Racecourse Association & Ors v Office of Fair Trading [2005] CAT 29 (2 August 2005) [2005] CAT 29

CAT
Racecourse Association & Ors v Office of Fair Trading [2005] CAT 29 (2 August 2005)
[2005] CAT 29 · 2005-08-02
[1]Burden of proof .......................................................................................... 58[2]The relevant market .................................................................................. 60[3]Did the MRA infringe the Chapter I prohibition: introductory ........... 65[4]Did the MRA amount to “collective selling”? ......................................... 68[5]Was the collective selling, or negotiation, “necessary” for the creation of the new product the rights for which the Courses were selling? ....... 69[6]Did the MRA have an anti-competitive effect? ....................................... 79[7]Did any price increase affect competition? .............................................. 91[8]Was the OFT entitled to find the price structure resulting from the collective negotiation restricted incentives for non-price competition between the Courses? ................................................................................ 92[9]Was the OFT’s section 2 analysis affected by the fact that the rights related to the transmission of recordings of sporting competitions? ....95[10]Was the OFT wrong in refusing an exemption under section 4? .......... 95 XII RESULT .......................................................................................................... 95 i I INTRODUCTION 1. These are two appeals against a decision of the Office of Fair Trading (“the OFT”) dated 5 April 2004 under section 31 of the Competition Act 1998 (“the Act”) that the sale of certain media rights under an agreement dated 2 May 2001 infringed the Chapter I prohibition imposed by section 2 of the Act and did not qualify for an individual exemption under section 9. The appellants are:(i) The Racecourse Association (“the RCA”) and entities owning 29 racecourses in Great Britain, represented by Mr Christopher Vajda QC and Mr Sam Szlezinger (references to “the RCA appellants” are to all those appellants); and(ii) the British Horseracing Board (“the BHB”), represented by Mr David Vaughan QC and Miss Maya Lester. The OFT was represented by Mr Rhodri Thompson QC and Mr Julian Gregory. 2. The RCA is a private company that was established in 1907 and represents the interests 58 of the 59 racecourses in Great Britain. At the time with which we are particularly concerned the RCA represented the interests of all 59 courses. Each such course owner is a member of the RCA and holds one voting share for each course it owns (some courses are in common ownership). The RCA undertakes marketing, administrative and representational functions for the courses. It owns Racecourse Technical Services Limited, which provides starting stalls, computer photo-finish equipment, CCTV coverage and the operation of a television outside broadcast fleet. Each course owner owns the media rights for events at its course or courses, although historically the RCA has also been responsible for negotiating agreements for the exploitation of certain of those rights on behalf of its members, including those for broadcasts to off-course licensed betting offices (“LBOs”). It owns 10% of Satellite Information Services Limited (“SIS”) as trustee for the courses. SIS provides televised coverage of horseracing to LBOs. 3. The BHB is the governing body of British racing. Its responsibilities include racing’s financial and administrative wellbeing in the interests of all participants, including racecourses, breeders, trainers, owners, jockeys and racegoers. It has a duty to ensure the proper financing of the industry. 1 4. The agreement of 2 May 2001, called a “Media Rights Agreement” (“the MRA”), was made between Attheraces Holdings Limited (“Holdings”), Attheraces Plc (“ATR”), the RCA and 49 of the 59 racecourses (“the Courses”). 5. Holdings is a joint venture company formed by Arena Leisure Plc (“Arena”), British Sky Broadcasting Group PLC (“BSkyB”) and Channel Four Television Corporation (“Channel 4”). At the time the MRA was made they owned Holdings’ issued shares equally. Arena is a listed company which owns six of the Courses. BSkyB is a listed company with subsidiaries engaged mainly in television broadcasting in the UK and Ireland. BSkyB creates channels that are broadcast via digital satellite (“DSat”) and are offered for redistribution via cable and digital terrestrial television (“DTT”), including pay per view services, and is also involved in the operation of a website and internet service provider and the provision of a fixed-odds betting service. Channel 4 is a statutory corporation which broadcasts a wide range of programmes in the UK (including sports programmes) via analogue and digital television. Its main channel is Channel Four. 6. ATR is a wholly-owned subsidiary of Holdings and was the vehicle used by the joint venture to acquire various media rights from the Courses under the MRA. The rights which the Courses granted to ATR included rights referred to in the OFT’s decision as “the Non-LBO bookmaking rights”. It was their sale that the OFT held infringed the Chapter I prohibition. The OFT described these rights in paragraph 56 of its Decision as:
“… the rights licensed by the Courses necessary to permit [ATR] to supply programming covering British horseracing to UK bookmakers other than LBO’s … for distribution in combination with betting services.”
More fully, these rights were picture rights which, in combination with betting rights and data, could be used to allow interactive betting using television or the internet. They only became commercially valuable with the development of interactive betting technology enabling punters to place off-course bets at home or at work rather than at an LBO. That was potentially attractive to broadcasters, who could use the rights to offer punters a new service combining audio visual pictures of horseracing together with the possibility of betting on a televised race through a linked website. If a punter 2 used the linked website to bet, the broadcaster could obtain a new source of income in the way that a traditional bookmaker does. The broadcaster could thereby profit from the cost of acquiring the pictures through betting income. So the exploitation of these rights was of interest to broadcasters. It was also of interest to racecourses, which could obtain a new source of income by licensing them. 7. ATR’s principal business was the exploitation of these rights (we use the past tense as it has since purported to terminate the MRA). It did so primarily through broadcasting a basic pay-TV channel and running a website, which together facilitated the provision of fixed-odds betting services and the placing of pool bets. It appointed three bookmakers as fixed-odds bookmakers on the channel and website; and it provided the pool betting service in conjunction with the Horserace Totalisator Board (“the Tote”), a statutory body with an exclusive right to run pool betting on horseraces, ATR’s right being to “co-mingle” stakes taken by it with the Tote’s pool. ATR sub-licensed certain of the acquired rights (in particular, the coverage of races on terrestrial television). The consequence of ATR’s purported termination of the MRA is that live British horseracing is no longer available on the channel or website under the arrangements envisaged by the MRA. 8. No British racecourse had granted any such interactive bookmaking rights prior to the MRA: the technology had not yet developed sufficiently for there to be a demand for them. Several well-known courses had, however, licensed rights to terrestrial broadcasters, such as the BBC and Channel 4. Since 1989, the courses had also licensed their audio visual rights so as to enable punters to watch races in LBOs. In 1999, the first negotiations took place between the courses and broadcasters in respect of a proposed deal covering the licensing of both the traditional terrestrial broadcasting rights and the new interactive rights. These resulted in the MRA. 9. The OFT found that the MRA effected a collective sale by the Courses of the Non-LBO bookmaking rights that infringed section 2 of the Act and did not qualify for exemption under section 9.

(a) the collective sale of the rights:

“has the effect of appreciably preventing, restricting, or distorting competition in the UK in the market for the supply [of] the Non-LBO 3 Bookmaking Rights by: (i) increasing the price for these rights; and (ii) restricting incentives within this market to improve the Courses’ output” (Decision, paragraph 342); and (b) that, although the notified arrangement “as a whole, improves production and distribution and promotes technical and economic progress, while allowing consumers a fair share of the resulting benefit”: “collective selling is not indispensable to attaining the benefits resulting from the Notified Arrangement. Collective selling also affords the possibility of eliminating competition with respect to a substantial part of the products in question (namely the supply of the Non-LBO Bookmaking Rights)” (Decision, paragraph 447) II THE CHAPTER I PROHIBITION 10. Part I of the Act is headed “Competition” and Chapter I is headed “Agreements”
. Section 2, sub-headed “The prohibition”, provides:
“2. Agreements etc preventing, restricting or distorting competition (1) Subject to section 3, agreements between undertakings, decisions by associations of undertakings or concerted practices which – (a) may affect trade within the United Kingdom, and (b) have as their object or effect the prevention, restriction or distortion of competition within the United Kingdom, are prohibited unless they are exempt in accordance with the provisions of this Part. (2) Subsection (1) applies, in particular, to agreements, decisions or practices which – (a) directly or indirectly fix purchase or selling prices or any other trading conditions; (b) limit or control production, markets, technical development or investment; (c) share markets or sources of supply; (d) apply dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage; (e) make the conclusion of contracts subject to acceptance by the other parties of supplementary obligations which, by their nature or according to commercial usage, have no connection with the subject of such contracts. 4 (3) Subsection (1) applies only if the agreement, decision or practice is, or is intended to be, implemented in the United Kingdom. (4) Any agreement or decision which is prohibited by subsection (1) is void. (5) A provision of this Part which is expressed to apply to, or in relation to, an agreement is to be read as applying equally to, or in relation to, a decision by an association of undertakings or a concerted practice (but with any necessary modifications). (6) Subsection (5) does not apply where the context otherwise applies. (7) In this section ‘the United Kingdom’ means, in relation to an agreement which operates or is intended to operate only in a part of the United Kingdom, that part. (8) The prohibition imposed by subsection (1) is referred to in this Act as ‘the Chapter I prohibition’.”
[11]Section 2 is modelled on Article 81(1) of the EC Treaty. Section 60 requires that questions arising under the Act are to be determined, so far as possible, and having regard to any relevant differences, in a manner consistent with Community law. Sections 4 and 9 (in a sub-part of Chapter I headed “Exemptions”) provided at the material time, so far as relevant, as follows:
“4. Individual exemptions (1) The Director may grant an exemption from the Chapter I prohibition with respect to a particular agreement if – (a) a request for an exemption has been made to him under section 14 by a party to the agreement; and (b) the agreement is one to which section 9 applies. (2) An exemption granted under this section is referred to in this Part as an individual exemption. … 9. The criteria for individual and block exemptions This section applies to any agreement which – (a) contributes to – (i) improving production or distribution, or (ii) promoting technical or economic progress, 5 while allowing consumers a fair share of the resulting benefit; but (b) does not – (i) impose on the undertakings concerned restrictions which are not indispensable to the attainment of those objectives; or (ii) afford the undertakings concerned the possibility of eliminating competition in respect of a substantial part of the products in question.”
