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

United Kingdom Competition Appeals Tribunal


You are here: BAILII >> Databases >> United Kingdom Competition Appeals Tribunal >> Skyscanner Ltd v CAT [2014] CAT 16 (26 September 2014)
URL: http://www.bailii.org/uk/cases/CAT/2014/16.html
Cite as: [2014] CAT 16

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


Neutral Citation [2014] CAT 16
Case No.: 1226/2/12/14

IN THE COMPETITION
APPEAL TRIBUNAL

Victoria House
Bloomsbury Place
London WC1A 2EB
26 September 2014

B e f o r e :

PETER FREEMAN CBE QC (HON)
(Chairman)
BRIAN LANDERS
STEPHEN WILKS
Sitting as a Tribunal in England and Wales

____________________

SKYSCANNER LIMITED Appellant
- supported by -
SKOOSH INTERNATIONAL LTD Intervener
- v -
COMPETITION AND MARKETS AUTHORITY Respondent
- supported by -
BOOKING.COM B.V.
EXPEDIA, INC.
INTERCONTINENTAL HOTELS GROUP PLC



Interveners

____________________

APPEARANCES
Ms Kassie Smith QC (instructed by Maclay Murray & Spens LLP) appeared for the Appellant.
Ms Kelyn Bacon QC and Mr David Bailey (instructed by the General Counsel, Competition and Markets Authority) appeared for the Respondent.
Mr Duncan Sinclair and Mr Samar Abbas (instructed by Shoosmiths LLP) appeared for Skoosh International Ltd.
Mr Tim Ward QC and Ms Jessica Boyd (instructed by Freshfields Bruckhaus Deringer LLP) appeared for InterContinental Hotels Group Plc.
Mr Josh Holmes (instructed by King & Wood Mallesons LLP) appeared for Expedia, Inc.
Mr Alistair Lindsay (instructed by Slaughter and May) appeared for Booking.com B.V.
Heard at Victoria House on 28 - 29 July 2014

____________________

HTML VERSION OF JUDGMENT
____________________

Crown Copyright ©

    CONTENTS

    I. INTRODUCTION 10
    II. THE HOTEL ONLINE BOOKING INVESTIGATION AND THE COMMITMENTS 10
    A. The Investigation 10
    B. The OFT's Competition Concerns 12
    C. The Commitments 19
        (i) Statutory Framework 19
        (ii) The Final Commitments 23
    III. PRINCIPLES APPLICABLE ON AN APPLICATION FOR JUDICIAL REVIEW 29
    IV. DID THE OFT FAIL TO TAKE INTO ACCOUNT RELEVANT CONSIDERATIONS? (GROUND 2) 44
    A. The Law: the OFT's Duty to Consult and to Consider 45
    B. The OFT's Initial Investigation 51
    C. The OFT's Consultation 53
        (i) The First Consultation 54
        (ii) The Second Consultation 59
    D. The Parties' Submissions 67
    E. Analysis 74
        (i) Price Transparency 75
        (ii) Inter-brand Competition 81
        (iii) Responses to the Consultations 83
        (iv) The OFT's Response 84
        (v) The Request for Supporting Evidence 86
        (vi) Our View 90
        (vii) The OFT's Discretion 94
        (viii) The OFT's Approach 97
        (ix) Conclusion 100
    V. DID THE OFT ACT ULTRA VIRES BY ACCEPTING COMMITMENTS WHICH POTENTIALLY HARM COMPETITION? (GROUND 3) 101
    A. The Law 104
        (i) Illegality 104
        (ii) Irrationality 106
        (iii) Relationship between Legality and Rationality 108
    B. The Parties' Submissions 110
    C. Analysis 119
        (i) The Law 119
        (ii) The Illegality Argument 121
        (iii) The Irrationality Argument 155
        (iv) Other Arguments 160
    VI. DID THE OFT ACT ULTRA VIRES BY IMPOSING REQUIREMENTS ON THIRD PARTIES? (GROUND 1) 163
    A. The Parties' Submissions 165
    B. Analysis 168
        (i) Other Cases 169
        (ii) Conclusion 172
        (iii) General Comment 173
    VII. CONCLUSION 174

    I. INTRODUCTION

  1. This is an appeal under section 47(1)(c) of the Competition Act 1998 (the "Act") against a decision of the Office of Fair Trading (the "OFT") dated 31 January 2014 to accept commitments, pursuant to section 31A(2) of the Act, to remove certain discounting restrictions for online travel agents (the "Decision"). It is the first time the Tribunal has been called upon to consider a commitments decision taken under section 31A of the Act. The appeal is brought by Skyscanner Limited ("Skyscanner"), which operates a price comparison website allowing consumers to search for and compare flight, hotel and car hire deals globally.
  2. By an Order made on 1 May 2014, Skoosh International Ltd ("Skoosh") was granted permission to intervene in support of Skyscanner; Booking.com B.V ("Booking"), Expedia, Inc. ("Expedia") and InterContinental Hotels Group Plc ("IHG") were granted permission to intervene in support of the Competition and Markets Authority (the "CMA").
  3. The Decision followed the opening of an investigation into the online supply of room-only hotel accommodation bookings by online travel agents ("OTAs") and the issuance of a Statement of Objections. By its Decision, the OFT accepted commitments (the "Commitments") from Booking and Expedia, both OTAs, and from IHG, a hotel group, together with Booking's ultimate parent company, priceline.com, and Hotel Inter-Continental London Limited (the "Commitment Parties").
  4. Skyscanner operates a "price comparison" or "meta-search" site (we were told the terms are inter-changeable). Meta-search sites display prices offered by third parties, and thereby assist consumers to compare pricing. After searching for a hotel room on Skyscanner's site, for example, consumers are directed to third party websites for the booking to take place. Skyscanner contracts with hotels and OTAs for the inclusion of their offerings in Skyscanner's meta-search results.
  5. Although Skyscanner had operated in the travel business for 10 years, it did not enter the hotel meta-search business until September 2013 when it acquired a hotel meta-search company.
  6. Skyscanner appeals against the Decision on three grounds. In summary, they are as follows:
  7. (a) In making the Decision, the OFT failed to take into account properly or at all the representations that Skyscanner made to it on the impact the Decision would have on the meta-search sector and/or on inter-brand competition (Ground 2).

    (b) By putting in place the Commitments without considering the potential anti-competitive consequences that they may have, the OFT acted contrary to the policy and objects of the Act (Ground 3).

    (c) The Decision was ultra vires because the Commitments had the effect of requiring third parties to act in line with them, even though those third parties had not offered commitments and the OFT had not accepted commitments from those third parties (Ground 1).

  8. Skyscanner therefore contends that the OFT acted unlawfully in making the Decision and that it should be quashed.
  9. The Decision was taken by the OFT, but this organisation ceased to exist on 1 April 2014. Accordingly, its successor - the CMA - is the appropriate respondent in these proceedings.
  10. For the reasons that follow, we find that Skyscanner's appeal succeeds and that the Decision cannot stand.
  11. II. THE HOTEL ONLINE BOOKING INVESTIGATION AND THE COMMITMENTS

    A. The Investigation

  12. In September 2010, the OFT opened an investigation into the online supply of room-only hotel accommodation bookings by online travel agents, following a complaint by Skoosh, a small OTA. In July 2012, the OFT issued a Statement of Objections to the Commitment Parties, which alleged that they had infringed Chapter I of the Act and Article 101 of the Treaty on the Functioning of the European Union ("TFEU"). The Statement of Objections alleged that Booking and Expedia each entered into separate arrangements with IHG which restricted each OTA's ability to discount the rate at which room-only hotel accommodation bookings were offered to consumers. Room-only rates can be distinguished from package rates (where a hotel room booking is supplied together with another product, such as car hire) and "opaque" bookings (where the hotel is not revealed until the booking is complete), neither of which were subject to the same discounting restrictions.
  13. The Commitment Parties did not formally respond to the Statement of Objections, although they did make submissions on the OFT's competition concerns for the purposes of offering commitments, including in respect of potential efficiencies associated with the restrictions on discounting. However, in order to address the competition concerns set out in the Statement of Objections, the Commitment Parties offered commitments that they would modify their behaviour in accordance with certain principles. In essence, these principles would allow OTAs to offer discounts from headline room-only rates to members of closed groups and to advertise the availability (but not the level) of such discounts to all consumers. The OFT consulted on these initial commitments in August 2013. The Commitment Parties subsequently offered revised commitments and the OFT accordingly conducted a second public consultation in December 2013. Skyscanner responded to this second consultation. Further details of the consultation process are set out in the context of Ground 2 below. By its Decision, the OFT adopted the final Commitments.
  14. B. The OFT's Competition Concerns

  15. In the Statement of Objections, the OFT alleged that Booking and Expedia had each entered into agreements with IHG and the Intercontinental London-Park Lane hotel ("ILPL") (the "Price Agreements") under which they agreed to offer hotel accommodation bookings at ILPL at a day-to-day room rate set and/or communicated by ILPL and not to offer room bookings at a lower rate, for instance by funding a promotion or discount from their own margin or commission. The OFT's provisional view was that these arrangements had the object of preventing, restricting or distorting competition and, therefore, they each breached the Chapter I prohibition and Article 101 of the TFEU.
  16. In addition to the discounting restrictions, the OFT identified rate parity obligations in the Price Agreements. Rate parity obligations are also known as "most favoured nation" (or "MFN") clauses. They ensure that the retail rates for hotel room bookings provided by hotels to OTAs are no less favourable than the lowest retail rate displayed by other online distribution outlets. This means that the OTA cannot be undercut.
  17. The OFT set out three theories of harm regarding the Price Agreements in the Statement of Objections:
  18. (a) restrictions on discounting limit price competition and increase barriers to entry;

    (b) rate parity obligations are capable of reinforcing and exacerbating any prevention, restriction or distortion of competition arising from discounting restrictions; and

    (c) to the extent that similar discounting restrictions and rate parity obligations are replicated in the market, then any prevention, restriction or distortion of competition is further reinforced and exacerbated.

  19. These theories of harm were presented somewhat differently in the Decision. In particular, the Decision did not identify rate parity obligations as a distinct competition concern and identified no separate harm to competition arising from them. In the Statement of Objections, the OFT stated that the existence of rate parity obligations was capable of reinforcing and exacerbating any prevention, restriction or distortion of competition arising from discounting restrictions, but that the OFT had not investigated the extent to which rate parity obligations were capable of doing so, making it clear that the OFT had made no findings in this respect. However, in the Decision the OFT stated that the focus of its investigation had been on discounting and that it had made no assessment of whether rate parity provisions may give rise to a breach of the Chapter I prohibition and/or Article 101 TFEU.[1] Skoosh placed some significance on the apparent difference in approach between the Statement of Objections and the Decision.
  20. As to the first theory of harm set out in the Statement of Objections, the OFT explained in the Decision that the restrictions on discounting a hotel's room-only rate had the following effects:
  21. (a) there was likely to be limited, if any, competition on the offer of room rates to consumers between OTAs and between OTAs and the hotel's direct online sales channel for those hotel accommodation bookings (that is, intra-brand competition); and

    (b) they may create barriers to entry to the extent that they prevented new OTAs from entering the market, and/or achieving sufficient scale (with discounted rates for room-only hotel accommodation).[2]

  22. As to the third theory of harm set out in the Statement of Objections, the OFT said in the Decision that it understood similar discounting restrictions to be potentially widespread in vertical distribution arrangements in the industry.[3] To the extent that similar discounting restrictions were replicated, then any prevention, restriction or distortion of competition arising would be wider reaching.[4] However, the OFT did not investigate the extent to which similar discounting restrictions were replicated in the market and made no findings in that respect.[5]
  23. The non-confidential version of the Statement of Objections was disclosed to Skyscanner with a copy provided to the Tribunal on 24 July 2014, shortly before the hearing in these proceedings. This followed a disclosure application made by Skyscanner and our Ruling granting the application ([2014] CAT 12). Prior to that date, the CMA, the intervening Commitment Parties and Skoosh had all seen the document. Unsurprisingly, the intervening Commitment Parties have been keen to point out that the Statement of Objections is strongly disputed. While we appreciate that the contents of the Statement of Objections were provisional and did not represent an established position, we have found the document necessary for our understanding of the nature of the OFT's competition concerns as at 31 July 2012.
  24. C. The Commitments

