affiliates, but not AkzoNobel itself, and receiving commissions from each supplier) (paragraph 11.92(b) and (c)). Agreements with other suppliers are arranged by the relevant BU (paragraph 11.92(c)).
  84. Contracts entered into by a number of AkzoNobel subsidiaries across Europe were signed on behalf of the subsidiary by the same director (paragraph 11.90).
  85. (v) The Commission's conclusions on the functioning of the AN Group
  86. On the basis of these primary facts, the Commission reached the following conclusions:
  87. (a) Neither the identity of the contracting AN Group entity, nor the formal corporate structure of the AN Group, reflected how, in substance, strategic and operational decisions were made within the AN Group (paragraph 11.90).
    (b) Referring in particular to the absence of a strategic plan for individual subsidiaries, the Commission recognised that there was a distinction between the corporate structure of the AN Group and its operational structure, and concluded that those arrangements reflected a structure in which decision-making was centralised within the Group (paragraph 11.91).
    (c) Referring to the purchasing arrangements noted in paragraph 60 above, the Commission observed that significant aspects of such arrangements were centralised (see Report, paragraph 11.93).
    (d) The arrangements revealed by a review of the Board minutes for one year of companies carrying on business in the United Kingdom were not determinative of whether AkzoNobel was carrying on business in the United Kingdom (paragraph 11.94).
    (e) The internal governance arrangements set out in the AkzoNobel Authority Schedule (summarised in paragraphs 54 and 55 above) showed that, far from being peripheral as AkzoNobel had claimed, the participation of AkzoNobel through ExCo was extensive and included the approval of operational decisions (paragraph 11.97).
  88. The Commission's overall conclusion as to these arrangements (which it characterised at several points in the Report as common among large corporate groups) is stated in paragraph 11.98 of its Report:
  89. "The arrangements described by AkzoNobel in its submission to us [the Commission] and in the Authority Schedule are complex. The Group carries out operations in the UK and business operations are part of a SBU, BU and BA. We have observed that AkzoNobel NV has structures in place such that the operations of the Group's various business activities are ultimately controlled by it. While appreciating that there are several steps of upward referral before the functional member of ExCo or AkzoNobel NV takes a decision, the structure in place, in our view, is one in which the operations within the Group are centrally monitored and directed which limits autonomy within the BUs and SBUs in practice. In our view, the organizational structure and arrangements we have described above, including the relevant business units, is the means through which AkzoNobel NV carries on business, including in the UK."

    B. The parties' submissions

  90. The essence of the dispute between the parties may be encapsulated in two issues. First, is section 86(1)(c) of the Act to be interpreted in its own terms, independently of the substantive purpose of the merger control provisions, even if that fetters the Commission's order-making powers (as AkzoNobel contends) or in a teleological fashion so that its construction is consonant with achieving the substantive purpose of the merger control provisions (as the Commission and Metlac contend)? Secondly, is the organisation and operation of the AN Group (as set out by the Commission in the Report) such that business activities carried on in the United Kingdom are those of the relevant operational subsidiaries but not AkzoNobel itself (as AkzoNobel contends) or such that there are business activities carried on in the UK by AkzoNobel itself, notwithstanding that it has chosen, formally, to establish a complex group of companies (as the Commission and Metlac contend)?
  91. AkzoNobel's Notice of Application stated its case in four submissions, namely that:
  92. (a) The language of section 86(1)(c) refers to "a person" (it is common ground that that includes both legal and natural persons) carrying on business in the UK, rather than a group of companies. It is not, therefore, satisfied by mere attribution of the activities of the AN Group to AkzoNobel. On the contrary, it requires the business activity in question to be that of AkzoNobel itself. In that respect, section 86(1) is to be contrasted with other parts of the Act, which refer to concepts such as "enterprises", "interconnected bodies corporate" and "associated persons", and which do envisage attribution of the kind that (AkzoNobel says) the Commission has made.
    (b) That interpretation of section 86(1) is consistent with prior competition legislation, notably the Fair Trading Act 1973 ("FTA") and the Restrictive Trade Practices Acts 1956 and 1976 (collectively, "RTPA" and, individually, "RTPA 1956" and "RTPA 1976").
    (c) That interpretation is also consistent with the treatment under the general law of a concept that is used in many other contexts, namely that the activities of a subsidiary may be attributed to its parent company in only extremely limited circumstances. AkzoNobel referred in particular to the leading case on the enforcement of foreign judgments, Adams v Cape Industries plc [1990] Ch. 433 (CA), which we discuss in detail below.
    (d) That interpretation reflects the resistance to an effects-based jurisdiction that the UK has consistently asserted.
  93. In its Defence, the Commission submitted that AkzoNobel had misinterpreted the basis for the conclusions reached. The assessment in the Report had been based not on the attribution to AkzoNobel of the activities of its operating subsidiaries but rather on a consideration of AkzoNobel's own activities. It accepted that the FTA was relevant prior legislation but said that, in the absence of any relevant case applying that Act, it did not advance the issue. It rejected the relevance of the RTPA and the broader case law such as Adams. Rather, the Commission maintained that section 86 must be interpreted in a way that is consistent with achieving the substantive purpose of the merger provisions in the Act, namely to enable the Commission to control mergers that are found to give rise to an SLC in a UK market: acceptance of the restricted interpretation for which AkzoNobel contended would permit merging parties to structure their transactions in a way that artificially avoided fully effective control by the UK authorities.
  94. It was common ground that the Commission had jurisdiction to consider the Transaction and determine whether it constituted an RMS, and, if so, whether it gave rise to an SLC. In argument, Mr Ward emphasised, however, that section 86 is clearly intended to constitute an additional threshold or check on the Commission's order-making powers, to fetter its jurisdiction to make orders with extra-territorial effect. If Parliament had intended to confer upon the Commission the broad power to make such orders wherever it finds an SLC on a UK market, Parliament would have used appropriately broad language in section 86(1) (employing concepts such as those noted in paragraph 64(a) above), or indeed would have omitted section 86 altogether. In any event, AkzoNobel submitted that it would be wrong to proceed on the basis that a restricted interpretation would leave the UK authorities powerless to deal with cases such as the Transaction. On the contrary, the UK authorities have at their disposal a range of investigatory and remedial powers that could be employed in such cases and could have been employed in relation to the Transaction, without the need to resort to an enforcement order within the meaning of section 86. As it is, the Commission has adopted an approach to section 86(1)(c) that wrongly elides a case of normal parent company control over its subsidiaries with the concept of carrying on business in the UK, in a way that conflicts with the general legal standards articulated in Adams. If taken to its logical conclusion, that construction of section 86(1)(c) would mean that AkzoNobel itself, as opposed to the AN Group, carried on business on a world-wide basis. Indeed, Mr Ward emphasised that, since the Commission had found the arrangements in the AN Group to be typical of large multi-national companies, that logical conclusion would extend to all ultimate holding companies.
  95. Mr Beard's primary case for the Commission was that whether a person was carrying on business in the UK was a matter for factual assessment and that that assessment had to be made bearing in mind the purpose for which the Act was enacted. He maintained that Adams was simply immaterial to the interpretation and application of section 86 since it was concerned with a different question, namely whether a parent company, Cape, was present in another jurisdiction by or through its subsidiaries. That was not, he said, the Commission's case here, although he added that, if Adams were applicable, the Commission's findings might be sufficient to satisfy the Adam criteria.[7] Mr Beard stressed that, whilst section 86(1)(c) requires a personal connection to the UK, it is sufficient that the relevant person is "involved in commercial activity in the United Kingdom".[8] In the present case, that requirement was said to be amply satisfied by AkzoNobel's extensive and active involvement in the operation and direction of the relevant business activities in the United Kingdom, as detailed in the Report (paragraphs 11.88-11.99).
  96. The fact that that finding may entail the conclusion that AkzoNobel is to be treated as carrying on business world-wide is a matter as to which the Commission is "entirely agnostic".[9] Its only concern is the effective enforcement of the Act in the UK. As to that, Mr Beard stressed that the alternative measures identified by Mr Ward, such as the offering by AkzoNobel of undertakings under section 82 of the Act, would not permit the Commission to deal comprehensively with the anti-competitive effects of the Transaction in the way required by the Act.
  97. C. The Tribunal's conclusions