III THE APPEALS TO THIS TRIBUNAL[12]The appeals raise the following questions:(a) was the OFT correct in finding that there was a relevant product market in the supply of Non-LBO bookmaking rights;(b) was there a horizontal agreement and/or concerted practice by the RCA and/or the Courses collectively to sell their Non-LBO bookmaking rights;(c) if there was a collective sale, was it “necessary” for the launch of the interactive betting services on digital television and the internet, so that there was no negative impact on competition;(d) if any collective sale was not “necessary”, did it have an appreciable anti-competitive effect by: (i) increasing the price of the relevant rights to ATR; and (ii) restricting non-price competition;(e) was the OFT’s section 2 analysis affected by the fact that the rights related to the transmission of recordings of sporting competitions; and(f) (if the OFT was correct to find an infringement of section 2), was the notified arrangement entitled to an exemption pursuant to the section 9 criteria? 6 IV FUNDING OF RACING BY BETTING AND OTHERWISE[13]Before 1960 it was only possible to place “on course” bets on horseracing. LBOs were first licensed by the Betting and Gaming Act 1960. LBOs were closely controlled by the new legislation, later consolidated in the Betting, Gaming and Lotteries Act 1963. To compensate racecourses for the expected loss of attendance, the legislation provided for a deduction (“the levy”) to be made on punters’ stakes or winnings and repaid to the courses. The Horserace Betting Levy Board (“the Levy Board”) was created under the 1963 Act to take responsibility for its collection from bookmakers and the Tote. British racecourses became heavily dependent on their income from the levy. In 2002, it provided £93m.[14]Nearly all LBOs have an information facility provided by SIS, which includes audio information on the betting markets, commentaries, results and general sporting news. In 1987, the RCA entered into an agreement providing for SIS to broadcast live television pictures of racing (with commentaries) in LBOs. This agreement was extended in 1992 for a ten-year term. In May 2002, the racecourses (apart from those few that had previously sold their rights to GG Media Limited) entered into an agreement with British Afternoon Greyhound Services Limited (“BAGS”) providing for the continuation of the SIS service until January 2005. The sale of picture rights to LBOs represents the largest off-course income for racecourses after the levy: the BAGS agreement generates a total of about £24m a year and the GG Media agreement (which also licenses other rights) about £5m a year.[15]British racecourses also enjoy an income stream from the sale of data relating to their fixtures, including lists of runners and riders: bookmakers need this information in order to be able to take bets on a race. This data is exploited not by the racecourses, but by the BHB through its central racing database. The Rules and Orders of British racing require prospective runners to notify the Racing Calendar Office and the supplied information becomes part of that database.[16]Racecourses operate subject to the Rules and Orders of Racing promulgated by the BHB and the Jockey Club, the sport’s governing bodies. The BHB’s Rules empower its directors to determine the date and time of any racing fixture, to control 7 the number of fixtures a racecourse can hold and to determine the programme and conditions of any race. British racing must also conform to the European Pattern Committee decisions which determine the dates on which group races are held. There are also further constraints on the ability of a racecourse to alter the date, time or type of fixtures it organises, or to organise additional fixtures. Constraints are imposed by the size of the horse population and practical turf management issues. The Levy Board makes payments for fixtures only if they are held at times acceptable to the LBOs, i.e. when there are no competing fixtures. The Levy Board criteria applicable at the time of the MRA provided that it would pay for at least two, but usually not more than four, fixtures per session (i.e. afternoon, evening etc). As a result, only a small number of races are actually run simultaneously – on Bank Holiday Mondays and the occasional weekend. On the rare occasions when the BHB has permitted simultaneous races, it has been driven by spectator demand, not by betting requirements.[17]In March 2000, the then Home Secretary informed the House of Commons that the Government intended to abolish the levy. His stated view was that “the arrangement under which racing receives income from bookmaking should become a matter for settlement between the parties on a commercial basis.” The Government asked the BHB to prepare a plan for the future funding of racing without the levy. The BHB’s plan, submitted to the Government on 13 October 2000, noted that British racing’s ability to generate income without a levy was:
“… dependent upon Racing being able to identify a product or products for which it can charge the betting industry in the same way that a seller charges a buyer in any commercial transaction.”
The BHB advocated, amongst other things, that British racing’s media rights should be combined and sold as one package, to develop a unified strategy for the industry and provide a sufficient, dependable and sustainable income stream. This was the background against which the MRA came to be signed. We will outline the course of the negotiations, but must first explain the make-up of the 59 racecourses at the material time. 8 V THE RACECOURSES[18]Twelve of the 59 became known as “the Super 12”. They were leading courses whose races were most often shown on terrestrial television and which formed themselves into a group at the outset of discussions in 1999. They are: Aintree, Cheltenham, Epsom Downs, Haydock Park, Kempton Park, Newmarket, Sandown Park, Ascot, Doncaster, Goodwood, Newbury and York. The first seven were owned by Racecourse Holdings Trust (“RHT”), which is ultimately owned by the Jockey Club and which re-invests its profits into racecourses and racing. RHT also owned five further courses, which had less, but still significant, exposure on terrestrial television: Huntingdon, Market Rasen, Nottingham, Warwick and Wincanton. The Super 12 had 34.5% of the off-course betting revenue for the year July 1999 to June 2000, such revenue being of critical interest to any organisation proposing to exploit the courses’ interactive rights. The 12 courses owned by RHT had between them a 26.2% share of that revenue. Each of the five Super 12 courses not owned by RHT was separately owned and had between them a 15.1% share.[19]Six of the 59 courses were owned or controlled by Arena: three all-weather courses (Lingfield Park, Southwell and Wolverhampton), Worcester, Folkestone and Royal Windsor. These courses shared about 19% of annual racing fixtures and had a 16.2% share of the off-course betting revenue.[20]Five of the 59 were owned or controlled by Northern Racing Group (“Northern”), a listed company: Bath, Brighton, Hereford, Newcastle and Uttoxeter. Northern was affiliated with The Chepstow Racecourse plc, which owned Chepstow racecourse. Including Chepstow, Northern’s share of the off-course betting revenue was 10.6%. Excluding Chepstow, it was 8.8%.[21]In August 2000, 17 of the 59 courses became known as the “Terrestrial Rights Group” (“TRG”) courses. Their fixtures accounted for nearly 90% of races broadcast on analogue terrestrial television. They comprised the Super 12, Ayr, Chepstow, Chester, Newcastle and Uttoxeter. They were determined to maintain their terrestrial coverage. We refer further to the TRG below when outlining the course of events leading up to the MRA. 9[22]Each course other than those owned by RHT, Arena and Northern was separately owned. Seven such courses (Ascot, Doncaster, Newbury, York, Goodwood, Ayr and Leicester) had shares of the off-course revenue ranging from 3.9% (Ascot) to 2% (Leicester). Another 27 had shares ranging from 1.8% (Catterick) to 0.2% (Cartmel).[23]The overall picture at the time the MRA was signed was therefore that three corporate groups (RHT, Arena and Northern) between them owned 23 courses, representing 53% of off-course betting revenue (51.3%, if Chepstow is excluded). All these courses signed the MRA, as did a further 26 single-owned courses, making up the 49 signatory Courses. The remaining ten courses (Exeter, Fakenham, Hexham, Kelso, Leicester, Perth, Sedgefield, Stratford-upon-Avon, Taunton and Towcester) signed an alternative agreement with GG Media.[24]It is worth looking at the profile of one of the tail of single-ownership courses which together accounted for 47% of betting revenue. We were referred to Pontefract, which had a 1.3% betting revenue share. It stages just 16 flat racing days per year and is thus one of the smaller courses, although a relatively successful one. Prior to the MRA negotiations, it had granted its LBO rights to SIS. Racecourses generally were at this time barely profitable, with only six of the 59 declaring dividends in 1999. After the MRA, in 2002, Pontefract granted its LBO rights to BAGS. Save for one particular race, Pontefract had no terrestrial television coverage. For the year to 31 March 2003, it had “race day income” (admission fees, on-course betting fees etc) of £687,929, levy payments of £79,550 and media rights income (from ATR and BAGS) of £544,343, of which £168,343 came from ATR, including a second instalment of £101,070 on the “up front” payments due under the MRA. Of the yearly ATR payments, 40% were applied to increasing prize money so as to attract better horses – as agreed with the BHB at the time of the MRA – with most of the balance and the up-front ATR payments being used to improve facilities at the course. Without its media rights and levy income, Pontefract would have made a loss for the year. 10 VI EVENTS LEADING UP TO THE MRA[25]In 1999, the racecourses began to consider how to license their interactive rights (the Non-LBO bookmaking rights), meaning picture rights which, in combination with betting rights and data, could be used to allow interactive betting via the internet or television. During the following months, the courses negotiated and discussed the future exploitation of their media rights (both terrestrial and interactive) with several companies. In August 1999, the Super 12 was formed. On 13 October 1999, Channel 4 made an outline bid to the Super 12 with a view to forming a “global partnership” for the exploitation of the media rights, including interactive services. Channel 4 had been a traditional terrestrial broadcaster but it was, or claimed to be, unwilling to continue to exploit terrestrial rights unless it could also exploit the interactive rights and share in the profits they could generate (Michael Jackson of Channel 4 confirmed this in a letter of 17 January 2001 to the courses).[26]The RCA acted as a negotiator for the 59 courses during most of the subsequent 17 or so months of discussions. It held an EGM on 16 December 1999, of which minutes were sent to members on 21 January 2000. Mr Deshayes, for the Super 12, is recorded as saying that:
“… the core objective of the Consortium [Super 12] racecourses was to maximise the value of their media rights and in doing so to unlock the potential of the racing industry. Their vehicle to achieve this would be collective and pro-active management and exploitation of new opportunities. Their aim was to improve the telling of the racing story and hence to generate added value. They saw a brighter future through decisive action, which would include working closely with the RCA, and using the incremental revenues for the long-term benefit of the whole industry. … Mr Deshayes sought to reassure Members by confirming that the Consortium racecourses were keen to work closely with the RCA and with other courses and that they had no intention of becoming any type of closed shop.”