    (i) Statutory Framework

  25. Section 31A of the Act provides that, for the purposes of addressing the competition concerns it has identified, the CMA (previously OFT) may accept, from such person or persons concerned as it considers appropriate, commitments to take such action (or refrain from such action) as it considers appropriate:
  26. "31A Commitments
    (1) Subsection (2) applies in a case where the CMA has begun an investigation under section 25 but has not made a decision (within the meaning given by section 31(2)).
    (2) For the purposes of addressing the competition concerns it has identified, the CMA may accept from such person (or persons) concerned as it considers appropriate commitments to take such action (or refrain from taking such action) as it considers appropriate.
    (3) At any time when commitments are in force the CMA may accept from the person (or persons) who gave the commitments-
    (a) a variation of them if it is satisfied that the commitments as varied will address its current competition concerns;
    (b) commitments in substitution for them if it is satisfied that the new commitments will address its current competition concerns.
    (4) Commitments under this section-
    (a) shall come into force when accepted; and
    (b) may be released by the CMA where-
    (i) it is requested to do so by the person (or persons) who gave the commitments; or
    (ii) it has reasonable grounds for believing that the competition concerns referred to in subsection (2) or (3) no longer arise.
    (5) The provisions of Schedule 6A to this Act shall have effect with respect to procedural requirements for the acceptance, variation and release of commitments under this section."
  27. Section 31B of the Act provides that the effect of commitments under Section 31A is as follows:
  28. "31B Effect of commitments under section 31A
    (1) Subsection (2) applies if the CMA has accepted commitments under section 31A (and has not released them).
    (2) In such a case, the CMA shall not-
    (a) continue the investigation,
    (b) make a decision (within the meaning of section 31(2)), or
    (c) give a direction under section 35,
    in relation to the agreement or conduct which was the subject of the investigation (but this subsection is subject to subsections (3) and (4)).
    (3) Nothing in subsection (2) prevents the CMA from taking any action in relation to competition concerns which are not addressed by commitments accepted by it.
    (4) Subsection (2) also does not prevent the CMA from continuing the investigation, making a decision, or giving a direction where-
    (a) it has reasonable grounds for believing that there has been a material change of circumstances since the commitments were accepted;
    (b) it has reasonable grounds for suspecting that a person has failed to adhere to one or more of the terms of the commitments; or
    (c) it has reasonable grounds for suspecting that information which led it to accept the commitments was incomplete, false or misleading in a material particular.
    (5) If, pursuant to subsection (4), the CMA makes a decision or gives a direction the commitments are to be treated as released from the date of that decision or direction."
  29. The CMA's power to accept binding commitments is intended to allow it to resolve cases more quickly and efficiently by avoiding the need for a full investigation, thereby enabling the CMA to use its limited resources for a broader range of enforcement purposes. There are also obvious benefits for the parties themselves, most notably avoiding an infringement decision against them.
  30. The OFT issued Guidance on its use of the power to accept binding commitments.[6] This Guidance states that the OFT (now CMA) is likely to consider it appropriate to accept binding commitments only in cases where:
  31. (a) the competition concerns are readily identifiable;

    (b) the competition concerns are fully addressed by the commitments offered; and

    (c) the proposed commitments are capable of being implemented effectively and, if necessary, within a short period of time.[7]

    (ii) The Final Commitments

  32. The OFT decided that this was an appropriate case for it to accept commitments under section 31A of the Act. In particular, the OFT decided that the Commitments addressed its competition concerns by allowing for greater discounting freedom, albeit with some "residual restrictions" in respect of discounting.[8]
  33. By the Commitments, the Commitment Parties agreed to modify their behaviour in accordance with certain principles. The main element of the Commitments is the removal of the complete prohibition on discounting room-only rates by OTAs and its replacement by limited discounting to closed groups of consumers.
  34. The Commitments apply to the arrangements between the Commitment Parties, and (in the case of IHG) other OTAs or (in the case of Booking and Expedia) other hotels. The relevant principles the Commitment Parties have signed up to can be summarised as follows:
  35. (a) OTAs and hotels may offer discounts off the headline room rates in UK hotels to any EEA resident who has joined a "closed group" and made a previous booking which has become non-refundable with that OTA or hotel at the headline rate. OTAs may discount up to the level of their commission or margin only. (Principle 18)

    (b) OTAs cannot publicise information about the specific level or extent of discounts outside the closed group. However, they may publicise information regarding the availability of discounts on their own websites, price comparison websites and meta-search sites, for example. (Principle 19)

    A closed group is essentially a membership group which consumers actively choose to join. While consumers can see the discounted rates once they have joined the closed group, they cannot take advantage of those rates until they have made one full price booking.

  36. The full text of the Commitments is available online at Annexe 1 of the Decision.[9] For our purposes, it is sufficient to set out the hotel online booking principles (the "Principles"):
  37. "18. OTAs shall be free to offer Reductions in respect of Hotel Rooms at Hotel Properties located in the UK that are:
    (a) available to and redeemable by Closed Group Members who have made at least one Prior Booking with that OTA;
    (b) no greater than the level of commission earned by that OTA for the relevant Hotel Property by reference (at the choice of the OTA) to:
    (i) the level of commission for the particular transaction in respect of which a Reduction is being offered; or
    (ii) the aggregate commission earned for the relevant Hotel Property over the course of a time period determined by the OTA but not exceeding one year, starting from the Effective Date or such later date as the OTA chooses; and
    (c) available to EEA Residents in respect of Hotel Rooms at Hotel Properties located in the UK.
    19. OTAs may publicise information regarding the availability of Reductions in a clear and transparent manner, including to price comparison websites and meta-search sites, subject to the following:
    (a) OTAs cannot publicise Specific Information about Reductions for any IHG Room to consumers who are not Closed Group Members, including on OTAs' own public websites and via price comparison websites and meta-search sites; and
    (b) any Other Hotel may prevent OTAs from publicising Specific Information about Reductions to consumers who are not Closed Group Members, including on OTAs' own public websites and via price comparison websites and meta-search sites.
    20. IHG and/or any Other Hotel contracting with an OTA is entitled to require from that OTA such information as may reasonably be required to enable IHG or the Other Hotel to assess and verify compliance with paragraphs 18(a), 18(b) and 18(c). However, IHG and/or any Other Hotel may not impose any method of accounting on any OTA which may restrict, limit or impede the OTA from operating on the basis of arrangements which are consistent with the Principles.
    21. OTAs shall not enter into or enforce any most favoured nation or equivalent provision as regards Reductions offered by Hotels to their respective Closed Group Members who have made at least one Prior Booking directly with that Hotel provided that:
    (a) such Reductions are only available to EEA Residents in respect of Hotel Rooms at Hotel Properties located in the UK; and
    (b) the Hotel does not publicise Specific Information about Reductions to consumers who are not Closed Group Members, including on the Hotel's own public website(s) and via price comparison websites and meta-search sites.
    22. For the avoidance of doubt, the commitments do not in any way restrict:
    (a) the ability of Hotels to set the Headline Room Rates for their respective Hotel Rooms; or
    (b) benefits available to members of OTAs' and Hotels' existing loyalty schemes prior to the Effective Date."
  38. The Commitments will remain in force for two years (the duration was reduced from a proposed three years following the second consultation), i.e. until 31 January 2016, whereupon they will lapse unless renewed.
  39. The CMA emphasised that the Commitments are a set of "minimum" obligations as, while the Commitment Parties are bound to permit discounting at least on the terms set out in the Commitments, they remain entitled to include provisions permitting discounting on more generous terms. Hotels therefore remain free to allow OTAs to discount more deeply.[10] Skyscanner argued, however, that such arrangements were unrealistic. While there was some mention of possible deeper discounting by one of the Commitment Parties at the hearing, we saw no evidence that this practice was likely to occur on any significant scale and do not consider this point further.
  40. III. PRINCIPLES APPLICABLE ON AN APPLICATION FOR JUDICIAL REVIEW

  41. Section 47 of the Act provides for third party appeals. Subsection (1)(c) provides that a person who does not fall within section 46(1) or (2) may appeal to the Tribunal with respect to a "decision of the CMA to accept or release commitments under section 31A, or to accept a variation of such commitments other than a variation which is not material in any respect." Subsection (2) specifies that a person may appeal under subsection (1) only if the Tribunal considers that he has a "sufficient interest" in the decision with respect to which the appeal is made, or that he represents persons who have such an interest. It is not disputed that Skyscanner has a sufficient interest to bring this appeal.
  42. The Tribunal's powers on an appeal brought under section 47(1)(c) are set out in paragraph 3A of Schedule 8 of the Act. Pursuant to paragraph 3A(2), the Tribunal must, by reference to the grounds of appeal set out in the notice of appeal, determine the appeal by applying the "same principles as would be applied by a court on an application for judicial review".
  43. The judicial review principles to be applied by the Tribunal are well-established. While the parties were largely agreed on the principles themselves, there were some differences of emphasis.
  44. In Merger Action Group v Secretary of State [2008] CAT 36, at [59], the Tribunal stated as follows:
  45. "The grounds on which an administrative act or decision can be called into question by judicial review are well-established i.e. the traditional grounds of illegality, irrationality and procedural impropriety. These principles were elaborated upon by Lord Diplock in Council of Civil Service Unions v Minister for Civil Service [1985] AC 374, at 410:
    "By "illegality" as a ground for judicial review I mean that the decision-maker must understand correctly the law that regulates his decision-making power and must give effect to it. […] By "irrationality" I mean what can by now be succinctly referred to as "Wednesbury unreasonableness" […]. It applies to a decision which is so outrageous in its defiance of logic or of accepted moral standards that no sensible person who had applied his mind to the question to be decided could have arrived at it. […] I have described the third head as "procedural impropriety" rather than failure to observe basic rules of natural justice or failure to act with procedural fairness towards the person who will be affected by the decision. This is because susceptibility to judicial review under this head covers also failure by an administrative tribunal to observe procedural rules that are expressly laid down in the legislative instrument by which its jurisdiction is conferred, even where such failure does not involve any denial of natural justice."
  46. Many applications for judicial review made to us are based on the second ground, irrationality. The Tribunal has considered its role in judicial review proceedings in relation to this ground on numerous occasions (particularly in BAA v Competition Commission (No. 2) [2012] CAT 3). We consider the relevant principles from those cases to be as follows:
  47. (a) The OFT must take reasonable steps to acquaint itself with the relevant information to enable it to answer each statutory question posed for it.[11]

    (b) The extent to which it is necessary to carry out investigations to achieve this objective will require evaluative assessments to be made by the OFT, as to which it has a wide margin of appreciation as it does in relation to other assessments to be made by it.[12]

    (c) The standard to be applied in judging the steps taken by the OFT in carrying forward its investigations to put itself into a position properly to decide the statutory questions is a rationality test.[13]

    (d) The Tribunal should not intervene merely because it considers that further inquiries would have been desirable or sensible. It should intervene only if no reasonable competition authority could have been satisfied on the basis of the inquiries made.[14]

    (e) A rationality test also applies to determine whether the OFT has a sufficient basis in light of the totality of the evidence available to it for making the assessments and in reaching the decisions it did.[15]

    (f) It is not the function of the Tribunal to trawl through the contested decision with a fine-tooth comb to identify arguable errors; the decision must be read as a whole.[16]

    These principles apply equally to the CMA.

  48. In relation to the other two grounds, illegality and procedural impropriety, we set out the principles governing our consideration of them at the appropriate place below, i.e. in relation to Ground 2 for procedural impropriety, and to Ground 3 for illegality. We also discuss the relationship between illegality and irrationality at paragraph 108 below.
  49. Several of the parties have drawn our attention to the judgment of the Court of Justice in C-441/07 P European Commission v Alrosa Co Ltd [2010] ECR I-5949, and the accompanying Opinion of Advocate General Kokott. Alrosa concerned a 2006 decision of the European Commission (the "Commission") to accept commitments from De Beers, the world's largest supplier of rough diamonds, pursuant to Article 9 of Regulation 1/2003. By the commitments, De Beers agreed to reduce its purchases of rough diamonds from Alrosa, the world's second largest supplier of rough diamonds, by a specified amount over three years, leading to a complete cessation by 2009. Alrosa successfully challenged the commitments in the General Court. In particular, Alrosa objected to the Commission's rejection of less onerous commitments offered jointly by it and De Beers. The General Court allowed this appeal, holding inter alia that the Commission should have examined whether less onerous commitments would have been sufficient.
  50. The Commission appealed against the General Court's judgment. In allowing the appeal, setting aside the General Court's judgment and substituting its own judgment against Alrosa, the Court of Justice considered in particular the proportionality assessment required under the Article 9 commitments procedure, the Commission's discretion to accept commitments and Alrosa's right to be heard in the proceedings. It held as follows:
  51. (a) Contrary to what the General Court held, the proportionality principle applied differently in a commitments context as compared to where the Commission made an infringement decision pursuant to Article 7 of Regulation 1/2003. Where Article 9 commitments were concerned, proportionality was limited to verifying that the commitments addressed the competition concerns expressed to the undertaking. Notably, the Court of Justice considered that undertakings who offered commitments consciously accepted that their concessions may go beyond the measures the Commission could impose pursuant to Article 7.

    (b) The General Court wrongly encroached on the Commission's discretion to accept commitments by substituting its own assessment of complex economic circumstances (in that it sought to assess whether less onerous commitments would suffice) in a situation where it was only entitled to review the lawfulness of the Commission's assessment.

    (c) As Alrosa was not an addressee of the statement of objections the commitments were intended to address[17], it was not an "undertaking concerned" and did not benefit from the corresponding procedural rights. Instead, Alrosa had the less extensive rights of an interested third party pursuant to Article 27(2) of Regulation 1/2003.