    (i) Scope of the Tribunal's inquiry
  98. It is necessary first to establish the scope of the inquiry that the Tribunal must make in relation to Ground 1. Mr Ward urged us to look at the primary facts found by the Commission with respect to AkzoNobel's activities rather than at the "few generalised conclusions it reached on [the basis of] the facts that it found" (Transcript, Day 2, p. 54, lines 13-15). He urged on us that this did not amount to a challenge to the Commission's factual findings but rather to the conclusions that it drew from its findings of fact.
  99. We do not accept that distinction in the present case. AkzoNobel's challenge alleges an error of law on the part of the Commission. As such, the challenge has to be assessed on the basis of both the primary facts found by the Commission and its analysis of, and conclusions on, those facts. If AkzoNobel wished to challenge the substance of the Commission's conclusions (for example, by alleging that the primary facts found did not support the conclusions drawn as to the degree of centralisation and participation by AkzoNobel in decision-making), it could, and should, have framed an explicit application for review on Wednesbury grounds.
  100. It is clear that the decision whether a person carries on business in the UK (or, for that matter, any other jurisdiction) requires the exercise of factual and legal judgment based upon the analysis of primary facts that are commonly detailed and complex. It is furthermore clear that that judgment may involve, as it does in the present case, decisions as to questions of degree and as to the weight to be attached to particular elements in the analysis.
  101. At one point, however, Mr Beard appeared to suggest that whether a person was "carrying on business in the United Kingdom" within the meaning of section 86(1)(c) does not raise a legal question but is rather a matter for assessment by the Commission in respect of which it enjoys a margin of appreciation.[10] If, by that, he intended to suggest that the Commission could not make an error of law in applying section 86(1)(c), such that its assessment is exclusively reviewable on Wednesbury grounds, we disagree. In our judgment, there are two aspects to the assessment under section 86(1)(c):
  102. (a) The first is the factual assessment made by the Commission of what arrangements are actually in place. In accordance with the judicial review principles we set out in Section III. above, that is primarily a matter for the Commission, in relation to which it has a margin of appreciation in accordance with well-established Wednesbury principles. As we have already indicated in paragraph 71, that assessment includes both the finding of primary facts and the conclusions to be drawn from those facts.
    (b) The second aspect, however, is a question of law and that is whether the findings made by the Commission are sufficient to ground a finding that the relevant person carries on business in the UK. In this respect, we see no distinction between this case and the well-known principle that, whilst the criteria to be applied by a decision-maker may be matters of law, the weight to be attached to them is principally a matter for the decision-maker.
  103. Having considered the submissions of both parties on this issue, it is clear that the question for the Tribunal in this case is whether the conclusions reached by the Commission are sufficient to ground a finding that AkzoNobel, as a matter of law, was carrying on business in the UK for the purposes of section 86.
  104. (ii) The first issue: interpretation of section 86(1)(c) by reference to the specific purpose of the merger control regime
  105. The jurisdictional conditions set out in section 86 of the Act have been present, in substantially identical terms, from the inception of competition law in the United Kingdom in 1948,[11] notwithstanding the significant amendments to the substantive provisions of the legislation (including, notably, the introduction in the Act of a competition-specific SLC test for mergers in place of the general public interest standard that had applied under the prior legislation). Nevertheless, the present case appears, somewhat surprisingly, to be the first occasion on which the interpretation of the jurisdictional conditions has been put in issue.
  106. Jurisdictional considerations did arise under the separate legislative regime for the control of cartels and other restrictive agreements that was carved out of the Monopolies and Restrictive Practices (Inquiry and Control) Act 1948 by the RTPA 1956 (subsequently extended under powers given in the FTA and consolidated in the RTPA 1976). That system has since been replaced by the Competition Act 1998 which introduced a system designed substantially to replicate the corresponding provisions of EU law (Articles 101 and 102 of the Treaty on the Functioning of the European Union). Although AkzoNobel sought in its Notice of Application to rely upon the RTPA, and in particular the decision of Stamp J in Registrar of Restrictive Trading Agreements v Schweppes (No.2) [1971] 1 WLR 1148, Mr Ward did not press the point at the hearing and was, in our view, correct not to do so. It is sufficient to say that the differences between section 86(1) of the Act and the jurisdictional standards applicable under the RTPA mean that the latter do not materially assist the interpretation of section 86(1) in the present case.
  107. In these circumstances, section 86(1)(c) falls to be interpreted in accordance with normal principles of construction. The term "carrying on business in the United Kingdom" is essentially undefined by the Act. There is no dispute that section 86(1)(c) is concerned, in broad terms, with commercial activities. The (somewhat circular) definition of "business" in section 129(1) of the Act does not, however, assist in the resolution of the central questions that arise here: what criteria suffice for these purposes to connect any activities of a diverse corporate group to a particular legal person or to a particular jurisdiction?
  108. As we have noted above, there was a substantial debate, both in the pleadings and at the hearing, as to whether and to what extent section 86(1) should be interpreted so as to act as a fetter upon the Commission's order-making powers, having regard to the protective purposes of the merger control powers. The use of the term 'fetter' in this context carries with it a prejudicial connotation that is not, in our view, helpful. The plain fact is that the Commission's exercise of its order-making powers is dependent on the satisfaction of both substantive conditions (relating to the characteristics and consequences of the transaction) and personal conditions (relating to the characteristics of the addressee of the decision). That being so, in some cases, section 86(1) necessarily precludes the making of an order that the Commission would, but for the presence of that section, be empowered and wish to make: if Parliament had intended otherwise (in order to secure that transactions that satisfy the substantive conditions are always subject to fully-effective control), it could simply have excluded section 86(1) from the Act. In that connection, we recall that (as noted in paragraph 74 above) Parliament has enacted legislation dealing with merger control on three occasions and, despite significant modifications to the substantive conditions, has on each occasion retained the jurisdictional conditions now to be found in section 86(1) of the Act in substantially unaltered form.
  109. The essential question, therefore, goes to the precise content of section 86(1)(c). The Commission's case in that respect was expressed in various forms as the proceedings developed. So far as this particular case is concerned, the Commission has consistently relied upon its finding in the Report that AkzoNobel is active in the operation and direction of the business in the United Kingdom. That, however, does not represent the outer limit of its view of what amounts to "carrying on business". Mr Beard submitted at the hearing that it is sufficient that a person is involved in a commercial activity in the United Kingdom. That submission reflects the proposition advanced in the Commission's Skeleton Argument that, where a parent arranges its group's affairs in functional business units, over which it exercises extensive control, and a number of different legal persons are involved in the operation of a business that involves selling goods in the UK, all of those persons are involved in carrying on business in the UK. We also note that, in its Defence, the Commission reserved its position as to whether the controller of an enterprise must, for the purposes of the Act, be considered to be carrying on that enterprise and the business constituted by it.
  110. Mr Beard essentially sought to justify a broad 'commercial involvement' standard as one that is sufficient to give meaning to section 86(1)(c) whilst ensuring, so far as possible, that the substantive objectives of the merger control provisions in the Act are attained. In view of the narrower 'active operation and direction' basis upon which he based his case in these particular proceedings, we are not required to reach a concluded view on the broader standard. Nonetheless, in view of the possible importance of the issue and in deference to the arguments of Counsel, we think it appropriate to express our general views on the proposition.
  111. The implications of a 'commercial involvement' standard would be far-reaching. In principle, it would capture any company that forms part of the supply chain for goods or services that are ultimately supplied in the United Kingdom, even though the activities of that company (including any relevant supply transaction) take place wholly outside the United Kingdom. On its face, that extended interpretation would conflate instances of trading in the United Kingdom and instances of trading with the United Kingdom. Furthermore, jurisdiction would not be dependent on a connection between that activity and the substance of the transaction at hand: there is no linkage in that sense between the substantive and personal conditions. Similarly, jurisdiction would not be limited to companies within the corporate groups that are party to the transaction at hand: in principle, as Mr Beard pointed out at the hearing,[12] the Commission's order-making powers extend to third parties. We do not exclude the possibility that the scope of the personal conditions could be extended, for example to include all interconnected bodies corporate, or all associated persons (as defined in section 127 of the Act), or all members of an undertaking (as that term is understood in the Competition Act 1998). That, however, is a matter for Parliament. We do not consider that Parliament can be taken to have intended that section 86(1)(c) should be given such an extended meaning (however rarely the order-making powers may in fact be exercised in such circumstances) without clear language to that effect.
  112. We turn now to the central debate about the interpretation of section 86(1)(c) in the context of the 'active operation and direction' basis upon which the Commission relies in relation to the order directed to AkzoNobel in this case. Mr Beard's invocation of the broader economic purposes of the Act is unashamedly an attempt to escape any strictures that may arise from the treatment of the carrying on business concept under the general principles of company law. Mr Ward resisted that attempt on two grounds: first, it is simply wrong because Parliament has expressed itself unambiguously; and, secondly, there are in any event alternative remedies and procedures available that would at least ameliorate any enforcement gap that arises.
  113. In our judgment, the appeal to the economic purposes of the Act and the apparent irony in that context of allowing technical legal concepts to limit the achievement of those purposes is, in the present context, misconceived. It is, of course, true that the subject-matter of the Act comprises the assessment and regulation of economic issues but that subject-matter is realised through a legally constituted framework of procedure and enforcement. That framework expressly incorporates the concept of a legal person and necessarily brings with it the general principles of company law that bear upon the interpretation and application of that concept. It would no doubt be open to Parliament to dispense with or modify those general principles in light of the particular subject-matter of the Act but it has not done so. That being so, there can be no special dispensation from those general principles, in the absence of any statutory provision to the contrary, simply because the substance of the issues under consideration is economic. Whether those general principles in fact apply in the present case (as Mr Ward maintains) to invalidate the Commission's decision is a matter for discussion in the context of the second issue. At this juncture, we simply conclude that Mr Ward's argument cannot be rejected out of hand by an appeal to the particular purposes of this legislation.
  114. That conclusion is supported by the fact that provisions substantially the same as section 86(1) appear in a range of different legislative contexts (including the Act itself, in which section 86(1) is applied to the Commission's powers in Market Investigation References by virtue of section 164(2)(a) of the Act). Mr Ward did not, however, seek to rely upon that fact to resist Mr Beard's argument, nor were we referred to any case in which an equivalent provision has been considered. In those circumstances, we do not attach decisive weight to the existence of these provisions. That said, we do observe that it would be strange to find that a company is to be regarded as "carrying on business in the United Kingdom" for one purpose but not for another, a possibility that would necessarily be created by placing the degree of emphasis upon the special nature of the purposes of the Act for which Mr Beard contended (and which would presumably be required in every other statutory context where similar provisions appear).
  115. Although not strictly necessary in light of our conclusion on Mr Ward's first point, we nonetheless express our views on the relevance of the various alternative powers that are available to the UK authorities. Mr Ward did not argue, as we understood him, that these alternatives are a complete answer to the Commission's concerns about the enforcement gap that would be created in those circumstances. Mr Ward's points were rather that (i) the presence or absence of an enforcement gap was not strictly relevant to the construction of section 86 and (ii) in any event, the gap was not nearly as large or as problematic as the Commission appeared to suggest. Mr Beard's response was that, whilst these alternatives may well exist, they do not remove the enforcement concerns and those concerns militated strongly in favour of the construction of the Act for which the Commission argued and which it should in any event be given on normal principles.
  116. In this respect, whilst we have rejected Mr Beard's principal argument in favour of a broad and purposive interpretation of section 86(1)(c), we do not base that conclusion on the availability of the alternative powers to which Mr Ward referred. Parliament has deliberately conferred upon the Commission a wide range of remedial powers from which it is required, in relation to any RMS, to select such powers as are appropriate to comply with its statutory duty under sections 35(4) and 41(4) of the Act "to achieve as comprehensive a solution as is reasonable and practicable" to the detriments presented by that RMS. Plainly, that duty cannot require or justify the exercise of a remedial power that is not available in relation to that RMS, for example by reason of section 86(1)(c): in such circumstances, the comprehensive solution must be found amongst the alternative powers available to the Commission. It is equally plain that, whilst the existence of alternative powers may affect the selection of a remedy in a specific case (under the well-established Tesco principles of proportionality), it cannot affect the general availability of a power to the Commission. That is the case, a fortiori, where the alternative powers are less effective than the power under consideration. Those alternatives, which we discuss in the following paragraphs, include other powers available to the Commission in the event that it identifies an SLC following a reference to it and other powers available to the OFT instead of making a reference to the Commission.