Captain Lees (clerk of the Leicester course) had attributed to him that:
“Whilst he felt it made commercial sense for the Consortium courses to work together in negotiating their network television contracts, he said he would be most unhappy with any arrangement whereby it fell to 12 courses to determine 11 levels of income to the other 47. In this context, he looked to the RCA for strong leadership to ensure that the interests of all courses were fully represented.”
[27]Mr Angus Crichton-Miller of the RCA wrote to Mr David Hillyard, the managing director of RHT on 24 December, referring to the EGM. He wrote:
“Whilst my impression is that you have accepted that the big money can only come from betting income, and that the vast bulk of that will derive from a ‘wall-to-wall’ service, I can imagine that the broadcasters are proving seductive on the ‘merging of the media’ argument. What I would put to you is that if you feel a decent deal can only be done by selling a broader range of rights, should you not then make it a 59 racecourses deal? That is increasingly the way the other racecourses see it.”
[28]On 4 January 2000, Arena (using the name “Attheraces”) entered the scene as a bidder in competition with Channel 4. It circulated a discussion document setting out plans for its own internet service offering pari-mutuel gaming along with live video feeds of UK horseracing, planned to be launched in autumn 2000. The document emphasised the unique opportunity offered by the internet and its derivatives for racecourses to participate in betting turnover on their races. It explained that Arena’s intentions were that all courses would participate with it in a joint venture. Revenue was to be distributed back to the courses, which could make their own investment decisions; and the entity providing the service would be owned by all courses in proportion to the relative value they brought to it. All 59 courses were intended to have an equity stake, although the structure would recognise the prestige value of the Super 12 (and Ascot in particular) and also the value and importance of the course groups with high numbers of fixtures (those owned by Arena, Northern and RHT). The paper outlined a suggested equity split between Arena and all 59 courses. Arena made clear that, whilst it intended to drive the process forward, it did not expect to own a controlling interest:
“all key strategic decisions should be fully discussed and agreed by UK racecourses as a whole.”
Its proposal was, in general, that Ascot, RHT, the remaining Super 12 courses and Northern would receive 4% of gaming revenue and the remaining 33 courses would receive 3% of revenue on their races. Arena assured the courses that its proposal provided a: 12 “unique opportunity to shape the destiny of UK horse racing. The industry is currently in a state of flux. [Arena] allows the industry to begin the process of reclaiming control of its highly valuable product from bookmakers, SIS and other third parties and exploit it more fully for the good of UK racing as a whole. It allows all 59 racecourses to have access to the potential benefits of online gambling, while recognising the importance to UK racing of a select grouping of prestigious racecourses and influential racecourse groups.” The paper proposed an “aggressive” timetable for the way forward: January 2000 was for discussion; the Super 12, Arena and Northern were to sign up on 1 February 2000; and the remaining courses were to sign up on 1 March 2000, via the RCA. Arena never sought to follow up this timetable.[29]Arena’s proposal was, therefore, made to all 59 courses collectively and was intended to involve all of them – not because of any sense of altruism by Arena but because the proposed venture required an exposure to the betting public of as comprehensive a racing service as possible: the more races shown, the greater the betting volume and the greater the betting revenue. Any bidder for the interactive rights required a minimum “critical mass” of coverage in terms of volume and quality of racing if its service was to be viable.[30]Channel 4 was in the meantime still pursuing its negotiations with the Super 12. On 11 January 2000, Mr Scott (Channel 4) wrote to Ms Ellen (United Racecourses) in advance of a meeting the next day with the Super 12. He emphasised the virtues of a deal between the Super 12 and Channel 4, saying:
“It is generally accepted that Channel 4 offers the best terrestrial televised coverage of racing in Britain. We have consistently expressed the view that the integration of terrestrial television, a new digital racing channel, interactive services, the internet and international distribution is the key to success in taking Racing into the next age. A strong commercial terrestrial television partner with broadcasting and creative skill, and one million regular viewers is a vital driver into the new media areas.”
He made clear that Channel 4 was also looking to involve racecourses other than the Super 12, saying:
“As you know we have consistently avoided tying up any deals with third parties until we have an agreement with the Consortium. I want to reiterate 13 that we believe that this is the best sequence events [sic] for the Consortium. There may be other parties offering large advances for particular rights but these deals can unquestionably still be achieved by a strong partnership of Channel 4 and the Consortium. Indeed we believe that better deals can be done by bringing together our strengths, pooling our knowledge and information and doing the best possible deals for the Consortium and the Joint Venture Company.”
[31]We presume Channel 4 knew there were other actual or potential competing bidders, of which Arena was one. The letter made clear that Channel 4 was anxious that the Super 12 would commit itself to a joint venture with it.[32]On 31 January 2000, Channel 4 made what it called “a significantly improved financial offer”, involving a commercial relationship also with the BBC (the other key terrestrial broadcaster), NTL (with its extensive technology skills) and the Racing Post (said to be the most authoritative racing publication). The new “subject to contract” offer was of a guaranteed minimum of £221m over ten years for the Super 12’s terrestrial and interactive rights, emphasising that Channel 4’s continued participation in televising racing was dependent on its having an “equitable and financial stake in these new [interactive] services.” £108m of the £221m was referable to non-terrestrial rights and so was largely referable to the interactive rights. Whilst the offer related only to the Super 12’s rights, it made it clear that additional rights payments were to be offered to the other racecourses and that the new interactive channel intended to carry races from them: the covering letter said that Channel 4’s “intention of launching a comprehensive racing channel ensures our proposal will appeal to other racecourse owners.” The minutes of a Super 12 meeting of 3 February 2000 record that at about this time NTL had confirmed that £1.5m per year (£15m over a ten-year term) would be offered to the other 47 courses for their interactive rights.[33]On 4 February 2000, Mr Hillyard (RHT) outlined to Mr Crichton-Miller (RCA) his understanding of the Channel 4 offer, saying it was “available to ALL British racecourses and, as soon as the Consortium [Super 12] has agreed the detail of the new contract, we would very much want you to lead the discussions and negotiations on behalf of all other RCA members who wish to participate.” He said: 14 “The headline sum involved of over £200 million together with a further sum of £25 million in respect of marketing support is substantially more than any previous arrangement involving racecourses’ media rights and, in addition, there is a revenue sharing arrangement from which the racecourses should derive further income and which, in due course, should be developed to provide even more to the participating racecourses. The headline sum also ignores the very significant investment that our new partners will make in the technical architecture in order that we can access these new income streams. It would be foolish to anticipate the outcome of any discussion at this early stage when, inevitably, emotions are running quite high but, whatever the outcome, the reality is that this deal will open up a new and developing market place and will set a bench mark that will enable negotiations by or on behalf of other racecourses to achieve significant new revenues.”[34]A Racing News article on 5 February 2000 was to the effect that the Super 12 had announced the makings of a deal with the Channel 4 consortium worth about £22.5m a year – thrice their previous earnings from terrestrial television alone. It said that, according to Channel 4, “the Super 12 tracks have given an undertaking to bring the remaining 47 racecourses into the daily digital mix.” It reported that the smaller courses outside the Super 12 were “getting together in a mood of self-preservation amid fears they could get left behind in the multi-million-pound discussions over media rights …. ”.[35]On 7 February 2000, Mr Crichton-Miller (RCA) wrote to all course managers, saying it needed to be established what was on offer from Channel 4, after which all courses needed to consider whether acceptance was in their best interests. He said the RCA would “orchestrate” this, whilst getting the courses involved in all discussions and negotiations.[36]On 15 February 2000, the Super 12 and Channel 4 entered into a “Lock-Out” Exclusivity Agreement and draft Heads of Agreement under which they agreed to negotiate exclusively until 30 April 2000. The Heads provided an agreed basis for negotiation of an agreement for the exploitation of the Super 12’s media rights (both terrestrial and interactive). By clause 3.13 of the draft Heads of Agreement, the parties agreed to put a proposal to the other UK courses and that:
“… as soon as is reasonably practicable they shall jointly finalise and implement a strategy for approaching such [other] racecourses and [the 15 Channel 4 consortium] shall disclose to [the Super 12] such financial assumptions and projections as are reasonably necessary to facilitate this process. [The Channel 4 consortium] will allocate an additional amount of up to £1.5 million per annum to achieve this objective.”