  52. AG Kokott's Opinion sets out a thorough analysis of the Article 9 commitments process and a number of the parties have sought to rely on it. The following part of AG Kokott's Opinion drew particular interest from the parties:
  53. "51. Article 9 of Regulation No 1/2003 is characterised by a concern for procedural economy. The Commission resolves the competition problems identified by it without first establishing an infringement in cooperation with the undertakings concerned on the basis of their voluntary commitments. In the context of a decision under Article 7, on the other hand, it would possibly have to identify remedies itself, which would require it to undertake much more extensive and lengthy investigations and also a fuller assessment of the facts.
    52. The distinctive features of Article 9 of Regulation No 1/2003 affect the examination of the proportionality of decisions on commitments adopted under that provision in two ways.
    53. First, higher demands are to be made in the context of Article 9 of Regulation No 1/2003 as regards the appropriateness of the commitments which have been made binding. If such commitments are not manifestly appropriate for eliminating the competition problems identified by the Commission, the Commission is entitled to reject them. Only in this way is it possible to meet the objective of Article 9 of Regulation No 1/2003, which is to ensure a quick and effective resolution of the competition problems while avoiding a considerable investigation and assessment effort on the part of the Commission. The Commission is not required to agree to commitments the appropriateness of which could be assessed only after a thorough examination by the Commission.
    [...]
    57. However, in this connection the Commission is required to take into consideration only alternatives which are equally appropriate as the commitments offered to it with a view to resolving the competition problems identified. Both the commitments actually offered and any alternatives to those commitments must therefore be manifestly appropriate for resolving the competition problems.
    58. [...] In accordance with the spirit and purpose of Article 9 of Regulation No 1/2003, the assessment of alternatives is not intended to require any extensive and lengthy investigations or evaluations. In proceedings under Article 9 the Commission need not take into consideration alternatives whose appropriateness could not be established with sufficient certainty without such efforts.
    [...]
    60. The general interest in finding an optimum solution from the point of view of speed and procedural economy justifies restricting the choice of possible measures in the context of Article 9 of Regulation No 1/2003. Undertakings which offer commitments consciously accept that their concessions may go beyond what the Commission itself might impose on them following a thorough examination in a decision under Article 7 of Regulation No 1/2003. In return, with the termination of the antitrust proceedings initiated against them, they are quickly given legal certainty and can avoid the finding of an infringement of competition rules which would be detrimental to them and possibly an impending fine.
    61. Third parties will also generally benefit from the fact that an undertaking makes relatively far-reaching concessions to the Commission in order to avoid a decision imposing a prohibition. As the present case clearly illustrates, however, commitments under Article 9 of Regulation No 1/2003 may sometimes work to the detriment of the interests of a third party. This is the case in particular where the third party has relied on the continued existence of a practice of a dominant undertaking which gives rise to concerns from the point of view of competition law. However, such reliance deserves at most limited protection, having regard to the general interest in undistorted competition."
  54. We are required by section 60 of the Act to ensure that so far as possible (having regard to any relevant differences between the provisions concerned) questions relating to competition within the United Kingdom are dealt with in a manner which is consistent with the treatment of corresponding questions arising in EU law. Although the power to accept binding commitments may be regarded as a matter of enforcement and procedure rather than the substance of the law, there are broad similarities between the commitments regimes of the EU and the UK and any judicial authority at EU level is likely to be useful and relevant in interpreting the UK commitment powers. The Court of Justice's decision in the Alrosa case is the only significant judicial assessment of the EU commitments regime and we have considered it closely. It is not "on all fours" with the case we have to decide, and some of its more important observations relate to matters not pleaded in our case. Thus the lengthy consideration given by both the Advocate General and the Court to comparing the principle of proportionality in commitments and infringement cases is not applicable here, as Skyscanner is not claiming the OFT's decision breached that principle. Also the discussion of the position of Alrosa as a third party is specific to the EU procedural regime.
  55. The issue in Alrosa was whether the commitments went too far, and much of the discussion was about whether the Commission should have accepted, or at least investigated, less onerous commitments. Moreover, the appellant, Alrosa, who was claiming that less onerous commitments should have been accepted, was itself a party to the arrangements to which the Commission objected and stood to benefit from their continuance. This is the context for the Advocate General's reference (at [61]) to third party rights deserving "at most limited protection" and to much, if not all, of the discussion about proportionality.
  56. Nevertheless, we can deduce the following propositions from Alrosa, in particular, from the Advocate General's opinion, which are useful and relevant for our decision:
  57. (a) Commitments play an important role in competition enforcement providing "a more rapid solution to the competition problems identified by the Commission instead of proceeding by making a formal finding of infringement." (Judgment at [35])

    (b) Commitments will be easy to assess and obvious in their likely impact (AG Opinion at [53]), they will not require great investigation and assessment and in this sense rest on "procedural economy".

    (c) Commitments may go further in their scope than could be established by an infringement decision. This reflects the fact that they are offered voluntarily by the parties to avoid further detailed argument and dispute. (AG Opinion at [60]; Judgment at [48])

    (d) Procedural economy requires that it should not be necessary to conduct a detailed assessment and that the appropriateness of the commitments to address the Commission's concerns should be clear (or "manifest"). (AG Opinion at [53])

  58. As to the appropriate level of judicial scrutiny to which commitment decisions should be subject, the Advocate General in Alrosa, in considering whether the General Court had intruded too far into the Commission's margin of assessment, said:
  59. "77. The existence of a margin of assessment in economic matters does not mean that the Community judicature must refrain from reviewing the Commission's interpretation of information of an economic nature. Rather, it has the power to examine the material lawfulness of Commission decisions with a view to ascertaining that the facts have been accurately stated and that there has been no material error of assessment. It must not only establish whether the evidence relied on is factually accurate, reliable and consistent but also examine whether that evidence contains all the information which must be taken into account in order to assess a complex situation and whether it is capable of substantiating the conclusions drawn from it."

    AG Kokott's description of the scope of review by the EU Courts is broadly analogous to that of the Tribunal in a judicial review.

  60. Bearing in mind that the commitments process is meant to provide a rapid solution and to look forward, rather than to make a condemnation of past conduct, the OFT (now CMA) must be allowed a fair degree of discretion in its assessment of the appropriateness of the commitments to meet the concerns it has expressed. Too heavy a degree of judicial scrutiny would have the effect of making the obtaining of commitments no more rapid and advantageous in terms of time and effort than a normal infringement decision, which would not be consistent with the purpose of the Act in this respect. This is perhaps why commitment decisions are subject under the Act to judicial review whilst infringement decisions are subject to full merits appeal.
  61. Nonetheless, the OFT (now CMA) clearly cannot be given a completely free hand to conclude any arrangement that it likes with alleged infringers merely to suit its own administrative convenience. In our view, the normal mechanism of judicial review, as explained above (see in particular paragraphs 31 – 33), is sufficient and appropriate to provide the necessary degree of scrutiny of commitment decisions.
  62. IV. DID THE OFT FAIL TO TAKE INTO ACCOUNT RELEVANT CONSIDERATIONS? (GROUND 2)

  63. This ground is based on procedural impropriety, and was pleaded as Ground 2. Skyscanner argues that the OFT failed to take into account properly or at all the representations that Skyscanner made on the impact the Commitments would have. This challenge centres on the extent to which the OFT placed itself in a position to consider the appropriateness or otherwise of the proposed commitments, how the OFT conducted the consultation process and how it considered the points made by Skyscanner and others in response to the consultations.
  64. A. The Law: the OFT's Duty to Consult and to Consider

  65. The procedural requirements which apply to commitments are set out in Schedule 6A to the Act. In particular, paragraph 2(1) of Schedule 6A provides that:
  66. "Before accepting the commitments or variation, the CMA must–
    (a) give notice under this paragraph; and
    (b) consider any representations made in accordance with the notice and not withdrawn."
  67. Therefore, the OFT was under an explicit statutory duty to consider representations made in response to its consultation notice.
  68. In determining whether the OFT's consultation process was fair, we have in mind the guiding principles set out in R v London Borough of LBC ex p Gunning (1985) 84 LGR 168, and endorsed by Lord Woolf MR (as he then was) in R v North and East Devon Health Authority ex p. Coughlan [2001] QB 213 at [108]:
  69. "To be proper, consultation must be undertaken at a time when proposals are still at a formative stage; it must include sufficient reasons for particular proposals to allow those consulted to give intelligent consideration and an intelligent response; adequate time must be given for this purpose; and the product of consultation must be conscientiously taken into account when the ultimate decision is taken: R v Brent London Borough Council, Ex p Gunning (1985) 84 LGR 168." (emphasis added)
  70. It is well established that what fairness requires is context specific (see, for example, Groupe Eurotunnel v Competition Commission [2013] CAT 30 at [167]). Accordingly, we are required to have regard to factors such as the nature and impact of the decision, the legislative framework and the purpose and practicalities of the situation (R v Secretary of State for Education, ex parte M [1996] ELR 162 at 206 - 7).
  71. While it was common ground that the OFT was not required to address every competitive issue that might potentially arise from the conduct under investigation, it did have to take into account all relevant considerations. In this regard, we were referred to Lord Keith's judgment in the oft-cited Tesco Stores Ltd v Secretary of State for the Environment [1995] 1 WLR 759 at 764:
  72. "It is for the courts, if this matter is brought before them, to decide what is a relevant consideration. If the decision maker wrongly takes the view that some consideration is not relevant, and therefore has no regard to it, his decision cannot stand and he must be required to think again. But it is entirely for the decision maker to attribute to the relevant considerations such weight as he thinks fit, and the courts will not interfere unless he has acted unreasonably in the Wednesbury sense (Associated Provincial Picture Houses Ltd. v Wednesbury Corporation [1948] 1 KB 223)"
  73. As this extract also makes clear, it is not for us to interfere with the weight the OFT attached to Skyscanner's representations.
  74. B. The OFT's Initial Investigation

  75. The OFT began its investigation into the arrangements between IHG, Expedia and Booking in September 2010, following a complaint from Skoosh. It sent the parties a Statement of Objections in July 2012, having conducted an investigation lasting some two years. It is significant for our purposes that the OFT categorised the potential restrictions of competition it had identified as "restrictions by object", that is to say restrictions that by their very nature were so harmful to competition that it was not necessary to examine their effects on the market in question. The Statement of Objections also envisaged that fines could be imposed. The OFT did not set out in the Statement of Objections any detailed market analysis as a basis for its possible future findings, although it did provide, in Annexe 1, a review of the relevant market for the purpose of establishing the basis of relevant business turnover in the event of the imposition of a fine.
  76. It would be reasonable to assume that the absence of any detailed market analysis made it more difficult for the OFT to assess not only the efficiency arguments advanced by the Commitment Parties in favour of the agreements under investigation (see paragraph 11), but also the likely impact on the market of the proposed commitments when these were offered by the Commitment Parties at a later stage. This would in turn have made it harder for the OFT to appreciate the significance of the objections raised by Skyscanner and others to aspects of the proposed commitments when these became public.
  77. C. The OFT's Consultation

  78. As mentioned above, the OFT conducted two consultations in relation to its online hotel booking investigation. Skyscanner did not participate in the first consultation, but did respond to the second consultation, in which the OFT gathered views on a revised form of commitments. We summarise the consultation process in further detail below.
  79. (i) The First Consultation

  80. Shortly after the Commitment Parties offered initial commitments on 7 August 2013, the OFT published its first consultation on 9 August 2013 under the title: "Hotel online booking: Notice of intention to accept binding commitments to remove certain discounting restrictions for Online Travel Agents and Invitation to comment". The consultation document invited interested third parties to make representations on the initial commitments, including the submission of any relevant evidence. However, the accompanying OFT press release did not mention that the proposed commitments included a restriction on the disclosure of specific information about the level and extent of the reductions. Rather, it emphasised the expected effect on the level of discounting.
  81. The OFT also specifically contacted a number of interested parties to notify them of the first consultation. There has been some disagreement between Skyscanner and the CMA over the extent to which the OFT sought Skyscanner's views in the first consultation. Nonetheless, the Tribunal sees little importance in the point, save to say that the OFT submitted a contact form for the commercial department (which Skyscanner says was the wrong part of the business) on Skyscanner's website during the first consultation. It did not receive a response. We note that, at the time the OFT attempted to contact Skyscanner, Skyscanner did not offer meta-search services for hotel rooms, having only acquired a Spanish meta-search company offering such services in September 2013 (see paragraph 5 above) and the possible impact of the proposed commitments on meta-search services for hotel room would not previously have been of concern to Skyscanner.
  82. The first consultation ran for five weeks and received 36 responses from independent hotels, hotels chains, OTAs, industry associations and other interested parties. While no meta-search providers responded to the first consultation, the meta-search sector was referred to in two responses. The comments made were as follows:
  83. Response 1: "Meta-search is the primary mechanism deployed by OTAs (including the smaller OTAs) to drive increased sales. By not allowing meta-search engines to display discounted comparison prices, there is, by definition, no open price competition as there will be no user-friendly way of the consumer shopping around for the best price. This is a retrograde step in a world where consumers rely heavily on technology to seamlessly trawl the prices of competing suppliers of identical products in order to find the best deal."
    Response 2: "What will this do to the operation of price comparison sites? Will they now all have to offer log-ins for people who are members of the closed groups? And if they were to do that, how would the price comparison sites know whether or not the visiting consumers were or were not bona fide members of the closed group? Furthermore, if closed groups proliferate, how will the price comparison sites cope with them all? And if the price comparison sites cannot cope with them all, these rules emasculate the potency of the price comparison sites, which is not good for competition."
  84. In addition, the OFT engaged Opinion Leader, a market research agency, to carry out consumer survey research in relation to the proposed commitments. This was a small study involving 30 consumers, split into four focus groups. The briefing slides used for the focus groups did not mention the role of meta-search sites. The results of this survey referred to in the Decision are described at a general level only. At paragraphs 31 – 32 of Annexe 3 to the Decision, the OFT records the following results from the consumers surveyed in the consumer focus groups:
  85. (a) the majority were open to joining multiple closed groups; and

    (b) most were not overly concerned with the idea of having to make one full price booking in order to receive discounts.