  117. So far as the powers (other than enforcement orders as defined in section 86(6) of the Act) available to the Commission are concerned, Mr Ward suggested first that the Commission could make behavioural orders with respect to the conduct of members of the AN Group within the UK. There may be cases in which such orders would be a sufficient remedy but, equally, there must be cases in which the Commission reasonably concludes that prohibition of the RMS is the only adequate remedy for the SLC identified. Indeed, this case appears to illustrate that possibility: despite its suggested alternative remedies, AkzoNobel has not challenged the reasonableness or proportionality of the prohibition order as such.
  118. Mr Ward's second suggestion was that the Commission could accept final undertakings under section 82 of the Act and apply to the civil courts to enforce them under section 94 of the Act. In that respect, we accept Mr Beard's argument that the unavailability of a final order-making power could have a significant bearing on the willingness of a party to offer final undertakings. Furthermore, the Act specifically provides a mechanism for the enforcement of undertakings where they are not fulfilled. That mechanism is provided by section 83 but that section is also subject to the 'fetter' of section 86. As such, any undertakings offered would only be enforceable under section 94 by means of enforcement proceedings in court. That, as Mr Beard submitted, would circumvent the mechanism that Parliament specifically enacted for the Commission to enforce undertakings given to it and may also be less effective, both as a means of enforcement for the Commission and as an inducement to the party giving undertakings to comply with them. Finally, as this case illustrates, the power to accept final undertakings would not address the situation in which prohibition is the only effective and proportionate remedy: there is no reason to believe that a party would voluntarily abandon a transaction that the Commission has no power to prohibit.
  119. AkzoNobel's more radical suggestion was that the enforcement gap in this case could have been avoided if the OFT had made use of its power under Article 22 of the Merger Regulation to request investigation of the Transaction by the European Commission. Had such a request been made and accepted, the Merger Regulation would undoubtedly have afforded the European Commission power to prohibit the Transaction should it have found that it gave rise to competitive harm and that prohibition was the appropriate remedy. There are, no doubt, cases where such a request may be the appropriate course for the OFT to take. That cannot, however, limit the proper scope of the Commission's jurisdiction under the Act. As Mr Beard observed in argument, the OFT can only request, not require, review by the European Commission; the scope of the Commission's order-making power cannot be limited by reference to the fact that the OFT, rather than referring a transaction to the Commission under the Act, could refer it under the Merger Regulation, a reference that the European Commission may refuse. Furthermore, Article 22 of the Merger Regulation only applies where there is an effect on trade between Member States. It is, therefore, inapplicable to international transactions where the trade affected is solely between the UK and non-European countries. Taking those two factors together, it cannot have been Parliament's intention that the power to make an enforcement order under the Act should be limited – or indeed interpreted – by reference to the availability of Article 22 in certain international transactions.
  120. We accordingly conclude that, for the reasons stated above, section 86(1)(c) is not to be given a special interpretation by reason of the particular purposes of the merger control regime established under the Act.
  121. (iii) The second issue: interpretation and application of section 86(1)(c) in accordance with general company law principles
  122. AkzoNobel maintains that the "carrying on business" criterion is to be interpreted in accordance with the general company law principles on the attribution of activities of subsidiaries to parent companies. It relies in that respect upon the Court of Appeal's judgment in Adams. As we have already noted, that case is the leading authority on the circumstances in which a foreign court is entitled to take jurisdiction over an overseas company[13] such that the English court should recognise a judgment entered by the foreign court against that company. The potential relevance of that judgment to the present case arises because it was accepted that a foreign court only has jurisdiction over an overseas company (absent some special statutory language[14]) if it is to be treated as 'present' in the territory of that court which, in turn, depends on whether the overseas company is to be treated as carrying on its own business in that territory either through its own servants and agents, or through a representative.
  123. The facts in Adams were complex. In summary, the case arose from litigation brought in the United States by employees and ex-employees of companies in the Cape Industries Group in respect of personal injuries allegedly arising from exposure to asbestos dust. The claimants brought proceedings in the US Courts against, amongst others, two members of the Group that were incorporated in England and maintained no physical presence in their own names in the United States, namely Cape (the ultimate parent company) and Capasco (a wholly-owned subsidiary of Cape responsible for worldwide marketing). Those defendants allowed default judgments to be entered against them. The claimants brought proceedings in England to enforce the default judgments on the footing that the US Courts had jurisdiction over Cape and Capasco by virtue of their presence in the United States through three companies that were responsible for the sale and marketing of the Group's products in the United States. The first, N.A.A.C., was incorporated and physically present in the US. The other companies were organised in furtherance of a scheme to reduce the Group's exposure to litigation and taxation in the United States: one was A.M.C., a Liechtenstein company whose shares were held by a nominee on trust for a Group subsidiary, and the other was C.P.C., a US company owned by the executive director of N.A.A.C. In essence, the scheme involved the liquidation of N.A.A.C. with the US sales and marketing activities being assumed by A.M.C. and C.P.C.
  124. In the result, Scott J dismissed the action and his decision was upheld by the Court of Appeal. The essential elements of the Court of Appeal's judgment, so far as it is relevant to the present case, appear from the headnote to the report of the judgment (at p.436):
  125. "(1) … [A]n overseas trading corporation was likely to be treated by the English court as present within the jurisdiction of the courts of another country only where either such a corporation had established and maintained at its own expense in that other country a fixed place of business and had carried on from there its business for more than a minimal period of time through its servants or agents or through a representative; that in either of these two cases presence could only be established where the overseas corporation's business, whether together with the representative's own business or not, has been transacted at or from the fixed place and, in order to ascertain whether the representative had carried on the corporation's business or his own, it would be necessary to investigate his functions and his relationship with the overseas corporation … .
    Quaere. Whether residence without presence will suffice.
    (2) … [O]n the facts, C.P.C. was an independently owned company and by the liquidation of N.A.A.C. and the creation of A.M.C. and C.P.C. Cape wanted sales of asbestos from its subsidiaries to continue in the United States but intended, by any lawful means which it was entitled to do, to reduce the appearance of its, or its subsidiaries', involvement there and to reduce the risk of its being liable for United States taxation or subject to the jurisdiction of the United States courts, and that, accordingly, since it was not a mere façade concealing the true facts it was not appropriate to pierce the corporate veil; … furthermore, since it was accepted that N.A.A.C. was incorporated so as to assist in marketing and to be a marketing agent of the Cape group in the United States and, since a substantial part of its business at all material times was, in every sense, its own business, it did not act as an agent of the Cape group; … in any event, since N.A.A.C. had no general authority to enter into contracts binding Cape or Capasco and third parties no such transactions were ever entered into by N.A.A.C.; … accordingly, the defendants were not present in the United States of America through N.A.A.C., C.P.C. or A.M.C. ..." (citations and internal references omitted)
  126. With that introduction, we consider the significance of the Court of Appeal's judgment for the present case in greater detail. No case, other than Adams, was cited to us in relation to this issue. Nevertheless, we are well aware that the issues raised in Adams have been the subject of extensive consideration in other authorities, albeit in different contexts, both before and after that judgment. Whilst we do not consider that those authorities alter the conclusions to be drawn from a proper consideration of Adams, we make reference to some of the authorities so that the application of Adams in the present case can be seen in its proper context.
  127. Any analysis must start with the well-known axioms established by the decision of the House of Lords in Salomon v A Salomon Ltd [1897] AC 22. A duly-incorporated company is a legal person distinct from its shareholders, even if it has only one shareholder who, therefore, has exclusive control of the company. A company's separate personality constitutes a 'veil of incorporation' from which it follows that the company's business is its own, not that of its shareholders, and that the company's rights and liabilities are its own, not those of its shareholders. The continuing vitality of the Salomon principles is evident from the judgment of Lord Neuberger in VTB Capital v Nutritek International Corp [2013] UKSC 5 at paragraphs 118 et seq. and from the subsequent judgment of the Supreme Court in Prest v Petrodel Resources Limited [2013] UKSC 34 (which was handed down after the hearing in this case).
  128. It has been conventionally accepted that, nonetheless, there are exceptional circumstances in which the Courts are entitled to lift or pierce the veil of incorporation. After close consideration of the legitimacy of that position, the Supreme Court confirmed in Prest that, in principle, such exceptions do exist. Lord Sumption, giving the leading judgment, concluded (at paragraph 35) that:
  129. "… [T]here is a limited principle of English law which applies when a person is under an existing legal obligation or liability or subject to an existing legal restriction which he deliberately evades or whose enforcement he deliberately frustrates by interposing a company under his control. The court may then pierce the corporate veil for the purpose, and only for the purpose, of depriving the company or its controller of the advantage that they would otherwise have obtained by the company's separate legal personality."
    Lord Neuberger adopted Lord Sumption's conclusion: see paragraph 81. We note, however, that a majority of the Supreme Court, whilst endorsing Lord Sumption's analysis, did not wholly exclude the possibility that exceptions may also be made in other unspecified but rare circumstances.[15]
  130. It is important also to mention at the outset that the Salomon principles do not preclude the Courts taking into account the reality of a company's ownership where that can be done without violating the Salomon principles. To give two specific examples, we refer to the judgments of the Court of Appeal in Atlas Maritime Co v Avalon Maritime, The Coral Rose (No. 1) [1991] 4 All ER 769 and Chandler v Cape plc [2012] EWCA Civ 525. In The Coral Rose (No. 1), the Court held that, whilst it would be wrong to find a principal/agency relationship between a creditor and a debtor which was a shell company whose sole activity was sponsored, funded and controlled by the creditor (a proposition described by Staughton LJ, at p.779E, as "revolutionary doctrine"), the reality of the relationship meant that payment to the creditor would not be in the ordinary course of business so that release of the debtor's assets from a Mareva injunction (now more commonly referred to as a freezing order) for that purpose would be inappropriate. In Chandler, the Court held that a parent company was liable to the ex-employees of a liquidated subsidiary in respect of personal injuries suffered by them as a result of conduct on the part of the parent in assuming responsibility for aspects of the group's health and safety policy. The Court held that, having assumed that responsibility, the parent owed a duty of care directly to the employees of its subsidiary. We note that at paragraph 69 of her judgment, Arden LJ (with whom Moses and Macfarlane LJJ agreed) stressed that the decision had nothing at all to do with piercing the veil of incorporation. She continued at paragraph 70 that the "question is simply whether what the parent company did amounted to taking on a direct duty to the subsidiary's employees."
  131. In the specific context of a jurisdictional assessment, there appear to be three grounds (considered by the Court in Adams) upon which a company not itself directly present in the jurisdiction might nonetheless be treated as present by reason of the direct presence of a subsidiary, namely that (i) the subsidiary has carried on the parent's business within the jurisdiction as the agent of the parent, or (ii) the veil of incorporation between the parent and the subsidiary should be pierced, or (iii) the parent and the subsidiary should be treated as part of a single economic unit.
  132. It was not the Commission's case, either in the Report or before us, that AkzoNobel's UK subsidiaries were its agents, nor that they were merely sham companies or façades (still less that there had been any impropriety[16]) such that the veil should be pierced. The Court of Appeal's discussion of the principal/agency question is relevant because Mr Ward relied upon the Commission's failure fully to consider the non-exhaustive and non-determinative check-list of factors identified by the Court of Appeal in that context as an indication of the Commission's non-compliance with general corporate principles:[17] in particular, Mr Ward stressed the fact that the Commission had disregarded the UK-based subsidiaries' lack of authority to bind AkzoNobel despite the importance attached to that factor in Adams. We do not consider, however, that this point can bear the weight that Mr Ward sought to put on it. It is quite clear from the Court of Appeal's judgment (at pp.
    530F – 531B of the case report) that the factors are only pertinent to the question whether a principal/agency relationship exists. They are not necessarily pertinent to the existence or consideration of the other grounds upon which jurisdiction might be established. Specifically, the fact that the subsidiary lacks contracting authority and has not in fact purported to exercise any such authority would be wholly consistent with the existence of a single economic unit.
  133. It is to the single economic unit argument (considered at pp. 532D – 539C of the case report) that we now turn. The Commission has not sought to rely expressly on this argument: indeed its position is somewhat closer to that expressed in paragraph 70 of Arden LJ's judgment in Chandler. Nonetheless, we think the argument is better addressed than avoided, because it seems the most pertinent of the exceptions mentioned in Adams, and, as Mr Beard observed in passing,[18] if Adams is relevant then the findings made by the Commission might well meet the requirements of that case law. A close consideration of the reasoning in Adams is, therefore, required.
  134. The Court of Appeal first directed its consideration to the question whether, as a matter of general principle, English law would admit that argument in circumstances where justice so required. The Court concluded that, whilst it had some sympathy with the claimants' arguments, it had no discretion in law to accede to them, holding that:
  135. "... [S]ave in cases which turn on the wording of particular statutes or contracts, the court is not free to disregard the principle of Salomon v A Salomon & Co Ltd [1897] AC 22, [1895–9] All ER Rep 33 merely because it considers that justice so requires. Our law, for better or worse, recognises the creation of subsidiary companies, which though in one sense the creatures of their parent companies, will nevertheless under the general law fall to be treated as separate legal entities with all the rights and liabilities which would normally attach to separate legal entities.
    In deciding whether a company is present in a foreign country by a subsidiary, which is itself present in that country, the court is entitled, indeed bound to investigate the relationship between the parent and the subsidiary. In particular, that relationship may be relevant in determining whether the subsidiary was acting as the parent's agent and, if so, on what terms. In Firestone Tyre and Rubber Co. Ltd. v. Lewellin [1957] 1 W.L.R. 464 … the House of Lords upheld an assessment to tax on the footing that, on the facts, the business both of the parent and subsidiary were carried on by the subsidiary as agent for the parent. However, there is no presumption of any such agency. There is no presumption that the subsidiary is the parent company's alter ego. … If a company chooses to arrange the affairs of its group in such a way that the business carried on in a particular foreign country is the business of its subsidiary and not its own, it is, in our judgment, entitled to do so. Neither in this class of case nor in any other class of case is it open to this court to disregard the principle of Salomon v A Salomon [1897] AC 22 merely because it considers it just so to do." (pp. 536H – 537D, emphasis added).
  136. The Court then considered the specific circumstances of the relationship between Cape/Capasco and N.A.A.C. In that consideration, the Court accepted that there was some broad support in the evidence for the submission that Cape ran a single integrated mining division with little regard to corporate formalities as between members of the group. That said, there was no challenge to Scott J's finding that the corporate forms applicable to N.A.A.C. as a separate entity had been observed. Furthermore, there was no challenge to his findings that (a) the corporate financial control exercised by Cape over N.A.A.C. in relation to the level of dividends and the level of permitted borrowing was no more and no less than was to be expected in a group such as the Cape group and (b) N.A.A.C.'s annual accounts were properly drawn on the basis that N.A.A.C.'s business was its own business. The claimants did, however, submit that Cape's control extended to the day-to-day running of N.A.A.C. The details of that control were considered in the unreported appendix to the judgment. The conclusion of that consideration was that the claimants' submission failed because it had not been established that day-to-day control was exercised by Cape rather than by N.A.A.C.'s own executive director (Mr Morgan), as explained in the following extract:
  137. "A degree of overall supervision, and to some extent control, was exercised by Cape over N.A.A.C. as is common in the case of any parent-subsidiary relationship – to a large extent through Dr Gaze [a director of Cape]. In particular, Cape would indicate to N.A.A.C. the maximum level of expenditure which it should incur and would supervise the level of expenses incurred by Mr Morgan. Mr Morgan knew that he had to defer in carrying out the business activities of N.A.A.C. to the policy requirements of Cape as the controlling shareholders of N.A.A.C. Within these policy limits … the day-to-day running of N.A.A.C. was left to him." (p.538C-D)
  138. On the basis of that analysis, the Court rejected the claimant's contention that Cape and Capasco were to be regarded as present in the United States by virtue of the single economic unit argument.
  139. The career of the single economic unit argument following Adams has been, to say the least, chequered. Hobhouse LJ in Ord v Belhaven Pubs [1998] EWCA Civ 243 observed that the Court of Appeal had recognised that the concept was "extremely limited indeed". Most recently, Flaux J in Linsen International v Humpuss Sea Transport [2011] EWHC 2339 (Comm) at paragraph 126 dismissed the argument completely:
  140. "Although the single economic unit argument has reared its head again in other cases, it has done so only to be firmly rejected (for example by Hobhouse LJ in Ord at 456i – 457f) and it forms no part of English law."
  141. The single economic unit argument has consistently foundered on the remarks of Robert Goff LJ on Bank of Tokyo v Karoon [1987] AC 45 to the effect that:
  142. "Counsel suggested beguilingly that it would be technical for us to distinguish between parent and subsidiary company in this context; economically, he said, they were one. But we are concerned not with economics but with law. The distinction between the two is, in law, fundamental and cannot here be bridged."
  143. Mr Beard suggested that it would be ironic to rely upon that distinction in support of a restrictive interpretation of a statute that is entirely focused on economic issues. Karoon cannot, in our judgment, be so readily set aside. On the contrary, it follows from paragraph 82 above that those remarks are fully applicable in the present case, notwithstanding its economic subject matter.
  144. Having said that, it is essential to consider the true scope and import of the single economic unit argument as discussed in Adams. At first blush, the Court of Appeal's reference to the "degree of overall supervision and to some extent control" exercised by Cape over N.A.A.C. and its detailed assessment of the degree of that control (summarised in paragraph 101 above) appear to suggest that the Court's conclusion might have been different had it accepted the claimants' case that Cape did indeed exercise day-to-day control over N.A.A.C.'s operations. Closer consideration convinces us, however, that that cannot be reconciled with the Salomon principles. It is clear from Salomon itself that the exercise of complete de facto control by a single shareholder over a company does not defeat the company's distinct legal personality or mean that its business and its associated rights and liabilities are to be treated as those of the shareholder. If that is true for a single natural shareholder, it must also be true for a single corporate shareholder.
  145. Whilst that excludes the attribution of the company's business to the shareholder, there is an important respect in which the situation of the one-man shareholder is distinguishable from that of the corporate parent in a large and complex group such as the AN Group. For a one-man shareholder of the kind discussed in Salomon, the entirety of the shareholder's business activities are conducted through the company; there is no meaningful sense in which it can be said that the shareholder has a business other than that of his company. By contrast, in the case of a large and complex group of companies, there is an entirely meaningful sense in which it is legitimate to consider whether the parent company carries on business in a way that is more than the simple sum of the activities of its operating subsidiaries.
  146. When the issue is framed in that way, the objections arising from the Salomon principles appear to us to fall away. The business activities under examination are those of the parent company as such. It is inherent in the concept of separate personalities established by the Salomon principles that those activities may adhere to the parent company independently of the businesses of the operating subsidiaries. There is, therefore, no question of conflating the businesses of the parent and subsidiary, and, most importantly, there is no question of fixing the parent company with the rights and liabilities of its subsidiaries.
  147. Consideration of that issue raises two questions in the context of section 86(1)(c): first, what activity of the parent company constitutes the carrying on of a business and, secondly, is that activity carried on in the United Kingdom? We are, of course, mindful of the fact that these questions cannot simply be answered by reference to the exercise of control over a subsidiary's business: were that approach to be adopted, it would be a clear breach of the Salomon principles. It is a question of fact and degree in each case whether the activities of the parent company are such as to be treated as carrying on business activities that are properly attributable to it as a legal person.
  148. In that context, we note the Commission's observation, repeated on several occasions (see paragraphs 11.91, 11.93 and 11.94 of its Report) that the arrangements in the AN Group are common, or at least not unusual, amongst large multi-national groups. We are not entirely sure what importance – if any – the Commission attached to that observation. It is, in our judgment, immaterial to the assessment of AkzoNobel's activities in the United Kingdom to consider whether those activities are commonplace or not. It is important, however, to emphasise that our judgment is limited to the facts of this case as found by the Commission and stated in the Report. In that respect, the Commission's finding that legal and operational structures within the AN Group diverge is of central importance. We do not intend to address other cases, including in particular (i) the case in which there is no distinction between legal and operational structures so that parental control is exercised wholly through the parent/subsidiary relationship and (ii) the case in which the parent company commences business in a country by establishing a subsidiary there.
  149. The first question set out in paragraph 109 above raises the logically prior question whether any activities of a parent holding company are to be regarded as carrying on a business. In that connection, we have already referred to the very broad concept of a business activity and the circularity of its definition in section 129(1) of the Act. Although the relevant part of the definition requires that the parent company's activity be "carried on for gain or reward", we see no reason why it should be confined to those specific activities that are the immediate generators of revenue. Such a construction would make little sense in view of the fact that the most obvious generators of revenue (the supply of goods or services) are identified as a separate instance of a business. In any event, it is hard to envisage what purpose the activities of a parent company do serve if it is not gain or reward. Equally, it does not make sense in our view to limit its scope by a form of reverse logic under which contracting authority is deemed to be an important jurisdictional locator so it must also be a defining characteristic of a business activity. On the contrary, it seems to us that, in principle, the notion of a business activity must be capable of including the functions characteristically performed by a parent holding company. One can, perhaps, test the proposition in this way: if we were asked to consider whether, as a matter of English law, AkzoNobel itself were carrying on business in the Netherlands by reason of its activities there as a holding company, we have no doubt that the answer would be in the affirmative.
  150. The decisive issue in relation to the first question, in our judgment, goes to the extent of the activities that are undertaken by the parent company. Viewed from that perspective, the first and second questions are closely connected because the extent of those activities likely has a decisive influence upon the issue of location. In the latter connection, we are mindful of the fact that in Adams there was no suggestion that the US Courts would be entitled to claim jurisdiction on the basis that Cape's own activities should be treated as extending to carrying on business in the United States, even though those activities included the operation of an integrated mining division that included the US subsidiaries. That omission may be attributed, we believe, to the requirement (apparently assumed in Adams but not present in section 86(1)(c) of the Act) that the overseas corporation should carry on business through a fixed place of business within the territory of the court asserting jurisdiction: whilst the Court reserved its position as to whether residence without presence would suffice, it did not decide the case on that basis or expand on what that concept would entail. We note, however, that the Court did refer (at p. 536F-G) to the saving for cases that "turn on the wording of particular statutes ...". If Parliament had intended a fixed place of business within the jurisdiction to be a necessary, or at least highly indicative, criterion for the 'carrying on business' test, it would have been a simple matter to include that requirement in section 86(1)(c), or indeed the definition of 'business' in section 129, of the Act.
  151. With those observations in mind, we now consider whether the Commission's decision discloses any error of law. That assessment is made on the bases that we have already stated namely that that the Commission's findings of primary fact and its analysis of those findings stand unchallenged. The Commission's Report does not expressly articulate the legal standards by reference to which it concluded that section 86(1)(c) is satisfied by AkzoNobel. Rather, those standards have to be inferred from the Commission's findings. The Commission's central conclusion was that the organisational and decision-making structure of the AN Group is based upon its functional units rather than its operating subsidiaries. Strategic decisions are made within the functional units, as evidenced by the absence of a strategic plan for subsidiaries. Contracting decisions are likewise made within the functional units: in that respect, Mr Mario Siragusa (who appeared for the Interveners) drew our attention to the Authority Schedule generally, section 10.7 of which specifies contracting authorities by reference to the functional units without making any reference to the subsidiaries located within those units. Similarly, other operational decisions are made within the functional units. Taken together, we are satisfied that the Commission was entitled, as a matter of law, to conclude that these activities constitute the carrying on of business within the functional units and that that activity extends to the UK. We are further satisfied that that conclusion does not violate the Salomon principles in view of the structure of the AN Group in which different functional units comprise disparate subsidiaries and decisions are made (as the Commission has found) within the functional units.
  152. An important aspect of the Commission's unchallenged decision is that, based on the Authority Schedule, decision-making within the AN Group is centralised through ExCo, which is an organ of AkzoNobel itself. It might be said that that decision is at variance with the distribution of decision-making authority between ExCo and the functional units. That issue is not, however, open to AkzoNobel in a challenge based solely on an error of law. In that context, it is important to appreciate that the language of section 86(1)(c) cannot be applied to a group of companies; it necessitates that the business activities are attributed to a legal person, or persons, within the group. The activities of AkzoNobel's functional units must be attributed to a legal person. Neither the ANPG SBU, nor the Industrial Coatings BU have separate legal personality so that the activities of those units cannot be attributed, for the purpose of section 86(1)(c), to them. They must, be attributed either to AkzoNobel itself or to the subsidiaries that are located within the units. In determining which of those attributions is correct, the Commission is in our judgment entitled, as a matter of law (consistently with section 86(1)(c) and without violating the Salomon principles), to consider, on the basis of the evidence available to it, whether the decisions made within the functional units are properly to be regarded as decisions made by the organs of the subsidiaries or decisions made by the functional units that are implemented through the subsidiaries. If the latter, then it may be the case – and this will be a matter for factual assessment – that the decisions of the functional units are in reality those of the ultimate holding company.
  153. We are satisfied that, given the nature and scope of the activities that have been attributed by the Commission to AkzoNobel, the Commission's Report discloses no error of law in concluding that they constitute business activities carried on by AkzoNobel in the United Kingdom.
  154. For these reasons, we unanimously conclude that Ground 1 of AkzoNobel's application for review must fail.
  155. V. GROUND 2: METLAC COMPETES MORE AGGRESSIVELY ON PRICE

