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

United Kingdom Competition Appeals Tribunal


You are here: BAILII >> Databases >> United Kingdom Competition Appeals Tribunal >> Celesio AG v Office Of Fair Trading [2006] CAT 9 (9 May 2006)
URL: http://www.bailii.org/uk/cases/CAT/2006/9.html
Cite as: [2006] CAT 9

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


    Neutral citation [2006] CAT 9

    Case No: 1059/4/1/06

    IN THE COMPETITION APPEAL TRIBUNAL

    Victoria House

    Bloomsbury Place

    London WC1A 2EB

    9 May 2006

    Before:

    Marion Simmons QC (Chairman)
    Professor Andrew Bain
    Vivien Rose

    Sitting as a Tribunal in England and Wales

    _____________________

    BETWEEN:

    CELESIO AG

    Applicant

    -v-
    OFFICE OF FAIR TRADING

    Respondent

    supported by

    BOOTS GROUP PLC

    -and-

    ALLIANCE UNICHEM PLC

    Interveners

    _____________________

    Mr Mark Hoskins and Miss Kelyn Bacon (instructed by Linklaters) appeared for the Applicant

    Mr Peter Roth QC and Mr Daniel Beard and Mr Julian Gregory (instructed by the Solicitor to the Office of Fair Trading) appeared for the Respondent

    Mr Nicholas Green QC and Miss Maya Lester (instructed by Slaughter and May and Allen & Overy LLP) appeared for the Interveners

    Heard at Victoria House on 11 April 2006

    _____________________

    JUDGMENT
     

    JUDGMENT (Non-confidential version)

    Note: Excisions in this judgment (marked "[…][C]") relate to commercially confidential information: Schedule 4, paragraph 1 to the Enterprise Act 2002.

    TABLE OF CONTENTS

    I INTRODUCTION 1
    II LEGISLATIVE FRAMEWORK 5
    III BACKGROUND 9
        The parties 9
        Investigations into the retail pharmaceutical sector 12
        Procedure before the OFT 17
    IV THE CONTESTED DECISION 34
        Proposed Undertakings 52
    V THE PROCEEDINGS BEFORE THE TRIBUNAL 55
        The original ground (a) in the notice of application 62
    VI THE DUTY OF THE OFT TO REFER 66
    VII THE STANDARD AND NATURE OF REVIEW 75
    VIII THE EVIDENCE 78
        A. CELESIO'S EVIDENCE 78
            The LECG evidence 78
            The evidence of Mr Justin Ash 82
        B. THE INTERVENERS' EVIDENCE 87
        C. THE OFT'S EVIDENCE 89
    IX THE PARTIES' SUBMISSIONS ON THE SUBSTANCE 91
        AMENDED GROUND 1 91
            Celesio's submissions 91
            The OFT's submissions 104
            The interveners' submissions 111
        AMENDED GROUND 2 120
            Celesio's submissions 120
            The OFT's submissions 126
            The interveners' submissions 128
        AMENDED GROUND 3 132
    X THE TRIBUNAL'S ANALYSIS 134
        The test for the admissibility of Pritchard 134
        Application of the admissibility test to Pritchard 144
        The Decision in respect of 4 to 3 areas 172
            - SLC in the context of weak competition in the relevant markets 173
            - Share of outlets evidence 176
        4 to 3 areas 180
        Amended grounds 2 and 3 193
    XI OTHER MATTERS 194
        Third party material 194
        Submissions by Celesio to the OFT during the investigation stage 196
        Celesio's disclosure of the Lloyds/Cohens & Scholes merger 198
        Evidence on pharmaceutical services regulation 199
    XII CONCLUSION 200

    I INTRODUCTION

  1. By a notice of application dated 21 March 2006 the applicant, Celesio AG ("Celesio"), applied, pursuant to section 120 of the Enterprise Act 2002 ("the Act"), for judicial review of the decision of the respondent, the Office of Fair Trading ("the OFT"), made on 6 February 2006 and published on 22 February 2006 ("the Decision"). The Decision was not to refer the proposed acquisition of Alliance UniChem Plc ("UniChem") by Boots Group PLC ("Boots") to the Competition Commission ("the CC") under section 33(1) of the Act, provided that suitable undertakings were given pursuant to section 73 of the Act to address the potential competition concerns outlined in the Decision.
  2. The Tribunal's power of review is set out in section 120 of the Act as follows:
  3. "(1) Any person aggrieved by a decision of the OFT … under this Part in connection with a reference or possible reference in relation to a relevant merger situation or a special merger situation may apply to the Competition Appeal Tribunal for a review of that decision.
    (4) In determining such an application the Competition Appeal Tribunal shall apply the same principles as would be applied by a court on an application for judicial review.
    (5) The Competition Appeal Tribunal may –
    (a) dismiss the application or quash the whole or part of the decision to which it relates; and
    (b) where it quashes the whole or part of that decision, refer the matter back to the original decision maker with a direction to reconsider and make a new decision in accordance with the ruling of the Competition Appeal Tribunal."
  4. It is common ground that Celesio is a "person aggrieved" for the purposes of section 120(1).
  5. For the reasons set out below, we dismiss Celesio's application.
  6. II LEGISLATIVE FRAMEWORK

  7. Section 33(1) of the Act provides:
  8. "The OFT shall, subject to subsections (2) and (3), make a reference to the [CC] if the OFT believes that it is or may be the case that:
    (a) arrangements are in progress or in contemplation which, if carried into effect, will result in the creation of a relevant merger situation; and
    (b) the creation of that situation may be expected to result in a substantial lessening of competition within any market or markets in the United Kingdom for goods or services."
  9. Section 36(1) of the Act provides, in so far as relevant:
  10. "(1) Subject to subsections (5) and (6) and section 127(3), the [CC] shall, on a reference under section 33, decide the following questions-
    (a) whether arrangements are in progress or in contemplation which, if carried into effect, will result in the creation of a relevant merger situation; and
    (b) if so, whether the creation of that situation may be expected to result in a substantial lessening of competition within any market or markets in the United Kingdom for goods or services."
  11. Section 73 of the Act provides, in so far as material:
  12. "(1) Subsection (2) applies if the OFT considers that it is under a duty to make a reference under section 22 or 33 …
    (2) The OFT may, instead of making such a reference and for the purpose of remedying, mitigating or preventing the substantial lessening of competition concerned or any adverse effect which has or may have resulted from it or may be expected to result from it, accept from such of the parties concerned as it considers appropriate undertakings to take such action as it considers appropriate.
    (5) An undertaking under this section-
    (a) shall come into force when accepted;
    …"
  13. The OFT has issued guidance on its approach to assessing whether a merger may give rise to a substantial lessening of competition ("SLC") and on the relationship between the test that it has to apply and the test to be applied by the CC if a reference is made – see Mergers – substantive assessment guidance (OFT 516, May 2003). In the light of the Court of Appeal's judgment in Office of Fair Trading and others v IBA Health [2004] EWCA Civ 142, [2004] All ER 1103, the OFT published a Guidance note revising 'Mergers – substantive assessment guidance' (OFT 516a, October 2004).
  14. III BACKGROUND

    The parties
  15. Celesio (formerly GEHE AG) is a major pharmaceutical distribution and retail company with its headquarters in Stuttgart, Germany. It operates three divisions: Celesio Wholesale, Celesio Pharmacies and Celesio Solutions. In the United Kingdom, Celesio is engaged in pharmaceutical wholesaling through its subsidiary, AAH Pharmaceuticals Limited ("AAH"). On the retail side, it owns a national chain of over 1,520 pharmacies through its subsidiary, Lloyds Pharmacy Limited ("Lloyds").
  16. Boots is a leading health and beauty retailer with 1,423 stores across the United Kingdom (1,350 of which contain a pharmacy). It also operates as a pharmaceutical wholesaler for its own outlets. In addition, Boots has an opticians business and has some manufacturing capability, in particular in contract manufacture of "specials" pharmaceutical products and beauty and personal care products.
  17. UniChem is an international pharmaceutical distribution and retail company. In the United Kingdom it is engaged in pharmaceutical wholesaling and retail pharmacy with a chain of 958 pharmacies (currently trading under the name Moss but being rebranded Alliance Pharmacy).
  18. Investigations into the retail pharmaceutical sector
  19. The nature and extent of competition in the retail pharmacy sector has been the subject of investigation on a number of occasions over the last decade. It appears that this experience established the starting point for the merging parties' discussions of the contested transaction in terms of defining the relevant product and geographical markets and also identifying the competition issues likely to cause concern. In particular the 1996 Report by the Monopolies and Mergers Commission ("MMC") UniChem PLC/ Lloyds Chemists plc and GEHE AG/ Lloyds Chemists plc, A report on the proposed mergers (Cm 3344, July 1996) ("the 1996 MMC Report") had divided the market into three different product categories: "ethicals" (medicines that can only be dispensed against a prescription), "P medicines" (medicines that can be dispensed without a prescription but only by a pharmacist or dispensing doctor) and General Sales List medicines ("GSL medicines"), which can be sold by any retailer. P medicines and GSL medicines both fall into the general category of "over the counter medicines" ("OTCs").
  20. In the 1996 MMC Report the MMC adopted a "fascia reduction test" for assessing the degree of market concentration in a locality. This test involves determining the number of different competitors present in a given area, counting all outlets operated by a single competitor as one fascia, and considering the effect of the merger on the degree of concentration in terms of the reduction of fascia. In that Report, the MMC considered overlaps between retail pharmacies within a one mile radius.
  21. Subsequent to the 1996 MMC Report, there have been a number of other OFT investigations into the retail pharmacy sector, notably the market investigation under section 2 of the Fair Trading Act 1973 on The control of entry regulations and retail pharmacy services in the UK (OFT 609, January 2003) ("the 2003 Report").
  22. Immediately prior to the time that the OFT was investigating the present transaction, it was also considering the proposed acquisition by Lloyds of 110 Cohens and Scholes retail pharmacies. That acquisition was cleared by the OFT in a decision of 29 November 2005, which was published on 7 December 2005.
  23. The appropriateness of a one mile radius for the purpose of ascertaining the relevant geographic market has generally been accepted by the OFT and undertakings active in the sector as reflecting the general willingness of consumers to travel about a mile to find a pharmacy. The discussion of the impact of a merger of retail pharmacy chains has tended to focus on the reductions in the number of fascias in the one mile radius circle.
  24. Procedure before the OFT
  25. The Boots/UniChem transaction met the jurisdictional thresholds of the European Community Merger Regulation (Council Regulation (EC) No. 139/2004, OJ 2004 L24/1) ("ECMR"), but, following a request from the parties, the European Commission made a decision on 30 November 2005, under Article 4(4) ECMR, to refer the case to the OFT.
  26. In accordance with section 34A of the Act, the OFT was required within 45 working days of the European Commission's decision to decide whether to make a reference to the CC under section 33 of the Act.
  27. Also on 30 November 2005, the merging parties lodged an informal merger notification with the OFT, followed, on 1 December 2005, by a meeting between the merging parties and the OFT. During the course of its eight-week investigation the OFT made six formal information requests of the merging parties.[1]
  28. Included in the information provided to the OFT by the merging parties were the following: an "Overview of regulatory framework"[2], "Boots Consumer Surveys – Analysis of Store Catchments Summary"[3], m of 2 to 1 local overlap areas (one mile radius)[4], maps of 3 to 2 local overlap areas (one mile radius)[5], and "Overview of Local Overlap Areas where the [merging parties] are located nearest to each other"[6].
  29. On 2 December 2005, the OFT sought comments from all interested third parties by the usual means of an "Invitation To Comment" placed on Reuters' Regulatory News Service. It also contacted various market participants and other affected or knowledgeable parties directly to seek their views. These included more than 60 Primary Care Trusts ("PCTs") or Health Boards and over 35 other consultees including: industry and consumer bodies; national pharmacy chain competitors; full line and short line wholesale competitors; pharmaceutical manufacturers; and the Department of Health.
  30. On 7 December 2005, the OFT sent Celesio, amongst other pharmacies, a pharmacy questionnaire, inviting responses by 13 December 2005.
  31. Twenty-eight third parties, including Celesio, provided responses to the OFT. Celesio sent a briefing paper to the OFT on 13 December 2005 setting out its preliminary concerns regarding the merger. The concerns raised at that stage focused on the strength of the vertically integrated pharmacy chain which would be created by the merger. Celesio pointed in particular to the strength of the Boots brand name and the other advantages that Boots enjoyed in the market which would, in Celesio's view, "combine to create a gulf between the merged entity and other smaller competitors, including Lloyds" (briefing paper, paragraph 3.1). This created, Celesio argued, a real risk of market exit leading to a reduction in competition and a loss of choice for consumers.
  32. On the question of local market definition, Celesio said:
  33. "We have not yet had the opportunity to carry out a full local analysis across the UK but are concerned that there may be areas where, even if there were 4 or more competitors remaining after the merger, some of these would provide a very weak constraint on [the merged entity]. We also believe that in a number of areas there will be fewer than 4 competitors. Moreover, …we believe that there would be a risk of market exit in a number of cases.
    Given the step change that the proposed merger would bring about at the retail level as described above, Celesio submits that it is incumbent upon the OFT to undertake a detailed appraisal of the impact on competition at a local level. We note that issues regarding not only the minimum appropriate number of local competitors but also the nature of the competition between them, as well as questions as to how local market[s] should be defined, have all been important features of the CC's inquiries into supermarkets (the 2004 inquiry into the acquisition of Safeway plc and the 2005 inquiry into the acquisition of certain Morrison stores by Somerfield). These issues also need to be considered very closely in relation to this proposed merger." (Briefing paper, paragraph 3.2.)
  34. Celesio's briefing paper concluded with the statement "Given the OFT's duties under the Enterprise Act 2002, we believe this is a case which should be referred to the Competition Commission for a detailed review. This is because the parties' proposals would give rise to major, permanent and anti-competitive changes to the wholesaling and retail pharmaceutical markets" (paragraph 5).
  35. Celesio met OFT officials on 14 December 2005 to discuss these issues, and on 19 December 2005 Celesio lodged its main submission in relation to the proposed merger and its response to the retail pharmacy questionnaire. In this main submission, Celesio asserted that although in most cases it is appropriate for the OFT to undertake an analysis of the number of fascias in local markets, using a one mile radius as a starting point, there were special features of the Boots/UniChem merger that meant that the fascia reduction test could not "capture or cure the merger's effects on competition in the retail market" (paragraph 2.1). The special features referred to by Celesio were the reduction in the number of competitors operating on a national scale; the large market share it was alleged the merged entity would hold in the supply of GSL medicines; Boots' ability to use its brand strength in the newly acquired stores; and the merged entity's purchasing power and pricing power post-acquisition.
  36. Celesio followed up its main submission with a supplemental submission on 6 January 2006 which provided additional information regarding the retail market.
  37. A further meeting took place between the OFT and the merging parties on 11 January 2006.
  38. On 16 January 2006, the OFT telephoned Celesio's legal representatives and indicated that they were "welcome" to make any additional submissions but that timing was tight.
  39. Having evaluated the evidence received from the merging parties and third parties, on 17 January 2006 the OFT sent the merging parties an "Issues Letter" outlining the hypotheses supporting the case for a reference to the CC.
  40. On 20 January 2006, the Deputy Director of Mergers at the OFT chaired an "Issues Meeting" with the merging parties to allow them to respond to the hypothetical "case against". The merging parties' final submissions, including divestment remedy proposals, were made on 19 and 23 January 2006.
  41. An internal "Case Review Meeting" was held by the OFT on 25 January 2006 at which the relevant issues, including Celesio's submissions, were considered by senior OFT officials. Later that day, at a "Decision Meeting", also attended by the Chief Executive and the Director of Competition Enforcement, the OFT decided to refer the transaction to the CC unless satisfactory divestment undertakings were given by the merging parties in respect of localities where there would be a reduction in fascias from 2 to 1 or 3 to 2 as a result of the merger.
  42. The Decision was then drafted and subsequently adopted on 6 February 2006. A non-confidential version of the Decision was published on the OFT's website on 22 February 2006.
  43. IV THE CONTESTED DECISION