  86. The CMA explained to us that, in light of the responses to the first consultation, the consumer survey evidence and further developments (such as the OFT's consideration of the relevant sector and its communications with the Commission and other competition authorities), the OFT asked the Commitment Parties to amend the initial commitments in certain respects. Among the amendments sought by the OFT were clarifications that hotels could offer discounts in an equivalent manner to OTAs. The duration of the proposed commitments was also shortened from three to two years in order to reduce the risk that they might cause competitive distortions by shifting consumers away from hotels' own booking websites to OTAs. As requested, the Commitment Parties offered a revised set of commitments on 27 November 2013.
  87. (ii) The Second Consultation

  88. The OFT published its second consultation on 20 December 2013. Interested third parties were invited to make representations on the revised commitments by 17 January 2014. Again, the accompanying press release made no reference to the restriction on disclosure or any possible effect on meta-search websites.
  89. We note that the concerns raised by two respondents to the first consultation about the impact of the proposed commitments on the meta-search sector were not mentioned in the OFT's summary of the responses to that consultation in Annexe 4 to the second consultation. Similarly, those concerns were not raised in the second consultation itself. The only reference to meta-search sites in the second consultation was in the context of the duration of the commitments, where the OFT noted that online travel agency services, including hotel online booking, is a growing sector and is characterised by frequent introduction of new technology or platforms and the expansion of search websites into the travel sector.
  90. The OFT received nine written responses to the second consultation, including one from Skyscanner.
  91. Throughout the proceedings, the CMA and the intervening Commitment Parties emphasised that Skyscanner responded on 16 January 2014, being the last day of the second consultation. This point was mentioned in both the Defence and the intervening Commitment Parties' Statement of Intervention, and several times at the hearing. However, the CMA confirmed at the hearing that it was not going to take any point with regard to the fact that the response came in on the last day of the second consultation. Further, Ms Bacon QC, for the CMA, stated that the date of receipt of the response was not a factor that had detracted from the weight given to that response by the OFT. That must be right. A consultation period is just that and it is of course not uncommon for responses (whether pivotal or not) to be received on the final day.
  92. In its response, Skyscanner expressed its concern that, although the proposed commitments may encourage intra-brand competition by allowing OTAs to offer discounts, Principle 19 - which allows hotels to prevent OTAs from publicising specific information about discounts to consumers outside the OTA's closed group (including meta-search sites) - could have a negative effect on inter-brand competition as consumers will be unable to use meta-search sites to compare the actual room prices and discounts offered by different hotels. In particular, Skyscanner made the following points:
  93. "Although the commitments specifically permit price comparison sites to be notified where Closed Group discounts are available, we would be unable to include actual discounted prices within our search results. The effect of this would be a lack of clarity for consumers in respect of the pricing available for hotel rooms, and inaccurate results for members of Closed Groups.
    Even where Skyscanner is able to indicate to consumers that discounts may be available if they join a Closed Group, this information would not be meaningful for consumers. They would be required to click through to the OTA website and log in and search for the pricing data on the OTA site again, hindering their ability to easily and accurately compare pricing. [...]
    Although it is clearly encouraging that competition within the online travel industry is being considered, Skyscanner does not believe that the commitments proposed enhance competition within the travel sector. The commitments have the potential to undermine the value afforded to consumers by both meta-search sites and search engines within the travel sector, and disrupt a distribution channel which specifically encourages competition within the travel sector. This disruption will only increase as the use of Closed Groups proliferates, ultimately obstructing active participation in the travel sector by distribution models other than OTAs and hotels, discouraging potential new entrants to the meta-search market, thus negatively affecting competition further."
  94. The OFT met Skyscanner on 20 January 2014, after the second consultation had closed, to discuss its concerns. There is some dispute between Skyscanner and the CMA as to what was said at that meeting. The main areas of disagreement concern the extent to which the OFT invited Skyscanner to provide evidence of its concerns prior to the Decision being taken and the impression given that the Decision was in final form and imminent. The Decision was issued on 31 January 2014.
  95. The concerns expressed to the OFT about the impact of the commitments on meta-search are summarised at paragraphs 66 - 67 of Annex 3 to the Decision:
  96. "66. A meta-search site respondent submitted that the closed group and advertising restrictions envisaged by the Final Commitments would result in a lack of clarity for consumers with regard to price, and inaccurate search results for members of closed groups. It noted its belief that consumers are frequently driven by price, particularly in the case of making a hotel booking. Therefore, it considered that the Final Commitments have the potential to undermine the value of meta-search sites and search engines to consumers.
    67. Furthermore, a meta-search site respondent requested that the Final Commitments be extended to relate to meta-search sites as well as OTAs and hotels, in those cases where a transactional booking does not take place on the meta-search website. It also requested that the Final Commitments allow discounts to be available after a prior purchase of any product offering, not only where there has been a previous hotel room booking."
  97. Later in Annexe 3, the OFT purports to respond to those concerns as follows:
  98. "74. The Final Commitments allow hotels to prevent OTAs from publicising information regarding the specific level of discounts for a particular room to consumers who have not joined their closed groups, for example on price comparison websites and meta-search sites. There are similar restrictions on hotels publicising such information about the specific level of discounts from the headline rate they offer for a particular room without MFN provisions being enforced. However, OTAs and hotels are free to publicise information on the general availability of discounts in a clear and transparent manner, including to price comparison websites and meta-search sites (that is, to members and non-members). The OFT remains of the view that the Final Commitments, including the provisions relating to advertising, are sufficient to address its competition concerns, which relate to intra-brand competition and barriers to entry for OTAs.
    75. The focus of the OFT's investigation has been restrictions on OTAs' discounting off the Room-Only Rate set by a hotel. The OFT has not investigated meta-search sites in this case and has made no assessment of whether similar restrictions may exist in this area. The OFT therefore does not consider it appropriate to extend the Final Commitments to meta-search sites, but also notes that transactional bookings, in relation to which discounts could be offered, do not take place on meta-search sites." (emphasis added)

    D. The Parties' Submissions

  99. Skyscanner contends that the OFT failed to take into account (either properly or at all) the points it made during the second consultation; the OFT did not consider the potential negative effects of the publication restrictions set out in Principle 19 on inter-brand competition as consumers will be unable to use meta-search sites to compare the actual room prices and discounts offered by different hotels. Therefore, according to Skyscanner, the OFT's procedure was flawed and unlawful in that it failed to take into account relevant considerations and/or the OFT acted in breach of paragraph 2(1)(b) of Schedule 6A to the Act.
  100. More specifically, Skyscanner argues that the OFT failed to consider conscientiously its representations in breach of both its statutory duties and the Gunning principles set out above (see paragraph 47). In considering the OFT's alleged procedural impropriety, Skyscanner asked us to have regard to the following factual matters:
  101. (a) Skyscanner raised real and plausible concerns about the commitments the OFT proposed to accept;

    (b) the OFT was required pursuant to section 31A of the Act to satisfy itself that the proposed commitments were appropriate to address its competition concerns;

    (c) the OFT had been investigating the online hotel booking market for three years and might have been expected to be able to recognise the difference between discounting into closed groups subject to disclosure restrictions and discounting generally, and also the importance of price transparency in this market;

    (d) the OFT was an expert regulator that had (i) the expertise to consider the issues raised during the consultation process and (ii) the ability to seek further evidence if necessary; and

    (e) the proposed commitments introduced a particular form of discounting within closed groups, which – so far as they existed before the Commitments – were limited and different in form. But, most importantly, they introduced a new restriction on price disclosure.

  102. Moreover, Skyscanner argues that it was wrong for the OFT to have rejected its representations without further consideration simply on the basis that they were not supported by "evidence".
  103. The CMA rejects this: it maintains that the OFT took Skyscanner's concerns seriously and invited Skyscanner to provide evidence to verify those concerns; Skyscanner could have provided, for example, information about the effects of pre-existing closed groups on the use of its website. However, Skyscanner provided no such evidence. In any event, the CMA contends that the OFT considered the issue of inter-brand competition in considerable detail, but concluded that there was no evidence that the Commitments would or might distort inter-brand competition between OTAs.[18] With regard to the alleged harm to meta-search sites, the CMA says there was no evidence of this before the OFT and that the harm was too remote and indirect to be relevant to the appropriateness of the Commitments.
  104. The intervening Commitment Parties emphasise that Skyscanner is only a third party respondent to the consultation, not (as was Skoosh) a complainant in the OFT's original investigation. However, we do not see that this should affect the significance the OFT would attach to a material point raised by a respondent on such a clearly relevant issue as price transparency, which has a potential bearing on the commitments' impact on a significant part of the market for online hotel booking.
  105. As to the reasoning in paragraphs 74 - 75 of Annexe 3 to the Decision (see paragraph 66 above), which refers to the Commitments' impact on meta-search, Skyscanner describes this as a complete failure to engage with the points it had made and an attempt to avoid the issues by simply asserting that the Commitments were sufficient to address the OFT's competition concerns without offering any real reasoning. The CMA, by contrast, says that paragraphs 74 - 75 further demonstrate that the OFT did consider Skyscanner's concerns before making the Decision, but that it concluded that those concerns did not affect its view that the Commitments were appropriate to address its competition concerns.
  106. The CMA also refers to meetings the OFT had with Skyscanner both before and after the Decision was taken, at which it says the OFT made clear that it had no evidence that the Commitments would harm meta-search.
  107. E. Analysis

  108. The OFT conducted what it clearly considered to be an appropriate process, as required by the Act, involving two rounds of consultation, and took considerable pains to respond to the comments it received. It modified the proposed commitments in response to the first consultation responses and appears to have satisfied itself not only that the proposed commitments were sufficient to meet its concerns as set out in the Statement of Objections, but also that there were no material objections from third parties. Yet we are obliged to conclude that, despite all these efforts, the process was in one important respect defective in that it failed conscientiously to address the objections raised by two respondents to the first consultation and by Skyscanner in response to the second. These objections concerned the possible effect of the proposed commitments on price transparency, and on meta-search websites.
  109. (i) Price Transparency

  110. In the Statement of Objections, the OFT recognised the benefits of price transparency in the market for the supply of online hotel bookings. It also referred to the low costs of searching on price comparison websites, which are essentially meta-search sites. The following extracts are instructive:
  111. "(B)ooking of hotel accommodation by using the Internet has increased considerably over the past decade and is now the most commonly used method of booking hotel accommodation." (Annex 1, paragraph 1.8)
    "The Internet allows for a much swifter search and comparison across a wide variety of choice factors including price, dates, quality and location." (Annex 1, paragraph 1.15)
    "The Internet brought about price transparency across the market, enabling consumers to identify the best deal, i.e. the lowest price for any given hotel room, at very low search costs [...]
    The OFT notes that, in that sense, the distribution arrangements resulting in rate or price parity undermine the benefits of the transparency and enhanced search functions brought about by the Internet and the possibilities offered by e-commerce." (paragraphs 1.14 – 1.15)
    "The OFT considers that consumers seeking to book hotel accommodation online are likely to use search terms relating to location, date and any preference over the price and quality of the hotel. Internet search engines produce details of both hotel websites and OTAs matching at least some of the desired criteria." (Annex 1, paragraph 1.10)
    "The OFT considers that online consumers are typically price sensitive, given the ease and relative cost of observing different prices of hotel accommodation online and therefore the level of demand for a particular online option would change significantly with a change in price, for example between an OTA and hotel website. This is supported by a recent industry report, according to which about 60 per cent of internet users who had stayed in a UK hotel said that price was the factor that most influenced their decision to stay at a particular hotel." (Annex 1, paragraph 1.10, footnote 761)
    "Given the high degree of price transparency and the low costs associated with price search on the internet through price comparison sites, certain OTAs may want the option to offer discounted prices for Room-Only hotel accommodation [...]." (paragraph 2.119)
  112. There is also a reference, at paragraph 2.42 of the Statement of Objections, to Expedia's annual Form 10-K report to the US Securities and Exchange Commission, in which Expedia identified as a risk factor:
  113. "the continued emergence and relative traffic share growth of search engines and metasearch engines as destinations sites for travelers (sic)".
  114. Although the OFT did not undertake a detailed market analysis in the Statement of Objections (as it had provisionally identified the restrictions as being "by object") the OFT's brief description of the relevant market in Annexe 1 (from which several of the extracts from the Statement of Objections at paragraph 75 above are taken) referred to a Mintel report on UK hotel bookings and to a number of previous decisions of the Commission to support its view of a dynamic, developing market where consumers used price comparison methods such as internet search engines to compare what was on offer.
  115. However, little attention was given to meta-search sites such as Skyscanner in the Statement of Objections. The OFT appears to have limited itself, at that stage, to search engines such as Google. Indeed, in diagrams prepared for the Statement of Objections, the OFT set out various ways in which a consumer could book a room at Hotel Inter-Continental London Limited. These included: (i) directly on IHG's website; (ii) on Expedia or Booking's websites; or (iii) by searching on Google. Meta-search sites were not referred to.
  116. We were also directed by Skyscanner to a number of materials which suggest that meta-search is a growing sector which is used by a large number of consumers. For example, the PhoCusWright European Consumer Travel Report (Fourth Edition, September 2013) indicates that over a third of travellers in all European markets use meta-search sites to compare travel products and, in the UK, about 20 percent of customers use a travel search engine specifically for hotel bookings.
  117. The comments made in relation to the dynamic and developing hotel online booking market in the Statement of Objections suggest that the OFT was generally aware of the role of the internet in increasing price transparency and the importance of price transparency in generating competition in dynamic markets, thereby increasing competition for the benefit of consumers. It should in our view also have been aware of the role played by operators such as Skyscanner in promoting price transparency.
  118. (ii) Inter-brand Competition