[37]Also in February 2000, it was agreed that negotiations with 42 other courses would be carried out via the RCA. There were of course 47 other courses but, in addition to its Super 12 courses, RHT owned five other courses which also had significant terrestrial coverage. These were also likely to sign up to the Channel 4 proposal and so, in practice, it was only the other 42 which were outside the fold (subsequent references to “the 42” are to courses other than the Super 12 and the five additional RHT courses; references to “the 47” are to all courses other than the Super 12).[38]On 29 February 2000, Mr Savill (Chairman of the BHB and of Plumpton racecourse) wrote to Mr Hutchinson (Ripon racecourse), to explain why he believed all sections of racing would benefit from the outline deal between Channel 4 and the Super 12. He wrote:
“I think it is important first of all to realise that the racecourses in question have previously always negotiated their own media rights for terrestrial television. I am quite sure that for them to have got closer together to negotiate as a group is one of the reasons why the income stream that will flow to those courses is substantially greater than it has previously been. … You are right to identify that a number of details are yet to be negotiated. Some of these are very important details and, in my opinion, the most important is to ensure that the product of the other 47 courses is available on the Racing Channel so that these courses can benefit from the development of new betting mediums. It is unclear at this point as to exactly what the offer will be to the other 47 courses but you have my assurance that I shall only support the deal as Chairman of Plumpton if it is attractive. More money into Racing, which this deal guarantees, can only benefit those with their investment and livelihoods firmly rooted in our sport. This includes racecourses, owners, trainers, breeders, jockeys, blacksmiths, farriers, transport companies and the many other businesses that depend on a healthy financial structure for British Racing.”
[39]Mr Hutchinson replied on 6 March, saying: 16 “First of course I agree that it is sensible to band together to get a better deal – the 6 Courses hosting the Classics and the then United Racecourses banded together to negotiate the original Channel 4 contract. What I would have liked to have seen was 59 Racecourses banding together not just 12. I believe that more money could have been wrung out of the media by someone negotiating in an orderly manner on behalf of all the Courses and without creating this unsavoury divide which endowing 12 Courses with Super status has inevitably caused.” He added that such endowment of the Super 12 courses would create two-tier racing, leading to sponsors favouring those courses and to a reduction of support for the smaller ones. He said, “It cannot be for the benefit of Racing as a whole if 47 courses only receive sufficient from the Levy (or its successor) to put on minimum value Races and for the majority of the prize money available only to be on offer at twelve courses.”[40]By 6 March 2000, Hawkpoint Partners had, with the RCA’s support, been appointed as advisers to the 42 with regard to the exploitation of their media rights. Channel 4’s stance at this stage was, or was perceived by the RCA to be, that any deal with the Super 12 and RHT was not conditional on the 42 also joining in. This appears from Mr Crichton-Miller’s letter of 6 March 2000 to the chairmen of the 42. He indicated that Channel 4 would be dealing directly with “the RCA/the ‘42’, and it would be an independent decision by the ‘42’ as to whether they wanted to join or not.” Channel 4 had not yet made an offer to the 42 and Mr Crichton-Miller asked their chairmen:
“to make no commitment to Channel 4 or any other offer until all the ‘42’ have been able to assess the options available. … The ‘42’ owe it to each other to avoid fragmentation and so maintain the considerable strength they have. This is not to say that the Channel 4 route is the wrong one, but one can only take that view after a calm assessment of the alternatives.”
In a further letter of the same day to the same chairmen, Mr Crichton-Miller also noted that: “I am glad to report that the Area Meetings demonstrated an unanimous determination amongst the 42 to stick together and exploit their considerable leverage. If everyone keeps their nerve and sense of purpose I am confident a remunerative conclusion will be reached.” 17[41]Whatever the RCA’s perception of Channel 4’s stance, Mr Scott (Channel 4) wrote on 10 March 2000 to each of the 47. He said they had reached an outline agreement with the Super 12 and were:
“… extremely keen to work with all 59 courses in achieving a profitable and exciting future for racing. This is a position we have made clear to the [Super 12] from the beginning of our discussions and they have always supported this strategy.”
He said that developments in the media environment meant there was a major opportunity for all courses to realise a greater share of the global revenue they generated. There were two goals – raising the profile of British racing and expanding its coverage on television and into the new media sectors – which were best realised “through an integrated approach involving all racing. Fragmented coverage of racing on British media and internationally will result in less effective revenue generation and a much greater risk of competition from other sports.” He outlined Channel 4’s plans and said it “would like to include the whole of British racing” in them. He invited the courses’ representatives to presentations at which Channel 4’s plans would be explained and proposed three dates for such presentations, on 27, 28 and 29 March 2000. In an interview with the Racing Post on 24 March 2000, Mr Brook (Channel 4’s director of strategy and development) was reported as emphasising that the benefits of the deal would go beyond the confines of the Super 12, that “We see this as a partnership that secures the future of racing for everyone in the UK” and that “We are interested in forming a partnership with all of the courses, and that’s why we have arranged to make a presentation to them. Right from the start, it has always been our plan to work out a deal that’s best for racing, and also to work with as many courses as possible.”[42]Despite Channel 4’s apparent enthusiasm to involve “all racing”, it appears that the Super 12 did not regard their proposed deal with Channel 4 as conditional on the participation in it of the other 47. Mr Deshayes (managing director of Newmarket and Super 12 spokesman) was reported in “The Times” on 28 March 2000 as saying that “We have got the deal we wanted … It never has been dependent on the other 18 courses coming in with us and the new digital racing services on Channel 4 will definitely start in November.”[43]Between 27 and 29 March 2000, Channel 4 made its presentations to the 47. It said it would offer a global sum of £1.2m a year for their Non-LBO rights. It explained that it had already outlined its overall business plan and would be giving its details “to each individual course in a separate letter.” In a written summary of questions and answers at the presentations it said that:
“The long-term revenue opportunity is in the growth of interactive betting revenue. By sharing these revenues equally between the courses and the media consortium we ensure both parties share the same objective of maximising revenue generation.”
As to what rights were being sought, for how long and whether exclusive rights were required on the internet, it replied:
“All media rights in all territories on an exclusive basis for a ten year term. Exclusivity is crucial in order to maximise the value of having a single brand which is extensively cross promoted on terrestrial television. … The Internet is an extremely crowded and competitive environment. Competitive services have been proved to be those that have terrestrial TV promotion and exclusive content. If online rights were to be non-exclusive, we would not be maximising the power of TV promotion and it would result a [sic] huge number of competing sites, with no effective promotion and at great risk of losing audiences to other sports sites. This would reduce the revenue the courses would be able to share in.”
It said that it had made a commitment of £2.5m a year for the ten-year term “to market and promote the whole of racing – not just the Group of 12. This commitment will be to promote both the new services we are creating but [sic] also racing attendance.” As to why the proposed deal offered the Super 12 over £200m over the ten-year term, with the other courses receiving proportionately less even though they represented the majority of coverage and betting revenue, Channel 4 replied that a significant proportion of the payments to the Super 12 represented the value of worldwide terrestrial television rights, with the equivalent rights to the smaller courses being of much smaller value. But it also explained that: 19 “We are offering the smaller courses the same 50/50 split of interactive betting revenue as has been agreed with the Group of 12. It is through the share of this revenue that we believe all courses will realise the value of the rights they retain. By sharing this revenue equally on a course by course basis, we ensure the fairest possible terms, the greatest incentive on both sides to revenue maximise and a no risk deal for the smaller courses.”[44]The 47 regarded Channel 4’s offer as too low. They considered the presentations to have been aggressive and rude and felt they were being “offered crumbs” and “left out.” Mr Hillyard (RHT) conveyed this to Mr Scott (Channel 4) on the telephone on 30 March 2000. There was a meeting of the Super 12 on the same day, the minutes of which summarised the 47’s response as being “a mixed reaction with a better reaction from the terrestrial non-Consortium [Super 12] tracks. There was a degree of anger and frustration with many of the smaller tracks having had their expectations raised.” Part of the minutes read:
“3.7 [Mr Deshayes – Newmarket] reported that Andrew Brann of C4 had stated that if the non-Consortium racecourses wanted additional money then this should be met by the Racing Consortium [the Super 12]. ALL agreed not to go down this avenue at this stage. 3.8 [Mr Townley – Active Rights Management] stated that in his view the Media Consortium would go ahead even if no further racecourses were signed up. The premium brand UK racing content of the Racing Consortium on the interactive channel could form part of a sports betting channel. [Mr Kershaw – Newbury] added that the US betting market had not been very interested in the Dubai World Cup and prefer domestic racing – therefore the value of non-Consortium UK racing in the US market may not be as great as the smaller courses expect. ALL agreed the added value that the non-Consortium courses would bring to a racing channel would need to be carefully considered.”
[45]On 31 March 2000, Channel 4 sent each of the 47 the details of its commercial proposition. We have seen a copy of the letter to Huntingdon, which we presume was in standard form. It stated that Channel 4 had “always believed that a single racing proposition, which can embrace all British racecourses, will have greater impact and generate more revenue that a number of fragmented racing services.” It said that the Channel 4 consortium offered a number of “unique and significant benefits to racing.” These were: 20 “The creation of a single racing service, consistently branded and available across all platforms, and driven by the huge promotional advantage that terrestrial television coverage brings. This promotion will ensure that our service will have far more impact and awareness amongst viewers and punters than a number of competing services. In addition, the Media Consortium will commit £25m to marketing over a 10 year period to promote racing and racecourse attendance. Revenue generated by these new services (revenue in which racecourses share directly) will therefore be greater than that generated by a number of fragmented services. Certainty in a world where many sports are challenging for air-time, that Racing will be sport at the top of the BBC and Channel 4 agenda for at least the next 10 years. We will make this service available internationally to offer the first real opportunity for British racecourses to share in the global betting market. The media consortium will work with all British racecourses to ensure that they take greater control of the long term future of their sport and ensure that they share directly in the future growth of the sport.” The terms proposed to Huntingdon were payments of(i) a rights fee of £28,100 for each year of the term (the first three years’ fees, totalling £84,300, being payable on signature), and(ii) a 50% share of the interactive betting and international TV revenue received by the media consortium attributable to that course. Like terms were offered to each of the other 46. This proposal valued the rights in aggregate at £1.5m a year over ten years. In exchange, each course was to give Channel 4 exclusive audio and audio-visual rights (terrestrial and interactive) to all racing at the course over a ten-year term. Racing from each course was to feature on the new racing channel, which would be available in most digital homes in Britain, but not to LBOs: each course would be free to renew its LBO contracts. The letter explained that Channel 4 was discussing deals with the Tote, as the principal bookmaker within the new service, and with other bookmakers. The deals involved the payment to Channel 4 of a percentage share of the gross revenue bet through its new service. The Channel 4 consortium would not be taking any costs before the 50/50 revenue split: costs would come out of its 50% share. Each course was invited to respond to the proposal by 1 May. 21[46]The Super 12 held a meeting on 10 April 2000. Three minuted items were:
“2.2 [Mr Deshayes – Newmarket] reported that the Media Consortium [Channel 4] felt disappointed that the Racing Consortium [the Super 12] had been unable to deliver the other racecourses to the Consortium. All agreed it was very difficult to sell the proposition without the Media Consortium’s business plan and details of the programming on the new racing channel. 2.3 The Media Consortium were not prepared to give additional guarantees to the other racecourses and had called on the Racing Consortium to fund any shortfall in minimum guarantees from the deal the Media Consortia had offered to it. All agreed that the Racing Consortium were not prepared to do this at this stage and that more information was needed from the Media Consortium to make an informed decision. … 2.13 All agreed that now that the nature of the deal had changed, with the Racing Consortium being asked to consider funding the offer to the other racecourses out of its minimum guarantee, it was essential to have access to the media partners’ business plan.”