  44. The Decision is 25 pages long. It covers a number of issues that are not relevant to the matters raised by this application, for example the vertical issues arising in relation to, on the one hand, UniChem's wholesale operations and Boots' retail operations and, on the other, UniChem's wholesale and retail operations and Boots' manufacturing operations. In this judgment we focus on the parts of the Decision most relevant to the issues raised by Celesio in its challenge.
  45. The Decision describes the retail pharmacy sector, referring to earlier investigations, in particular the 1996 MMC Report. It adopts the definition of the three different product categories – ethicals, P medicines and GSL medicines – as the relevant product market (paragraphs 8 to 9). The OFT also concludes on the basis of the evidence before it that the geographic scope of its assessment should be considered to be a one-mile radius around each of the relevant pharmacies (paragraph 16). Neither the product categories nor the geographic scope were in issue in the present case.
  46. Turning to "horizontal issues", the OFT first considered the retail pharmacy aspects of the merger in so far as they concern provision of services to consumers. Looking at the extent and significance of national overlap between the merging parties, the OFT noted that the retail pharmacy sector in the UK is highly fragmented, there being large numbers of independent pharmacies. At the national level, the merging parties' share of supply by numbers of outlets would be 19 per cent with the increment from the merger being 8 per cent (paragraph 26; Table 1).
  47. The parties' shares of supply by value of different product categories were also noted by the OFT. The shares were set out in Table 2 as follows:
  48.             Dispensing     GSL medicines P medicines
      Revenue (£m) Share (%) Revenue (£m) Share (%) Revenue (£m) Share (%)
    Boots [1,000-1,200][C] [10-20][C] [300-500][C] [15-25][C] [100-200][C] [25-35][C]
    UniChem [700-900][C] [0-10][C] [0-100][C] [0-10][C] [0-100] [0-10]
    Boots/ UniChem [1,700-2,100][C] [15-25][C] [300-600][C] [20-30][C] [100-300][C] [30-40][C]

  49. The OFT stated that in relation to GSL medicines, the largest competitors to Boots were Tesco and Superdrug, with [5-15] [C] per cent and [0-10] [C] per cent of supply respectively. The increment to Boots arising from the merger with UniChem is very limited, with GSL medicines only accounting for about [less than 5] [C] per cent of UniChem's pharmacy turnover (paragraph 28).
  50. In relation to P medicines, the OFT noted that supermarket shares were more limited. Boots was significantly larger than its competitors with a [25-35] [C] per cent share. Lloyds and UniChem were its largest competitors, holding [0-10] [C] per cent and [0-10] [C] per cent share respectively. The OFT considered that whilst the merger would remove the third largest supplier in this segment, and would strengthen Boots' position, there were a number of reasons why competition at a national level would not be reduced. The parties' evidence suggested that UniChem was not a particularly aggressive competitor to Boots in this segment, P medicines accounting for only [less than 5] [C] per cent of UniChem's revenues. Supermarkets were said by the parties to be the [most significant individual competitors] [C] in both P and GSL medicines; the merger would not alter this position. Moreover, Boots had provided [...][C]. Supermarket pharmacies had reduced prices for a basket of P medicines by 30 per cent shortly after the ending of resale price maintenance ("RPM") in 2001. Finally, the OFT noted that around 40 per cent of P medicines have a GSL alternative (paragraphs 29 to 30).
  51. The OFT considered whether competition between pharmacies took place primarily at a national or at a local level. Having noted that the merging parties had provided evidence to suggest that the vast majority of their pricing policies are set nationally, the OFT stated:
  52. "the extent to which national chains face each other in local areas may well be a factor in determining the extent to which they can influence each other's decision making at the national level, e.g. on pricing policies" (paragraph 14).
    On the question of how far local competition influences national policies, the OFT noted (in paragraph 31 of the Decision) that the parties had argued that:
    "… on the whole, Boots and UniChem do not tend to be located close to each other; Boots stores tend to be based in high street locations, while UniChem's pharmacy stores fit the "community pharmacy" model and are generally found in residential areas, health centres or smaller shopping centres. … However, both appear with relatively high frequency in rural towns… Overall, the evidence suggests limited competition between the parties in respect of the pricing of P medicines."
  53. The OFT concluded that "[l]ooking at the sector overall, the shares of supply held by the supermarkets and others counteract any suggestion that adverse competitive effects could arise simply by the merger resulting in a reduction from three to two truly national pharmacy chains" (paragraph 32).
  54. The OFT then turned to local area overlaps. Given its importance to this case we set out the relevant paragraphs of the Decision (footnotes omitted):
  55. "33. One third party submitted that a fascia approach is not appropriate in this case, and that local area analysis based on shares of supply would be more appropriate. This was on the basis that Boots was a significantly different type of retail pharmacy which meant that it should not be treated on the same terms as any other fascia. The OFT considers that this could suggest more that Boots is not a close competitor to other types of pharmacies rather than that a fascia approach would not be appropriate in this case. No other comments were made by third parties to suggest that the fascia test was inappropriate.
    34. In 1996 when it looked at UniChem/GEHE/Lloyds, the MMC only examined two-to-one overlaps on a one mile basis – and even in these areas concluded that the customer benefits of the mergers would in each case outweigh any reduction in competition. However, at the time of the MMC inquiry, resale price maintenance was still in place in OTC (GSL and P) medicines such that scope for competition, especially on price, might be expected to have been more limited than it is today.
    35. The OFT report on control of entry regulations (see footnote 4) – published after the removal of RPM – found that competition, particularly price competition, between pharmacies and between pharmacies and other retailers was still muted. It did however find some correlation between local concentration and quality of service on matters such as extended opening times, provision of consultation areas and home delivery.
    36. There is no price competition between retail pharmacies in the supply of ethicals, which account for [80-90] [C] per cent of UniChem's turnover ([20-30] [C] per cent of Boots turnover) and a number of retail outlets other than pharmacies supply GSL medicines and non-pharmaceutical products. Price competition between retail pharmacies in the product categories we are considering is therefore mainly limited to P medicines. A number of third parties suggested that competition between pharmacies on service levels was also limited because NHS service contracts specify national minimum standards for pharmacies. The NHS contract is restrictive in terms of its imposition of service standards and required opening times for instance. Such restrictions serve to limit the parameters of competition. In fact, third parties suggested that the main form of competition takes place on initial choice of location – a factor of competition which, on the basis of third party comments, still appears to be highly regulated by PCTs.
    37. However, the parties' internal documents clearly show that there is some competitive interaction between different retail pharmacies. Both Boots and UniChem monitor a number of different price and service quality variables and compare themselves to competitors. Possible aspects of competition mentioned in Boots' consumer research documents include product availability, store layout, ease of shopping and shopping environment. At a local level, waiting times for obtaining a prescription and the provision of deliveries seem to be key measure of competition too. For instance, a Boots' internal document […][C].
    38. Even though locality appears to be the main factor determining customers' choice of pharmacy, there is nonetheless some evidence to suggest that a reduction in pharmacy fascia numbers could bring about a reduction in competition, whether by lowering service/ quality levels, by reducing choice (on a quality level) or, less so, affecting prices, particularly for P medicines.
    39. On the question of what is the appropriate level of competition in a locality, differing views have been provided to the OFT. One third party told us that it would expect to face at least two competitors within a one-mile radius. The parties, on the other hand, state that there is no realistic prospect of a substantial lessening of competition arising in any local area due to the limited nature of competition in the sector. This is particularly true in this case, they submit, because Boots and UniChem have different business models and therefore should not be considered to be particularly close competitors."
  56. The OFT then considered whether the possibility of new entry might mitigate any local competition concerns and concluded that views were mixed as to whether changes in the control of entry regulations had made market entry easier. It concluded further that individual consumers did not possess buyer power.
  57. The OFT stated its conclusions on local overlap analysis in the following terms:
  58. "Conclusion on local overlap analysis
    43. In looking at local area overlaps the starting point of our analysis has been the 1996 MMC report which found that a reduction in pharmacy fascia from two to one in a one mile radius would [not][7] be against the public interest. The OFT notes, however, that this was before RPM on OTC medicines was removed, suggesting that the scope for competition in 1995 might have been more limited than it is currently. On the other hand, UniChem is a relatively small player in the OTC sector and this could suggest a similar approach being taken here. Additionally, the fact that there has been further deregulation in the form of the relaxation of entry rules could militate in favour of a less intrusive approach, although it is noted that the evidence on this issue provides far from conclusive backing to such an approach. The parties' internal documents do, however, show that a level of competitive interaction between retail pharmacies does exist and that new entry will usually prompt some competitive response. Such evidence would clearly suggest that a reduction in pharmacy fascias from two to one within a one mile radius could lead to a reduction in the incentives to compete and have a detrimental effect on matters such as the standard and quality of the services provided by the pharmacies concerned (over and above the levels stipulated by the NHS contract).
    44. In many respects the pharmacy sector is unique in terms of being a highly regulated retail sector where matters such as entry and even opening hours are, to a large extent, controlled by the local PCTs. However, a level of competition does exist and the OFT has sought to ascertain whether it is possible to establish, to a reasonably conclusive degree, the number of fascia required to maintain adequate competition in a local area. To this end, in addition to the two to one analysis referred to above, the OFT looked in detail at those local areas where the merger will result in a reduction in competitor fascias from three to two. Even in these areas, switching of customers between Boots and Unichem stores may be high, and for a large proportion of customers, Boots and Unichem would be the closest competitors. This may arise particularly in those localities where the merging parties' pharmacies are located close together. In such a situation the competitive scenario post merger does not significantly differ from that in the two to one areas.
    45. This has to be considered alongside the high barriers to entry in this sector combined with what the parties' internal documents say on the issue of competition – see, for instance, the references at paragraph 37 above to the monitoring of various competitor price and service quality variables. On balance, the OFT believes that a substantial lessening of competition may go beyond those areas outlined at paragraph 43 and may also arise where fascia are reduced from three to two within a one mile radius.
    46. Any higher reduction in fascia number than this (e.g. four to three or higher) could also give rise to a lessening of competition but on the basis of the evidence before the OFT, it believes that this cannot be expected to be substantial. One major competitor suggested that it would be usual to face two other competitors within a local area. Moreover, the CC's 1995 report considered only two to one fascia reductions. While there have been small changes to the market since then (primarily the removal of resale price maintenance on OTC medicines) which might suggest some small increase in the scope for potential competition, these changes do not support an argument that reductions from four to three fascias might give rise to competition concerns."
  59. The OFT then examined the effect of the merger on the provision of NHS and PCT services (paragraphs 50 to 55) and wholesaling (certain third parties having argued that in so far as Boots self-supplies its top lines, and has a national supply network, it ought to be considered a potential competitor in full-line wholesale supplies to third parties) (paragraph 56). In neither case did the OFT consider that the merger may be expected to result in a SLC. The OFT also rejected a number of possible concerns put forward by third parties in relation to vertical issues and wholesale supply.
  60. In the section headed "Third party views" the OFT recorded the results of its consultation exercise in the following terms:
  61. "80. A large exercise to consult third parties was undertaken in this case and, as can be seen above, a number of concerns were raised. Some PCTs were concerned about local competition in their particular areas and about the parties' willingness to engage in, and negotiating strength in respect of, PCT initiatives.
    81. A number of competitors also expressed concerns about the merger, however, some of these appeared directed at the fear that a combined Boots / UniChem would be a stronger competitor, which may indeed have a positive effect on competition, and it is noted that not all competitors were concerned by the merger. A suggestion was also made that the complexity of the issues raised meant that they were not capable of being resolved by a first stage investigation."
  62. The next section is headed "Assessment". The OFT stated:
  63. "82. The parties overlap in the provision of retail pharmacy services. In terms of national competition, no concerns arise. However, on a local level on the basis of a one mile radius around both Boots and UniChem pharmacies there are 38 areas where the merger would result in a two to one reduction in the number of competing pharmacies ('fascias') and a further 61 areas where it would result in a reduction in the number of fascias from three to two. The evidence considered during this assessment clearly shows that a reduction in pharmacy fascias from two to one in a local area is, despite the restrictive terms of the NHS contract, expected to result in a substantial lessening of competition (SLC). Such an SLC could take the form or reductions in quality or the level of service provided (over and above the levels stipulated in the contract). There may also be an impact on pricing, particularly of P medicines. The evidence on whether an SLC would arise in the case of three to two overlaps is less conclusive but, on balance, the OFT takes the view that it may be the case that the merger may be expected to result in an SLC within these three to two overlap areas given the high barriers to entry present in this market as a result of the control of entry regulations. …
    85. Given the conclusions at paragraph 82 above, the OFT believes that it is or may be the case that the merger may be expected to result in a substantial lessening of competition within a market or markets in the United Kingdom."
  64. The OFT then explained that where it is under a duty to make a reference under section 33(1) of the Act, it may instead accept undertakings in lieu. Boots and UniChem were prepared to consider divesting pharmacies where the OFT found there to be a SLC in the following circumstances: (i) where there is a 2 to 1 or 3 to 2 overlap on the basis of a one-mile […][C] overlap, […][C] they would divest one pharmacy; and (ii) […][C].
  65. The OFT further noted the parties' preparedness to arrange for any divestments to be made in 'packages' to address the OFT's concerns about the practicability of considering and consenting to a large number of individual disposals (paragraphs 86 to 88).
  66. The OFT considered that such an undertaking would address all of the potential adverse effects arising from the merger; accordingly, the OFT decided to exercise its discretion under section 73(2) of the Act to negotiate undertakings in lieu of reference (paragraphs 89 to 90).
  67. Under the heading "Decision", the OFT concluded as follows:
  68. "91. The merger will therefore be referred to the Competition Commission under section 33(1) of the Act, unless Boots gives suitable undertakings pursuant to section 73 of the Act to address the potential competition concerns outlined above."
    Proposed Undertakings
  69. On 14 March 2006, the OFT gave notice, pursuant to paragraph 2(1) of Schedule 10 to the Act, of proposed undertakings offered by Boots pursuant to section 73 of the Act. The notice and proposed undertakings were published on the OFT's website.
  70. The OFT considered that the proposed undertakings were appropriate to remedy, mitigate or prevent the competition concerns identified in the Decision, and was therefore minded to accept them.
  71. In the notice of 14 March 2006, the OFT invited interested parties to make their views known by no later than 29 March 2006.
  72. V THE PROCEEDINGS BEFORE THE TRIBUNAL

  73. Celesio's grounds of review, as set out in paragraph 3 of its notice of application, were as follows:
  74. (a) The OFT incorrectly based its assessment of competitive concerns on the local retail market solely on a "fascia test". The OFT thereby erred in its reasoning, and failed to consider material factors relevant to the identification of the competitive concerns arising from the merger.
    (b) Even if the fascia test was sufficient to identify the competitive concerns arising from the merger on the local retail market, the OFT's conclusion in this case that a reduction from 4 to 3 fascias would not give rise to a SLC was incorrect in fact and law, and inadequately reasoned.