  119. It appears to us that the OFT was concentrating on the need to secure a greater degree of discounting as between hotels and OTAs and much less on the need to maintain, if not to promote, price transparency generally. The OFT did consider inter-brand competition, but did so from the point of view of possible barriers to entry for OTAs, increased switching costs for consumers arising from the requirement to join closed groups and protection for new OTAs from "predatory" deep discounting by established OTAs. The OFT satisfied itself on the basis of the consumer research by Opinion Leader (see paragraph 57 above) that the closed group concept was acceptable because consumers were not concerned by having to make one full-price booking to receive discounts through a closed group and would be "open" to multi-homing (i.e. joining multiple closed groups). The risk of harmful deep discounting could be addressed by the imposition of a discount "cap".[19]
  120. Ms Smith QC, counsel for Skyscanner, subjected the Opinion Leader survey to some trenchant criticism, pointing to its qualitative nature, small sample size and the questions asked, which were leading in some cases. It is not for us to judge the merits or otherwise of this kind of qualitative research, save to say that generally a high degree of caution should be exercised in basing hard and fast conclusions on a single qualitative survey of some 30 people. What is clear is that the focus groups were not asked whether they used meta-search websites or whether their use of them would be affected in any way by the proposed commitments. Indeed, meta-search or price comparison websites do not appear to have been mentioned at all in this exercise. Generally, in its consideration of inter-brand competition, the OFT does not appear to have attached any great significance to the role of meta-search and price comparison websites in promoting price transparency.
  121. (iii) Responses to the Consultations

  122. We have referred above to: (i) the two responses to the first consultation, which warned of the possible harm to meta-search and price comparison websites; and (ii) Skyscanner's response to the second consultation, which made the same point, at somewhat greater length. Skyscanner's focus was on the restriction on disclosure of actual discounts outside the closed group. It said this would mean Skyscanner and other price comparison websites would be unable to show actual discounts available, and that this would in turn prevent consumers from making easy price comparisons. Whether the OFT was in a position adequately to evaluate these responses, and how it in fact dealt with them are the essence of Skyscanner's complaint under this ground.
  123. (iv) The OFT's Response

  124. The OFT's response to the two earlier comments was at best opaque. Skyscanner drew our attention to the OFT's consideration of the responses to its first consultation in Annexe 3 to the Decision.[20] Although meta-search sites were not explicitly referred to, the OFT appears to have had this point in mind when it described the concerns raised about the impact of the proposed commitments on the structure of the market and, in particular, on consumers' ability to shop around:
  125. "Some respondents raised concerns about consumer confusion regarding the availability and level of discounts. For example, it was submitted that because price differences would only be visible to closed group members, rather than publicly, it would not be easy for consumers to compare the effective prices being offered by OTAs and hotels via their closed groups. It was also queried whether consumers would be able to figure out which website will offer the best long-term deal."[21]

    However, there is no mention of these concerns in the subsequent section, which purports to provide the OFT's response to the points raised about the impact of the commitments on the structure of the market.[22]

  126. We now consider what the OFT did to evaluate the important issues of price transparency and the impact on meta-search sites in relation to competition in this market.
  127. (v) The Request for Supporting Evidence

  128. The CMA put to us that, far from closing its mind to Skyscanner's concerns, the OFT considered them very carefully and explored them further at the meeting, convened three days after the end of the consultation period. As we have noted, there is some dispute as to the precise content of that meeting but for our purposes we accept that there was a serious discussion of Skyscanner's concerns. However, the OFT indicated in clear terms to Skyscanner that it could not take the concerns any further in relation to the proposed commitments without evidence of possible harm to meta-search in general and Skyscanner in particular.
  129. We do not think this was a fair position to take. We have heard argument about where the burden of proof lies in such cases, and agree that a situation of this kind is more akin to a regulatory decision than to an infringement or exemption finding. The CMA denied strongly that the OFT disbelieved Skyscanner and emphasised that it considered Skyscanner's point to be plausible. However, the CMA stressed throughout that in a consultation and decision making exercise of this kind it would not be inclined to follow up an objection that was made without supporting evidence. Mr Rasmussen, who was the OFT's Project Director for the investigation into the hotel online booking sector, said in his first Witness Statement for the CMA:
  130. "[...][T]he case team and I thought that Skyscanner's arguments were insufficiently substantiated for them to carry great weight."
    "In the absence of any evidence of any harm to meta-search sites the OFT did not consider that it was necessary to carry out further analysis of that issue before accepting the Final Commitments".
  131. As to what the OFT meant by evidence, this was explained to us by Ms Bacon, for the CMA, in the following terms:
  132. "I think sometimes, if you use the word "evidence" it obscures the real point which is that the OFT was talking about the overall material in front of it, and some of that material would have consisted of submissions which may have referred to existing facts. Other of that material would have consisted of submissions as to what might have happened, and one can imagine, for example, that if you were making a prediction as to what might occur, you could put in an economic model and that would be regarded as evidence."
  133. Mr Rasmussen put the point in the following terms:
  134. "Skyscanner's submission might have been more persuasive if it had, for example, explained which hotels and OTAs it had arrangements with and what proportion of rates and hotels it already had access to, and then provided data on observed effects of pre-existing closed group schemes on traffic through its site or on the extent to which prices varied and how this might be expected to change under the commitments. Skyscanner might also have conducted its own consumer research...."

    (vi) Our View

  135. We find this quite unsatisfactory. The CMA seems to be saying that, to be taken seriously, a submission in response to a consultation must be accompanied by some material to provide a veneer of substance. Mr Ward QC, for the intervening Commitment Parties, impliedly criticised the Skyscanner submission for its brevity and lack of supporting evidence, describing it as "a very, very flimsy submission". We disagree. If a consultation response raises an important and obvious point of principle, it is for the authority to examine it further. This is particularly so where the authority has not carried out an analysis of the economic effects of the practices which it proposes to address with its commitments decision and where that decision itself may generate its own economic effects within the market.
  136. In any case, it is not clear what evidence Skyscanner could reasonably have produced in this case. Not only was Skyscanner itself a recent entrant to the hotel online booking sector, but what was at issue was the potential effect of proposed commitments. The OFT had itself described the "pre-existing closed group schemes" as "much smaller, niche distribution channels – hidden/opaque, membership or package – which do not allow for Room-only price comparison" so it is not clear to us that much could be gleaned from "observed effects" in relation to them. No relevant economic model and no specific consumer research were available to Skyscanner at that time. In these circumstances, it does not seem unreasonable for Skyscanner to say, in effect, "we have the following plausible concerns about the likely effect of the proposed commitments, and this is something we think you, the authority, should consider before proceeding further". No doubt Skyscanner could have provided some "material" in the sense described by Ms Bacon, but in these particular circumstances this would have amounted to no more than descriptive padding to the Skyscanner response with no substantive content.
  137. Without wishing to add to the CMA's burdens in cases of this kind, it is not acceptable for it to say that when an interested party, operating in the market under consideration, raises a point that puts in question an essential feature of proposed commitments, the authority will not act on it without supporting material provided by the party raising the point. Of course the objection cannot be fanciful or frivolous, but the OFT accepted Skyscanner's point as plausible. In this instance, Skyscanner was not in a position to provide "hard" evidence based on its past experience and, in any event, the concern was about potential effects in a new and previously untried situation.
  138. To the extent that the OFT could reasonably have felt the need for additional material, this could relatively easily be obtained and verified by the OFT itself. As we have already said, we are concerned that by pursuing its investigation on the basis that it had identified restrictions "by object" the OFT may have deprived itself of the ability properly to appreciate the significance of the role of operators such as Skyscanner, even though it had initially acknowledged the importance of price transparency as a force for competition and was aware, at least, that meta-search operators existed. In our view, faced with objections of the kind raised by Skyscanner, the OFT should have felt obliged to investigate them further before taking a decision.
  139. (vii) The OFT's Discretion

  140. Counsel for the intervening Commitment Parties rightly warned us not to encroach on the OFT's discretion to attach what weight it thought appropriate to any particular consideration. He put to us that this was not a matter of whether the OFT had had regard to a material consideration but what weight it had attached to it, as explained by Lord Hoffmann in the Tesco Stores case. (We referred to Lord Keith's observation to similar effect in the same case at paragraph 49 above.) The OFT, Mr Ward said, had clearly considered Skyscanner's point, but had attached little weight to it. We do not agree. Skyscanner's claim is that the OFT failed to take its objections into account "properly or at all". There is a difference between the weight attached to a consideration (to which Lord Hoffmann's and Keith's comments are directed) and the manner in which it was taken into account. Take into account means, in our view, properly take into account, in particular by not imposing unreasonable or unnecessary additional requirements, in this case by asking for supporting evidence which, in the nature of the situation, was unlikely to be available to Skyscanner. The approach put forward by Mr Ward simply provides an authority with an easy option of appearing to take a material consideration into account whilst in reality not doing so at all.
  141. As to whether the OFT did consider Skyscanner's objection, in Annexe 3 of the Decision, the OFT stated that OTAs and hotels were free to publicise information on the general availability of discounts and repeated what Ms Smith described as the "mantra" that the OFT was satisfied that its concerns had been met. As Ms Bacon accepted, this did not address the objection that specific information could not be disclosed. We disagree that not addressing an objection is the same as giving little weight to it. And we do not think that, in this case, the OFT can be said to have given Skyscanner's objection proper consideration.
  142. At the hearing, the CMA sought to address this point by reference to adequacy of reasoning, arguing that the reasons given in a decision could be supplemented by other means, in this case meetings and correspondence. However, Skyscanner's complaint is not based on inadequacy of reasons in the Decision. Instead it claims that the OFT did not properly consider the objection. Nothing that was said in meetings or correspondence responded specifically to Skyscanner's point. All that it was told was that it should provide evidence to back up its objection, which we have already considered as an unfair position for the OFT to have adopted.
  143. (viii) The OFT's Approach

  144. It would be very easy to conclude that the OFT found the Skyscanner objection inconvenient because it threatened to upset a carefully constructed edifice that the OFT believed would, over time, introduce greater discounting as between OTAs and hotels. Whether or not this is so, it was explained to us at the hearing that the restriction on disclosure of actual discounts was essential to the closed group model given the continued prevalence of rate parity obligations. We should note, however, that this was not explained to Skyscanner in correspondence or meetings at the time and does not appear either in the Decision or in the CMA's or the intervening Commitment Parties' pleadings.
  145. In so far as it may have thought Skyscanner's concern about harm had validity, the OFT considered this could be kept under review and if necessary addressed either on expiry of the Commitments after two years or, if the "evidence" was overwhelming, before then. The OFT may genuinely have intended this and the CMA may indeed be receptive to further evidence. But the OFT seems to have taken the view that the Decision itself, which Ms Bacon stressed was "just the start" of the move to greater competition, could not be delayed or fundamentally altered at this late stage without imperilling the entire exercise and the benefits to competition that the OFT envisaged.
  146. The threshold for review of a commitments decision is in any case relatively high. There must be reasonable grounds to: (i) believe that there has been a material change of circumstances; (ii) suspect that the commitments have been breached; or (iii) suspect that the information which led to the commitments being accepted was incomplete, false or misleading in a material particular.[23]
  147. (ix) Conclusion

  148. Accordingly, we find that the OFT failed properly to consider or conscientiously to take into account the objection to the proposed commitments raised by Skyscanner and others. This objection centred on the restriction on disclosure of specific price information outside the "closed groups" established as part of the commitment arrangements. It is not clear to us whether this was because the OFT had closed its mind to the point, or whether it was unable to appreciate the potential significance for price transparency, and hence for competition, of what was being said. In either case, the OFT failed properly to investigate a plausible point further and instead insisted on more evidence or supporting material from Skyscanner itself. We find that in so doing the OFT acted unfairly and that the process by which it subsequently reached its decision was procedurally improper. We therefore uphold Skyscanner's appeal on this ground.
  149. V. DID THE OFT ACT ULTRA VIRES BY ACCEPTING COMMITMENTS WHICH POTENTIALLY HARM COMPETITION? (GROUND 3)

  150. We now consider Skyscanner's challenge to the substance of the decision. This is a challenge based on illegality and, in the alternative, irrationality, and was pleaded as Ground 3. Skyscanner argues that the OFT was required to exercise its powers to accept commitments rationally and so as to promote the policy and objects of the Act. By adopting the Commitments without considering their potentially anti-competitive consequences, it says the OFT acted contrary to the policy and objects of the Act, which were to promote competition for the benefit of consumers and/or acted irrationally. Skoosh as intervener supports these contentions but also raises some more far-reaching points.
  151. Skyscanner's main allegation is that the Commitments create a new market equilibrium which is potentially worse than the existing situation. It argues that, while the Commitments may encourage intra-brand competition, they will have a negative effect on inter-brand competition. In particular, the efficient functioning of the market will be harmed as consumers will be unable to use meta-search sites to compare actual room prices and discounts offered by different hotels. Therefore, the Commitments restrict competition to the detriment of consumers.
  152. The CMA sought to persuade us that Grounds 2 and 3 made essentially the same point. While we recognise that these grounds are related and that there is a certain amount of overlap, we nonetheless see the two grounds as covering different aspects of Skyscanner's claim. The way the OFT responded to the objections from Skyscanner and others affected not only the procedural impropriety considered under Ground 2, but also whether the Decision itself was based on sufficient and appropriate evidence and hence goes to its substance.
  153. A. The Law

    (i) Illegality

  154. A decision may be unlawful under the "illegality" head of review, as set out by Lord Diplock in the Council for the Civil Service Unions case (see paragraph 32 above), where an authority acts in breach of the statute conferring the relevant power or duty. For example, an authority's action may be unlawful if there is no legal basis for its decision, or where the authority misinterprets the instrument relevant to the function it purports to perform. It is also clear that a decision of a public body may be illegal where it frustrates the purpose of the empowering legislation, as Lord Reid said in Padfield v Minister of Agriculture [1968] AC 997 at 1030B - D:
  155. "Parliament must have conferred the discretion with the intention that it should be used to promote the policy and objects of the Act; the policy and objects of the Act must be determined by construing the Act as a whole and construction is always a matter of law for the court. In a matter of this kind it is not possible to draw a hard and fast line, but if the Minister, by reason of his having misconstrued the Act or for any other reason, so uses his discretion as to thwart or run counter to the policy and objects of the Act, then our law would be very defective if persons aggrieved were not entitled to the protection of the court."
  156. Skyscanner's illegality challenge focuses on the OFT's duty to promote the purpose of the Act, and its duty to exercise its statutory powers for the purpose for which they were conferred.
  157. (ii) Irrationality