These minutes suggest that Channel 4 wanted the Super 12 to get the other courses on side. The evidence of Richard Johnston, the managing director of RHT, is positive that the Super 12 never gave anyone any undertaking to deliver the 47.[47]On 11 April 2000, Mr Crichton-Miller (RCA) wrote to Mr Scott (Channel 4) informing him that the 42 had rejected Channel 4’s proposal, saying they had asked the RCA to pursue all negotiations on their joint behalf. He said that Channel 4’s points of contact with the 42 should be the RCA or Hawkpoint.[48]On 16 May 2000, Mr Penrose (Arena’s finance director) had a discussion with Mr Derby (Ascot). Mr Penrose revealed that Arena and Sky had been talking, with Sky wanting an “all 59 racecourse interactive channel and have the eyeballs and kit to deliver revenues.”[49]On 17 May 2000, there was a meeting between Mr Scott (Channel 4) and the Super 12. The minutes reflected that the constitution of the Channel 4 consortium was not yet final, that Channel 4 was in a dialogue with Sky and that Channel 4’s stance was that its deal with the Super 12 was “contingent on the other UK racecourses joining. The Media Consortium’s critical date in order for a racing channel to be 22 launched this year was in fact mid-July.” Mr Deshayes followed this up with a letter of the same day to Mr Scott. He recorded his understanding that Mr Scott had made it clear that the Channel 4 consortium regarded the Super 12 as having “provided an undertaking to you to deliver the other racecourses and that it was the view of the Racing Consortium [the Super 12] that the other racecourses could indeed be delivered for a figure of £1.5 million a year.” He said that neither the Channel 4 proposal nor the draft Heads of Agreement had imposed such an obligation on the Super 12 and that although the Super 12 had been delighted with the £1.5m figure, doubts had been raised at the time as to whether the other courses could be brought in at this level. He added that the Super 12 continued “to be as keen as ever to work with you to find a 59 course solution and are happy to explore every reasonable option.” He said:
“We have not closed our minds to the possible allocation of some of the existing minimum guarantees but we still need further help in understanding your business plan and we still do not understand why, if others are entering the Media Consortium, the terms that will apply to such entry will not accommodate the other racecourses.”
We infer, therefore, that at this stage Channel 4 wanted – and needed – a package involving not just the Super 12 courses, but rather more: ideally, all 59.[50]In June 2000, Arena re-entered the scene as a competing bidder. We mention that it had a subsidiary, Arena Online, which had developed software to provide an interactive pool betting service based on horseracing (the “Trackplay” system). It allows users of PCs, digital television, mobile telephones and personal digital assistants to:(i) see live video pictures of horseracing,(ii) have access to racing information, and(iii) place bets, including pool bets. Trackplay was later to provide ATR’s pool betting service, Arena becoming part of the ATR consortium. On 19 June 2000, Arena made a presentation to the RCA and representatives of the 42. Arena made it clear at that meeting that their aim was in fact to bring together all 59 courses, although they followed the meeting up with a “subject to contract” offer of the same date merely to each of the 42 for their media rights for a ten-year term (excluding the LBO rights the subject of the current agreement with SIS). The offer was of a guaranteed cash sum of £81.3m for the rights period. £71.1m was to be paid over 10 23 years, to be shared between the 42 pro-rata to their respective shares of UK off-course betting revenue for the preceding year. The balance of £10.2m was to be paid over the same term, to be shared between those of the 42 whose races were televised by terrestrial television on a like pro-rata formula. The offer included a provision, in certain events, for payments in excess of the guaranteed sums. The 42 were to have 5% of the equity of the licensee company. The offer was to remain open for acceptance until 7 July 2000. It was conditional on various matters, including acceptance by courses accounting together for not less than 50% of UK off-course betting revenue. The offer was almost 600% higher than Channel 4’s (£12m for the non-LBO rights over ten years). Although Arena had made it plain on 19 June 2000 that they wanted to deal with all 59 courses, at this stage the Super 12 were (or were presumed by Arena to be) in exclusive negotiation with Channel 4, which may provide the explanation as to why the offer was just to the 42. The RCA appellants contended that this offer was nothing more than a spoiling tactic by Arena.[51]On 22 June 2000, Mr Crichton-Miller (RCA) wrote to Arena thanking them for the presentation, declining the offer on behalf of the 42 and saying Hawkpoint would be in touch and that the 42 looked forward to a revised offer. On the same day, he wrote to the 42 saying that the RCA and their advisers were focusing on gaining access to the business plans of the highest offerors; subject to that, refining and improving the offers; developing a “Go-it-alone” option; and discussing with Channel 4 the possibility of a 59 course deal.[52]On 3 July 2000, Channel 4 made a revised “subject to contract” offer to the 47, conditional on all accepting it. It was made by Go Racing Limited (which was to be the rights acquisition vehicle) and was increased so as to compete with the Arena bid. The offer was again to acquire the rights for a ten-year term. There was a guaranteed payment of a minimum of £75.5m: £7.5m for each of the first five years and £7.6m for each subsequent year. The first three years’ payments were to be made on signing. £5.5m of the total payment was to be allocated to the 47’s terrestrial rights and the balance to their other rights. The split between the 47 was to be on a pro-rata basis, essentially like that proposed by Arena. Additional payments were to be made if certain minimum thresholds were exceeded. The offer was conditional on:(i) the 24 acceptance by the 47 by 13 July 2000, and(ii) the acquisition of the rights of the Super 12.[53]On 4 July 2000, Mr Scott (Channel 4) wrote to Mr Hamilton-Fairley (Premium TV, which was a subsidiary of NTL, a member of the Channel 4 consortium) saying the net effect of the increased offer was:
“… to add £4.5 million cost to our business plan. I know that to be unwelcome, but feel we have to do so if our bid is to be in the same ballpark as Arenas.”
On the same day, there was a meeting between Richard Johnston (RHT) and others, including Mr Penrose of Arena. Mr Johnston’s note records that Arena’s then proposal was for the acquisition of the rights of all 59 courses.[54]At a meeting on 6 July 2000 with Mr Johnston (RHT) and others, Channel 4 set a deadline of 13 July 2000 for all 47 courses to accept its offer, or at least for an agreement in principle that the 59 courses would work together to finalise a deal.[55]On 11 July 2000, Mr Scott (Channel 4) and Geoffrey Hamilton-Fairley (NTL) wrote jointly to Mr Deshayes (RHT), saying:
“As you know, we have always been committed to a solution which involves all of the UK racecourses – it is only in this way, we believe, that the venture has real prospects of delivering the proposition both we and your consortium have been committed to for the last six months or more. As you also know we have become increasingly frustrated by the factionalising which seems to be endemic in this industry and the many contortions we have had to go through in an attempt to come up with something which will appeal to all parties. … The financial terms which it seems we would have to offer in order, possibly, to secure the rights to the other 47 courses mean that this venture is no longer a viable proposition for us. … Your proposal was that the media consortium should take on another £9m of fixed costs to bring the bid to £75m. That we feel unable to do both because of the additional financial cost and also following your advice last week that such a bid was likely to fail within our timescale. We also have severe doubts about whether it would provide the basis for drawing in the Arena and Northern courses, both of which we believe to be crucial to a full solution. 25 We have therefore come to the view that it would be better for all concerned if the media consortium were to withdraw its offer. In withdrawing we would like to think that the racing industry (or its major groupings) might come back to us with a coherent and united product and that we could still be part of such a solution. … Although we have lost confidence in the current RCA tender process, we are as committed as ever to working with the ‘Super 12’ to secure a stronger racing narrative and a viable overall future for racing.”
[56]On the same day, Channel 4 issued a press release announcing it was pulling out of the negotiations (including with the Super 12) until the 59 courses came back with a united position. David Brook said in the release:
“Throughout our year long negotiations, Channel 4, Premium TV and the Super 12 courses have shown remarkable consensus on the best way forward for racing. We remain convinced that a multimedia rights solution involving all UK racecourses is in the long-term interest of everyone involved in the sport. Unfortunately, the factionalism that seems endemic in some parts of the industry has made it impossible to secure the necessary involvement of the UK’s remaining 47 courses…. Although this deal is dead, if Britain’s racecourses can regroup and come back to us with a united and coherent position then Channel 4 and Premium TV would still hope to be involved in a future deal for television and online horse-racing rights.”