    (c) The defects in the OFT's assessment of the local retail market vitiate the OFT's conclusions that no competitive concerns arise at a national level. The OFT's conclusions on this issue are also not substantiated by the evidence.

    (d) If any of points (a) to (c) above are correct, the undertakings proposed by the parties are insufficient and inappropriate to remedy the competitive concerns arising from the merger.
  75. The notice of application annexed an expert report by Dr Jorge Padilla, Managing Director of the European Competition Policy Practice of LECG and other supporting materials including Celesio's pre-decision submissions to the OFT and a number of authorities. In preparing the notice of application Celesio had not seen either the evidence submitted by Boots and UniChem to the OFT, or the OFT's Issues Letter.
  76. A case management conference was held on 28 March 2006. In view of the urgency of the matter the Tribunal directed that formal pleadings be dispensed with and that the OFT should file and serve a skeleton argument which would stand in place of the defence. Likewise, Boots and UniChem, who had been given permission to intervene in support of the OFT by way of an Order dated 24 March 2006, were to file a skeleton argument in place of a statement of intervention.
  77. The OFT indicated prior to the case management conference that it intended to challenge the admissibility of Dr Padilla's report. At the case management conference counsel for Boots and UniChem indicated that they too intended to instruct an expert to counter the points raised by Dr Padilla. The Tribunal did not make a preliminary determination as to admissibility but it emphasised that any expert evidence submitted by Boots and UniChem should be limited to matters of principle rather than giving detailed analysis of factual matters. The parties accepted that the Tribunal could look at the material de bene esse and rule on it at the same time as giving its judgment on the substance. In the event, much of the expert evidence considered by the Tribunal de bene esse was not relevant given the more limited scope of the issues on which Celesio relied at the final hearing.
  78. In accordance with the Tribunal's Order of 28 March 2006, the OFT filed its skeleton argument and evidence in support, including a witness statement by Mr Simon Pritchard, Director of the Mergers Branch at the OFT, on 3 April 2006.
  79. On 4 April 2006, the interveners filed their skeleton argument and evidence in support, including an expert report by Mr Simon Baker of RBB Economics; a witness statement from Mr Michael Oliver, the Company Secretary of Boots; and a witness statement from Mr Steven Duncan, the Retail and Pharmacy Director, and a main board director, of UniChem.
  80. On 7 April 2006, Celesio filed its skeleton argument in reply, together with an additional expert report from Mr Justin Coombs of LECG and a witness statement from Mr Justin Ash, the Managing Director of Lloyds.
  81. The original ground (a) in the notice of application
  82. Before summarising the parties' evidence and principal submissions, it is convenient to deal at the outset with ground (a) in the original notice of application: that the OFT had incorrectly based its assessment of competitive concerns on the local retail market solely on a "fascia test".
  83. When the OFT and the interveners filed their pleadings and evidence, Celesio saw for the first time the requests for information and the evidence given to the OFT by the merging parties, together with the Issues Letter and the merging parties' responses. Having had the opportunity to consider this material, Celesio did not pursue this first ground of challenge.
  84. The grounds pursued by Celesio at the hearing, as set out in its skeleton argument at paragraph 1, were therefore confined to the following:
  85. "(a) the reasons given in the Decision are not capable of sustaining the OFT's conclusion that a reduction from 4 to 3 fascia in local retail markets would not give rise to a SLC;
    (b) this error of assessment in relation to local retail markets necessarily vitiates the OFT's conclusion relating to the national retail market; and
    (c) it further follows that the OFT is not entitled to accept undertakings in lieu of a reference pursuant to s.73(2) of the Enterprise Act 2002."
    In this judgment we refer to these issues as amended ground 1, amended ground 2 and amended ground 3 respectively.
  86. The oral hearing in these proceedings took place on 11 April 2006.
  87. VI THE DUTY OF THE OFT TO REFER

  88. Celesio submits that the following principles apply regarding the exercise of the OFT's judgment under section 33(1)(b) of the Act:
  89. (a) The OFT must refer the proposed merger to the CC where it believes that it "is or may be the case that" the merger may be expected to result in a SLC. The words "may be the case" denote a lower degree of likelihood than an expectation, but exclude the purely fanciful. The threshold for the making of a reference to the CC is therefore a low one.
    (b) The OFT has a margin of judgment "between the fanciful and a degree of likelihood less than 50 per cent". However that margin of judgment does not imply that the OFT has a wide discretion; rather, that the degree of likelihood of a SLC will vary according to the circumstances.
    At the hearing Celsio submitted that if the prospects of there being a SLC are between fanciful and 50 per cent or above, then there is an obligation on the OFT to refer the proposed merger to the CC.
  90. The OFT does not accept Celesio's submission that where the prospects of there being a SLC are between fanciful and 50 per cent or above, then there is an obligation on the OFT to refer the proposed merger to the CC. In that regard, the OFT submits that between fanciful and 50 per cent, it is a matter of judgment for the OFT as to whether there is sufficient likelihood of SLC so as to warrant a reference to the CC. The OFT further submits that the degree of likelihood will vary with the circumstances.
  91. The interveners submit that the issue for the OFT is one of factual judgment and that the OFT has a margin of judgment or evaluation of facts, which is a wide margin between a fanciful likelihood of a SLC and a likelihood of 50 per cent, and that judgment must be made by the OFT on the balance of probabilities.
  92. We have referred to sections 33 and 36 of the Act at paragraphs 5 and 6 above. Section 33 of the Act provides that:
  93. "The OFT shall, subject to subsections (2) and (3), make a reference to the [CC] if the OFT believes that it is or may be the case that –
    (a) arrangements are in progress or in contemplation which, if carried into effect, will result in the creation of a relevant merger situation; and
    (b) the creation of that situation may be expected to result in a substantial lessening of competition within any market or markets in the United Kingdom for goods or services."
  94. If the OFT makes a reference then under section 36 of the Act the CC shall decide the following questions:
  95. "(a) whether arrangements are in progress or in contemplation which, if carried into effect, will result in the creation of a relevant merger situation; and
    (b) if so, whether the creation of that situation may be expected to result in a substantial lessening of competition within any market or markets in the United Kingdom for goods or services."
  96. The distinction between these sections is that the OFT's role is to consider whether it "believes that it is or may be the case that" a relevant merger situation "may be expected to result in a substantial lessening of competition" whereas the CC "shall decide" whether the relevant merger situation "may be expected to result in a substantial lessening of competition".
  97. The following principles can be ascertained from the judgment of the Court of Appeal in IBA Health, cited above:
  98. i. Section 33 imposes a duty on the OFT to refer if the requirements set out in the section are met.
    ii. The OFT's investigation is a first-stage investigation to consider whether a reference is necessary.
    iii. The belief must be reasonably held by the OFT: it must be reasonable and objectively justified by relevant facts.
    iv. The phrase "believes that it may be the case" imports a lower degree of likelihood than paragraph (b) in sections 33(1) or 36(1) would by itself involve. As Sir Andrew Morritt V-C said in IBA Health at paragraphs 47 and 48:
    "47. That lower degree of likelihood might, for example, exist in circumstances where the work done by the OFT did not justify any positive view, but left some uncertainty, and where OFT therefore believe that a substantial lessening of competition might prove to be likely on further and fuller examination of the position (which could be undertaken by the Competition Commission).
    48. At the other end of the scale it is clear that the words "may be the case" exclude the purely fanciful because OFT acting reasonably is not going to believe that the fanciful may be the case. In between the fanciful and a degree of likelihood less than 50 per cent there is a wide margin in which OFT is required to exercise its judgment. I do not consider that it is possible or appropriate to attempt any more exact mathematical formulation of the degree of likelihood which OFT acting reasonably must require."
  99. Above "fanciful" the OFT has to make a judgment as to the point at which it comes to believe "that it is or may be the case" that the relevant merger situation "may be expected to result in a substantial lessening of competition" so as to trigger its duty to refer the matter to the CC. As this Tribunal stated in UniChem v Office of Fair Trading [2005] CAT 8 at paragraph 172, the Vice Chancellor in paragraph 48 of IBA Health "was not implying that the OFT had a wide discretion, only that the degree of likelihood of a SLC will vary widely depending on the circumstances".
  100. It follows that we reject Celesio's bald submission that the OFT is always under an obligation to refer a merger to the CC where the prospect of there being SLC is greater than fanciful.
  101. VII THE STANDARD AND NATURE OF REVIEW

  102. The parties agree that the relevant principles are to be found in the following cases: IBA Health, cited above, UniChem, cited above, and Somerfield plc v Competition Commission [2006] CAT 4.
  103. The starting point for the Tribunal's review in this case is section 120(4) of the Act, which provides that the Tribunal "shall apply the same principles as would be applied by a court on an application for judicial review". The Tribunal should apply the principles which emerge from the judgment of the Court of Appeal in IBA Health which were followed by the Tribunal in its judgments in UniChem and Somerfield.
  104. As to the intensity of review, we refer to what was stated by the Tribunal at paragraph 57 of the Somerfield judgment.
  105. VIII THE EVIDENCE

    A. CELESIO'S EVIDENCE

    The LECG evidence
  106. The LECG evidence comprises the report of Dr Jorge Padilla and the subsequent report of Mr Justin Coombs. Much of the material in Dr Padilla's report had been directed to ground (a) in the notice of application which was not pursued at the hearing. At the hearing, Celesio only relied upon the reports for two matters.
  107. First, Celesio relied on Appendix B to Dr Padilla's report. This comprises a bar chart showing how many outlets (rather than fascias) there are in those local areas where the merger would lead to a reduction of fascias from four to three but where there are in fact currently more than four outlets. The bars in the chart also show the proportions of those outlets owned by the merging parties, by Celesio and by other retail pharmacists. The chart illustrates that the merging parties' share of the total number of outlets in some of the 4 to 3 areas was over 50 per cent.
  108. Secondly, Celesio relies upon Appendix A to the report of Mr Justin Coombs, which shows that on Celesio's count there are 41 areas where Boots and UniChem are located closest to each other and there will be a 4 to 3 fascia reduction as a result of the merger.
  109. The OFT did not object to Celesio adducing this evidence.
  110. The evidence of Mr Justin Ash
  111. The witness statement of Mr Ash sought to address particular factual errors contained in the witness statement of Mr Simon Pritchard (which, for convenience, we refer to as "Pritchard"), filed on behalf of the OFT, and statements made by Slaughter and May on behalf of the merging parties in their response to the Issues Letter.
  112. Having noted paragraph 57 of Pritchard, which mentioned the merging parties' observation that since the 2003 Report opening hours have become regulated and consultation areas mandatory under the new pharmacy contract, Mr Ash stated (at paragraphs 8, 12 and 15 respectively):
  113. "The parties' submission that consultation areas are mandatory under the new pharmacy contract is incorrect.
    … there is no obligation upon a pharmacy to provide Advanced Services under its pharmacy contract. Therefore, while it is true that consultation rooms to a certain standard are required for the provision of Advanced Services by those pharmacies which wish to provide such services, it is not the case that consultation rooms are mandatory for all pharmacies.
    With regard to the regulation of opening hours, these have always been regulated. Under the new regulations core hours must be declared to the PCT but discretion remains as to when these hours of opening occur and as to whether a pharmacy contractor wishes to open for hours additional to the core hours. These are classified as supplementary hours…"
  114. Mr Ash's evidence was the subject matter of correspondence between the Tribunal and the Celesio. By a letter dated 10 April 2006, Celesio informed the Tribunal that:
  115. "Celesio felt that it was important to identify and correct these errors, both by way of background to the hearing before the Tribunal, and in case the matter is to go any further (e.g. upon remittal to the OFT).
    However, …Celesio does not intend to make any free-standing submissions in reliance on Mr Ash's witness statement".
  116. During the hearing on 11 April 2006, counsel for Celesio stated that:
  117. "We have made the point that we actually think that there is an error of fact, but if I lose on all my submissions today and that is the only thing left, I will not be submitting that we nonetheless win on that point."
    (transcript, p 7)
  118. In these circumstances Mr Ash's witness statement is not relevant to our review of the Decision.
  119. B. THE INTERVENERS' EVIDENCE

  120. Except for the expert report of Mr Baker, all of the material provided by the interveners for the hearing was submitted to the OFT at the administrative stage. The Tribunal was not referred, at the hearing, to Mr Baker's report.
  121. Since no reliance has been placed on Mr Baker's report, the question of its admissibility does not arise in this case.
  122. C. THE OFT'S EVIDENCE

  123. The OFT relied on the witness statement of Mr Pritchard. Given its importance we set out the relevant paragraphs of Pritchard below (footnotes and cross-references omitted):
  124. "Relevance of local overlaps to national issues

    37. The OFT is well aware of the significance of local issues in retail markets in which there are national policies on pricing or non-pricing aspects of competition (such as service standards). As noted at paragraph 14 of the Decision, national policies may be affected by the extent to which chains face each other locally across the country. Two issues are relevant in this regard: first, the degree of overlap as a proportion of each parties' national estate; second, the degree of competition concerns in those overlap areas.
    38. In the case of Boots and UniChem, the degree of overlap of estates in areas where local concerns arise is particularly low. As noted, UniChem and Boots have 958 and 1,350 pharmacy outlets respectively, and they are each other's only local rivals (on a 1-mile radius) in only 38 areas, and two of three in only 61 areas. The sum total of 2:1 and 3:2 overlap areas is around 10% of either party's estate, i.e. each party faces close geographic competition from the other in such areas only a fraction of the time. In such circumstances, it could not be said that the low national shares were masking a potential unilateral effect at national level deriving from local-level competition between the parties.

    39. This is in line with evidence from Boots' internal documents which indicate that, on the whole, Boots and UniChem stores tend not to be located close to each other: Boots stores tend to be in high street locations, whereas UniChem's pharmacy stores fit the "community pharmacy" model and are generally found in residential areas, health centres or small shopping centres. Boots' own analysis of catchment areas breaks down pharmacy coverage across classes of shopping centre area. Boots appears most frequently ([…][C] % of stores) in "major regional centres" with an average catchment population of […][C]; a category which accounts for only […][C]% of "Moss" (the brand of most of Alliance Pharmacy stores) portfolio. Moss' most frequent location is "small district centres", with an average catchment population of […][C]; whereas only […][C]% of Boots' stores fit this profile. While both appear with relatively high frequency in rural towns, on balance the OFT took the view that the evidence did not support the proposition that the parties were particularly similar as concerned the geographic scope of their store portfolio.
    40. As to the degree of competition concern in local areas, this goes to the heart of the local analysis conducted in this case; the OFT decided that local divestment in 2:1 and 3:2 areas addresses concerns at the local level, and in turn those that may arise at the national level, though increased incidence of problematic overlaps within the merging parties' national store portfolio.

    IV THE OFT'S CONTESTED ASSESSMENT OF LOCAL ISSUES

    41. The Decision's assessment of local competition issues is the focus of two grounds of Celesio's application. The OFT's analysis of local competition issues is set out in the Decision at paragraphs 33 to 49, which draws on and should be read in the light of its more general analysis of the retail pharmacy sector in paragraphs 8 to 16 and 26 to 32.
    42. As noted above, in accordance with OFT guidelines, the OFT takes an assessment of market concentration as an initial starting point before a full assessment of competitive effects. Typically, therefore, the OFT proceeds as follows. First, it identifies the measure(s) of concentration most appropriate in the circumstances of the case. Second, it carries out a full assessment of competitive effects within a framework set by the concentration measure adopted. In this case, the OFT concluded that a fascia test was the most appropriate measure of concentration on the facts, and its assessment of competitive effects was carried out within a framework set by the fascia test, i.e. the OFT first analysed competitive effects in local areas with a 2:1 fascia reduction, and then moved on to consider 3:2 and 4:3 areas.