  158. We referred to the relevant principles at paragraphs 32 and 33 above. Of particular importance here are principles: (a) by which the authority must take reasonable steps to acquaint itself with the relevant information to enable it to answer the statutory questions posed for it; and (e) that a rationality test applies to determine whether the authority has sufficient basis in the light of the information available to it for making the assessments and reaching the decisions it made. The Tribunal explained these principles in the BAA case at [20(3)] – [20(4)]:
  159. "(3) The CC, as decision-maker, must take reasonable steps to acquaint itself with the relevant information to enable it to answer each statutory question posed for it [...]: see e.g. Secretary of State for Education and Science v Tameside Metropolitan Borough Council [1977] AC 1014, 1065B per Lord Diplock; Barclays Bank plc v Competition Commission [2009] CAT 27 at [24]. The CC "must do what is necessary to put itself into a position properly to decide the statutory questions": Tesco plc v Competition Commission [2009] CAT 6 at [139]. The extent to which it is necessary to carry out investigations to achieve this objective will require evaluative assessments to be made by the CC, as to which it has a wide margin of appreciation as it does in relation to other assessments to be made by it: compare, e.g., Tesco plc v Competition Commission at [138]-[139]. In the present context, we accept Mr Beard's primary submission that the standard to be applied in judging the steps taken by the CC in carrying forward its investigations to put itself into a position properly to decide the statutory questions is a rationality test: see R (Khatun) v Newham London Borough Council [2004] EWCA Civ 55; [2005] QB 37 at [34]-[35] and the following statement by Neill LJ in R v Royal Borough of Kensington and Chelsea, ex p. Bayani (1990) 22 HLR 406, 415, quoted with approval in Khatun:
    "The court should not intervene merely because it considers that further inquiries would have been desirable or sensible. It should intervene only if no reasonable [relevant public authority – in that case, it was a housing authority] could have been satisfied on the basis of the inquiries made."
    (4) Similarly, it is a rationality test which is properly to be applied in judging whether the CC had a sufficient basis in light of the totality of the evidence available to it for making the assessments and in reaching the decisions it did. There must be evidence available to the CC of some probative value on the basis of which the CC could rationally reach the conclusion it did: see e.g. Ashbridge Investments Ltd v Minister of Housing and Local Government [1965] 1 WLR 1320, 1325; Mahon v Air New Zealand [1984] AC 808; Office of Fair Trading v IBA Health Ltd [2004] EWCA Civ 142; [2004] ICR 1364 at [93]; Stagecoach v Competition Commission [2010] CAT 14 at [42]-[45]."
  160. It follows that the authority must not only take proper account of the material before it but must also ensure that it has all the material before it that is relevant and necessary for its decision. This reflects, as we have noted, the approach adopted by Advocate General Kokott in the passage already cited (see paragraph 47) where she refers to the court having to "examine whether that evidence contains all the information which must be taken into account in order to assess a complex situation and whether it is capable of substantiating the conclusions drawn from it".
  161. (iii) Relationship between Legality and Rationality

  162. It is well known that the grounds for judicial review are overlapping and merge into one another. As Lord Irvine LC remarked in Boddington v British Transport Police [1999] 2 AC 143 at 152 E-F:
  163. "the various grounds for judicial review run together. The exercise of a power for an improper purpose may involve taking irrelevant considerations into account, or ignoring relevant considerations; and either may lead to an irrational result."
  164. Accordingly, a decision that frustrates the purpose of the statute and is therefore "illegal" may also be irrational. Nonetheless, it is not necessary to show irrationality to establish illegality.
  165. B. The Parties' Submissions

  166. As mentioned above, Skyscanner argues under this ground that the OFT acted contrary to the policy and objects of the Act by putting in place the Commitments. Primarily, this is a Padfield illegality challenge. In the alternative, Skyscanner says the Decision was irrational.
  167. Skyscanner's case is that the implicit policy and objects of the Act are to promote competition for the benefit of consumers. We note that the OFT's successor body, the CMA, has an explicit statutory duty in almost exactly the same terms pursuant to section 25(3) of the Enterprise and Regulatory Reform Act 2013. Perhaps unsurprisingly therefore, there was no dispute as to the policy and objects of the Act.
  168. Skyscanner contends that the OFT was not entitled to ignore the potential anti-competitive consequences of its proposed commitments (through reducing price transparency), even if those commitments might be "sufficient" to address other competition concerns (restrictions on discounting). Here Skyscanner refers to the harm which may be caused to meta-search sites by the restriction on disclosure of actual discounted prices, which Skyscanner says is the whole point of meta-search sites and is the way in which they facilitate competition (i.e. by enhancing price transparency, reducing search costs and increasing inter-brand competition). This harm extends to consumers, who will face increased search costs. Ms Smith QC, for Skyscanner, explained this at the hearing as follows:
  169. "Without the advertising restriction as a consumer you go to one website, you say, "I want a hotel [for the][...] 1st August, Paris", and you get all the prices offered by all the OTAs for that hotel. If you want to see discounts available, post the commitments, you go on to Skyscanner, you see all the prices, the headline room rates available, you see next to each of the OTAs' websites:
    "Discounts may be available", "Discounts will be available", "Discounts may be available if you join a closed group".
    So you go to Booking.com, you click on it, you join its closed group and you see what discount is available [...] That situation does clearly increase search costs. It clearly has that impact. It clearly reduces price transparency.
    There is a further point and it is this: if you make your first purchase with Booking.com, what incentive do you even have to look at what discounts may be available on [an alternative OTA's site]? You know that even if you go to [that alternative OTA's site] and you join their closed group and they are offering a better discount than would be available to you on your second purchase on Booking.com, you cannot take advantage of that discount today. You cannot take advantage of that discount until you make your second purchase on [the alternative OTA's site]. You do not even know what discount might be available for your second purchase, let alone when you are going to make your second purchase. So once you have been caught by the closed group for Booking.com, what incentive is there to even go and look at the other closed groups, even to join the other closed groups. [...] There is not any. Before the advertising restrictions were put in place, you would have had access to all this information in one place."
  170. The CMA repeats its response to Ground 2 here to say that Skyscanner provided no evidence that the Commitments harm competition and, therefore, the OFT had no reason to investigate the unsubstantiated assertion further. In particular, the CMA says that the OFT had no evidence that consumers would stop using meta-search sites because of the Commitments. It emphasises that the Commitments permit meta-search sites to publicise information regarding the availability of discounts. Further, the CMA reasons that the OFT relied on consumer survey evidence to support its view that consumers would be willing to "multi-home" and join multiple closed groups. Finally, the CMA says that that there was no evidence before the OFT to suggest that consumers would be worse off compared with the situation before the Commitments, where there was limited, if any, price competition between OTAs. Nonetheless, the CMA says the OFT recognised that there was uncertainty surrounding the exact consequences of introducing (limited) price competition and, accordingly, it limited the duration of the Commitments to two years, required the Commitment Parties to report annually and was open to receiving evidence as to the effect of the Commitments.
  171. Skoosh's intervention in support of Skyscanner relates primarily to this ground of appeal. While Skoosh went somewhat further than Skyscanner in its statement of intervention and skeleton argument, the two parties appeared to be more closely aligned with each other's position at the hearing. Nonetheless, we are conscious that, as a meta-search site and an OTA respectively, Skyscanner and Skoosh have slightly different interests in this litigation. This divergence is apparent in Skyscanner's focus on the publicity restrictions (Principle 19) and Skoosh's focus on the residual discounting restrictions (Principle 18).
  172. First, Skoosh argues that the OFT misdirected itself in relying on the "procedural economy" lying behind the commitments process.
  173. Further, Skoosh says that the OFT made the following errors which led it to act ultra vires contrary to the object and purpose of the Act and/or irrationally:
  174. (a) The OFT adopted as the counterfactual the position before the Commitments (being a position which it had considered to involve restrictions by object in the Statement of Objections), when it should have instead had regard to the position of a market without any restrictions.

    (b) The OFT failed to consider Article 101(3) TFEU, which it should have done as the Commitments contained restrictions of competition. Skoosh argues that the "residual restrictions" on discounting in the Commitments - as a form of pricing restriction - have as their object the restriction of price competition in all cases except for closed group members. This amounts to an object restriction caught by Article 101(1) TFEU and is therefore automatically void pursuant to Article 101(2) TFEU unless justified under Article 101(3) TFEU. In the alternative, Skoosh says the residual restrictions are so obviously an actual or likely restriction by effect that they required further analysis by the OFT.

    (c) The OFT accepted commitments which were manifestly inappropriate (contrary to the purpose and object of the legislative regime) and contained clear and unjustified restrictions of competition.

  175. The CMA denies that the Commitments contain restrictions of competition by object. Rather, it says that their purpose is pro-competitive as they seek to ensure the effectiveness of the discounting permitted by the Commitments in light of the features of the market. Referring to the economic and factual context of the online hotel booking market, the CMA explained that the OFT reasonably expected that the availability of discounts to consumers who had joined a closed group and made at least one full-price booking would increase competition between OTAs, and between OTAs and hotels, as they would compete to attract customers to their closed groups.
  176. Finally, Skoosh refers to the OFT's treatment of rate parity clauses in the Statement of Objections (see paragraphs 13 - 15 above).
  177. C. Analysis

    (i) The Law

  178. We have described the legal framework governing the OFT's (now CMA's) power to make commitment decisions. Essentially, the CMA may accept commitments for the purposes of addressing the competition concerns it has identified. We heard argument as to whether such commitments need to be "appropriate" or "manifestly appropriate" or merely "sufficient" and whether they need to "address" or to "fully address" the CMA's concerns. We are content to adopt the CMA's own formulation, in the Guidelines to which we referred earlier (see paragraph 22). That is, commitments will normally be accepted where the competition concerns are readily identifiable and fully addressed by the commitments; and the proposed commitments are capable of being implemented effectively and, if necessary, within a short period of time.[24]
  179. We have mentioned the dearth of judicial authority here and at the EU level on what is a relatively new administrative practice. We have said that whilst it is not our function to adopt an over-intrusive scrutiny of commitment decisions, which are an important and useful contribution to overall competition enforcement, the CMA cannot have a completely free hand and is subject to normal considerations of judicial review.
  180. (ii) The Illegality Argument

  181. We first consider the Decision before deciding whether, as claimed, it frustrates the purpose of the Act in that it does not promote competition to the benefit of consumers.
  182. a. The Decision
  183. The Decision represents a balance struck between the OFT's initial position as set out in the Statement of Objections and the strongly opposing views of the Commitment Parties.
  184. The OFT's starting point was that the Price Agreements previously in force, which severely limited discounting by OTAs: (i) infringed the Chapter I prohibition and Article 101(1) TFEU; (ii) did not fall within the Vertical Agreements Block Exemption[25]; and (iii) were presumed unlikely to fulfil the requirements of Article 101(3).[26] The OFT regarded the restrictions as amounting to resale price maintenance and therefore infringements "by object". It also contemplated imposing a penalty.
  185. The Commitment Parties considered, by contrast, that some limitation on discounting was essential to protect a certain level of return for hotels (and presumably also OTAs). Further, unrestricted discounting would be harmful to consumers overall. In this regard, after the publication of the Statement of Objections, the Commitment Parties made submissions to the OFT in respect of potential efficiencies associated with the restrictions on discounting that were the subject of the OFT's Statement of Objections.[27]
  186. In relation to the efficiency arguments put forward by the Commitment Parties, the OFT obviously thought there was some validity in these arguments as it recognised, at paragraphs 6.53 – 6.54 of the Decision, that unrestricted discounting may potentially have harmful effects:
  187. "[F]reedom by OTAs to discount hotel accommodation without any restrictions may potentially have harmful effects by reducing the incentives of hotels to deal with OTAs (or to limit the number of OTAs that they deal with) thereby potentially damaging inter-brand competition, and chilling innovation in the development of new business models." (6.53)
    "Risks could also be created by requiring a greater degree of price freedom than provided for by the Final Commitments ...might jeopardise the possible realisation of efficiencies put forward by the Parties." (6.54)
  188. In the end, however, it took no position on the efficiency arguments as, in a commitments decision, it did not have to.[28]
  189. What the OFT did do, and this is the essence of the Decision, was accept Commitments which, although they "do not allow for unrestricted discounting", may be expected "to result in greater price competition, where there may currently be none or it may be significantly restricted, as well as lowering barriers to entry".[29]
  190. In a telling passage in the summary of responses at Annexe 3 to the Decision, the OFT says:
  191. "Based on its assessment of the evidence available, the OFT has sought to strike the right balance in terms of the intervention needed in this sector to address its competition concerns".[30]
  192. At the hearing, this was put in terms of "half a loaf is better than no bread" (i.e. some discounting, with restrictions, is better than no discounting at all).
  193. b. Striking a balance
  194. We now consider whether it was legitimate for the OFT to seek to strike a balance in terms of the appropriate level of intervention in this case.
  195. First, much has been made of the Alrosa case. However, the commitments in Alrosa did not strike a balance between two positions. They were instead a complete cessation of the conduct objected to. The parallel here would be if the Commitment Parties had agreed to abandon all agreements which restricted discounting. Instead they offered a system of limited discounting under certain conditions, including a limitation to closed groups of customers and a restriction on disclosure of actual discounts outside the closed group.
  196. Despite the arguments put forward by Mr Sinclair for Skoosh, we see no reason why the OFT should not, in the exercise of its judgement, "strike a balance" when accepting commitments. However, the OFT must be vigilant that the conditions it has set for itself in its own guidance as to effectiveness and ease of implementation are observed. It must also ensure that when striking a balance in this way (which involves giving some preference to a comparison with pre-existing rather than possible alternative conditions of competition) it takes proper account of material points drawn to its attention and avoids obvious error.
  197. We now consider Skoosh's argument that the OFT used the wrong counterfactual.
  198. c. The "Counterfactual"
  199. Skoosh argued that the OFT used the wrong counterfactual when it compared the conditions of competition before the Commitments with those following them. Skoosh did not think the OFT could proceed in this way, and ought to have set its sights higher. The CMA contended that the OFT not only considered the pre- and post-Commitments positions - which it said was a relevant exercise - but also considered the situation that would exist absent the conditions of the Commitments. In any event, the CMA rejected Skoosh's "rigid" distinction between the two counterfactuals (i.e. the situation before the Commitments and a situation with no pricing restrictions at all).
  200. The term "counterfactual" is a term used in competition analysis to describe the basis of comparison for any particular assessment. In this case, the OFT clearly compared the situation which would be created by the Commitments with the then present situation. Indeed, the Decision states expressly: "the OFT considers that the Final Commitments address its competition concerns by allowing for a greater degree of price competition than currently exists".[31] Contrary to Skoosh's submission, the OFT also considered a situation in which there was unrestricted discounting (what Skoosh described as the position without the restrictions). This can be seen in the OFT's analysis of the efficiency arguments put forward in favour of limited discounting, which we have extracted in paragraph 125 above.
  201. While it took no final view on the point, the OFT evidently had regard to a situation in which there were no discounting restrictions. Accordingly, we cannot accept Skoosh's argument that the OFT's analysis was flawed due to its use of the wrong counterfactual.
  202. d. The Residual Restrictions
  203. We heard much argument as to whether the so-called "residual restrictions" (i.e. the conditions governing the allowance of discounts in Principle 18) were themselves restrictive of competition. Mr Sinclair, for Skoosh, argued that Principle 18, which inter alia sets limits on the amount of discounting, and requires the use of closed groups, itself infringed Article 101(1) if not by object, then by their effect, which the OFT had in any case declined to investigate. We have explained at paragraphs 127 - 128 how the OFT saw these restrictions as being essentially pro-competitive within the context of "the right balance in terms of intervention needed." Whether or not we agree with this, with all due respect to Mr Sinclair, we do not need to decide this point, as it is not Skyscanner's contention. Skyscanner is essentially concerned with the restriction on disclosure contained in Principle 19.
  204. e. The Restriction on Disclosure
  205. We now consider Skyscanner's argument that OFT acted contrary to the purposes of the Act in accepting Commitments which contained the Principle 19 restriction on disclosure.
  206. Reasons for the Restriction on Disclosure