    A. Introduction

  156. AkzoNobel's second ground of review relates to the Commission's findings on price competition. At 9.12 of the Report, the Commission set out the following conclusion:
  157. "... [W]e are of the view that it is unlikely that Valspar and PPG would replicate the constraint that Metlac currently provides in relation to B2E because they do not compete as aggressively on price as Metlac..."
  158. AkzoNobel argued that finding was unlawful because the Commission failed to take a decision that was supported by the evidence, failed to carry out sufficient enquiries and failed to have regard to material considerations. Given the centrality of this conclusion to the Commission's SLC finding, AkzoNobel submitted that the decision prohibiting the Transaction cannot stand and the Report must be quashed.
  159. In reaching its conclusion that Metlac competes more aggressively on price than PPG or Valspar, the Commission considered, inter alia:
  160. (a) competition in the coatings industry – including evidence relating to market structure, market development (growth and innovation), product-customer overlaps, switching and pricing (see Chapter 8 of the Report); and
    (b) the competitive effects of the Transaction – including unilateral effects, the loss of actual and potential competition in the relevant market, and efficiencies (see Chapter 9 of the Report).
  161. In its analysis of: (i) whether Metlac is a significant competitive force (see Report, paragraphs 8.127 to 8.168); and (ii) switching in the metal packaging coatings market (see Report, paragraphs 8.169 to 8.202), the Commission relied on evidence of customer views and of pricing. AkzoNobel's challenge under its second ground of appeal centres on the Commission's use of this evidence.
  162. The Commission drew on three sources of evidence in its analysis of Metlac's competitiveness on pricing:
  163. (a) a subset of the responses provided to the BKartA's survey of customer views and pricing in the coatings market as part of that authority's investigation of the Transaction under German law;
    (b) responses to the Commission's survey of customer views (which were gathered by various means, including by written questionnaire, oral hearings and written follow-up questions); and
    (c) the Commission's own pricing data.
  164. AkzoNobel submitted that (i) the Commission's analysis of, (ii) the reliance it placed on, and (iii) the conclusions it drew from, each of these sources was flawed, such that the Commission's findings on price competition are fatally undermined.
  165. We note the wide-ranging analysis conducted by the Commission, of which customer views and pricing formed part. We have already accepted that, as Mr Beard emphasised, the Report must be read as a whole and not analysed as if it were a statute (see paragraphs 39(d) and 40 above).
  166. B. The BKartA customer survey

  167. As part of its investigation into the Transaction pursuant to German law, the BKartA carried out a postal survey of customer views (see BKartA decision, paragraph 11).[19] Respondents to this survey also provided pricing data to the BKartA. The Commission obtained a subset of these responses[20], which it analysed as part of its assessment of price competition in the coatings markets it was concerned with.
  168. It will be recalled that the BKartA cleared the Transaction, unlike the Commission. It is important to note, however, the relevant legal tests being applied by those two national competition authorities were not the same. The BKartA was required, pursuant to the German merger control law then in force, to consider whether the Transaction could be expected to create or strengthen a dominant position on the relevant markets that would not be outweighed by improvements of the conditions of competition.[21] In contrast, section 36(1) of the Act requires the Commission to determine whether the Transaction may be expected to result in an SLC. The latter clearly sets a lower threshold for intervention by the Commission than that which was applied by the BKartA.
  169. In addressing the German dominance test, the BKartA considered Metlac's pricing. The BKartA's report refers to customers' views of Metlac's pricing and pricing data provided by customers (see, for example, paragraphs 82, 103 – 104 and 108). This data was potentially a valuable additional source of evidence for the Commission. It was, therefore, appropriate for the Commission to consider this evidence, with a critical eye, in its own investigation of the Transaction pursuant to the Act.
  170. AkzoNobel made two principal challenges to the Commission's use of the responses to the BKartA's survey. First, that the "partial" set of BKartA customer responses used by the Commission skewed its conclusions. This, AkzoNobel asserted, amounted to a failure to have regard to a material consideration. Secondly, AkzoNobel contended that the Commission reached a different conclusion to the BKartA in relation to price competition. In failing to establish why its analysis was at variance to that of the BKartA, AkzoNobel alleged that the Commission failed to conduct sufficient enquiry. It is to be noted that at no point did AkzoNobel argue that the Commission was in some way bound by the decision the BKartA came to.
  171. (i) The Commission's alleged reliance on a partial set of pricing data from the BKartA's survey
  172. The Commission obtained a partial set of responses to the BKartA's survey from customers. In its analysis of pricing data (as opposed to customer views), the Commission relied on a subset of these responses. AkzoNobel submitted that this partial use of responses to the BKartA's questionnaire skewed the Commission's conclusions, amounting to a failure to have regard to a material consideration.
  173. In particular, AkzoNobel argued that the B2E pricing data provided in response to the BKartA survey and considered by the Commission was overwhelmingly based on the data of one customer, which accounted for most of Metlac's UK turnover. According to AkzoNobel, by failing to look at B2E pricing for a broader range of customers – particularly those who used other suppliers – the Commission introduced a serious flaw into its analysis in that a customer supplied by Metlac was likely to consider it was getting a good deal or would, presumably, have chosen a different supplier. Indeed, it may have been that customers used other suppliers precisely because Metlac's prices were too high.
  174. In this regard, AkzoNobel emphasised that, of the four main B2E customers, the Commission only had data from two of them. Of those two, one customer ([…][(]) did not see Metlac as a low-priced supplier (see paragraph 8.135 of the Report). We note, however, that the pricing data provided by that customer shows that Metlac was in fact the cheapest supplier in four out of five instances (see Report, Appendix K, Table 1).
  175. When analysing the pricing data provided to the BKartA (and made available to the Commission), the Commission selected those responses which contained data regarding Metlac's prices in the B2E market segment. The Report makes it clear at paragraph 8.139 that the Commission was conscious of the risk that customers who did not provide pricing data for Metlac's B2E products (because they did not purchase them) may have used alternative suppliers based on price considerations. Further, the Commission took additional steps to investigate the position, including by analysing the customer views provided in response to both the BKartA's and its own customer surveys (including oral hearings and written follow-up questions), and by conducting its own pricing analysis.
  176. Accordingly, we consider that AkzoNobel's argument that the Commission failed to have regard to a material consideration by relying on only a subset of pricing data provided to the BKartA fails.
  177. (ii) The alleged divergence between the BKartA and the Commission's findings on price competition
  178. AkzoNobel further contended that the subset of responses relied on by the Commission led it to reach a different conclusion to that reached by the BKartA with regard to pricing, namely that Metlac was the lowest priced competitor, whereas the BKartA stated that Metlac's prices were "not always" lowest.
  179. Using its survey results, the BKartA made the following statements regarding pricing:
  180. "103. Some purchasers complained in the purchaser survey that the undertakings Akzo, PPG and Valspar offer their products for a high price. Other purchasers stated that there is no intensive competition among these companies. When looking at the prices the undertakings Akzo, PPG and Valspar did not receive good marks.
    104. However the analysis of the individual prices reveals that the prices of Metlac are not always the best compared with Akzo, PPG and Valspar ... Thus the information regarding individual prices does not confirm that Metlac is always the cheapest supplier." (emphasis added)[22]
  181. AkzoNobel read the above paragraphs as signifying an important mismatch between the views expressed by customers and the data provided in response to the BKartA's survey. This is important, according to AkzoNobel, because it demonstrated that the views of customers could not be regarded as self-proving. But it is not the case that the customer views summarised in paragraph 103 of the BKartA's decision pointed in a different direction to the pricing data addressed in paragraph 104. In our judgment, there is no inconsistency between the BKartA's summary of customer views, at its paragraph 103, and what is stated at paragraph 104, namely that Metlac is sometimes, but not always, the cheapest supplier.
  182. Mr Ward argued that the Commission failed to establish why its analysis was so clearly at variance with that of the BKartA and, therefore, that its reasoning fails on judicial review grounds. He appears to be relying on the BKartA's statement at paragraph 104 (as set out above) that the pricing analysis did not confirm that Metlac is "always the cheapest supplier". That argument only holds up, however, if the Commission actually found that Metlac was always the cheapest supplier. That is plainly not the finding the Commission made. On the contrary, and in similar terms to those used by the BKartA, the Commission's conclusion on prices included the following: "[w]hile we observed that Metlac is not always the lowest price supplier, for a number of products and customers Metlac has been charging prices lower than its competitors." (see Report, paragraph 8.168)[23] Accordingly, this argument fails.
  183. (iii) The overall approach of the BKartA and the Commission
  184. Whilst it was common ground that the BKartA's analysis was conducted pursuant to a different legal regime, Mr Ward nevertheless appeared to suggest that the BKartA's analysis was more thorough than the Commission's. In light of the BKartA's clearance of the Transaction under the different German competition law test, there was an implicit inference that the Commission's overall conclusion may therefore be defective. This inference is not persuasive.
  185. Whilst the merits or otherwise of the BKartA's reasoning – and indeed its conclusion – are outside the scope of this judgment, we note that, in fact, the Commission's investigation was, at least, as rigorous as the BKartA's and in some respects more so. In particular, we note that the Commission's in-depth investigation was carried out over a period of 32 weeks compared to a period of only 17 weeks in Germany. Moreover, the BKartA's pricing analysis was set out over a handful of paragraphs in its 30 page report. By contrast, the Commission devoted some 18 pages and an entire annex to pricing issues, in the context of a 150 page report (excluding annexes).
  186. More telling than this quantitative comparison, however, is the conduct of the respective inquiries. The BKartA sets out the course of its inquiry at paragraphs 8-25. The matters set out there can be contrasted with those set out by the Commission in paragraphs 4-14 of Appendix A to the Report. It is instructive to read those side by side. We note in particular that, whilst both authorities sent out detailed questionnaires to large numbers of purchasers and suppliers of coatings, and received multiple rounds of written submissions, the BKartA's investigation was carried out principally on paper. The Commission, in addition to the undoubtedly important paper exercise, also held nine oral hearings with selected third parties,[24] made site visits to both Metlac and AkzoNobel's premises, and held hearings with each of those companies. That, taken together with the in-depth examination of those data sources and the comprehensive reasoning behind the Commission's conclusions led us to conclude that the Commission's inquiry went somewhat further than the BKartA's.
  187. For the reasons given above, we are satisfied that the Commission's treatment of responses to the BKartA's survey was not irrational.
  188. C. The Commission's customer survey