    43. This section elucidates the OFT's reasoning in relation to local issues in the light of the arguments advanced by Celesio in its application, and is structured as follows.
    • First, I discuss the evidence which suggested that there existed only a limited scope for local competition among retail pharmacies in the UK. This conclusion served as a point of departure for the OFT's local area analysis, including contested elements of it.
    • Second, in response to Celesio's first ground of application, I explain why Celesio mischaracterises the role played by the fascia test, and why, contrary to Celesio's submissions, it was the most appropriate measure of concentration in the circumstances of the case.
    • Third, in response to Celesio's second ground of application, I discuss the OFT's reasons for its conclusions in respect of local areas featuring various reduction in fascia, i.e., in respect of 2:1, 3:2, and 4:3 areas, and why the OFT committed no error of assessment in concluding that the statutory reference test was not met in respect of 4:3 areas.

    LIMITED SCOPE FOR LOCAL PHARMACY COMPETITION

    44. In its application, Celesio argues, among other things, that the OFT should have concluded that the statutory reference test was satisfied in relation to 4:3 areas in the light of its approach in other cases relating to other sectors. It is important to remember that the retail pharmacy market is in many respects unusual and, although some limited competition appears to exist within it, the extent and intensity of that competition is considerably less than in many other markets. For example, paragraph 44 of the Decision notes that:
    "In many respects the pharmacy sector is unique in terms of being a highly regulated retail sector where matters such as entry and even opening hours are, to a large extent, controlled by the local PCTs".

    Previous cases relating to the local retail pharmacy market

    45. The OFT was already aware of the limited nature of competition within the retail pharmacy market when it came to carry out its local area analysis in this case, and this awareness informed its analysis and conclusions. As described at paragraphs 34 to 36 and following of the Decision, the OFT considered the evidence from the MMC's 1996 report and the OFT's MPI report on Control of Entry Regulations ("the MPI report"). Also relevant in the context of this challenge by Celesio is the evidence gathered during, and Celesio's submissions in Lloyds/ C&S, its own recent merger case. Finally aspects of the OFT's analysis of national issues are relevant also at the local level.
    The MMC Report
    46. As noted in paragraph 34 of the Decision, in UniChem/ GEHE/ Lloyds, the MMC restricted its examination of potential merger concerns to 2:1 areas. It noted that competition between retail pharmacies was muted and concluded that even in 2:1 areas any detrimental competitive effects would be small and outweighed by the beneficial effects of the merger. However, at the time of the MMC inquiry, resale price maintenance ("RPM") was still in place in P medicines.

    The OFT's pharmacy report

    47. Following the lifting of RPM for P medicines in 2001, the OFT examined the retail pharmacy sector in the MPI report. The OFT found only muted competition between pharmacies. The econometric analysis commissioned by the OFT and carried out by Frontier Economics ("Frontier") as part of this study revealed no statistically observable relationship between the prices, in 2001, charged by a pharmacy on P medicines, and any (tested for) measure of local concentration. In other words, the number of pharmacies in an area did not, as one might expect, affect discounting (or lack thereof) on P medicines.
    48. The analysis performed by Frontier on behalf of the OFT did find a degree of correlation between local concentration and quality of service (Decision, para 35):

    • pharmacies are more likely to be open before 9 am if they face a higher number of community pharmacies (CPs) per GP in their locality, or if they are a closest pharmacy to a GP;
    • pharmacies are more likely to offer a consultation area if there are more supermarket pharmacies in their locality; and
    • when a pharmacy faces no other pharmacy within 5 km, it was less likely to offer home delivery.

    49. In relation to this analysis, in Boots/UniChem the merging parties pointed out that these relationships are open to a number of interpretations, and did not necessarily indicate the existence of significant competition. This is acknowledged by Frontier itself. Further, the parties noted that, since the Frontier report, two out of three of these aspects of quality of service had now become regulated under the new pharmacy contract (see paragraph 57 below). However, notwithstanding these uncertainties and limitations, the OFT was unable to rule out the possibility that there was some degree of competition in local markets and we proceeded on that cautious basis.
    Lloyds/ C&S
    50. In Lloyds/ C&S, Celesio and the target argued that competition was limited. Their briefing paper to the OFT […][C] states that, in analysing local overlaps, the "nature" of pharmacy competition should be noted. Celesio argued as follows:

    "(a) The bulk of a pharmacy's sales relate to POM Medicines. NHS dispensing contracts are regulated and so there is no price competition between local pharmacies in relation to the supply of POM medicines.

    (b) Other P Medicines can be sold only under the supervision of a pharmacist and may be subject to price competition. In relation to these medicines, however, local pharmacies (i.e., within the 1 mile radius) are not the only other alternative suppliers. As these are not prescription medicines, Lloyds finds that pharmacies outside the immediate neighbourhood location could exert price pressure on a local pharmacy. In any event, P medicines account for only a small proportion of a typical local pharmacy's sales.
    (c) For GSL medicines (or OTCs) and non-pharmaceutical products, competition also comes from non-pharmacy outlets including supermarkets, newsagents, petrol station forecourts, etc.
    (d) The provision of other pharmacy services (such as diagnostic services) is complementary to that provided by GPs. The quality of such services is in many cases subject to control by the local Primary Care Trust (PCT) under the pharmacy contract.
    (e) Regulatory safeguards exist in relation to the maintenance of standards by pharmacists."

    51. Celesio (Lloyds) thus characterised the UK pharmacy sector as one where limited competition occurs and where such competition that does occur is limited to non-price factors such as the availability and quality of consulting rooms (as stated at para. 1.4 of the Lloyds/ C&S Merger Notice). […][C], Celesio (Lloyds) submitted that the highest degree of concentration by a fascia count was a 4:3 overlap. The briefing paper states that Lloyds does not consider that the acquisition of C&S will lead to any lessening of competition in the supply of retail pharmacy products or services in any of the locations where the parties overlap in part because the merging parties would be constrained by "regulatory incentives to maintain service quality, so as to ensure competition in all affected local markets".
    52. Third party inquiries in Lloyds/ C&S corroborated Celesio's position, indicating that competition between retail pharmacies is limited.
    Evidence in Boots/UniChem considered above in relation to national competition

    53. Finally, the OFT's findings in Boots/UniChem discussed above in relation to national competition were of relevance to its analysis of effects on local competition. As noted above, Boots and UniChem each has a national commercial strategy which determines the nature of its competitive offering, i.e. Boots positions itself and competes as a general health and beauty retailer, whereas UniChem is more of a local pharmacy. To a large extent, at least in relation to price competition on P medicines and service offerings, all pharmacies exert similar levels of competitive constraint, subject to their precise geographic position. However, to the extent that Boots and UniChem have distinct competitive offerings, they are not close competitors. This is true at a national level, but is also relevant when carrying out a full assessment of the degree of competitive constraint imposed by the parties on each other in local markets.
    54. For all of these reasons, the OFT's point of departure when it came to analyse local competitive effects in the Boots/UniChem case was that the scope for local competition among pharmacies was relatively limited.
    Local evidence in the instant case

    55. As part of its detailed effects analysis in this case, the OFT sought further evidence as to the extent of competition in local areas.
    Evidence suggesting no material non-price competition
    56. Consistent with Celesio's position in Lloyds/C&S, the merging parties and third parties submitted and provided evidence to support the conclusion that non-price competition between retail pharmacies was limited, because national minimum standards reduce the scope for competition (Decision; para 36.
    57. In particular, the merging parties submitted that national minimum standards had increased over time. They noted that, since the OFT's MPI report opening hours have become regulated and consultation areas mandatory under the new pharmacy contract. These parameters were notable as being two of only three aspects of pharmacy service in respect of which Frontier, on behalf of the OFT, had found possible evidence of local competition.

    Evidence suggesting material non-price competition
    58. Notwithstanding the above the OFT requested a substantial body of internal documents from the merging parties likely to be relevant to the scope for non-price competition. In the event, such documentary evidence proved to be key in providing evidentiary support for the OFT's ultimate conclusion that there was some, albeit limited, competitive interaction among pharmacies and thus between the parties pre-merger. The following points emerged from the evidence.

    • Both Boots and Alliance Pharmacy monitor and benchmark themselves against rivals on a number of price and non-price variables. Possible parameters of competition, not considered by the OFT's MPI report, but mentioned in Boots' consumer research documents, included:
    • product availability;
    • store layout;
    • "ease of shopping";
    • "shopping environment".

    • Boots' internal guidance […][C]:
    • […][C]
    • […][C].

    59. On balance, the documentary evidence did not suggest substantial competitive interaction between retail pharmacies, certainly relative to the majority of local (or national) markets that the OFT has occasion to consider in merger review. Indeed, the competitive conditions in the retail pharmacy sector in particular because of the high level of regulation are obviously very different from (i.e. less lively than) those in other markets, including some of those Celesio urges by way of comparison in its N[otice] o[f] A[pplication]. However, this evidence from the internal documents of the merging parties was considered sufficient to eliminate the hypothesis that there was no or so little competition at local level such that, per se, no retail pharmacy merger in the UK could give rise to a substantial lessening of competition.
    Evidence on scope for price competition
    60. Finally, the parties and other market participants in this case affirmed the previous conclusion on scope for price competition, as being limited to discounting in P medicines only (aside from GSL, where no issue arises).
    THE OFT'S DUTY TO REFER IN RESPECT OF PARTICULAR LOCAL MARKETS

    The OFT's fascia reduction analysis

    99. Once a fascia analysis is adopted, the next step is to class areas according to the relevant reduction of fascia: 2:1, 3:2, 4:3, 5:4 areas and so on. Each step in concentration reflects the loss of one independent consumer choice in an area. The basic intuition is that the lower the number of choices to begin with, the more significant the loss of any single choice is likely to be.
    100. In the Boots/UniChem case, the OFT considered it appropriate to use an iterative approach beginning with the most concentrated markets: the 2:1 or post-merger monopoly areas, where consumer choice is effectively eliminated. If further competitive effects analysis reveals ongoing concerns for 2:1 areas, the OFT next considers 3:2 areas, and the iterative process continues until no concerns arise at a given fascia reduction. The rationale is relatively simple: if effects analysis for an area of a higher degree of concentration fails to raise concerns (say 2:1), this will tend to obviate the need for further iterative effects analysis at lesser degrees of concentration (3:2 and beyond). In this case, beginning with the most highly concentrated areas is particularly appropriate given the peculiarly limited scope for competitive interaction over and above regulated price and service levels in local pharmacy markets.

    101. There was no evidence to suggest any particular level of concentration above 2:1 where a "step change" to observable competitive (price or non-price) variables occurred when comparing one degree of fascia concentration to another. For example, there was no statistical support for a proposition that prices were higher or service levels lower in an area with three pharmacy fascia than in areas with four. This was the conclusion of the OFT study, which tended to be supported by UniChem's internal econometric analysis provided in this case.

    102. That the OFT's analysis developed incrementally in this way is evident from paragraphs 43 to 46 of the Decision. However, this development is not recognised by Celesio in paragraph 81 of its application, where it states that the OFT's "reasons for excluding 4:3 areas are set out in a single paragraph of the Decision, paragraph 46". That is not the case. Rather, the OFT's conclusion on 4:3 areas was reached in the light of its overall assessment of the level of competition in the market, informed by evidence referred to throughout the Decision. A better summary of this aspect of the Decision is set out in paragraph 43(g) of the application, where Celesio states that the OFT concluded that a reduction in fascia numbers of 4:3 or higher could not be expected to give rise to a substantial lessening of competition "on the basis of the evidence before the OFT" (Celesio there using the language of paragraph 46 of the Decision).

    The OFT's assessment of competitive effects in respect of 2:1 areas

    103. The OFT assessment of the potential for a substantial lessening of competition ("SLC") therefore began in respect of 2:1 local areas. While the OFT had regard to all the evidence in making its assessment of 2:1 local areas, its reasoning was fairly straightforward.
    104. By way of summary, it is common ground among all affected parties that for many years competition in the local retail pharmacy sector has been modest, in particular since 1996 when the MMC had described such competition as being "muted". Notwithstanding the subsequent ending of RPM on OTC medicines and some relaxation of entry barriers, this background remained relevant as a great deal of evidence in the case supported the conclusion that price competition on P medicines and non-price competition (retail proposition, including service levels) remained limited, in large part as a result of regulatory controls [Decision paragraph 35 to 36].

    105. However, although the evidence was not entirely conclusive, the internal documents of the merging parties supported the conclusion that there existed some, albeit limited, competition. A 2:1 merger would, by definition, eliminate all such existing competition. As for the threat of competition from entrants, there was insufficient evidence to conclude that the relaxation of the entry requirements had been effective in reducing barriers to entry. Accordingly, the OFT decided that it is the case that the merger may be expected to result in a substantial lessening of competition in 2:1 areas.

    106. It may be noted that this conclusion was consistent with the OFT's view in Lloyds/ C&S that a 2:1 in Mexborough warranted the sending of an Issues Letter. The matter was never resolved in the Decision because the parties removed this overlap from their transaction in lieu of proceeding with the CRM process. Celesio had, however, argued that the 2:1 overlap did not give rise to a substantial lessening of competition.

    The OFT's assessment of competitive effects in respect of 3:2 areas
    107. In contrast, the OFT's assessment of whether the statutory reference test was satisfied in respect of 3:2 areas was much more finely balanced. As stated in paragraph 44 of the Decision, the OFT "looked in detail" at 3:2 areas, considering all the evidence available to it.
    108. Obviously, 3:2 areas are very significantly different from 2:1 areas in that competition, creating consumer choice, is not eliminated within the area, and the merger does not create a local monopolist. If, as the evidence suggested, demand for pharmacies is driven primarily by convenience and customers were willing to travel up to 1 mile for a pharmacist, the remaining competitor in the isochrone represents an effective consumer choice which the merged company must necessarily take into account.

    109. The OFT developed concerns in relation to 3:2 areas only because, following our scrutiny of the relevant maps, we could identify a minority of 3:2 areas that closely resembled 2:1 areas (i.e. where the third competitor was within 1 mile, but significantly further away from the two parties than they were to each other).

    110. However, such concerns as existed in relation to this minority of 3:2 areas were marginal for two reasons:
    • First, because the basis for the concern, namely that these areas were similar in some respects to 2:1 areas, ran contrary to the proposition underpinning the OFT's geographic market definition, namely that consumers would travel up to 1 mile to visit a pharmacy, which was supported by considerable evidence and numerous parties within the industry;
    • Second, given the very limited nature of competition in the market, there was a serious question as to whether the presence of both merging parties as two fascia in that market, along with a third fascia, were really making conditions substantially more competitive than with two fascia.