  207. We were told at the hearing that the main reason why the restriction on disclosure was so important was because of the prevalence of rate parity obligations or MFNs, which are often triggered by display of public rates for hotel rooms. If the discounted rates were visible across the board, the hotels would be required to honour their MFNs and offer the same rates to other OTAs. Mr Rasmussen, of the CMA, explained in his second witness statement that this created a risk that hotels offering MFNs would refuse to deal with OTAs offering discounted public rates given the potential cost of honouring the hotel's MFN. In particular, he said:
  208. "A mechanism was proposed to allow OTAs and hotels to offer discounts that would not be publicly available and thus not subject to MFNs. This mechanism was the Closed Group. For this to be a potentially appropriate way of addressing the OFT's concerns, the OFT wanted to ensure that being a member was sufficiently easy so as not to discourage consumers from signing up."
  209. The OFT's dilemma appears to have been that, to avoid the problems created by the prevalence of rate parity obligations, it had to ensure that the prices were not "publicly available", whilst at the same time being "sufficiently easy" for the public to find. As prices on price comparison websites were inherently "publicly available", disclosure by them had to be prohibited.
  210. It is not entirely clear whether it was the OFT or the Commitment Parties who originally proposed this restriction, but the OFT adopted it and the CMA has defended the need for it before us. Restricting disclosure of discounted rates is, as Ms Bacon argued, an essential feature of the particular model of limited discounting through closed groups that the OFT accepted as sufficient to meet its concerns about the complete absence or at most very limited level of discounting under the previous arrangements.
  211. Effect on Skyscanner

  212. Skyscanner contends that faced with this restriction on disclosure, particularly as it becomes more prevalent as the CMA intends, it will no longer be able to show the actual discounts available on its website. There was some argument before us as to whether the disclosure restriction would necessarily harm Skyscanner's meta-search business (i.e. whether an indication that a discount was available at a particular location would be sufficient) and indeed whether consumers would suffer overall as a result. For the reasons that follow, we are satisfied that the restriction on disclosure is likely to harm Skyscanner's business, but we cannot be so certain about the overall effect on competition and consumers.
  213. On the question of harm to Skyscanner's business, Ms Bacon sought to convince us that it was relatively straightforward for Skyscanner's customers to operate its website under the new regime. The Commitment Parties were permitted to disclose that, in general terms, a particular OTA or hotel offered discounts. Skyscanner could then indicate in its search results that discounts were available at particular hotel for particular dates when purchased on a certain OTA/hotel website. It was then, so the CMA said, comparatively straightforward for the Skyscanner user to click through to that OTA/hotel website, join the closed group if he or she had not already done so, and see the actual discounted price. It was not necessary to make a full price booking through the closed group to see what discount was available.
  214. We are not, however, convinced. It cannot seriously be contested that requiring would-be customers to conduct additional exercises and searches is a more complex and hence less straightforward exercise than making the actual discount immediately visible and that Skyscanner is right to fear that its consumers will find this less attractive. Moreover, the specific explanation of the ease with which consumers might use meta-search despite the restriction on disclosure being operative does not appear to have been discussed in correspondence or at any meeting with Skyscanner.
  215. In our view, Skyscanner is much better placed than the OFT to judge whether or not its way of doing business is likely to be affected and we accept Skyscanner's claim that the ease of access for consumers to specific price information will have changed substantially for the worse as a result of the Commitments and that the restriction is self-evidently harmful in that respect.
  216. Effect on Competition

  217. The CMA rightly points to the need to consider the effect of the Commitments on competition and consumers overall, as opposed to the effect on the interests of an individual competitor, such as Skyscanner. As Ms Bacon put it to us, the publicity restriction "does not damage the consumer because of the increased competition because there is discounting competition". That argument would perhaps carry more weight if the OFT had not itself, in the Statement of Objections and elsewhere, pointed to the importance for competition in the online hotel booking market of price transparency and price comparison websites. Nevertheless, we agree that weighing the possible benefits of intra-brand competition in terms of more discounting against possible harm to inter-brand competition from less price transparency is a matter of judgement. We also accept that it is theoretically possible (although we rather doubt it) that the Commitments might, in the light of this weighing in the balance, be found to benefit consumers "in the round". We would merely observe that, whilst it is of course correct that the interests of Skyscanner do not automatically coincide with the interests of the consumer, in this particular case price comparison websites fulfil a very particular purpose in promoting price transparency and the relative interests of competitor and consumer are much more likely to be the same here than may generally be the position.
  218. Rate Parity Obligations

  219. It was also put to us that by not pursuing its case against rate parity obligations, the OFT had "pulled its punches" in some way contrary to its statutory duty. The CMA, however, referred us to Principle 21 of the Commitments, which obliges the Commitment Parties to amend MFN provisions if and when they would otherwise prevent the discounting allowed by the Commitments. The CMA contends that this demonstrates that the OFT did take the MFN issue into account in its Decision. Other national competition authorities have taken a somewhat different approach. For example, it was explained to us that the German competition authority, investigating similar issues in Germany, had moved against rate parity obligations themselves. It is clear that MFNs often raise complex issues for competition assessment, as can be seen by the CMA's findings in the private motor insurance market investigation.[32] These issues are beyond the scope of this judgment and we do not find it necessary to decide whether the OFT was right or wrong in not identifying MFNs as a standalone competition concern.
  220. We consider rate parity obligations here only as part of the underlying reasoning for Principle 19. The CMA explained at the hearing that the restriction on disclosure was needed, for the most part at least, because of the continued prevalence of these obligations. The paradox of allowing one potential restriction of competition to continue because of the continuance of another looks at first sight worrying. But the CMA emphasised that, even in the Statement of Objections, the OFT did not identify rate parity obligations as in themselves unlawful, and was concerned more with their effect of exacerbating the restrictions on discounting that were its main concern. Whilst we appreciate that the OFT was obliged to recognise the continued existence of rate parity obligations and take this into account in framing the Commitments, this did not give the OFT a completely free hand to reach any kind of commitments decision without properly considering the likely consequences of the obligations contained in it.
  221. What the Decision Said

  222. The Decision itself does not address the potential consequences of the restriction on disclosure with any specificity. By repeating the assertion in the second consultation response that consumers will receive generalised information about the availability of discounts, it leaves unanswered the complaint that consumers need specific pricing information. There is no mention at all of the effect on meta-search in the body of the Decision. In the summary of responses to the second consultation in Annex 3, we find two paragraphs (66 and 74, extracted at paragraphs 64 - 66 above) that refer to Skyscanner's complaint. The OFT's response to Skyscanner's concern in paragraph 74 noted that "OTAs and hotels are free to publicise information on the general availability of discounts in a clear and transparent manner, including to price comparison web sites and meta-search sites (that is, to members and non-members)". Further, the OFT confirmed its view that the Commitments, including the provisions as to advertising, are sufficient to address its competition concerns, which relate to intra-brand competition and barriers to entry for OTAs.
  223. Ms Bacon, for the CMA, argued strongly that the OFT did not consider that the restriction on disclosure would harm the meta-search model or competition generally because it "did not have anything indicating that because the actual rate was not able to be displayed on the meta-search site...consumers would be driven away from those sites". We have discussed in relation to the previous Ground of Appeal the fairness of that position, bearing in mind the points put to the OFT by Skyscanner and others. It seems to us to be incorrect to say that the OFT did not have "anything"; it had the objections raised by Skyscanner and others, which it had declined to investigate properly.
  224. Conclusion on Illegality

  225. In allowing restrictions on the disclosure of specific price information, the OFT has limited the availability of the kind of information used by meta-search and other price comparison websites to enable consumers to make direct and immediate comparisons of actual prices of available hotel rooms. Restricting access to such information may be seen as an obvious error and were this an appeal "on the merits" we might well wish to quash the decision on these grounds. But this is an application for judicial review and the grounds on which we can intervene are more limited.
  226. Skyscanner claims that the OFT, by taking the decision, has breached its duty to promote competition and has therefore acted illegally. Attractive though this argument may be, we cannot accept it. This is because assessing whether or not a particular restriction on conduct restricts competition – even where the restriction appears to be "self-evidently harmful" – is a matter of appreciation and expert judgement. As Ms Bacon has pointed out, there may be (although we cannot necessarily see them) arguments that even the restriction on disclosure is not, when considered in the round, anti-competitive. Somewhat reluctantly, therefore, we have come to the view that this is a case where expert appreciation is needed and, as this is an application for judicial review, we must refrain from substituting another assessment for that of the OFT.
  227. There must also be some doubt as to the precise definition and scope of the OFT's legal duty that Skyscanner has asserted and the CMA has accepted existed. It is at best a high-level requirement that would need to be pinned down and applied in a particular context. To show that a particular measure is in breach of the duty requires precisely the sort of assessment that we have said should have been carried out in this case. It is not possible to state with a sufficient degree of certainty on the information available to us that the restriction on disclosure necessarily amounts to a restriction of competition, even though it may indeed appear to be so. Therefore, we are unable to find that the OFT infringed any overriding duty not to frustrate the purposes of the Act.
  228. For the above reasons, Skyscanner's appeal on the particular ground of illegality fails.
  229. (iii) The Irrationality Argument

  230. We now consider Skyscanner's alternative claim that the Decision was irrational. We described the relevant legal principles at paragraphs 106 - 109. In essence, Skyscanner must show, in order to succeed here, that no reasonable authority could have taken the decision that the OFT took on the basis of the evidence before it.
  231. We have already concluded that the OFT failed properly to consider Skyscanner's objection to the proposed commitments. The CMA has argued that Skyscanner's complaints under Grounds 2 and 3 are essentially the same and all arise from whether the OFT acted reasonably or not. To an extent we agree with this view, but we disagree that Skyscanner has no case under irrationality. To the contrary, we find that the OFT acted unreasonably not only in failing properly to consider the objection (Ground 2) but also in coming to a decision that effectively ignored the point Skyscanner and others had raised in relation to the potential impact of Principle 19 on meta-search and competition more generally.
  232. The CMA argued strongly that whether or not a point required further investigation or whether commitments addressed its competition concerns were entirely matters for the competition authority and were not susceptible to judicial review. We disagree. Whilst the authority enjoys a substantial margin of appreciation in exercising its judgement, where it makes a decision that raises obvious competition concerns that have on its own admission not been fully assessed, the Tribunal can and should intervene. This is entirely consistent with the approach of Advocate General Kokott in paragraph 77 of her opinion in the Alrosa case to which we have referred.
  233. In this case, whilst we agree that if the OFT had investigated the possible impact of the restriction on disclosure it might conceivably have come to the conclusion that the restriction did not harm competition, it failed to do so. In coming to the view that the OFT did not acquaint itself with the information needed to answer the statutory questions posed to it and, accordingly, failed to take into account matters it ought to have taken into account, we have regard to the following considerations:
  234. (a) the OFT had been aware in the investigation of the importance of price transparency in promoting inter-brand competition and had referred to this in its Statement of Objections;

    (b) in response to its first consultation, two respondents had warned of the likely damage to meta-search operators that the proposed commitments would cause;

    (c) in response to its second consultation, Skyscanner had made the same point in even clearer terms;

    (d) the OFT attached importance to price transparency in the Statement of Objections (as we cover at length in paragraph 75 above), but then failed to sustain consideration of the importance of the role of meta search operators in promoting price transparency in the focus group qualitative research conducted by Opinion Leader, in the two consultations and accompanying press releases, and in the Decision itself; and

    (e) when Skyscanner repeated its plausible concerns about the potential harm to competition that the restriction on disclosure might cause, at the meeting on 20 January 2014, the OFT failed to postpone or alter its proposed decision in any significant way.