  189. The Commission gathered customer views using a survey, oral hearings and follow-up questions. AkzoNobel submitted that there were three overarching flaws in this evidence, which undermined the Commission's findings on price competition:
  190. (a) sampling bias - by focussing on the responses of Metlac customers, there was a bias in the customer sample;
    (b) leading questions - the wording of the Commission's customer survey presupposed answers which would support an SLC finding; and
    (c) conflation of evidence from a separate market - the Commission erroneously used evidence from the FCG market to support findings in the distinct B&B market.
    (i) The alleged sampling bias
  191. AkzoNobel submitted that the Commission was guilty of "sampling bias" as it allegedly focussed on Metlac's own customers, who were more likely to consider Metlac's prices to be the lowest (and for those prices to have in fact been the lowest for those customers) in the market for the products they purchased. AkzoNobel argued that this sample of customers introduced an inherent bias into the Commission's inquiry which was more likely to lead to a finding that Metlac competes more aggressively on price than Valspar and PPG in B2E. Such an approach, AkzoNobel stated, is contrary to the Commission's own good practice guide for the design and presentation of consumer survey evidence in merger inquiries.[25]
  192. AkzoNobel referred in particular to the Commission's statement that:
  193. "Metlac was often mentioned as the lowest-priced supplier by customers that purchase a substantial share of their demand from it" (paragraph 7 of Appendix K to the Report; emphasis added).
    AkzoNobel's case was that this was hardly a startling conclusion. It argued that, in only taking into account the views of Metlac's customers, the Commission failed to have regard to a material consideration, namely, the views of customers not supplied by Metlac and the pricing of products not supplied by Metlac. In AkzoNobel's view, it was particularly important to analyse those tenders which Metlac had participated in but not been awarded. AkzoNobel submitted that it was not open to the Commission to conclude that Metlac's prices were in fact consistently low across the market based only on the views of its customers.
  194. We do not agree that this is in fact what the Commission did and, as such, we reject as unfounded this aspect of AkzoNobel's challenge. First, it must be recalled that there are only four buyers of B2E products of any significance in the UK. That made the target population from which a sample could be drawn exceptionally small to begin with. Secondly, it is, with such a small population, almost impossible to carry out any probative statistical analysis. This means that customer views (albeit that these must be properly tested) take on a renewed significance that might not be seen in more diverse markets. Thirdly, whilst the Commission does accord prominence to the views of Metlac's customers in the Report, it is not accurate to say that it ignored the views of other buyers. In particular, we note that the Commission sent detailed questionnaires to 15 large customers, all of which were supplied by AkzoNobel and 14 of which bought from Metlac (see Report, Appendix K, paragraph 12).[26] The Commission records that it actually received more responses from AkzoNobel's customers than it did from Metlac's (Report, Appendix K, paragraph 14). It further records that it considered whether the views were biased towards existing customers of Metlac alone and concluded they were not (Report, Appendix K, paragraph 15).
  195. It is our view that, on that basis, it cannot be said that the Report suffered from some inherent sampling bias. The Commission plainly took care to gather views from a range of customers and was alive to the dangers of placing too much emphasis on the views of Metlac's customers alone. Beyond that, and in accordance with the applicable judicial review principles set out in Section III. above, the relative weight to be placed on the views it received was principally a matter for the Commission. In any event, and as will be seen when we come to consider the Commission's own pricing analysis in Section V.D. below, the Commission sought to cross-check the conclusions it drew based on customer views against empirical pricing data.
  196. (ii) The alleged use of leading questions
  197. AkzoNobel argued that the Commission used leading questions in its customer survey. AkzoNobel focused its attack principally on the following question:
  198. "I. Please provide us with details of any instances in the last 5 years where Metlac's lower pricing is a factor you have used in pricing discussions with other metal packaging coatings suppliers, in order to successfully drive a lower price from those other suppliers. If you have supporting evidence of these discussions ... please provide [this] with your response.
    II. If this conduct occurs frequently, please estimate the proportion of your pricing discussions with metal packaging coatings suppliers where Metlac's lower pricing is referenced ... Alternatively, if Metlac's pricing is not used as a negotiating factor with your other suppliers any more frequently than any other metal packaging coatings supplier's prices are used to negotiate a lower price, please indicate this." (emphasis added)
  199. This question led the Commission to state at paragraph 8.191 if the Report that:
  200. "the evidence indicated that some B&B and FCG customers claim to use Metlac's low pricing as a 'stick' to reduce the price offered by their other suppliers, and that they do not use other customers' prices in the same manner or to the same extent".
  201. It was AkzoNobel's submission that this was a hopelessly leading question. In particular, Mr Ward emphasised that the question proceeds from the contested basis that Metlac's pricing is "lower" than other suppliers' and, perhaps more fundamentally, appears to presuppose that only Metlac's prices are used as a 'stick' to drive down those of other suppliers. Whilst Mr Ward did accept that the Commission had asked respondents to provide evidence of their use of Metlac's prices as a 'stick' and that Part II of the question asked respondents to indicate if Metlac's prices were used for that purpose no more often than other suppliers' prices, he argued that this did not come close to ameliorating the leading nature of the question.
  202. Mr Beard explained in argument, however, that this question formed no part of the main questionnaire sent out by the Commission. On the contrary, it was an entirely separate question sent out by email (although we do not regard the medium of its communication as determinative) to follow-up on a point that was raised in the oral hearings conducted by the Commission. During these hearings, a customer had specifically referred to using Metlac's lower pricing as a 'stick' to drive down the prices of other suppliers. Mr Beard also indicated that the statement at paragraph 8.187 of the Report that the Commission "sent a questionnaire to customers" with a view to establishing whether Metlac's pricing was used as a 'stick' in fact contained a typographical error. It should have stated that the Commission sent a "question". Mr Ward did not seek to challenge this clarification.
  203. Whilst it is clearly of the upmost importance that questions posed by the Commission in merger investigations are neutral and do not presuppose any particular answer, as per the OFT and Commission's guidance,[27] we do consider that the context in which this question was asked is highly relevant. We do not doubt that the question could have been phrased better. In our view, however, it was quite proper for the Commission to follow-up on this issue that arose in the oral hearings before it. Had a question phrased in this manner been put in the main questionnaire, we would have had some concerns about the manner in which the Commission conducted its enquiry. As a follow-up question in relation to a matter which had squarely arisen and formed only one strand of the Commission's analysis on pricing, however, it is less objectionable. We also consider it relevant that this was not a question posed to consumers, in the sense of laypeople, but to customers in the coatings industry, who are undoubtedly of some commercial sophistication.
  204. We are not persuaded that the less than perfect formulation of the question at issue undermines the Commission's entire conclusion on pricing, which is the subject of wider discussion and consideration in the Report.
  205. (iii) The alleged conflation of evidence from FCG to B2E
  206. AkzoNobel submitted that the Commission had impermissibly conflated evidence from the FCG market, where it found no SLC, with the B&B market (in particular, the B2E segment). As explained above, the Commission did not maintain the finding made in its Provisional Findings that the Transaction would give rise to an SLC in the FCG market. AkzoNobel argued that, when it came to the Report, the Commission simply used the information which had previously underpinned the FCG SLC to shore up its conclusions on the SLC in relation to the B&B market, which the Commission had defined as a separate market (see Report, paragraph 7.33). AkzoNobel identified this issue in particular in relation to customer views collected by the Commission, including in response to the purportedly leading question discussed above.[28]
  207. It was not disputed that, in order to find that the Transaction would give rise to an SLC on the B&B market, the Commission had to have evidence in relation to pricing in the B2E segment (being the only B&B segment in which Metlac is currently active). As can be seen from the Report, however, the Commission did have regard to evidence which was specific to the B2E market segment[29]. We also attach some significance to the highly concentrated nature of the downstream market for B2E products meaning that there was, necessarily, a limited set of data available to the Commission. That being the case, we consider that the Commission was entitled to take some account of what it found to be Metlac's general business model, namely, pricing aggressively whilst offering high-quality products, using what is perceived to be a low-cost operation based on production efficiencies (see Report, paragraph 9.4). That finding relied on information that related to FCG, as well as B&B. Plainly, whatever the paucity of data, it would not have been open to the Commission to find an SLC on the B&B market by reference only to FCG data. That is not what the Commission did, however. It used the FCG data to corroborate findings made on the basis of the limited B&B data available to it.
  208. We are persuaded that evidence from both the FCG and B&B markets was relevant to understanding Metlac's general business model, and the likelihood that Metlac would continue to pursue a low-price strategy in the B2E market segment (as well as in relation to the BE and B2I segments which it plans to enter). In our view, references to conduct on the FCG market must be read in the context of the rest of the Report, which contained a wide ranging and detailed analysis of competition in the B&B market.
  209. In our judgment, it was neither irrational nor unreasonable for the Commission to take account of evidence from the FCG market, as part of the overall picture of pricing behaviour in the B&B market.
  210. Mr Ward prayed in aid the decision of the EU General Court (formerly the Court of First Instance) in Case T-310/01 Schneider Electric SA v Commission [2002] ECR II-4071, which he submitted stood as authority for the proposition that concerns about competition in a particular market must be based only on evidence from that particular market. We agree with Mr Beard, however, that the General Court's decision was somewhat more nuanced than that. In Schneider, the European Commission was found to have engaged in a discussion of dominance across the EU without taking account of the individual national markets where the dominance findings had to be made. At paragraph 171 of its judgment, the Court held that:
  211. "the creation or strengthening of a dominant position on the national sectoral markets could, in this instance, be apprehended only on the basis of evidence of economic power relating to those markets, possibly supplemented by a consideration of transnational effects ..." (emphasis added).
    Furthermore, we consider that there is a significant distinction between findings of market power (of the kind that were in issue in Schneider), which must be market-specific, and findings as to a firm's likely behaviour (of the kind that are in issue here), where behaviour on one market may be a relevant indicator of the firm's likely behaviour on another market. On that basis, we do not agree that Schneider stands as authority for the proposition that a competition authority can never consider evidence relating to a neighbouring or related market (or indeed segment), albeit that that should not be used as more than a supplement.