[57]On 14 July 2000, Arena made a renewed “subject to contract” offer to the 42, in the form of two alternative options. The total guaranteed amount on offer under Option A was £94.6m, whereas that on offer under Option B was £97m, both amounts representing an increase on the June 2000 offer. The difference is explained by the fact that the latter offer included an additional £2.4m for the non-terrestrial rights. In addition, the structure of the offered payments differed between the two options. Both offers were conditional (inter alia) on acceptance by such of the 42 as accounted for not less than 50% of betting turnover, the same percentage as in the June offer.[58]On 31 July 2000, Arena made a “subject to contract” offer to all 59 courses to acquire their non-terrestrial (but not LBO) rights for £178m. The Chairman’s letter enclosing the offer explained that: 26 “b. Our offer of £178 million for non-terrestrial rights facilitates racecourses to conclude their own terrestrial deals which can reasonably be expected to exceed £150 million. On a like for like comparison, UK Racing should therefore be expecting to receive guaranteed sums of a minimum of £328 million, £88 million (37%) more than the previous offer we were all urged to accept as being ‘in the best interests of racing. … d. Central to our objective is a continuing improvement in the racing product ….” The Chairman expressed his “sincere wish that each and every racecourse joins with us in this exciting venture that brings a new minimum guaranteed sum of £178m into our sport, opportunities for significant additional amounts, and the ability to continue to maximise the potential from both terrestrial and SIS contracts.”[59]Arena therefore valued the non-LBO and terrestrial rights of the 59 courses at £328m. The offer was conditional on acceptances from racecourses accounting for:(i) not less than 80% of UK off-course betting turnover; and(ii) not less than 75% of fixtures televised terrestrially. Arena’s expressed intention was, however, that all the courses should participate.[60]An RCA meeting of racecourse representatives was held on 31 July 2000. One matter discussed was the pooling of the data rights which were thought to be held by the BHB. The courses mandated the RCA to continue discussions with Arena “on an improved basis”, pursue alternative offers and avenues, and attempt to conclude a deal with SIS for the continuation of the Racing Channel and to initiate negotiations for a terrestrial deal. The RCA’s recommendation to courses was not to accept the Arena offer, which did not extend to the courses’ terrestrial rights. As regards the push towards the negotiation of a deal for the licensing of the courses’ terrestrial rights, this led to the formation of the Terrestrial Rights Group (“the TRG”), comprising the 17 courses we have identified and the RCA. Its primary function, according to its terms of reference, was to: “1. Assume responsibility for negotiating UK and overseas terrestrial and all other media rights, except interactive rights by whatever platform and LBO picture rights, on behalf of those courses with current terrestrial rights, recognising that future terrestrial coverage may be different.” 27 The TRG courses already derived significant income from terrestrial coverage, which they were anxious to maintain. The new revenue stream offered by the interactive technology was also attractive to them, but probably less so than to the less well-known courses, which had less terrestrial coverage.[61]On 20 September 2000, following discussions with Channel 4, Arena revised its offer to a “subject to contract” one including terrestrial television rights (but still not the LBO rights). It was for £320m, valuing the terrestrial rights at £140m and the interactive rights at £180m (up from the £178m in the July offer). Arena informed Aintree in a letter of the same date that they:
“… acknowledge that the RCA appointed [TRG] will choose the appropriate terrestrial broadcaster of the Terrestrial Television coverage with our consent and that we will work with the [TRG] to achieve the best outcome for UK racing.” and added that: “Inevitably, due to the excessive length of these negotiations as we, and the RCA, have been endeavouring to produce a 59 racecourse solution, the timetable has slipped ….”
Arena’s revised offer was, by clause 8, conditional on various matters, including the acquisition of rights from such of the courses as accounted for not less than 80% of the total UK betting revenue for 2000 (the same as in the July offer).[62]On 22 September 2000, another bidder emerged. Carlton Communications Plc (“Carlton”) wrote to The Hon David Sieff (Newbury), referring to the Arena offer, and saying that Carlton wished to make an offer to all 59 courses for their terrestrial and interactive rights. Mr Murphy (Carlton) wrote that:
“Carlton’s extensive experience in broadcasting premier sports events together with our access to free TV, pay TV and the internet for distribution, means that we are uniquely placed to bring these assets to bear for the benefit of British racing. 28 With Formula 1 racing and UEFA Champions League we have shown how we can turn sports into major entertainment brands. We would like the opportunity to do the same with British horse-racing. We are confident that our financial proposals will achieve this and a [sic] deliver a better return to racing and the race courses than the Arena bid.”
[63]Carlton submitted its “subject to contract” offer on 9 October 2000 to each course. Its making had been preceded by discussions with the RCA which Carlton said had been encouraging. The offer, for both the terrestrial and interactive rights, was for a guaranteed £350m over ten years, £30m more than the Arena bid. It was conditional upon, inter alia, acceptance by:(i) courses capturing 80% of total television horserace viewing in 1999;(ii) at least seven of nine specified Super 12 courses; and(iii) courses capturing 70% of the 1999 off course betting revenue. Carlton’s business plan is not in evidence.[64]Between April and October 2000, expressions of interest in acquiring the interactive rights of the 59 courses had also been made by TVG (a US operation), Interactive Racing Media and SIS. They did not feature materially in the story and we say no more about them.[65]The time for acceptance of Arena’s bid expired on 10 October 2000, the day after the Carlton offer. On 11 October 2000, a consortium made up of Channel 4, BSkyB and Arena announced that within 14 days it would be making a competing bid to all 59 courses. The announcement stated that the consortium members were:
“all conscious of the requirement to maximise the value of UK horseracing by securing as wide a distribution of coverage as possible across all platforms ….”
The consortium, called “Go Racing”, sent its “subject to contract” offer to all 59 courses on 30 October. It was a global offer to all courses of a guaranteed minimum of £320m, made up of £250m for the 17 TRG courses and £70m for the 42 others. The payments were to be made annually, with the amounts payable to each of the TRG courses to be determined by them; and likewise as regards those payable to the others. By clause 2 of the bid letter, Go Racing committed itself to spending at least £80m during the 10-year term on “marketing and promoting the Rights and/or horseracing.” 29 This was an increase in the minimum marketing expenditure previously proposed by Arena and was intended to bridge the gap between the Arena bid and Carlton’s higher bid. Clause 7 stated that, in respect of various identified matters, the courses were to “act collectively” and that the RCA was to have authority to bind the courses: these matters included the negotiation of “definitive contracts” and any variations to the terms of the bid letter.

(i) all the TRG courses; and (ii) also from:

“such of the Non-TRG Courses which, together with the TRG Courses, account for not less than 70 per cent [of off-course betting revenue]”
The bid described the courses comprising (i) and (ii) as the “Required Minimum Courses.” Go Racing’s business plan on the basis of which it formulated this condition is not in evidence.[66]The “Go Racing” consortium later became Holdings, the joint venture company in which Arena, BSkyB and Channel 4 held equal shareholdings (see paragraph 4 above). The company which, under the MRA, was the eventual acquirer of the interactive rights of the 49 Courses was its subsidiary, ATR and from now on we will simply call the consortium “ATR”. Its plan remained one under which it would launch a new product making available to the viewer and punter a television channel dedicated to British horseracing at low cost. It would be available as part of the basic package of any pay-TV subscription at no additional costs to pay-TV subscribers as opposed to the premium rates normally charged for sporting events. It would permit interactive betting. The channel could be a low cost one because it was to be financed principally from betting income.[67]Both the Carlton offer and the ATR offer were for the courses’ terrestrial and interactive rights. The Carlton cash offer was materially higher than ATR’s. Carlton wrote to all courses on 18 November 2000, extolling the virtues of its bid. It regarded it as the best on the table (£350m against £320m) and asserted that it was “the only party able to act as an independent broker for the British racing industry” (probably an allusion to the fact that Arena was in the ATR consortium and owned six courses). Carlton also modified the conditions of its bid: it now required, as a minimum, 30 acceptance by all TRG courses apart from Chepstow, Doncaster, Lingfield, Newcastle and Uttoxeter; and by a total number of courses accounting for 65% of betting turnover. There is no evidence as to why it imposed this change.[68]The RCA held an EGM on 21 November 2000, and wrote to all courses on 28 November 2000. It said that RHT (with, we note, 26.2% of betting turnover) had agreed to enter into an exclusivity agreement with ATR until 15 December 2000 and it recommended all courses to do the same. The result was that 57 of the 59 courses did so, the exclusivity arrangement being renewed at regular intervals. Neither ATR nor any of the courses was committed to such renewals, although ATR always wanted them. The larger courses favoured the ATR offer as it guaranteed terrestrial coverage on Channel 4. It combined the extensive broadcasting experience of Channel 4 and BSkyB with Arena’s heavily promoted Trackplay interactive technology. Many of these courses regarded Channel 4’s involvement as crucial given its experience of and commitment to terrestrial broadcasting, and because it had also indicated that it could cease terrestrial broadcasting of racing unless it was allowed the opportunity to exploit interactive racing. This led to ATR’s bid being described by Mr Sporborg, RHT’s chairman, as the “only deal in town.” The larger courses were sceptical as to whether Carlton could offer equivalent terrestrial coverage. Many smaller courses, which had less or no terrestrial coverage, favoured the Carlton bid because it offered more money for their interactive rights; and some chose to negotiate instead with GG Media, another bidder.[69]GG Media differed significantly from Carlton and ATR in that it sought to acquire not just the courses’ interactive rights but also their traditional LBO rights. By April 2001, some 26 courses were interested in the GG Media terms (none was a Super 12 course, nor under common ownership with such a course, nor owned or controlled by Arena or Northern). In that month, GG Media sent letters to these courses inviting them to a presentation at which an offer for all their media rights (including LBO rights) would be made. GG Media’s solicitors made it clear that they were unwilling to hold 26 separate sets of negotiations with these courses and end up with 26 different contracts: they suggested that all courses interested in signing up with GG Media should attend a joint meeting at the offices of the solicitors acting for Exeter Racecourse (one of the 26). 31[70]On 1 May 2001, the RCA Board met to decide whether to make a recommendation in favour of either the ATR or the Carlton bid. The RCA Board knew that by then the Super 12 and RHT (representing courses with a 41.3% share of betting turnover) would accept the ATR bid, which the Board recommended as now being the only one that could be accepted by sufficient courses to satisfy the conditions of the competing bids. The RCA Board then sent a package of documents, including its recommendations and the MRA, to all courses that were not in exclusivity, or had not signed up, with GG Media.[71]On 3 May 2001, ATR issued a press release announcing it had obtained acceptance signatures from courses representing over 71% of UK off-course betting turnover. The signatory courses were those owned by the TRG, RHT, Arena and Northern. None of the independent courses had signed up at that time. The release urged them to do so by the deadline of 11 May 2001, an urging directed at limiting the number of defectors to GG Media. When the RCA sent the package to the racecourses, it did not know (nor did ATR) which, if any, of the independent courses would sign. In the event, no deal was finally concluded between the courses and ATR until June 2001. This was because of difficulties met by the RCA in negotiating a data licence with the BHB on ATR’s behalf and, subsequently, by ATR in negotiating this on its own behalf. The acquisition of such rights was essential to the exploitation of the Non-LBO bookmaking rights: without them, the interactive rights were virtually worthless. VII THE MRA[72]The MRA was signed by Holdings, ATR, the RCA and 49 racecourses (“the Courses”) out of the 59 (the remaining ten signing with GG Media). By the second recital, each signatory party acknowledged that: “… the provisions of this Agreement … are necessary for [ATR] to be able to fulfil its objective of realising the full potential value of British horseracing. [ATR] is committed to the development of British horseracing, whilst preserving its tradition and culture, through advertising and innovative promotion, including [via the provision of interactive betting services] …” 32[73]Clause 2.2 provided, so far as material, as follows:
“2.2 If any of the following conditions precedent are not satisfied: … 2.2.2 the receipt by [ATR] by no later than 6.00pm on 11 May 2001 of a copy of this Agreement duly signed by the RCA and copies of this Agreement and Confirmations, each duly signed by: (i) all of the TRG Courses, together with all of the Non-TRG Courses whose racecourses are under common ownership or control with any racecourse owned or controlled by the TRG Courses; and (ii) such of the Non-TRG Courses which, together with the TRG Courses, account for no less than 70 per cent of the total annual UK off-course betting revenue as determined by reference to the Betting Revenue percentages set out in Schedule 16; 2.2.3 the execution of an agreement by no later than 30 June 2001 with the BHB (and/or the RCA) for the provision of the information and Data referred to in Clause 3.4.1 on the terms set out in that Clause or on such other terms as are acceptable to [ATR] and the RCA (for the avoidance of doubt, the execution by the RCA of this Agreement shall not satisfy this condition precedent); … then this Agreement shall automatically terminate without any party owing any liability to the other …”
[74]Satisfaction of condition 2.2.2(i) required acceptance by not just the 17 TRG courses, but also by the five non-TRG courses owned by RHT: Huntingdon, Market Rasen, Nottingham, Warwick and Wincanton; and by the three non-TRG courses owned by Northern (Bath, Brighton and Hereford). This collection of courses totalled 25 and as Arena was also obviously going to sign up in respect of its own six (non-TRG) courses (so bringing the total to 31), the effect of the satisfaction of the clause 2.2.2(i) condition was that the clause 2.2.2(ii) condition would inevitably also be satisfied, because these 31 courses had more than a 70% share of betting turnover. ATR received acceptances from these courses by 3 May 2001 and, as referred to at paragraph 71 above, was able to issue a press release on that day saying that they represented over 71% of betting turnover. The release added that ATR believed that: 33 “… it is important to ensure that the best solution for British racing is achieved, with as many racecourses as possible of the 59 signing up. [ATR] is committed to achieving this and will be working with these important racecourses between now and the deadline of 11 th May 2001…. We hope to be able to welcome most of the remaining racecourses by 11 th May to ensure we deliver the best solution for racing.”[75]One inference from the clause 2.2 conditions is that they represented the minimum requirements of the quantity and quality of fixtures that ATR needed in order to launch its new channel: but it plainly wanted acceptances by more than just the minimum number of courses sufficient to meet the conditions. These conditions go to what was referred to in the OFT’s Decision and the argument as “the critical mass”, that is the rights from the minimum number of courses sufficient to make the new interactive venture a viable one. We also draw attention to condition 2.2.3, relating to the provision of data as to form, running arrangements, horse numbers etc and essential for anyone offering the betting service that ATR was proposing to offer.[76]By the MRA, the Courses granted various rights to ATR on a worldwide basis. In exchange, the Courses were to receive payments by ATR to the RCA, to be divided between them in accordance with a formula agreed between them and the RCA. The rights were granted for a ten-year term, subject to certain break rights exercisable after five years. Three classes of rights were granted:(a) access rights, namely rights for ATR and its licensees to attend all races at the Courses (and events relating to them held on the same day) with outside broadcasting units in order to produce audio and audio-visual coverage of the races and events (hereafter “races”);(b) media rights, comprising: (i) the exclusive right for ATR and its licensees to produce films, live feeds and audio coverage of races and any other material comprising moving representations of races where such material was derived from the access rights; and (ii) the exclusive right for ATR to distribute worldwide in any media, by itself or its licensees, all materials in relation to races filmed, recorded or produced by Racecourse Technical Services Limited (an RCA 34 company) together with all sound tracks and on-course commentary; and(c) other rights, including: (i) the non-exclusive right to advertise, promote and publicise British horseracing and the exercise of the rights at races; (ii) the like right to use information and data relating to races for exploitation and in connection with rights granted under the rights agreement; and (iii) the right, in conjunction with the distribution of this coverage, to offer fixed-odds and/or pool betting services and interactive functionality including e-commerce services. The all-important interactive rights were, therefore, included as part of these “other rights”. Some 20 categories of rights were also reserved, certain of which have now expired, but those that have not included the right to produce, transmit and/or otherwise to make available material for reception in LBOs in the UK and the Republic of Ireland.[77]As for licence fees, the ATR offer to all 59 courses had been £320m (£30m less than Carlton’s offer of October 2000). As only 49 signed, the total of the guaranteed payments was reduced to £307m, payable to the RCA over the ten-year term (to be split between the Courses) and apportioned as to £125m to terrestrial rights and as to £182m for all remaining rights (including the interactive rights, but excluding the LBO rights), with a total of some £222m going to the Super 12. (This apportionment reflects a (proportionate) change from the original £140m and £180m in the original ATR offer to the 59 courses, but ATR had in the meantime proposed a change in that apportionment to £125m and £195m). ATR never intended to exploit the terrestrial rights itself and entered into simultaneous sub-licence agreements with Channel 4 and the BBC, to whom it also passed the burden of the £125m licence fee. ATR also intended to recover a significant proportion of the £182m by granting sub-licences of overseas rights. It further agreed to spend at least £80m on marketing and promoting coverage of the races and British horseracing generally. This was in the nature of an obligation by ATR to market its own business: there was no guarantee that any money would be paid to the Courses. 35 VIII NOTIFICATION TO THE OFT[78]On 15 November 2001, Holdings (ATR’s parent), each of its three shareholders and the RCA jointly made a “Form N” notification of the agreement to the OFT under the former section 14 of the Act. It asked for a negative clearance or, alternatively, an individual exemption under section 4. The application related to the formation of ATR to bid for and exploit the racecourses’ media rights, the sale of such rights by the Courses “pursuant to [the MRA] negotiated on their behalf by the RCA”, and certain ancillary arrangements relating to the licensing and exploitation of the rights, the inter-related agreements relating to these matters all being (and hereafter called) the “notified arrangement”. It explained that ATR had acquired the rights for ten years from 1 July 2001, with a review of the position in the fifth year, under a commitment by ATR to share its revenues with the Courses on an agreed basis, including commitments to make minimum payments totalling £306m over the ten years and to spend at least £80m on promoting racing in Britain. ATR would exploit the rights by creating a new pay-television channel, to be dedicated to racing and distributed as a “basic-tier” channel via DSat, cable and DTT. Fixed-odds and pool betting was to be accessible from it via interactive television services, the pool betting services to be provided in conjunction with the Tote. ATR was also to operate a website featuring live video-streams of races and facilitating interactive fixed-odds and pool betting. No fee was to be charged to viewers of the channel or website. ATR was to derive its revenue from sub-licensing the terrestrial rights to certain races to the BBC, Channel 4 and BSkyB, but primarily from a share of the interactive betting revenues from bets placed by punters via the channel or the website. The applicants explained that Arena Online had developed the “Trackplay” system, which was to be used to provide and operate the pool betting services.[79]The applicants asserted that the notified arrangement had no, or no appreciable, effect on competition in any relevant market. They said there were three relevant product markets: the supply of rights to video programming; the supply of video programming to distributors; and the supply of betting and gaming services. They said that effective competition would remain on all relevant markets and that neither the Courses nor ATR had, or would have, market power in any relevant market. They asked, alternatively, for an individual exemption. They said the new “basic tier” 36 channel would lead to wider television coverage of British racing and from a wider range of courses, with the new website providing coverage for the first time on the internet of British racing. ATR’s services would facilitate the development of interactive betting, in particular pool betting. They said the notified arrangement:
“… will also contribute to supporting the economic development of the British racing industry at a time when it faces considerable short to medium term financial uncertainties, competition from other leisure pursuits and the need to find new, commercial sources of income from the planned abolition of the Levy Board and the privatisation of the Tote. It is hoped that the Notified Arrangement will create a ‘virtuous circle’ of technical and economic development, which will improve the entire industry;”
They said consumers would benefit from the introduction of the new and improved services, effective competition would continue on all markets and that:
“there are no restrictions on competition which are not indispensable to achieving the above benefits. The Courses had no feasible alternative to centrally negotiating the Rights Agreement in order to maximise the value of the Rights. Arena, BSkyB and Channel 4 had to create the [ATR] joint venture, as individually they could not have successfully bid for the Rights. The ten year duration of the Rights Agreement is objectively necessary to maximise the returns to British racing and to enable [ATR] to make a reasonable return on its substantial investments in acquiring and exploiting the rights, particularly given the novel, innovative and untried [ATR] Model;”
[80]Pages 25 to 29 of the notification explained the background against which the MRA came to be signed, to which the OFT did not refer in its Decision. The applicants, however, regarded this as important to an understanding of the notified arrangement. It included: the claim that British horseracing comprised an important section of the economy but faced significant challenges from uncertainty over its future funding and increased competition from other leisure pursuits; that it was, in comparison with racing overseas, significantly under-funded; that racecourses were, in general, barely profitable and were largely dependent on the levy, with only a limited amount of revenue coming from the exploitation of media rights; and it referred to the Government’s intention announced in March 2000 to abolish the levy and its statement that “the arrangements under which racing receives income from betting should become a matter for settlement between the parties on a commercial basis.” The applicants referred to the BHB plan, published in October 2000 ( The 37 Future Funding Plan for British Racing), which noted that British racing’s ability to generate income from betting turnover without the levy was:
“dependent upon Racing being able to identify a product or products for which it can charge the betting industry in the same way that a seller charges a buyer in any commercial transaction.”