  235. We therefore find that the procedural impropriety under Ground 2 finds its reflection in the irrationality of the Decision itself under Ground 3. The OFT took this without informing itself about the possible impact on price transparency of an obvious and clear restriction on disclosure of price information. In this way it failed to take account of a matter of which it ought to have taken account and acted as no reasonable authority should act. We therefore find that Skyscanner's appeal succeeds on this ground also.
  236. (iv) Other Arguments

  237. It remains for us to consider other arguments that are not central to this judgment on this ground of appeal. First we do not have to follow Mr Sinclair, counsel for Skoosh, into the detailed consideration of the architecture and legal hierarchy of Article 101(1) and (3) that he invited us to undertake. This is not an infringement finding and we agree with the CMA that this is more in the nature of a regulatory decision subject only to the normal control of judicial review. Nor, as we have said, do we have to decide whether the existence of so called "residual restrictions" in the Commitments was itself an infringement of the law and whether these were restrictions "by object" or "by effect". We also do not have to consider, as Mr Sinclair suggested at the hearing we should, whether a private action or defence could cover the same ground as this appeal and come to a different conclusion.
  238. Skoosh also asked us to apply the observations of Advocate General Kokott in the Alrosa case to the effect that, because commitments were a pragmatic solution without a finding of infringement, higher demands had to be made as to the appropriateness of the commitments, and in this sense they had to be manifestly appropriate (see Opinion at [53], discussed earlier at paragraph 40(d)). Attractive though this argument might appear, we do not have to adopt it. The Court of Justice did not endorse this aspect of the Advocate General's opinion and, in any case, Ms Bacon for the CMA accepted that the Commitments must, in the CMA's view, be "appropriate" and "fully address" its concerns, as the CMA claimed the Commitments did. We therefore see no additional assistance to be gained from this point.
  239. Skoosh also argued that the OFT had misapplied the requirement of "procedural economy" and had used this concept to "cut corners" and conduct an insufficient analysis, resulting in a decision which contained restrictions of competition. We have dealt with this argument in part in considering the basis of comparison that the OFT was entitled to make. For the rest, we do not find that, in the context of the commitments process, the OFT misapplied the requirement of procedural economy. The OFT was permitted to accept commitments that addressed its competition concerns without having to carry out a detailed analysis of every other competitive issue that might potentially arise from the conduct under investigation.
  240. VI. DID THE OFT ACT ULTRA VIRES BY IMPOSING REQUIREMENTS ON THIRD PARTIES? (GROUND 1)

  241. Skyscanner contends that the Commitments impose requirements on third parties who did not sign up to them. In doing so, Skyscanner argues that the OFT acted ultra vires its powers as set out in section 31A of the Act. This was originally pleaded as Ground 1.
  242. Skyscanner refers to section 31A(2) of the Act, which states that the OFT (now CMA) may - for the purposes of addressing the competition concerns it has identified - accept from such person (or persons) concerned, as it considers appropriate, commitments to take such action (or refrain from taking such action) as it considers appropriate. In essence, Skyscanner argues that the Commitments require compliance by other OTAs and hotels who were not investigated by the OFT and did not sign up to them.
  243. A. The Parties' Submissions

  244. Skyscanner contends that section 31A(2) of the Act does not give the OFT the power to require third parties who do not offer commitments to act in a particular way. Those third parties were not accused by the OFT of engaging in anti-competitive conduct so as to make them the subject of the OFT's investigation under section 25 of the Act. The potentially anti-competitive conduct identified by the OFT was that engaged in by the Commitment Parties only. They are the only parties which can be required to take action (or refrain from taking action) as a result of the Commitments. Skyscanner explains that the Commitments bind third parties in the following ways (see paragraphs 11 - 12 and 14 - 17 of the Commitments):
  245. (a) The Commitment Parties each agreed to clarify or amend their existing commercial arrangements with (i) other OTAs in the case of IHG, and (ii) other hotels in the case of Booking and Expedia, in order that those commercial arrangements comply with the Principles and do not contain any provisions that are incompatible with the Principles. Where a third party's consent is required to amend the previously agreed commercial terms, the Commitment Parties must use their reasonable endeavours to procure such consent.

    (b) Similarly, the Commitment Parties each agreed that their new commercial arrangements with either other OTAs or other hotels would comply with the Principles and would not contain any provisions which were incompatible with the Principles. Therefore, the only commercial arrangements available to other OTAs or other hotels contracting with the Commitment Parties for the duration of the Commitments are arrangements which comply with the Principles. To this extent, the Commitments do impose requirements on third parties.

  246. Skyscanner also argues that it was always the intention of the OFT that the Commitments would impose requirements on parties other than the parties that had offered them to the OFT. By way of example, Skyscanner refers to paragraph 3.4 of the Decision in which the OFT said that, although it had limited its investigation to a small number of major companies, the alleged practices were potentially widespread in vertical distribution arrangements in the industry. Further, the Commitments invite the Commitment Parties to notify the OFT of contractual arrangements of other hotels and/or OTAs which may be incompatible with the Principles - there would be little reason to do this unless the OFT was expecting third parties to comply with the Principles.
  247. In its Defence, the CMA makes the following points in response to Skyscanner's arguments:
  248. (a) It is an inherent feature of commitments accepted by the OFT (now CMA) and by the Commission that they have changed the behaviour of the person (or persons) that offered the commitments and thereby had an effect on third parties. This is the lawful and normal consequence of commitment parties having agreed not to enter into commercial arrangements on terms that appear to the OFT (now CMA) to give rise to competition concerns.

    (b) The OFT did not intend for the Commitments to affect dealings wholly between third parties and they do not do so. The purpose of inviting the Commitment Parties to notify non-compliant third party arrangements was simply to enable the CMA to monitor whether such agreements contained clauses of the type which gave rise to the OFT's investigation.

    (c) The Commitments are a minimum obligation and so do not prevent third parties from agreeing with the Commitment Parties that discounts may be disclosed and advertised outside a closed group. However, Skyscanner responds that - given the history - it is fanciful to suggest that the Commitment Parties would be happy to go further than they are required to by the Commitments.

    B. Analysis

  249. Skyscanner did not press this ground of appeal at the hearing. In our view, this was a wise decision. Having considered Skyscanner's arguments, we are not persuaded that the Commitments bind third parties, other than in the sense that third parties who deal with the Commitment Parties may be affected by them as the Commitment Parties will be unable to contract in terms which breach the Commitments. We do not see anything unusual in this. Indeed, the legitimate purpose of the Commitments is to change the Commitments Parties' behaviour in the market for the online supply of hotel room bookings. In our judgment, it is not ultra vires for the OFT to accept commitments which affect the terms upon which the parties to those commitments may contract with third parties.
  250. (i) Other Cases

  251. Parties offering commitments may agree not to enter into certain terms with third parties where, for example, it is that term which has given rise to competition concerns. It is entirely appropriate for the OFT (now CMA) to accept a commitment of that nature. The CMA has provided several examples of where this has happened in the past. For example:
  252. (a) In Repsol[33], the Commission concluded that non-compete clauses contained in distribution agreements for motor fuel between Repsol and service station operators may be creating a foreclosure effect on the retail market in Spain. Repsol proposed a series of commitments in which it agreed, among other things, not to sign a new exclusive supply contract with anyone that exceeded five years. This was a measure that affected third parties, but legitimately resolved the Commission's concerns. Accordingly, the Commission accepted Repsol's commitments.

    (b) In Supply of road fuels in Western Isles[34], the OFT accepted commitments from Certas Energy to address concerns that Certas might have abused a dominant position by entering into long-term exclusive agreements for the supply of road fuels to filling stations in the Western Isles. Certas agreed to terminate its existing contracts with the relevant filling stations and agreed to provide three options for continued supply, with varying pricing mechanisms and durations. These commitments had an effect on those filling stations since they could not continue their existing contracts and instead must choose one of the specified options.

  253. These examples demonstrate that it is normal for commitments to have an effect beyond the immediate parties to those commitments.
  254. At the hearing, Ms Smith for Skyscanner argued that Repsol, Road Fuels and the other examples of commitments affecting third parties can be distinguished from the present situation. This was because the effects in those cases were to prevent those third parties from engaging in anti-competitive behaviour, which was very different from our situation where third parties have to accept restrictions on their commercial freedom and ability to compete. In our view, this criticism concerns the substance of the Commitments themselves, which we have already considered in relation to the third Ground.
  255. (ii) Conclusion

  256. We find that the OFT did not act ultra vires its powers by accepting commitments which affect third parties. Accordingly, this Ground of Appeal fails.
  257. (iii) General Comment

  258. Finally, we must emphasise that we do not in any way wish to place unnecessary limits on the CMA's use of commitment decisions or to impose any unnecessary or overly intrusive degree of judicial oversight in an area that is essentially one where the CMA must exercise its judgement. We are willing to concede a large margin of appreciation to the CMA in cases of this kind (although we would emphasize that, ipso facto, commitment decisions are not instances where complex economic evidence and analysis need to be considered) to allow it to exercise its expert judgement. The Tribunal must, however, intervene under the normal principles of judicial review where there has been a clear error.
  259. VII. CONCLUSION

  260. For the reasons set out above, we unanimously conclude that Skyscanner succeeds in its appeal on Grounds 2 and 3, but fails on Ground 1. We therefore quash the Decision and remit it to the CMA with a direction to reconsider the matter in accordance with our Judgment. We do not consider it appropriate to make any further directions.





  261. Peter Freeman CBE, QC (Hon)





    Brian Landers





    Stephen Wilks
         




    Charles Dhanowa OBE, QC (Hon)
    Registrar
     




    Date: 26 September 2014

Note 1    Decision, paragraph 6.39    [Back]

Note 2    Decision, paragraph 1.4    [Back]

Note 3    Decision, paragraph 5.11    [Back]

Note 4    Decision, paragraph 5.11    [Back]

Note 5    Decision, paragraph 5.10    [Back]

Note 6    Enforcement (OFT 407) December 2004 (adopted by the CMA), Part 4    [Back]

Note 7    Ibid., paragraph 4.3    [Back]

Note 8    Decision, paragraphs 6.43 – 6.65    [Back]

Note 9    http://webarchive.nationalarchives.gov.uk/20140402142426/http://www.oft.gov.uk/shared_oft /ca-and-cartels/hob-annexe1%282%29.pdf    [Back]

Note 10    Decision, paragraph 6.62 (which refers to “a minimum standard only”)    [Back]

Note 11    BAA at [20(3)]; Secretary of State for Education and Science v Tameside Metropolitan Borough Council [1977] AC 1014, 1065B per Lord Diplock; Barclays Bank plc v Competition Commission [2009] CAT 27 at [24]    [Back]

Note 12    BAA at [20(3)]; Tesco plc v Competition Commission [2009] CAT 6 at [138]-[139]    [Back]

Note 13    BAA at [20(3)]; R (Khatun) v Newham London Borough Council [2005] QB 37 at [34]-[35]    [Back]

Note 14    Ibid, and R v Royal Borough of Kensington and Chelsea, ex p. Bayani (1990) 22 HLR 406 at 415 per Neill LJ    [Back]

Note 15    BAA at [20(4)]    [Back]

Note 16    BAA at [20(8)]    [Back]

Note 17    The Commission had issued two statements of objections. The first was addressed to both Alrosa and De Beers and concerned a potential breach of Article 101 TFEU. The second was addressed to De Beers only and concerned a potential breach of Article 102 TFEU. The commitments at issue in Alrosa were directed to address the latter.     [Back]

Note 18    Decision, 6.43 – 6.65 and paragraphs 9 – 39 of Annexe 3    [Back]

Note 19    Decision, Annexe 3, paragraphs 9 - 39    [Back]

Note 20    Decision, Annexe 3, pp. 1 - 15    [Back]

Note 21    Decision, Annexe 3, paragraph 20    [Back]

Note 22    Decision, Annexe 3, paragraphs 22 - 39    [Back]

Note 23    Section 31B(4) of the Act    [Back]

Note 24    Enforcement (OFT 407), paragraph 4.3 (which has been adopted by the CMA)    [Back]

Note 25    Regulation (EC) 330/2010 [2010] OJ L 102/1    [Back]

Note 26    Statement of Objections, paragraphs 4.161 and 4.283    [Back]

Note 27    These efficiency arguments are summarised at Annexe 2 to the Decision    [Back]

Note 28    Decision, paragraphs 6.53 and 6.55 and Annexe 3, paragraphs 43 and 44    [Back]

Note 29    Decision, paragraph 6.52    [Back]

Note 30    Decision, Annexe 3, paragraph 45    [Back]

Note 31    Decision, paragraph 6.61    [Back]

Note 32    https://www.gov.uk/cma-cases/private-motor-insurance-market-investigation    [Back]

Note 33    Decision of 12 May 2006    [Back]

Note 34    Decision of 24 June 2014    [Back]


BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII
URL: http://www.bailii.org/uk/cases/CAT/2014/16.html