    D. The Commission's pricing analysis

  212. AkzoNobel argued that it was particularly important for the Commission to conduct an objective pricing analysis in this case in order to confirm (or otherwise): (i) the results of its survey of customer views; and (ii) its analysis of the BKartA's survey of customer views and pricing data, both of which – as explained above – AkzoNobel considered to be defective. In addition to analysing a subset of the data submitted by customers to the BKartA on the prices of the five most important (in terms of purchased volumes) packaging coatings purchased by customers, the Commission conducted its own price comparison on a sample of products selected by AkzoNobel and Metlac. The Commission conducted this additional analysis following the publication of the Provisional Findings, as a 'cross-check' of its findings.
  213. To do this, the Commission asked each of Metlac and AkzoNobel to select a sample of three coatings in their portfolio. The other party then selected its functionally equivalent products. Valspar and PPG were asked to identify which of their own products they considered to be functionally equivalent to those selected by Metlac and AkzoNobel (see Report, Appendix K, paragraphs 41 and 42). The Commission could then compare the various companies' prices for those products. At the hearing, Mr Ward described this as "[a]bsolutely the right thing to be doing".[30]
  214. PPG and Valspar, however, raised a concern that the descriptions of the coatings selected by AkzoNobel could cover several different products. This meant that PPG and Valspar could not be certain they had identified their functionally equivalent products. Moreover, that fact might actually limit the validity of the price comparison because it would not be comparing like with like (see Report, Appendix K, paragraph 43).
  215. This issue did not arise – or arose only to a lesser degree – in relation to the products selected by Metlac. Nevertheless, the Commission attached limited weight to its price comparison analysis due to concerns that, even 'functionally equivalent' coatings may differ in some technical characteristics (see Report, paragraph 8.157). AkzoNobel submits that, rather than simply attaching limited weight to its pricing data, the Commission should have conducted further inquiries and analyses to ensure that it was comparing like with like. We do not agree. As the Tribunal said in Somerfield (at [176], cited above), the question of precisely where the line is drawn in determining when an inquiry has gone far enough is an issue for the Commission to evaluate. We agree with the view of the Tribunal in Somerfield that it would need a strong case indeed to show that the Commission had manifestly drawn the line in the wrong place. AkzoNobel's case falls short of that standard. The Commission analysed the data arising from its price comparison and took it into account, albeit according it little weight for the reasons it gave. That it decided not to carry out further analyses, with a view to potentially attaching more weight to those findings, is very far from a manifest error.
  216. AkzoNobel further submitted that, in any event, the Commission wrongly interpreted its pricing analysis as supporting its conclusion that Metlac was consistently the lowest priced supplier in the B2E segment. According to AkzoNobel, the pricing analysis in fact shows that Metlac was the cheapest in just one of the three B2E product categories the Commission collected data for, thereby directly contradicting the conclusion the Commission reached for B2E based on its own (in AkzoNobel's view, flawed) survey of customer views.
  217. We consider that this limb of AkzoNobel's challenge proceeds from the same flawed premise as its challenge based on the differences in outcome between the Commission's and the BKartA's inquiries. The Commission did not find that Metlac was always the lowest priced supplier and nor, as we have said, did it need to. Thus, the result in the B2E segment, that Metlac was cheapest for one out of three B2E products for which the Commission had data, was "broadly consistent with pricing evidence received from customers ... [that] [a]lthough not in all cases, Metlac tends to charge lower prices than its major competitors" (see Report, Appendix K, paragraph 51). This becomes clear when one takes into account the data set out in Appendix K, in particular, at paragraphs 47 to 50, and Tables 4 and 5. This shows that, for example, based on average prices in 2011, Metlac was the cheapest for 16 out of 26 products and, where it was the cheapest, it was so by some margin. We accept, of course, that that data does not relate exclusively to the B2E segment but the Commission had limited data to work with and this exercise was carried out as a cross-check to other data.
  218. Had the Commission based the SLC finding solely on this pricing data, the result of this challenge might have been different. In the event, however, the Commission considered a range of evidence regarding the competitiveness of Metlac's prices, most of which lent support to the conclusion that Metlac generally (but not always) offered low prices in the relevant markets. In attaching limited weight to its own pricing comparison, and balancing pricing information relied on by the BKartA with customer views, the Commission appropriately considered and weighed the available evidence. In our view the Commission did not commit a reviewable error and AkzoNobel's argument regarding the Commission's pricing analysis fails.
  219. E. The Commission's investigation of the reasons for any price differences

  220. AkzoNobel submitted that the Commission failed to properly investigate the reasons why Metlac is able to offer lower prices than PPG and Valspar. In particular, AkzoNobel argued that the Commission wrongly failed to enquire as to whether any price differences could be explained by the differing levels of technical assistance provided by the coatings supplier. Mr Ward sought to persuade us that the Commission was under a duty to do so, particularly as one plausible reason for Metlac's lower pricing was that it did not have a UK-based technical assistance team.
  221. The Report makes it clear that the Commission did consider whether pricing differences could be explained by factors such as the performance of a coating, its application and its performance on the production line (see paragraphs 8.145 – 8.146 of the Report). Indeed, the Commission asked customers to comment on any technical differences between the coatings supplied by various suppliers. With regard to service levels, the Commission had well in mind the possible importance of a local assistance team. This is demonstrated by the Commission's consideration of: (a) customer views regarding the need for local assistance at paragraph 7.30 of the Report; and (b) the quality of Metlac's service offering at paragraph 4 of Appendix K and paragraph 8.166 of the Report. The Commission found that Metlac was well-ranked by customers with respect to non-pricing factors, including technical support, and that it had seen no data suggesting that Metlac offers a low-price, low-quality product. Moreover, as the Commission records, whilst some customers indicated that support at their plants was an important factor, there was no consensus that this required the supplier to have local support teams.
  222. We hold that AkzoNobel has failed to make out a case that the Commission acted irrationally in not further investigating the reasons for pricing differences.
  223. F. Conclusion on Ground 2

  224. For the reasons set out above, it is our unanimous view that the Report discloses no judicially reviewable error in relation to the Commission's conclusion that Metlac competes more aggressively on price than Valspar and PPG, and AkzoNobel's challenge fails on this ground too.
  225. VI. GROUND 3: A LOSS OF COMPETITION IN INNOVATION

  226. AkzoNobel's third ground of review was that the Commission erred in finding that the Transaction would lead to a loss of competition in innovation when there was no evidence to support that conclusion. It submitted that the analysis in the Commission's Provisional Findings, which had underpinned that conclusion, was dropped from the Report. The Commission, therefore, made findings in the Report that were not supported by the evidence and failed to carry out sufficient enquiries.
  227. Strictly speaking, it may be unnecessary to resolve this ground since, as we read the Report, and taking it in its totality as we must, Mr Beard was in our view correct that the conclusion on innovation was "not crucial overall to the SLC finding. It was merely part of the supporting assessment ..."[31] As such, AkzoNobel having failed on its challenge to the Commission's pricing conclusion, it appears to us that, even if AkzoNobel now succeeds in relation to innovation that would not lead to the SLC finding being quashed. However, as the parties themselves dedicated very little time to this challenge (perhaps confirming our view that this ground is ancillary at best), we can also deal with it briefly and propose to do so for the sake of completeness.
  228. The essence of the Commission's conclusion in relation to innovation was that it would expect:
  229. "a weakening of rivalry in innovation, particularly when AkzoNobel and Metlac are head-to-head in the race to develop new formulations or minor changes to existing products (and this is also relevant to our views in relation to potential competition in B&B) (see Appendix G for more detail on innovation)." (see Report, paragraph 9.57)
  230. Mr Ward emphasised that, due to the change between the Commission's Provisional Findings and the Report, the evidential foundation for this conclusion had been stripped away. In its Provisional Findings, the Commission stated that the Transaction might lead to a severe weakening, or even removal, of Metlac's innovative and dynamic qualities (see Provisional Findings, paragraph 9.11). That finding was not maintained in the Report, however. In its Defence, the Commission submitted that, whilst it did not find that Metlac was a "particularly innovative supplier which would be 'lost' through the merger, it does not follow that the merger would have no impact at all on levels of innovation."[32]
  231. Mr Ward accepted that that finding was not per se irrational but argued that there had to be some evidential underpinning for it.[33] The essence of his submissions was that the Commission had simply not demonstrated that (i) post-Transaction, the merged group would compete less strongly with its pooled research and development divisions, and (ii) the effect of the Transaction would be to reduce the incentive to innovate. If the Commission were entitled to simply rely on it being a self-proving hypothesis that a four-to-three merger will be bad for innovation, then every such merger would founder on this ground.[34]
  232. If that were indeed what the Commission had found, we would be inclined to share Mr Ward's concern. Mr Beard rightly drew attention, however, to the particular current market context in which the Commission had made its assessment, something which, in our view, AkzoNobel's submissions failed to take proper account of.
  233. In the Report, the Commission notes that, although the basic technologies used in the production of coatings have existed for years, there is continual investment by the coatings suppliers in developing new products (see Report, paragraph 2.73). At present, however, the coatings industry is "on the cusp of the most significant change in decades" (Report, paragraph 8.204). It is expected that France, and in due course other countries, will ban the inclusion of a substance called Bisphenol-A ("BPA")[35] in coatings due to concerns relating to human health (see Report, paragraph 2.77). Since it is, apparently, almost impossible to ensure that coatings are entirely BPA-free, it seems that going forward the market will look to develop coatings that do not intentionally contain BPA ("BPA-NI").[36] As the import of cans (intentionally) containing BPA into France will be banned, this is expected to have pan-European implications for the coatings industry (see Report, paragraph 2.78).
  234. In a question that did not specifically refer to innovation, but was more general in nature, three customers ([…][(] (see Report, Appendix G, paragraphs 4 to 6)) specifically informed the Commission that they regarded Metlac highly for innovation, whilst two further customers highlighted superior product quality (see Report, Appendix G, paragraph 19 and footnote 11). Against a background in which several customers had already identified Metlac as an innovator, the impending need to produce BPA-NI products, a process taking place over years, appears to take on a renewed significance.
  235. The Commission records that PPG, Valspar, AkzoNobel and Metlac are all developing BPA-NI solutions for internal and external coatings (see Report, Appendix G, paragraph 23). Whilst recognising that the information available to it was relatively limited, the Commission noted that a number of customers were currently testing, and two ([…][(]) considering purchasing, Metlac's BPA-NI products in the B&B segment (see Report, Appendix G, paragraph 29). The Commission noted that the:
  236. "evidence provided to us indicated that Metlac, along with AkzoNobel, Valspar and PPG, was developing BPA-NI B&B coatings and we are of the view that a move to BPA-NI B2E coatings would, if anything, enhance Metlac's ability to further expand its B2E supplies, given its strong position in development of BPA-NI coatings." (see Report, Appendix G, paragraph 29)
  237. Mr Beard submitted that, with the above circumstances in mind, it is clear that there is a race to innovate, which sees different suppliers working with different customers and, as such, "it is fanciful ... for AkzoNobel to turn round and say, 'Well, actually, taking Metlac out of this wouldn't affect this race.'"[37] We agree that it is not possible to say, as AkzoNobel seeks to do, that the Commission's conclusion that the Transaction would likely result in a reduction of competition in innovation was without any evidential basis. Clearly, the Commission had considered some evidence on this point. The assessment of that evidence was a matter principally for the Commission and involved at least an element of economic prediction (especially in relation to BPA-NI). On that basis, it seems plain to us that the Commission was entitled to reach the conclusion it did.
  238. Finally, we consider that Mr Siragusa, appearing for the Interveners, was also correct to point out[38] that, while the Commission focused on innovation in relation to BPA-NI, it also found that Metlac was planning to enter the market for BE and B2I (see Report, paragraph 9.117). We consider it to be implicit in that finding that, as Mr Siragusa submitted, Metlac would need to innovate on those segments in order to attract customers since price is not the only factor taken into account by customers. Since AkzoNobel is already active in both BE and B2I, it may be that the Transaction would lead to a reduction in competition in innovation here also.
  239. In our judgment, therefore, the Commission had a sufficient evidential basis upon which to conclude that the Transaction might lead to:
  240. "... a weakening of rivalry in innovation, particularly when AkzoNobel and Metlac are head-to-head in the race to develop new formulations or minor changes to existing products (and this is also relevant to our views in relation to potential competition in B&B) ..." (see Report, paragraph 9.57; emphasis added)
  241. As such, it is our unanimous judgment that AkzoNobel's third ground of review fails also.
  242. VII. OVERALL CONCLUSION

  243. For the foregoing reasons, we unanimously conclude that AkzoNobel's application for review fails in its entirety.
  244. In its Notice of Application, AkzoNobel submitted that the review of the Report merited a high degree of urgency and we have, therefore, moved with all due expedition in considering the case. We are also aware that there are other proceedings afoot, in Italy at the least, in relation to the Transaction, which will be to some extent affected by the decision we have come to. For that reason, we consider it appropriate to abridge the time for filing any applications for permission to appeal pursuant to Rules 19(2)(i) and 58 of the Tribunal's Rules. Any application for permission to appeal this decision must be filed and served by not later than 4pm on the day falling three weeks after the date on which this judgment is handed down at a public hearing.