    111. In considering whether the merger reference test was met in relation to this problematic minority of 3:2 areas, the OFT took into account all the evidence before it in the case, summarised throughout the Decision. The Decision required careful judgement. On balance, the OFT did not believe that it was likely – that it "is the case" under s 33 of the Act – that the merger may be expected to result in a substantial lessening of competition, even in the 3:2 areas that, on closer examination of the relevant maps, resembled a 2:1 area. In other words, even in respect of these areas, the OFT considered that it was unlikely to result in an SLC. In effect, this was because there was an effective competitive choice for consumers within a 1-mile radius of the merged firm, acting as a disincentive and constraint on it to raise price or reduce service. However, the OFT believed that it "may be the case" that the merger may be expected to result in an SLC within the meaning of s 33 of the Act. The OFT did not feel able to conclude that the prospect of a SLC in these areas was merely fanciful. Rather, it considered the statutory reference test to be satisfied in respect of these problematic areas.
    112. Given this position, the OFT then considered whether it was possible to segregate the vast majority of 3:2 areas that were not marginally problematic from the minority that were. The OFT therefore devoted considerable time, including ultimately at the CRM meeting, to the question of whether it could reasonably apply a "secondary filter" to the 3:2 areas to discriminate in a robust manner between those few "may be the case" 3:2 areas from the remainder.

    113. After considering whether we could develop a principle sufficiently robust to account for classifying all 3:2 areas into "reference" and "no reference" categories, we determined that it was not possible to do so within the statutory timetable. Moreover, such an approach would tend, again, to question the fundamental premise of geographic market definition, which had substantial evidentiary support in this case (and had been used, for example, by the MMC).

    114. The OFT was therefore obliged to conclude, reluctantly, as a result of the limited number of 3:2 areas that were marginally problematic, its duty to refer applied "across the board" to every 3:2 area. I say reluctantly because we were aware that, depending on the undertakings in lieu proposals, that this might compel a reference in respect of areas where none was necessary, or result in divestments where none was really necessary. Nevertheless, this issue is consistent with our first-phase role in UK merger control, despite the extensive investigation and experience applied to this particular case.
    The OFT's assessment in respect of areas of lesser concentration (4:3 and beyond)

    115. In was against this background of a detailed analysis in respect of 3:2 local areas that the OFT came to assess 4:3 local areas. The Decision at paragraph 46 opens with the decisive point as to the OFT's positive belief:

    "Any higher reduction in fascia number than [3:2] (e.g. four to three or higher) could also give rise to a lessening of competition but on the basis of the evidence before the OFT, it believes that this cannot be expected to be substantial."
    116. This statement identifies the critical question for the OFT, namely whether any lessening of competition is substantial. As mentioned, our test is whether a merger "is expected to weaken rivalry to such an extent that customers would be harmed." We could not expect such harm in a 4:3 area.
    117. The OFT's conclusion in respect of 4:3 areas was, as in the case of other areas, reached after a detailed consideration of all the evidence in the case. In short, the OFT was already concerned that a finding of SLC, even on the lower "may be the case" standard, was marginal in respect of those limited number of 3:2 areas that arguably resembled 2:1 areas, but felt obliged to extend this lower level of belief across the entirety of the 61 3:2 overlaps. Once an additional effective competitor and choice for consumers within 1 mile of the merged firm was "added" to the competitive assessment, the merged firm would be disciplined by two effective competitors within 1-mile of its operations. Any prediction that harm would occur in the circumstances of local competition among pharmacies, became, in our judgment, plainly speculative and fanciful because it was unsupportable on the evidence. Hence the OFT's reference to its positive belief "on the basis of the evidence before the OFT" (Decision; paragraph 46). Having established a lack of concern in respect of 4:3 areas, there was no need to proceed further with the iterative approach.
    118. At the drafting stage, two supplemental observations were added to the key conclusion set out in the first sentence of paragraph 46 of the Decision.

    119. One relates to the MMC report, and was intended to add some historical context to the OFT's conclusions. This was merely that the MMC considered only 2:1 areas and concluded (as mentioned at paragraph 34 of the Decision), that the detriment to competition in a 2:1 area was sufficiently small that it was outweighed by the customer benefits. Given this analysis, and the fact that 4:3 areas are two fascia steps removed from 2:1 reductions, one might reasonably expect there to be evidence of a relatively large-scale increase in the degree of local pharmacy competition in the UK in order to justify a conclusion that a 4:3 area might raise concerns.

    120. As to the sentence recounting the observation of a competitor that it would be usual to face two other competitors within a local area, this is correct, but in fact did not form part of the OFT's reasoning on any aspect of the case, including as to our evaluation of 4:3 areas prior to the CRM, at the CRM, or at the Decision Meeting. Since it did not form part of the Decision, I recognise that this sentence should not have been included in the Decision."
  125. We consider the admissibility of Pritchard at paragraphs 134 to 169 below.
  126. IX THE PARTIES' SUBMISSIONS ON THE SUBSTANCE

    AMENDED GROUND 1
    Celesio's submissions
  127. Celesio notes that, at paragraphs 33 to 46 of the Decision, the OFT concludes that a SLC may arise where fascias are reduced from 2 to 1 (paragraph 43) and 3 to 2 (paragraph 45). Celesio submits that in relation to 4 to 3 reductions or higher, the Decision concludes that they could give rise to a lessening of competition, but the OFT did not believe that this could be expected to be substantial (paragraph 46).
  128. In Celesio's submission, having already accepted that a 2 to 1 reduction could "clearly" give rise to a SLC and that a 3 to 2 reduction may also give rise to a SLC because switching of customers between Boots and UniChem stores may be high and, for a large proportion of customers, Boots and UniChem would be the closest competitors, it was not open to the OFT merely to assert that the reduction in competition resulting from a 4 to 3 or higher reduction cannot be expected to be substantial.
  129. Celesio submits that the first sentence of paragraph 46 of the Decision (set out at paragraph 44 above) is a mere statement of the OFT's conclusion and is not a substantive reason justifying a conclusion that reductions of 4 to 3 or higher will not give rise to a SLC.
  130. In respect of the second and third sentences of paragraph 46 of the Decision, Celesio observes that in paragraphs 118 to 120 of Pritchard, the OFT has explained that these were supplemental observations and not reasons for its conclusion. Moreover, the evidence referred to in the second sentence did not form part of the OFT's reasoning on any aspect of the case and should not have been included in the Decision, and the reference to the 1996 MMC Report was only included so as to provide some historical context to the OFT's conclusions.
  131. Celesio submits that, in reality, the reasoning in paragraph 46 of the Decision amounts to no more than saying that because 4 to 3 involves a greater number of fascias than 3 to 2, it is less likely that there will be competition concerns. It submits that such an abstract assertion is unacceptable. Celesio submits that the Decision indicates that the OFT has failed to carry out any substantive analysis of the effect on competition of a 4 to 3 or greater reduction, and that the assumption upon which paragraph 46 rests is not consistent with the previous findings in relation to 2 to 1 and 3 to 2 fascia reductions.
  132. Celesio states that Pritchard relies on the suggestion that there was only limited competition in the relevant markets as a reason for excusing the OFT from conducting any detailed consideration of 4 to 3 reductions or higher. However, in Celesio's submission this argument reverses the usual logic in a competition assessment: if firms compete fiercely then the loss of one competitor out of four might not be a concern (if competition is fierce then three competitors may be enough), but that when competition is weak then the loss of one out of four competitors should be more, not less, likely to raise concerns. The perceived lack of competition should have led to more careful scrutiny of 4 to 3 reductions or higher, not less scrutiny.
  133. Celesio submits that it is significant that the OFT asked the merging parties to provide maps of each 2 to 1 and 3 to 2 overlap area and that it was by scrutinising these maps that the OFT concluded that there was cause for concern in relation to 3 to 2 reductions (Pritchard, paragraph 109). However, because no such maps were requested in respect of areas of 4 to 3 reductions or higher, the OFT precluded itself from conducting any specific analysis of such areas. It could only assess areas of 4 to 3 reductions or higher on the basis of the detailed material that it had in relation to 2 to 1 and 3 to 2 areas.
  134. Celesio notes that the OFT considered that some 3 to 2 areas gave cause for concern because, having looked at the individual maps of these areas, it was apparent that the third competitor was within 1 mile, but significantly further away from the two parties than they were to each other. The OFT considered that these areas closely resembled 2 to 1 areas and therefore believed that they gave cause for concern (Pritchard, paragraph 109).
  135. Celesio submits that, in principle, it is perfectly feasible that in any given 4 to 3 area (or higher), the two third party competitors could be within 1 mile of the Boots and UniChem stores, but both significantly further away from the merging parties' stores than they (the merging parties) are from each other but that the OFT did not even direct its mind to this possibility. In its oral submissions Celesio referred to the evidence as to market share by outlets set out at Appendix B to Dr Padilla's report (referred to at paragraph 79 above) to illustrate the kind of evidence that the OFT ought to have gathered and considered in relation to 4 to 3 areas.
  136. Celesio referred the Tribunal to paragraph 100 of Pritchard, which states that:
  137. "the OFT considered it appropriate to use an iterative approach beginning with the most concentrated markets: the 2:1 or post-merger monopoly areas, where consumer choice is effectively eliminated. If further competitive effects analysis reveals ongoing concerns for 2:1 areas, the OFT next considers 3:2 areas, and the iterative process continues until no concerns arise at a given fascia reduction."
  138. In Celesio's submission it is apparent that the OFT failed to apply its own "iterative approach" properly in this case. It is not suggested in either the Decision or Pritchard that there were no concerns in relation to 3 to 2 areas; accordingly, applying its own test, the OFT was obliged to move on and consider the 4 to 3 areas. However, it did not do so. Celesio submits that the OFT compounded matters by assuming that there were no SLC concerns in any 4 to 3 areas, therefore overlooking the fact that there might be SLC concerns in some 4 to 3 areas but not others.
  139. Celesio submits that it is striking that, even though the OFT accepts that it was required as part of its analysis to consider the position in relation to 4 to 3 areas, the OFT did not actively consider 4 to 3 areas, either as a class or individually, so as to come to a substantive conclusion that no SLC might arise. Its approach to 4 to 3 areas was based solely on presumptions and inferences said to arise from its general conclusions as to the level of competition in the relevant markets and its specific views in relation to 3 to 2 areas.
  140. In the light of the above, Celesio submits that the OFT has failed to discharge the onus upon it to show with a sufficient degree of certainty that it was clearly the case that no SLC could arise in relation to any 4 to 3 areas.
  141. The OFT's submissions
  142. The OFT submits that Celesio misunderstands the OFT's reasoning in so far as 4 to 3 fascia reduction areas were not considered to present a problem. Referring to Pritchard at paragraphs 99 to 102, the OFT submits that it did not decide a priori that 4 to 3 reduction was of no concern. It worked upwards, incrementally, from the worst case, i.e. it started with the largest potential effect on competition. Hence the OFT began by considering 2 to 1 reductions, and concluded that they may give rise to SLC. Then it considered 3 to 2 reductions and found that generally they were unlikely to give rise to SLC, but reached the view that in some areas it might just give rise to SLC; and the OFT could not determine to its satisfaction in which particular 3 to 2 areas that may or may not be the case. Given that the OFT only narrowly ("on balance") formed that view that it may be the case that the merger would be likely to give rise to SLC on 3 to 2 reductions, it was not necessary to consider in detail the next level of reduction, i.e. 4 to 3. It is the evidence supporting this reasoning that is referred to in the first sentence of paragraph 46 of the Decision.
  143. In its oral submissions the OFT also pointed to the facts that the focus was on a very small geographical area (a one-mile radius) and that the market is highly regulated with limited scope for competition anyway. The OFT drew attention to Pritchard at paragraph 117 and submitted that that paragraph is not in the least bit inconsistent with the Decision; nor does it place a gloss on the Decision. Contrary to Celesio's submissions, the OFT did not find that SLC "will" arise in 2 to 1 areas, simply that it found it likely in 2 to 1 areas and, on balance, that it may be the case in 3 to 2 areas, but they cannot be conclusive.
  144. In response to Celesio's submission that also in 4 to 3 areas Boots and UniChem stores may be in closest proximity, the OFT submitted that whilst that might be the case, it might equally be the case in 5 to 4 or 6 to 5 areas, but Celesio does not suggest that the OFT should have looked at all 5 to 4 or 6 to 5 areas. The OFT submitted that where one is examining a very small local area there is a point when the other fascia, albeit several hundred yards away, provide a sufficient constraint, particularly when the scope for competition is so limited in the first place. The question where to draw the line is, in the OFT's submission, precisely what falls within the OFT's margin of assessment in determining whether the lessening of competition is substantial.
  145. As to whether the OFT acted irrationally in exercising its judgment in the way that it did, the OFT notes Celesio's submission that when competition in the market is already weak, one should be more, not less, concerned about any reduction of competition. The OFT submitted that the assessment the OFT must conduct is not so simple. The question is "How much competition is lost?" rather than "How much was there in the first place?".
  146. In response to Celesio's submission that the OFT should have considered the parties' market shares in local areas by outlet numbers, the OFT submitted that Celesio's argument in this respect is illogical. If Celesio's approach is correct, then that should be applied equally in relation to 5 to 4 areas or even 6 to 5 areas, because in respect of both such areas one could certainly envisage the merger resulting in certain localities in more than 50 per cent share by outlets. Furthermore, this approach would have to be used as a filter to exclude from concern some of the 3 to 2 areas. The OFT refers to paragraph 78 of Pritchard, and submitted that Celesio's approach risks both over-inclusion and under-inclusion.
  147. As to the second and third sentences of paragraph 46 of the Decision, the OFT submits as follows:
  148. (a) The report of the view of a major competitor is set out but did not form any part of the OFT's reasons. The OFT accepts that this sentence therefore should not have been in the Decision: Pritchard, paragraph 120.
    (b) The 1996 MMC Report was relied on to the extent that the MMC had found that even in localities with a 2 to 1 reduction, it expected any detriment to competition to be small. The OFT noted, however, that this was before the removal of RPM on OTC medicines, but found no evidence that this change alone meant that a 4 to 3 reduction may give rise to competition concerns.
  149. The OFT submits that, far from the conclusion regarding a 4 to 3 reduction being inconsistent with the previous paragraphs in the Decision, it in fact followed from them and from the view formed regarding 3 to 2 areas.
  150. The interveners' submissions
  151. The interveners submit that the OFT's view that reductions from 4 to 3 fascias did not give rise to competition concerns was entirely logical. The OFT had decided that 2 to 1 areas were a concern and that 3 to 2 areas were a concern but only on balance, out of an abundance of caution. Having taken that view of areas where fascias were reduced from 3 to 2, the interveners submit that it was entirely rational for the OFT not to identify concerns in areas where fascias were reduced from 4 to 3 (or higher) and that Celesio's attempt to cast what was a lawful decision-making process as some species of public law error is misleading.
  152. The interveners submit that Celesio's challenge is to the OFT's exercise of judgment over facts on one issue, namely the incremental local competition added by a fourth fascia, primarily in the retailing of P medicines. In their submission, this contrasts with the point on which the applicant in UniChem succeeded, which related to the foundation for a key finding in the OFT's clearance decision concerning the nature of the applicant's distribution operations.
  153. The interveners submit that it is clear from the Decision that the decision-making process was lawful: the OFT consulted numerous third parties on the issue, turned its mind squarely to the correct question, considered a mixture of evidence on the point, reached a nuanced judgment that a 4 to 3 reduction could, at least in theory, give rise to a LC but not a SLC, and gave reasons for that view.
  154. The interveners submit that even if the Decision on its face contains inadequate reasons for reaching the OFT's conclusion on a 4 to 3 fascia reduction, it is clear that the OFT may amplify its reasons by evidence, and that the Tribunal's task is to examine whether the material ultimately before it, taken as a whole, disclosed grounds on which the OFT could reasonably have reached the conclusion it did.
  155. The interveners submitted extensive evidence to the OFT before the Decision was made. This demonstrates that there was considerable evidence before the OFT to support its conclusions, including on the limited scope for competition in retail pharmacy due to the tight regulatory framework, the complementary nature of the interveners' business models, the impact of deregulation and the evidence from the 2003 Report, which found no correlation between local concentration and prices of P medicines.
  156. The interveners submit that it is striking that Celesio submitted no evidence to the OFT at all that a reduction to 3 fascias in an area would give rise to competition concerns. It seems that no third party argued in favour of a 4 to 3 fascia rule. The reality, say the interveners, is that Celesio disagrees with the merits of the OFT's judgment on this point.
  157. The interveners submit that Celesio's argument that the OFT's conclusions on 4 to 3 areas are inconsistent with paragraphs 43 to 44 of the Decision is unpersuasive. The OFT in those paragraphs states that it has reached a judgment on the evidence before it (in particular the regulatory structure of the market and the fact that UniChem is a small player in OTC) that a 2 to 1 fascia reduction could give rise to a SLC, a 3 to 2 reduction could on balance do so (though less conclusively), and a 4 to 3 reduction could in theory give rise to a lessening of competition but not a substantial one. In the interveners' submission, there is nothing illogical or inconsistent in that reasoning.
  158. The interveners submit that Celesio's submissions in these proceedings are inconsistent with those it advanced before the OFT almost wholly at the same time in the Lloyds/Cohens & Scholes case. Celesio's submissions in Lloyds/Cohens & Scholes were effectively identical to those made by the interveners to the OFT in this case. Pointing to Celesio's written submissions to the OFT in Lloyds/Cohens & Scholes, the interveners submit that Celesio accepts that:
  159. (a) regulation governs a very large number of facets of competition;
    (b) there is no suggestion that P medicines are substantially, or at all, problematic;
    (c) there is no material problem at the local level: local retail competition is sufficiently strong that even 2 to 1 reductions are unproblematic in respect of both price and non-price competition.
  160. The interveners submit that the inconsistency in Celesio's position is highly relevant to the weight to be attached to the evidence submitted to the OFT. Celesio's submissions before the OFT did nothing to contradict the evidence and submissions it had received in this case. The interveners submit that the Tribunal should deprecate those parties who play "fast and loose" with different sets of submissions or in the same proceedings, both to ensure the protection of merging parties' procedural rights and to avoid undermining the integrity of the system.
  161. AMENDED GROUND 2
    Celesio's submissions
  162. Celesio submits that, given that the OFT's analysis of national retail markets in the Decision depends materially on its analysis of local retail markets, the inadequacy of the OFT's consideration of the local retail markets inevitably requires that the conclusions in relation to national retail markets should also be quashed. In that regard, Celesio referred the Tribunal to paragraph 14 of the Decision, which states:
  163. "…the OFT notes that the extent to which national chains face each other in local areas may well be a factor in determining the extent to which they can influence each other's decision making at the national level, e.g. on pricing policies."
  164. Celesio further submits that paragraphs 37 and 40 of Pritchard support its view that the Decision's analysis of the national retail market depends materially on the analysis of local retail markets.
  165. In those circumstances, Celesio submits that, to the extent that the OFT, having erred in its analysis of local markets, must reconsider its analysis of these markets, the OFT will therefore inevitably have to consider the impact of such further analysis on its conclusions relating to national retail markets. In that respect, Celesio also referred the Tribunal to paragraph 31 of the Decision. Celesio noted that the document upon which the analysis in that paragraph is based is a Boots internal document which does not identify areas in which Boots and Moss both have pharmacies but merely identifies the types of locations in which Boots, Moss and other fascias appear and does not provide any specific information about actual area overlaps. Celesio submits that in reaching the conclusion that Boots and UniChem "do not tend to be located close to each other", the Decision relies solely on the contentions of the merging parties and on a document which did not actually address the issue of overlaps in particular local areas.
  166. Moreover, Celesio submits that whilst the Decision relies on the areas in which Boots and UniChem have the most shops, it is evident from the numbers of shops in other areas that it is perfectly possible that there will be significant numbers of other local areas in which Boots and UniChem might both have shops, such as "small rural centres" and "small suburban centres".
  167. In its oral submissions Celesio referred the Tribunal to Appendix A to the report of Mr Coombs, referred to in paragraph 80 above, and submitted that this is an example of the type of evidence that the OFT should have looked at when analysing the national retail markets.
  168. In light of the above, Celesio submits that if its challenge to the OFT's analysis of horizontal local retail markets is successful, any remittal to the OFT must also require reconsideration of horizontal national retail markets.
  169. The OFT's submissions
  170. The OFT submits that paragraphs 26 to 32 of the Decision clearly consider national competition separately from local factors and, in those circumstances, the criticism made by Celesio of the analysis of local competition is not material to the way the OFT referred to local competition in its analysis of national competition. In particular the OFT submits that at paragraph 31 of the Decision it was looking at the general character of local competition as it applied across the country and not local competition in particular localities.
  171. In respect of Appendix A to the report of Mr Coombs, the OFT notes that this is a merger between Boots (which has 1,350 retail pharmacies) and UniChem (which has 958 retail pharmacies). Even on Celesio's figures there are only 41 local areas with a 4 to 3 reduction, involving 44 Boots pharmacies and 52 Moss pharmacies (belonging to UniChem), which equates to less than 5 per cent of either party's estate. The OFT submits that to say that an error of assessment of that number of cases vitiates its conclusion at the national level in respect of more than 1,000 stores is preposterous.
  172. The interveners' submissions
  173. The interveners submit that, in considering the points Celesio raises under this ground, the Tribunal must examine whether all the material that was before the OFT, taken as a whole, disclosed grounds on which the OFT could reasonably have reached the conclusion it did.
  174. This includes the considerable evidence on national competition submitted by the interveners to the OFT in relation to the impact of the regulatory framework and evidence of competition from several other major chains in the UK retail pharmacy sector, which Celesio itself recognised in its submission on the Lloyds/Cohens & Scholes merger.
  175. The interveners submit that the OFT properly considered that evidence alongside the evidence it received from third parties, and reached the reasonable, logical, justifiable, and correct view that the merger would not give rise to a SLC where national competition was concerned.
  176. In respect of Celesio's reliance on Appendix A, the interveners submit that the proposition that 41 local areas with a 4 to 3 reduction is sufficient materially to affect the analysis and the strategy of the merged company in the totality of its national estate is both remote and improbable. In those circumstances, the interveners submit that amended ground 2 discloses no reviewable error at all, let alone a material one.
  177. AMENDED GROUND 3
  178. As noted in paragraphs 52 to 54 above, pursuant to section 73 of the Act the OFT is engaged in seeking undertakings from the merging parties to cure the potential SLC identified in the Decision. The OFT accepts that, in the event that the Tribunal finds that Celesio is correct in respect of amended grounds 1 and 2 and the findings in the Decision as regards SLC are quashed, the OFT cannot accept the proposed undertakings in lieu of a reference to the CC.
  179. The interveners submit that Celesio has provided no ground for quashing or remitting the Decision to the OFT, and, accordingly, the interveners' undertakings remain appropriate to curing the potential SLC identified by the OFT in the Decision.
  180. X THE TRIBUNAL'S ANALYSIS