  245. The Hon. Mr Justice Norris








    William Allan








    Prof. Gavin Reid
         




    Charles Dhanowa O.B.E., Q.C. (Hon)
    Registrar
     





    Date: 21 June 2013

    ANNEX: GLOSSARY OF DEFINED TERMS

    Defined term Meaning
    Act Enterprise Act 2002
    AkzoNobel Akzo Nobel N.V. (incorporated in the Netherlands)
    AN Group The group of companies of which AkzoNobel is the ultimate holding company
    ANCI Akzo Nobel Coatings International B.V., a wholly-owned subsidiary of AkzoNobel (incorporated in the Netherlands)
    ANPG SBU AkzoNobel Packaging Coatings SBU
    Ardagh The Ardagh Group
    B&B Beer and beverages coatings
    B2E Beverage externals coatings
    B2I Beverage internals coatings
    BA Business Area
    BA Responsibles The four members of the Board of Management together with the four executives who have responsibilities for BAs, and who comprise the ExCo
    Ball Ball Corporation
    BE Beverage ends coatings
    BKartA Bundeskartellamt (the German Federal Cartel Office)
    BPA Bisphenol-A
    BPA-NI Bisphenol-A non-intent (i.e. coatings which do not intentionally contain BPA)
    BU Business Unit
    C&C Caps and closures coatings
    Can-Pack Can Pack S.A.
    coatings Metal packaging coatings
    Commission Competition Commission
    Crown Crown Holdings Inc
    EEA European Economic Area
    ExCo Executive Committee (of AkzoNobel)
    FCG Collective term for Food, C&C and GL coatings
    Food Food coatings
    FTA Fair Trading Act 1973
    GL General line coatings
    ICI Imperial Chemical Industries
    Merger Regulation Council Regulation 139/2004/EC on the control of concentrations between undertakings OJ L 24, 29.1.2004, p. 1
    Metlac Metlac S.p.A. (incorporated in Italy)
    Metlac Holding Metlac Holding S.r.L. (incorporated in Italy)
    Mortar Mortar Investments International Ltd, a wholly-owned subsidiary of AkzoNobel (incorporated in the UK)
    OFT Office of Fair Trading
    PPG PPG Industries, Inc.
    Report The Commission's report published on 21 December 2012 and entitled "A report on the anticipated acquisition by Akzo Nobel N.V. of Metlac Holding S.r.l."
    Rexam Rexam plc
    RMS Relevant merger situation, as defined in section 23 of the Act
    RTPA RTPA 1956 and 1976 together
    RTPA 1956 Restrictive Trade Practices Act 1956
    RTPA 1976 Restrictive Trade Practices Act 1976
    SBU Sub Unit
    SLC Substantial lessening of competition
    Transaction The proposed acquisition by AkzoNobel of the outstanding 51% of shares in Metlac Holding pursuant to an option held by ANCI
    Valspar The Valspar Corporation

Note 1    OJ L 24, 29.1.2004, p. 1.    [Back]

Note 2    See Articles 4(5) and 22 of the Merger Regulation.    [Back]

Note 3    The Commission explained that by prohibition it meant that “AkzoNobel (and any of its subsidiaries) would be prohibited from acquiring any additional shares, via exercise of the call option or any other means, in either Metlac Holding or Metlac, or any of their subsidiaries.” (Report, paragraph 11.9)    [Back]

Note 4    Lodged at the Tribunal on 17 January 2013, amended on 11 February and re-amended 18 April 2013, on each occasion with the permission of the Tribunal.     [Back]

Note 5    Transcript, Day 1, p. 14, lines 14 and 15.    [Back]

Note 6    Transcript, Day 2, p. 54, lines 21-24.    [Back]

Note 7    Transcript, Day 2, p. 7, lines 32-34.    [Back]

Note 8    Transcript, Day 2, p. 3, line 14.    [Back]

Note 9    Transcript, Day 1, p. 65, line 13.    [Back]

Note 10    Transcript, Day 1, p. 56, lines 29-32.    [Back]

Note 11    See section 10(4) of the Monopolies and Restrictive Practices (Inquiry and Control) Act 1948, which was applied by reference to mergers when they were brought within the legislative scheme (see section 6(5) of the Monopolies and Mergers Act 1965), and the subsequent restatement in section 90(3) of the FTA.    [Back]

Note 12    Transcript, Day 1, p. 61, lines 16-27.    [Back]

Note 13    In other words, a company incorporated under the laws of a country other than that of the court asserting jurisdiction.    [Back]

Note 14    The alternative exception recognised inAdams, arising from contract, is immaterial in the present context.    [Back]

Note 15    See the judgments of Lady Hale (with whom Lord Wilson agreed) at paragraph 92, Lord Mance at paragraphs 100 and 102, Lord Clarke at paragraph 103, and Lord Walker at paragraph 106.    [Back]

Note 16    The requirement for impropriety was stated by Munby J in Ben Hashem v Al Shayif [2008] EWHC 2380 (Fam) at paragraphs 163 and 164 (in remarks endorsed by Lord Neuberger in VTB Capital at para 128) that “it is necessary to show both control of the company by the wrongdoer(s) and impropriety, that is, (mis)use of the company by them as a device or façade to conceal their wrongdoing … at the time of the relevant transaction(s)”. Although Lord Sumption’s judgment in Prest expressed the test in different terms (see paragraph 95 above), we do not see any substantive difference that is material for present purposes.    [Back]

Note 17    Transcript, Day 2, p.56, lines 3-17.    [Back]

Note 18    Transcript, Day 2, p. 7, lines 32-34; see also Transcript, Day 2, p. 9, lines 13-15.    [Back]

Note 19    Decision of 24 April 2012 in Case B 3 – 187/11.    [Back]

Note 20    See paragraph 8.130 and footnote 132 of the Report, which explain that the Commission and the OFT were provided with copies of responses to the BKartA’s survey by some respondents, although others declined to provide their responses.     [Back]

Note 21    On 18 October 2012, the German Federal Parliament adopted a new merger control law that, among other things, replaces the dominance test applied to the Transaction with the EU ‘significant impediment to competition’ test applied under the Merger Regulation.    [Back]

Note 22    Unofficial translation of the BKartA’s decision provided by AkzoNobel.    [Back]

Note 23    See also, paragraph 8.144 and Appendix K, paragraph 39. The Commission also accepted that, when smaller suppliers were present, Metlac was the lowest-priced supplier less frequently than when it was in competition with only AkzoNobel, PPG and Valspar (Appendix K, paragraph 34).    [Back]

Note 24    The Commission addressed a full range of issues during these hearings, including customer views on pricing. We refer, for example, to the quote from the hearing with […][(] in the Report, Appendix K, paragraph 40.    [Back]

Note 25    Para 3.13 of which states: “Thought should be given as to whether the appropriate sample to provide views on a merger is all potential consumers of a product, the customers of all of the firms believed to be in the market, or only the customers of one or both of the merging parties. Screening should be documented explicitly, and the numbers of people 'screened out' at each stage recorded and reported.”    [Back]

Note 26    The Commission also notes that it received a number of responses from customers of PPG and Valspar.    [Back]

Note 27    See, in particular, paragraph 3.37 of the OFT and Commission’s joint publication: ‘Good practice in the design and presentation of consumer survey evidence in merger inquiries’.    [Back]

Note 28    Although this argument was presaged to some extent in the Amended Notice of Application, it only took on any real prominence in AkzoNobel’s Skeleton. In its Skeleton, the Commission responded to the new points raised but argued they were not open to AkzoNobel as they had not been pleaded. AkzoNobel appeared to accept that its Skeleton raised a new point and, therefore, provided a draft Re-Amended Notice of Application, with an application for permission to amend, shortly before the hearing. Whilst we consider amendments at such a late stage to be generally undesirable, and the Court of Appeal has recently reminded us in the context of the Civil Procedure Rules that late amendments must be justified on a much sounder basis than has hitherto been the case, it did seem to us in this case that the amendment added relatively little and that the matters raised had already been fully addressed by the Commission in its Skeleton. We therefore granted AkzoNobel permission to re-amend (see Transcript, Day 1, p. 2, lines 15-19).     [Back]

Note 29    See the Commission’s breakdown of the BKartA’s pricing data at paragraph 8.141 and in the particular emphasis on the views of customers in the B2E market segment, set out at paragraph 9.58. We note also the discussion of customer views at paragraphs 8 and 9 of Appendix K to the Report.     [Back]

Note 30    Transcript, Day 1, in camera session, p. 7, line 30.    [Back]

Note 31    Transcript, Day 2, p. 40, lines 20 and 21.    [Back]

Note 32    Defence, paragraph 193.    [Back]

Note 33    Transcript, Day 1, p. 49, line 15.    [Back]

Note 34    Transcript, Day 1, p. 49, lines 19-21.    [Back]

Note 35    The glossary to the Report has this to say on BPA: “BPA is a chemical agent which is present in epoxy resin used in many metallized coatings. BPA is produced by condensing two parts phenol with one part acetone. BPA is a weak hormone (oestrogen) and environmental endocrine disruptor.    [Back]

Note 36    The entry in the Report’s glossary for BPA-NI states: “A term used to describe coatings where no BPA has intentionally been included in the ingredients used to make the coating. BPA-NI rather than BPA free is the term generally used in the industry as other substances come into contact with the coating which could contaminate it, making it impossible for coatings manufacturers to guarantee that the coating is BPA free.” (emphasis in original)    [Back]

Note 37    Transcript, Day 2, p. 40, lines 8-11.    [Back]

Note 38    Transcript, Day 2, p. 50, lines 14-24.    [Back]


BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII
URL: http://www.bailii.org/uk/cases/CAT/2013/13.html