    The test for the admissibility of Pritchard
  181. Celesio notes that the admissibility of witness statements was considered by the Tribunal in Somerfield, cited above. Celesio draws attention to the following paragraphs of that judgment:
  182. "64. The application of this approach is a matter of fact and degree in each case. The general principle in [R v Westminster City Council, ex p Ermakov [1996] 2 All ER 302] was expressed by Hutchison LJ in these terms at 315h:
    'The court can and, in appropriate cases, should admit evidence to elucidate or, exceptionally, correct or add to the reasons; but should, consistently with Steyn LJ's observations in Ex p Graham, be very cautious about doing so.'
    67. Accordingly, we would anticipate that in most cases such as this supplementary witness evidence from the CC should be kept to a minimum. As with judicial review generally, any witness statements that are necessary should be closely cross-referred to the report under consideration, with any appropriate explanation of the relevance of the additional evidence, bearing in mind that it is the report, not the witness statement, that is the subject of the review. While it may well be helpful for a witness statement to elucidate technical matters contained in the report or respond to evidence submitted by the applicant, witness statements are not submissions and should not need to repeat or place any particular "gloss" on the report in question. …"
  183. Celesio submits that although the judgment in Somerfield was concerned with a review of a CC report, it is clear that the above statements of principle apply to judicial review, and hence to reviews under section 120 of the Act, generally.
  184. Celesio also relies on paragraph 106 of IBA Health, cited above, where Carnwath LJ noted that the Court may need to be "circumspect" to ensure that witness evidence "is not used as means of concealing or altering the true grounds of the decision". In Celesio's submission, it follows that, where a witness statement "glosses" a decision, or is inconsistent with it, the Tribunal must have regard to the reasons given in the Decision rather than the statements contained in the witness statement.
  185. In so far as there is inconsistency between the Decision and Pritchard, Celesio submits that it is the Decision which must be considered and that Pritchard is inadmissible.
  186. In its oral submissions the OFT accepts that Pritchard cannot be relied upon if and in so far as it puts forward reasoning that is inconsistent with the Decision. However, the OFT does not accept that Pritchard is inconsistent with the Decision. The OFT submits that it can rely on Pritchard by way of amplification or explanation of the reasoning process. In particular the OFT relies on the following dicta of Carnwath LJ in IBA Health:
  187. "ADEQUACY OF REASONS
    [102] Finally, although inadequacy of reasons is not a ground of challenge as such, it may be helpful to comment briefly on the tribunal's observations on this aspect.
    [103] The tribunal expressed concern at having to consider material outside the decision letter. It noted that the OFT was under a statutory duty to give its reasons, and referred to what it called the 'well-known principle' that 'the court should, at the very least, be circumspect about allowing material gaps to be filled by affidavit evidence or otherwise' (see para 257, citing R v Westminster City Council, ex p Ermakov [1996] 2 All ER 302 at 312 per Hutchison LJ). It commented (at para 258):
    'If a material element is not set out in the decision, it is very difficult for the reviewing court or tribunal to be satisfied that the matter was properly investigated or that the supplementary reasons did in fact form part of the decision-making process.'
    In other parts of the judgment, the tribunal criticised the failure of the OFT to set out all the underlying material (see paras 211, 252).
    [104] With respect, I think this concern, and the associated criticisms, were misplaced. The statutory duty to give reasons is an important one, but it is not the same as a duty to give a 'judgment' (such as that of a court) or a duty to make a 'report' (such as that of an inquiry inspector). Again reference to the textbooks might have assisted. The numerous cases on the subject lay down no general test, other than the requirement that reasons must be 'intelligible and must adequately meet the substance of the arguments advanced' (see Re Poyser and Mills' Arbitration [1963] 1 All ER 612 at 616, [1964] 2 QB 467 at 477-478, cited in De Smith, Woolf and Jowell p 465 (para 9-049, n 8) as 'the most frequently cited judicial articulation of the test'; see also Wade and Forsyth pp 916–919).
    [105] In a case such as the present, where the subject matter is complex and the supporting material voluminous, there is no statutory requirement for all the evidence to be set out in the decision letter. However when a challenge is made,  there is, as the tribunal noted, an obligation on a respondent public authority to put before the court the material necessary to deal with the relevant issues; 'all the cards' should be 'face upwards on the table' (see R v Lancashire CC, ex p Huddleston [1986] 2 All ER 941 at 945).
    [106] There is certainly nothing unusual, particularly in a case which has to be dealt with in a relatively short timescale, for the stated reasons to be amplified by evidence before the court. While in some areas of the law, the court may need to be 'circumspect' to ensure that this is not used as means of concealing or altering the true grounds of the decision, that does not arise in this case. As I understand it, no objection had been taken to any of the evidence being put before the tribunal (or to the additional evidence adduced in the Court of Appeal). The question for the tribunal was not whether the reasoning was adequately expressed in the decision, but whether the material ultimately before it, taken as a whole, disclosed grounds on which the [OFT] could reasonably have reached the decision it did."
  188. In respect of the Somerfield judgment, the OFT notes that Celesio failed to refer to paragraph 66, which states:
  189. "In our respectful view the remarks of Carnwath LJ in IBA Health, cited above, arise in the different context of the "short-form" clearance decisions given by the OFT under sections 22 and 33 of the Act, where the duty to give reasons arises not under section 38 but under the more general provisions of section 107 (in particular section 107(1), (4) and (5)). In our view the details and reasoning of the OFT's decision on preliminary investigations under sections 22 or 33 cannot be expected to be set out to the same extent as those of the CC's report following a formal in-depth inquiry."
  190. In the OFT's submission, paragraph 66 is crucial as it draws a very clear distinction between the situation that arises on a review by the CC, producing a very clear report over a much more extended timetable, and a judicial review of the OFT on a merger reference. The OFT submits that in a decision under section 33 of the Act, where the reasons may be quite understandably abbreviated and concise, evidence by way of amplification or explanation of the reasoning process is admissible.
  191. The test that we are applying for the admissibility of Pritchard is that set out by Carnwath LJ in IBA Health.
  192. We would mention, however, that under section 107 of the Act the OFT is statutorily required to publish its reasons. In our view, the purpose of the duty to give reasons for a decision under section 33 is the same as under section 38: first, to allow interested parties to know the justification for the measure so as to enable them to protect their rights, to know the outcome and to enable them to assess whether they have any ground for challenging an adverse decision; secondly, to enhance public confidence in the decision-making process through the knowledge that decisions supported by sound reasons must be given; and thirdly, that giving reasons concentrates the mind, so that the resulting decision is much more likely to be soundly based on the evidence than if they are not given (see Ermakov, cited above, at 309f – 310c).
  193. We would also stress that the remarks of the Tribunal at paragraph 67 of the Somerfield judgment apply equally to witness statements of the OFT. Whilst Pritchard did contain some cross-referencing to the Decision, it would have been helpful to the Tribunal if it had followed more closely the structure of and the paragraphs in the Decision. This would have made the Tribunal's task of assessing whether Pritchard simply elucidated the Decision or went further an easier one.
  194. Application of the admissibility test to Pritchard
  195. Celesio submits that the attempt in paragraphs 103 to 106 of Pritchard to play down the effect on competition of a 2 to 1 reduction is inconsistent with paragraph 43 of the Decision, which states that a 2 to 1 reduction could "clearly" give rise to a SLC, and paragraph 44 of the Decision, which found that a 3 to 2 reduction may also give rise to a SLC because switching of customers between Boots and UniChem stores may be high and, for a large proportion of customers, Boots and UniChem would be the closest competitors.
  196. In relation to 3 to 2 areas, Celesio submits that at paragraphs 109 to 114 Pritchard suggests that the OFT concluded that (a) there was no problem in relation to the "vast majority" of the 3 to 2 areas, but that (b) a minority of them were "marginally problematic". In Celesio's submission, this attempted gloss is inconsistent with paragraphs 44 to 45 of the Decision, which state that a reduction from 3 to 2 fascias may lead to a SLC in all such areas, not just a minority (although there may be a particular problem in those localities where the merging parties' pharmacies are located close together).
  197. Celesio further submits that, in contrast to Pritchard, the Decision finds that there is a problem in all 3 to 2 areas and that the potential effect on competition is sufficiently significant that it may give rise to a SLC (i.e. it is not merely "marginally problematic").
  198. Celesio submits that Pritchard is in part inconsistent with or at the very least a gloss on the Decision and therefore cannot be relied upon by the Tribunal.
  199. As set out above, the OFT accepts that Pritchard cannot be relied upon if and in so far as it is inconsistent with the Decision. However, the OFT does not accept that Pritchard is inconsistent with the Decision. The OFT submits it can rely on Pritchard by way of amplification or explanation of the reasoning process in particular since an OFT decision under section 33 of the Act may be somewhat abbreviated and concise.
  200. We note that in the appeal before us the OFT had to work within the statutory time limit of 45 days to produce its Decision under section 33. We note that the OFT ascribes eight weeks to its period of investigation during which it made six formal information requests of the merging parties under section 34B of the Act (Pritchard, paragraph 12), had meetings with the merging parties and received written submissions from them, invited comments from all interested third parties, directly contacted more than 60 PCTs or Health Boards and over 35 other consultees, and received 28 third party responses (Pritchard, paragraphs 13 and 14).
  201. The Tribunal notes that Pritchard sets the Decision into context. It explains that the OFT's existing awareness of the limited nature of competition within the retail pharmacy market informed its analysis of and conclusions on the local retail pharmacy market. Pritchard refers to the substantial experience which the OFT had developed in competition issues affecting the retail supply of products and services by pharmacies. It refers to the OFT having considered the 1996 MMC Report and to the 2003 Report, published in January 2003, which was supported by econometric analysis commissioned by the OFT and carried out by Frontier Economics. In addition Pritchard refers to the OFT's recent consideration, between October and December 2005, of the then proposed (and now completed) acquisition of a chain of Cohens and Scholes pharmacies by Lloyds (Pritchard, paragraphs 45 to 52 and 104).
  202. In relation to horizontal issues, Pritchard also refers to the experience which the OFT had gained from its consideration of the wholesale sector: in 2000, when it gave negative confidential guidance in respect of a proposed acquisition by UniChem of East Anglican Pharmaceuticals ("EAP"); in 2003, when it decided to refer the proposed acquisition by AAH of EAP; and in December 2004, when it considered the acquisition by Phoenix Healthcare Distribution Limited of EAP (Pritchard, paragraph 9).
  203. Pritchard explains the method of analysis which the OFT adopts when considering whether to refer a situation under section 33 of the Act. Pritchard refers to the OFT's Mergers – procedural guidance (OFT 526, May 2003) and its Mergers – substantive assessment guidance (cited above) under the Act (Pritchard, paragraph 42). Pritchard explains that the OFT takes an assessment of market concentration as an initial starting point before a full assessment of competitive effects.
  204. Pritchard states (in paragraph 42) that typically the OFT proceeds as follows:
  205. "First, it identifies the measure(s) of concentration most appropriate in the circumstances of the case. Second, it carries out a full assessment of competitive effects within a framework set by the concentration measure adopted. In this case, the OFT concluded that a fascia test was the most appropriate measure of concentration on the facts, and its assessment of competitive effects was carried out within a framework set by the fascia test, i.e. the OFT first analysed competitive effects in local areas with a 2:1 fascia reduction, and then moved on to consider 3:2 and 4:3 areas."
    and in paragraph 62 that:
    "the OFT uses concentration tests to provide an initial indicator of potential concerns before carrying out further and a more detailed investigation for the likely competition effects of a merger. In this case, the OFT adopted the fascia test as its initial concentration test. As well as providing an initial and rough guide to the level of likely concern in different local markets, the fascia test here also provided the OFT with a framework or structure in which it carried out its more detailed analysis, as the OFT analysed the different markets in "fascia-reduction groups", beginning with 2:1 local areas, and progressing to 3:2 and 4:3 areas."
  206. Pritchard then explains how the OFT applied its general approach to the Boots/UniChem case. In paragraph 67 it states:
  207. "…. the OFT carried out a full assessment of competition in local areas, and did not limit itself to fascia considerations. Rather, the OFT expressed the conclusions of its full competitive assessment in terms of different fascia reduction areas, and used the fascia test to provide a structure for its analysis, progressing from 2:1, to 3:2 and finally to 4:3 areas, as is evident from paragraphs 43 to 46 of the Decision."
    and in paragraph 99:
    "Once a fascia test is adopted, the next step is to class areas according to the relevant reduction of fascia: 2:1, 3:2, 4:3, 5:4 area and so on. Each step in concentration reflects the loss of one independent consumer choice in an area. The basic intuition is that the lower the number of choices to begin with, the more significant the loss of any single choice is likely to be."
    and in paragraph 100:

    "…, the OFT considered it appropriate to use an iterative approach beginning with the most concentrated markets: the 2:1 or post–merger monopoly areas, where consumer choice is effectively eliminated. If further competitive effects analysis reveals ongoing concerns for 2:1 areas, the OFT next considers 3:2 areas, and the iterative process continues until no concerns arise at a given fascia reduction. The rationale is relatively simple: if effects analysis for an area of a higher degree of concentration fails to raise concerns (say 2:1), this will tend to obviate the need for further iterative effects analysis at lesser degrees of concentration (3:2 and beyond). In this case, beginning with the most highly concentrated areas is particularly appropriate given the peculiarly limited scope for competition interaction over and above regulated price and service levels in local pharmacy markets."
    and in paragraph 102:

    "That the OFT's analysis developed incrementally in this way is evident from paragraphs 43 to 46 of the Decision… [T]he OFT's conclusion on 4:3 areas was reached in the light of the overall assessment of the level of competition in the market, informed by evidence referred to throughout the Decision."
    and in paragraph 103:
    "The OFT assessment of the potential for a substantial lessening of competition ("SLC") therefore began in respect of 2:1 local areas. While the OFT had regard to all the evidence in making its assessment of 2:1 local areas, its reasoning was fairly straightforward."

    and in paragraph 105:

    "However, although the evidence was not entirely conclusive, the internal documents of the merging parties supported the conclusion that there existed some, albeit limited, competition. A 2:1 merger would, by definition, eliminate all such existing competition. As for the threat of competition from entrants, there was insufficient evidence to conclude that the relaxation of the entry requirements had been effective in reducing barriers to entry. Accordingly, the OFT decided that it is the case that the merger may be expected to result in a substantial lessening of competition in 2:1 areas."
  208. Pritchard then turns to the OFT's approach in respect of 3 to 2 areas. It notes in paragraph 108 that 3 to 2 areas were significantly different from 2 to 1 areas in that competition, creating consumer choice, is not eliminated within the area, and the merger does not create a local monopolist. It refers to the evidence which suggested that demand for pharmacies is driven primarily by convenience and that if customers are willing to travel up to 1 mile for a pharmacist, the remaining competitor in the isochrone represents an effective consumer choice which the merged company must necessarily take into account.
  209. Pritchard then notes that the OFT developed concerns in relation to 3 to 2 areas only because, following its scrutiny of the relevant maps, it could identify a minority of 3 to 2 areas that closely resembled 2 to 1 areas (i.e. where the remaining competitor was within 1 mile of the merging parties but significantly further away from the merging parties than they were to each other). Pritchard states that the concerns were marginal because, first, the basis for concern ran contrary to the proposition underpinning the OFT's geographic market definition – that consumers would travel up to 1 mile to visit a pharmacy – and, secondly, given the very limited nature of competition in the market there was a serious question as to whether the presence of both merging parties (as two fascias) in the market, along with a third fascia, was really making conditions substantially more competitive than would exist with just two fascias (Pritchard, paragraphs 109 to 110).
  210. Pritchard sets out the OFT's considerations in relation to 3 to 2 areas as follows:
  211. "111. In considering whether the merger reference test was met in relation to this problematic minority of 3:2 areas, the OFT took into account all the evidence before it in the case, summarised throughout the Decision. The Decision required careful judgement. On balance, the OFT did not believe that it was likely – that it "is the case" under s 33 of the Act – that the merger may be expected to result in a substantial lessening of competition, even in the 3:2 areas that, on closer examination of the relevant maps, resembled a 2:1 area. In other words, even in respect of these areas, the OFT considered that it was unlikely to result in an SLC. In effect, this was because there was an effective competitive choice for consumers within a 1-mile radius of the merged firm, acting as a disincentive and constraint on it to raise price or reduce service. However, the OFT believed that it "may be the case" that the merger may be expected to result in an SLC within the meaning of s 33 of the Act. The OFT did not feel able to conclude that the prospect of a SLC in these areas was merely fanciful. Rather, it considered the statutory reference test to be satisfied in respect of these problematic areas.
    112. Given this position, the OFT then considered whether it was possible to segregate the vast majority of 3:2 areas that were not marginally problematic from the minority that were. The OFT therefore devoted considerable time, including ultimately at the [case review] meeting, to the question of whether it could reasonably apply a "secondary filter" to the 3:2 areas to discriminate in a robust manner between those few "may be the case" 3:2 areas from the remainder.
    113. After considering whether we could develop a principle sufficiently robust to account for classifying all 3:2 areas into "reference" and "no reference" categories, we determined that it was not possible to do so within the statutory timetable. Moreover, such an approach would tend, again, to question the fundamental premise of geographic market definition, which had substantial evidentiary support in this case (and had been used, for example, by the MMC).
    114. The OFT was therefore obliged to conclude, reluctantly, as a result of the limited number of 3:2 areas that were marginally problematic, its duty to refer applied "across the board" to every 3:2 area. I say reluctantly because we were aware that, depending on the undertakings in lieu proposals, that this might compel a reference in respect of areas where none was necessary, or result in divestments where none was really necessary. Nevertheless, this issue is consistent with our first-phase role in UK merger control, despite the extensive investigation and experience applied to this particular case."
  212. It is submitted by Celesio that Pritchard in paragraphs 112 to 114 gives evidence which may go beyond the Decision and is inconsistent with it. Celesio submitted that it is clear from the Decision that the OFT concluded that it may be the case that SLC may be expected in all 3 to 2 areas and not just where Boots and UniChem are close together. However, during the hearing Celesio accepted that paragraphs 44 and 45 of the Decision may be ambiguous. We have therefore carefully considered paragraphs 44 and 45 of the Decision, and in particular the last three sentences of paragraph 44 and paragraph 45. These read:
  213. "Even in these areas, switching of customers between Boots and Unichem stores may be high, and for a large proportion of customers, Boots and UniChem stores would be the closest competitors. This may arise particularly in those localities where the merging parties pharmacies are located close together. In such a situation the competitive scenario post merger does not significantly differ from that in the two to one area." (paragraph 44)
    "This has to be considered alongside the high barriers to entry in this sector combined with what the parties' internal documents say on the issue of competition – see, for instance, the references at paragraph 37 above to the monitoring of various competitor price and service quality variables. On balance, the OFT believes that a substantial lessening of competition may go beyond those areas outlined at paragraph 43 and may also arise where fascia are reduced from three to two within a one mile radius." (paragraph 45)
  214. In the sentences of paragraph 44 set out above, the OFT first observed that in 3 to 2 areas switching of customers between Boots and UniChem may be high and that for a large proportion of customers Boots and UniChem would be the closest competitors. It then observed that this (i.e. high switching of customers between Boots and UniChem and Boots and UniChem being the closest competitors) may arise particularly in those localities where the merging parties' pharmacies are located close together. Finally, it observed that "in such a situation" (i.e. a situation of high switching of customers and Boots and UniChem being the closest competitors for a large proportion of customers, which may arise particularly in those localities where the merging parties' pharmacies are located close together) the competitive scenario post-merger does not significantly differ from that in the 2 to 1 areas.
  215. In paragraph 45 the OFT noted that alongside high barriers to entry in this sector, the merging parties' evidence indicated that there was some scope for competitive behaviour. The OFT then stated its belief that on balance, a substantial lessening of competition may go beyond 2 to 1 areas and may also arise in 3 to 2 areas.
  216. It seems to us that on a proper reading of the Decision, the OFT is drawing a distinction between two categories of locality within the 3 to 2 areas: first, those localities where switching is high and for a large proportion of customers Boots and UniChem would be the closest competitors, which may arise particularly where the merging parties' pharmacies are located close together; and secondly, those where this is not the case. It seems to us that paragraph 45 follows on from paragraph 44 and accordingly in paragraph 45 the OFT was considering the first and not the second category of locality. It is only in the first category of locality that the competitive scenario post-merger "does not significantly differ from that in the 2 to 1 areas" and the OFT believes that on balance a substantial lessening of competition may also arise where fascias are reduced from three to two within a one mile radius.
  217. Paragraph 109 of Pritchard elucidates the Decision by explaining that the OFT's concerns in relation to 3 to 2 areas were in "a minority of 3:2 areas that closely resembled 2:1 areas" where the remaining competitor was within 1 mile of the merging parties but significantly further away from them than they were to each other. In our view, this is consistent with the penultimate sentence of paragraph 44 of the Decision, and in particular with the use of the word "may" in that sentence, since Pritchard elucidates that the OFT's concerns were confined, within the localities where the merging parties' pharmacies were located close together, to those localities where the remaining competitor was within 1 mile of the merging parties but significantly further away from the merging parties than the merging parties were from each other.
  218. In paragraphs 113 to 114 Pritchard explains that the OFT determined that it was not possible for it to develop a principle sufficiently robust to classify all 3 to 2 areas into "reference" and "non-reference" categories within the statutory timetable. For that reason it concluded "reluctantly" that as a result of the limited number of 3 to 2 areas that were marginally problematic, it was under a duty to refer the merger in respect of all 3 to 2 areas.
  219. We are satisfied that in the Decision the OFT makes a distinction within the 3 to 2 areas between the first and second categories set out above, and that it was only the first category that gave rise to a concern that there may be a SLC.
  220. At paragraph 46 of the Decision the OFT states:
  221. "Any higher reduction in fascia number than this (e.g. four to three or higher) could also give rise to a lessening of competition but on the basis of the evidence before the OFT, it believes that this cannot be expected to be substantial. One major competitor suggested that it would be usual to face two other competitors within a local area. Moreover, the CC's 1995 report considered only two to one fascia reductions. While there have been small changes to the market since then (primarily the removal of resale price maintenance on OTC medicines) which might suggest some small increase in the scope for potential competition, these changes do not support an argument that reductions from four to three fascias might give rise to competition concerns."
  222. Pritchard has explained that the second and third sentences of paragraph 46 were supplemental observations and not reasons for its conclusion. The evidence referred to in the second sentence did not form part of the OFT's reasoning on any aspect of the case and should not have been included in the Decision. The reference to the 1996 MMC Report (erroneously referred to in paragraph 46 of the Decision as "the CC's 1995 report") was intended to add some historical context to the OFT's conclusion (see Pritchard, paragraphs 118 to 120). It seems to us unfortunate that the OFT chose, at the time of drafting the Decision, to include two observations in paragraph 46 which were not being relied upon as reasons for its Decision.
  223. The OFT refers, in the first sentence of paragraph 46, to "the evidence before the OFT". It is unfortunate that the OFT only made reference obliquely to the evidence before it in paragraph 46 rather than identifying or summarising the evidence on which it was relying for its belief that any lessening of competition cannot be expected to be substantial in areas where there was a higher reduction in fascia numbers than 3 to 2.
  224. However, looking at paragraphs 43 to 46 of the Decision it appears to us that "the basis of the evidence" to which the OFT refers in paragraph 46 includes all the matters to which we have referred above when discussing paragraphs 43 to 45.
  225. In paragraph 117 Pritchard turns to the OFT's assessment in respect of areas of lesser concentration (4 to 3 and beyond) and the OFT's conclusion in paragraph 46 of its Decision. In paragraph 117 Pritchard states:
  226. "The OFT's conclusion in respect of 4:3 areas was, as in the case of other areas, reached after a detailed consideration of all the evidence in the case. In short, the OFT was already concerned that a finding of SLC, even on the lower "may be the case" standard, was marginal in respect of those limited number of 3:2 areas that arguably resembled 2:1 areas, but felt obliged to extend this lower level of belief across the entirety of the 61 3:2 overlaps. Once an additional effective competitor and choice for consumers within 1 mile of the merged firm was "added" to the competitive assessment, the merged firm would be disciplined by two effective competitors within 1-mile of its operations. Any prediction that harm would occur in the circumstances of local competition among pharmacies, became, in our judgment, plainly speculative and fanciful because it was unsupportable on the evidence. Hence the OFT's reference to its positive belief "on the basis of the evidence before the OFT" (Decision; paragraph 46). Having established a lack of concern in respect of 4:3 areas, there was no need to proceed further with the iterative approach."
  227. In the present case we are satisfied that it was appropriate for Pritchard to explain the matters which were the subject of amended ground 1 of the application and to elucidate in particular paragraphs 43 to 46 of the Decision. Whilst we consider that the OFT could perhaps, in the time available to it, have incorporated in the Decision a certain amount of the explanation provided by Pritchard as to the OFT's reasoning process, we consider that:
  228. (a) its content is not inconsistent with the Decision; and
    (b) it contains elucidation of the Decision.
  229. We have taken Pritchard into account only for the purposes of elucidation. On that basis we do not accept the submission of Celesio that Mr Pritchard's witness statement is inadmissible or should not be given any weight.
  230. The Decision in respect of 4 to 3 areas
  231. Against the background set out in Pritchard we turn to consider the OFT's reasons for reaching its decision not to refer 4 to 3 areas. However, we must first deal with the following points raised in Celesio's submissions:
  232. (i) SLC in the context of weak competition in the relevant markets; and
    (ii) Share of outlets evidence.
    - SLC in the context of weak competition in the relevant markets
  233. As set out at paragraph 96 above, Celesio suggested that where competition in the relevant markets is weak, there should be more, not less, scrutiny. In that regard, the OFT submitted that the question that arises is not how much competition was there in the first place but rather how much competition is lost as a result of the merger (see paragraph 107 above).
  234. As stated in paragraph 3.8 of Mergers substantive assessment guidance, cited above, "[t]he core concept of the substantial lessening of competition test is a comparison of the prospects for competition with and without the merger", which requires the OFT to look at the "possible loss of rivalry". We agree that the SLC test requires the OFT to consider what competition may be lost as a result of the proposed merger.
  235. We do not accept Celesio's submission that when competition in a particular market is weak this perceived lack of competition makes any further reduction resulting from the merger more serious and thus should have led to more careful scrutiny by the OFT in the present case of the 4 to 3 areas. We consider that the focus should be on the loss of competition, if any, which may be expected to result from the merger.
  236. - Share of outlets evidence
  237. Celesio only relies on Appendix B to Dr Padilla's report to the extent that, in Celesio's submission, it shows that the OFT failed properly to investigate the 4 to 3 areas. Celesio was not challenging the use of a fascia test as a starting point but submitted that the OFT should also have looked at share of outlets evidence on 4 to 3 areas. Celesio submits that Appendix B shows that in some 4 to 3 areas the merging parties' share in terms of the number of outlets exceeds 50 per cent.
  238. The OFT submits that as a means of identifying areas in which there may potentially be SLC problems a fascia analysis provides a better measure than a share of outlets analysis. This was supported by Pritchard at paragraph 71: "In particular, fascia analysis tends to capture the notion that it is the actual or threatened loss of custom to a rival, not owned by the same parent company, which spurs local competition between pharmacies…". The OFT also submits that a share of outlets analysis could mislead in that it could bring into consideration some areas in which there were several independent competitors and exclude from consideration some areas in which there was only a single independent competitor.
  239. It seems to us that the difficulty with Celesio's approach is not only that it would, in certain areas, be either over-inclusive or under-inclusive but also that it mischaracterises the SLC test, which requires the OFT to consider "the prospects for competition with and without the merger" and the "possible loss of rivalry". In this regard the Tribunal accepts that it was reasonable for the OFT, at least in the particular circumstances of the retail pharmaceutical sector, to take the view that the fascia test was a better measure for the market under consideration as it "emphasises the question best calculated to capture the merger effect – how many rival suppliers, and how many competitive consumer choices will remain in the area post merger?".
  240. In those circumstances we are satisfied that it was not unreasonable for the OFT to arrive at its Decision without considering share of outlets evidence.
  241. 4 to 3 areas
  242. It is important to have in mind the totality of the conclusions of the OFT as contained in the Decision (in particular paragraphs 46 and 82):
  243. (a) In terms of national competition no concerns arise.
    (b) On a local level on the basis of a one-mile radius around both Boots and UniChem pharmacies there are 38 areas where the merger would result in a 2 to 1 reduction in the number of competing pharmacies and a further 61 areas where it would result in a reduction in the number of fascias from 3 to 2.
    (c) The evidence clearly shows that a reduction in pharmacy fascias from 2 to 1 in a local area, despite the restrictive terms of the NHS contract, is expected to result in a substantial lessening of competition. Such SLC could take the form of reductions in quality or the level of service provided (over and above the levels stipulated in the contract). There may also be an impact on pricing, particularly of P medicines.
    (d) The evidence on whether a SLC would arise in the case of 3 to 2 areas is less conclusive but, on balance, it may be the case that the merger may be expected to result in a SLC within these 3 to 2 overlap areas.
    (e) Any higher reduction in fascia number than 3 to 2 (i.e. 4 to 3 or higher) could also give rise to a lessening of competition but the OFT believes that this cannot be expected to be substantial.
  244. Celesio's challenge to the OFT's conclusions is as follows:
  245. (a) The reasons given in the Decision are not capable of sustaining the OFT's conclusion that a reduction from 4 to 3 fascias in local retail markets would not give rise to a SLC;
    (b) This error of assessment in relation to local retail markets necessarily vitiates the OFT's conclusion relating to the national retail market;
    (c) It further follows that the OFT is not entitled to accept undertakings in lieu of a reference pursuant to section 73(2) of the Act.
  246. In considering whether Celesio's challenge is justified the starting point is the Decision itself.
  247. At paragraph 43 the OFT considered 2 to 1 areas. The evidence to which the OFT referred was as follows:
  248. (a) The 1996 MMC Report, which found that a reduction in pharmacy fascias from 2 to 1 in a one mile radius would not be against the public interest.
    (b) The fact that since that report RPM on OTC medicines has been removed.
    (c) UniChem is a relatively small player in the OTC sector.
    (d) Entry rules into the pharmacy sector have been relaxed (although the evidence as to the impact of this is inconclusive).
    (e) The merging parties' internal documents showed that:
    i. a level of competitive inter-action between retail pharmacies does exist; and
    ii. new entry will usually prompt some competitive response.
  249. Having considered this evidence the OFT concluded, as set out in paragraph 43, that a 2 to 1 reduction could lead to a reduction in the incentives to compete and have a detrimental effect on matters such as the standard and quality of the services provided by the pharmacies (over and above that stipulated by the NHS contract).
  250. At paragraphs 44 and 45 the OFT considered 3 to 2 areas. The points to which the OFT referred were as follows:
  251. (a) The pharmacy sector is unique in terms of being highly regulated.
    (b) A level of competition does exist in this area.
    (c) Switching between Boots and UniChem may be high for a large proportion of customers, particularly in those localities where the merging parties' pharmacies are close together.
    (d) There are high barriers to entry in this sector.
    (e) That there is some competitive interaction between different retail pharmacies. The OFT referred to the parties monitoring a number of different price and service quality variables and comparing themselves to competitors; to Boots' consumer research documents which mention product availability, store layout, ease of shopping and shopping environment as possible aspects of competition and, at a local level, waiting times for obtaining a prescription and the provision of deliveries as key measures of competition too (see Decision, paragraph 37 as elucidated by Pritchard, paragraphs 49 and 58).
  252. Elsewhere in the Decision the OFT referred to the following additional matters:
  253. (a) That the vast majority of the parties' pricing policies are set nationally (paragraph 14); and that following the lifting of RPM, price competition was still limited (paragraphs 31 and 35);
    (b) That in the P medicines sector, UniChem was not a particularly aggressive competitor to Boots (paragraph 29); that the parties consider that supermarkets are their most significant individual competitors in P medicines and GSL medicines (paragraph 29); and that, by numbers of stores, the supermarkets are gaining significant shares of supply (paragraph 32);
    (c) P medicines account for less than 5 per cent of UniChem's revenues (paragraph 29).
  254. Having considered this evidence the OFT concluded that in situations of high switching of customers between Boots and UniChem stores and Boots and UniChem being the closest competitors for a large proportion of customers, which may arise particularly in those localities where the merging parties' pharmacies are located close together:
  255. (a) the competitive scenario post-merger does not significantly differ from that in the 2 to 1 areas; and
    (b) on balance, a SLC may arise where fascias are reduced from 3 to 2 within a one-mile radius.
  256. Celesio submitted that the OFT had decided that it may be the case that SLC may be expected to result in all 3 to 2 areas. As we have explained above, and having regard to the Decision as a whole, we do not consider that this is the proper construction of paragraphs 44 and 45. We consider that the Decision, which is short-form and was prepared under strict time constraints, should not be read in a manner akin to a statute. In our judgment, as we explain at paragraphs 159 to 164 above, the Decision and the elucidation of it provided by Pritchard make it clear that the OFT's belief that "it may be the case" that a SLC may be expected to result from the merger was limited to a subset of locations within the 3 to 2 category.
  257. The OFT then turned to consider areas where the fascia reduction numbers were higher than 3 to 2. It considered that the reduction by one in the number of fascias in such areas:
  258. (a) could also give rise to a lessening of competition; but
    (b) on the basis of the evidence before the OFT it did not believe that such a reduction could be expected to be substantial.
  259. Turning to the grounds of appeal as set out in Celesio's skeleton argument we are satisfied that the reasons given in the Decision are capable of sustaining the OFT's conclusion that a reduction in fascia numbers at a lesser level of concentration (4 to 3 or higher) in local retail markets would not give rise to a SLC. In contrast to the 3 to 2 areas, in which only one independent competitor would remain after the merger, in all 4 to 3 areas the continued presence of two independent competitors would be expected to be sufficient to prevent any substantial reduction in the competitive interaction between the retail pharmacies in the area. As we have indicated above we have been assisted in our consideration of the reasons given in the Decision by the elucidation of those reasons given by Pritchard, in particular in this regard at paragraph 117.
  260. For those reasons, we also do not consider that it was unreasonable for the OFT to arrive at the Decision without considering the proximity of Boots and UniChem outlets in 4 to 3 areas.
  261. Accordingly, for all of the reasons set out above we reject Celesio's submissions contained in amended ground 1.
  262. Amended grounds 2 and 3
  263. At the hearing, counsel for Celesio accepted that the three amended grounds of appeal were linked and if Celesio failed on amended ground 1 it also failed on amended grounds 2 and 3. Given that we reject amended ground 1, the two subsidiary grounds of appeal do not fall to be determined.
  264. XI OTHER MATTERS

    Third party material
  265. In its skeleton argument Celesio submitted that Pritchard referred to third party material in so far as it supported the OFT's case but had not disclosed the third party material which, Celesio submitted, would inevitably include material which would undermine the OFT's case as well as support it. Celesio submitted that the evidence made available to the Tribunal was "skewed towards that which supports the OFT's case" (skeleton, paragraph 43).
  266. Celesio did not make any application for disclosure and declined to identify what material should have been disclosed by the OFT. At the hearing the OFT made it clear that it was not relying on third party evidence in defending the challenge to the Decision which it was meeting at the hearing, and on that basis Celesio accepted that its submissions on non-disclosure of third party material were not relevant to the points in issue in this appeal. In these circumstances we do not deal with this matter further. We do, however, have some concern as to the making of allegations such as those contained in Celesio's skeleton argument, for which there does not appear to us to have been any foundation. We have seen no basis for the forthright submissions (a) that the evidence was "skewed" towards that which supports the OFT's case and (b) that is it inevitable that the OFT had, but had not disclosed, third party material which undermined its case.
  267. Submissions by Celesio to the OFT during the investigation stage
  268. The interveners submitted that Celesio should not be entitled to make submissions at the appeal stage which it had not put forward at the OFT investigation stage. We have therefore considered the material which Celesio put forward at the investigation stage. We note that on 13 December 2005 Celesio raised a concern that there may be areas where even if there were four or more competitors remaining after the merger, some of these would provide a very weak constraint on the merged entity. In this document Celesio also stated its belief that in a number of areas there will be fewer than four competitors. Celesio submitted that given the step change that the proposed merger would bring about at the retail level, it was incumbent upon the OFT to undertake a detailed appraisal of the impact on competition at a local level. Celesio referred to the CC inquiry into supermarkets, which involved issues regarding not only the minimum appropriate number of local competitors and the nature of competition between them but also how local markets should be defined. Celesio suggested that these issues needed to be considered very closely in relation to the proposed merger.
  269. It seems to us that in the light of these submissions at the administrative stage, the interveners' criticism of Celesio is unjustified. We leave over for another case the extent to which a potential appellant can appeal to this Tribunal on grounds on which it did not rely at the administrative stage but on which it could have relied at that stage.
  270. Celesio's disclosure of the Lloyds/Cohens & Scholes merger
  271. It was suggested that Celesio had been disingenuous in its summation of the Lloyds/Cohens & Scholes case in the notice of application since it had not disclosed in its written materials in this appeal that one location, Mexborough, had been removed from that merger when the OFT became concerned that a 2 to 1 reduction would result in that area. Having regard to the particular circumstances relating to Mexborough, we do not consider on the materials shown to us that such a criticism of Celesio is justified.
  272. Evidence on pharmaceutical services regulation
  273. In the course of these proceedings it emerged that there is an error in paragraph 40 of the Decision. In this paragraph the OFT stated that the control of entry regulations for pharmacies were amended in 2005, in ways that were intended to facilitate the entry of new pharmacies. While correct as regards England there were in fact no such changes in Scotland, Wales or Northern Ireland. The error appears to have originated in documents supplied by the merging parties. It is not material to Celesio's application because the OFT did not rely on any easing of entry regulations in the Decision. We are, however, surprised that neither the parties, who conduct business in all four constituent parts of the United Kingdom, nor the OFT appear to have recognised that pharmaceutical services regulation is not uniform throughout the United Kingdom.
  274. XII CONCLUSION

  275. It follows that we unanimously dismiss Celesio's application for review.
  276. Marion Simmons Andrew Bain Vivien Rose

    Registrar 9 May 2006

Note 1   Information requests were issued on 7 December 2005, 21 December 2005, 6 January 2006, 10 January 2006, 11 January 2006 and 17 January 2006.    [Back]

Note 2   Annex 6 to the merging parties’ main submissions to the OFT on 30 November 2005.    [Back]

Note 3   Attachment G to the merging parties’ main submissions to the OFT on 30 November 2005.    [Back]

Note 4   Annex 1 to the merging parties’ main submissions to the OFT on 30 November 2005.    [Back]

Note 5   Annex 4 to the merging parties’ main submissions to the OFT on 30 November 2005.    [Back]

Note 6   Annex 5 to the merging parties’ main submissions to the OFT on 30 November 2005.    [Back]

Note 7   During the course of these proceedings the OFT clarified that the word “not” was inadvertently omitted from the Decision.    [Back]


BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII
URL: http://www.bailii.org/uk/cases/CAT/2006/9.html