They said that:
“The BHB advocated, amongst other things, that all British racing’s media rights should be combined into a rights package for sale to bookmakers and media companies for an initial 10 year period, with income to be shared between prize money, the racecourses, and the BHB (on behalf of other industry participants). The BHB considered at the time that combining and selling the rights as one package, would enable racing to: - develop a unified strategy and to unite as an industry; and - develop a sufficient, dependable and sustainable income stream The combining and exploitation of media and other rights in British racing as a single package was seen by the BHB and other industry participants as key to the future financial viability of British racing, although in fact the 59 racecourses’ media rights were ultimately sold in two distinct packages, to [ATR] (49 racecourses) and to GG-Media (10 racecourses).”
[81]The applicants explained how British horseracing faced increased competition from other leisure pursuits and that its future depended on its public profile being raised. “Racing must become an attractive, broad-based leisure activity, generating a dependable and long-term income stream, the principal source of which must inevitably be the betting industry.” They explained how, under the notified arrangement, ATR was to be able to offer services comprising enhanced terrestrial coverage “and services that are both new and innovative.” The latter involved the introduction of the new channel and website and the applicants explained how these would make available its new interactive betting opportunities, including pool betting in conjunction with the Tote. The applicants referred to the ATR business model, under which:
“… the ability to offer betting and interactive functionality (and to permit third parties to do the same) is integral to the right to distribute coverage of the races. The Model creates a “virtuous circle”: increased coverage of racing, linked to increased opportunities to place bets, leads to greater revenues being 38 paid to the Courses, enabling them to invest in improved facilities and prize money, thereby raising the standard of British racing, and attracting new interest in horse racing (and betting) among the UK public. Separating out the rights to offer betting and interactive functionality from the coverage rights [which we understand to be a reference to terrestrial coverage rights] would not create the same benefits for British racing.”
[82]Turning to the various agreements making up the notified arrangement, the applicants first described the MRA. They explained that it was conditional on various matters, including “the participation of certain courses accounting for not less than 70% of total UK off-course betting revenue for 2000.” They said of this:
“The participation of a minimum number of courses (measured by betting revenues rather than by number of meetings or races) is necessary for the Model to work: a channel dedicated primarily to British racing and funded largely by betting revenues, clearly needs sufficient programming, including live coverage of races, both to fill the available hours and to encourage viewers to bet.”
This “minimum participation” was the “critical mass”. The applicants also referred to ATR’s separate agreement with the BHB, one not forming part of the notified arrangement, concerning the grant of non-exclusive licences by the BHB to ATR to use the BHB database of racing information for ATR’s broadcasting and new media and bookmaking activities: without these rights, the grant of the Courses’ interactive rights to ATR would have been virtually worthless.[83]The applicants explained the benefits of the notified arrangement, being benefits including improved live coverage of UK horseracing on television; wider distribution; cheaper access to it for consumers; the provision of interactive functionality enabling the placing of bets and including a pool betting service; and the reversal of the then current decline in the fortunes of British racing. They explained why the notified arrangement did not result in any appreciable prevention, restriction or distortion of competition in any relevant market: on the contrary, it enhanced competition and benefited consumers. They referred to the manner of the sale of the Courses’ rights to ATR, the exclusive nature of the rights granted and the duration of the rights period, all of which they said were “objectively justifiable and necessary for the Notified Arrangement to be a success, and therefore do not prevent, restrict or 39 distort competition in any relevant market.” They said this of the manner of the sale of the rights by the Courses:
“… the [MRA] was negotiated by the RCA as the representative of its member racecourses, but each racecourse considered a number of other offers and some racecourses chose to accept an alternative offer from GG-Media. Further, an insufficient number of racecourses accepted [Carlton’s] offer, causing it to lapse, and each racecourse also decided not to accept a number of previous offers made by a variety of undertakings, including BSkyB, Channel 4 and ntl (jointly), Arena and TVG. It was only when [ATR] was established that a bid acceptable to the Courses could be made by the [Holdings] shareholders. There was, and is, no agreement or other arrangement between the members of the RCA to sell the Rights in any given manner, including to the same person to whom other racecourses sell their rights. The approach adopted by the RCA and the Courses in relation to the sale of the Rights therefore did not prevent, restrict or distort competition. The [MRA] was conditional upon acceptance by a minimum number of racecourses. The participation of a minimum number of racecourses was objectively necessary in order to create an attractive product capable of commercial exploitation by a purchaser, in particular with regard to the need to secure a revenue stream through interactive and internet betting. None of the racecourses (or groups of commonly owned racecourses) had sufficient rights to offer to a purchaser on an individual basis. Without a level of ‘critical mass’, no purchaser would have been able to make a commercially sustainable offer that would have been attractive to the racecourses. The Applicants also consider that the sale by the Courses of the Rights is justified on the basis of the ‘solidarity’ principle. As has been recognised by the European Commission, the collective selling of sports rights or the central negotiation of individually concluded contracts, together with the resulting redistribution of income, justifies this method of selling rights.”
[84]As regards the bidding process itself, the applicants said that: “… There was a competitive and fair bidding process, resulting in fierce competition between the two main contestants, [ATR] and [Carlton]. Subsequently another offer was made to certain of the RCA’s members by GG-Media. In the absence of the [ATR] joint venture, the individual Shareholders could not have successfully bid for the Rights on their own. In such a case, and in the absence of other bidders, [Carlton] would have faced no or reduced competition for the Rights and the Courses could be expected to have received a lower amount or less favourable terms than if there had been a competitive bidding process.” 40[85]In the section of the notification making out the case for (if necessary) an individual exemption, the applicants said that:
“There is no agreement or arrangement between members of the RCA which requires them to sell their media rights in respect of races collectively. All British racecourses (including the Courses) have been free to negotiate and to sell their rights on an individual basis, rather than through the agency of the RCA, and to decide whether to enter into the [MRA]. Indeed, 10 racecourses decided not to accept the [ATR] offer negotiated by the RCA, despite the RCA’s recommendation that the [ATR] offer was the best for its membership as a whole. It was necessary for the Rights to be sold pursuant to a centrally negotiated agreement in order to put together a package of rights sufficient to be attractive to purchasers and to allow the radical move away from the current method of funding British racing (i.e the Levy plus limited commercial revenues). The involvement of a significant number of courses was necessary in order to achieve an efficient sale of the Rights and the necessary ‘critical mass’ to the [ATR’s] product offering feasible.”
IX THE OFT’S RESPONSE TO THE NOTIFICATION[86]The OFT informed the applicants of its objections to the notified arrangement on 10 July 2002. On 8 April 2003, after a period of informal consultation (during which the applicants re-affirmed the integrity of the arrangement), the OFT issued a Rule 14 Notice concluding that one aspect of it infringed the Chapter I prohibition. It was issued to the RCA, ATR, Holdings’ three shareholders and to the Courses. On 18 June 2003, the BHB sought to intervene in the proceedings. After initial opposition by the OFT and the launch of judicial review proceedings, the OFT permitted it to do so.[87]In the meantime, Holdings and ATR had a change of heart about the MRA. It was proving to be commercially unsatisfactory and so they decided to retreat from their stance in the notification that the MRA should be given a negative clearance by the OFT or else a section 4 exemption. They instead chose to adopt the OFT’s stance that the MRA was anti-competitive, unlawful and void and thereby repudiated their fulsome assertions to the contrary effect in the notification. 41 X ATR’S PURPORTED TERMINATION OF THE MRA[88]Under clause 24.3 of the MRA, ATR was entitled to give notice to the RCA if the “Tote Takeout Rate” (as defined) reduced to less than 20% over any successive period of three months. This rate is the proportion of pari-mutuel betting turnover which the Tote retains and in which ATR shares when its punters place pool bets. ATR gave notice on 24 October 2003. That started a 90-day negotiation period. No agreement was reached and on 29 January 2004 ATR gave notice purporting to terminate the MRA. That notice expired on 29 March 2004. The RCA disputes its validity and there is an issue as to whether the MRA has been lawfully terminated, the RCA’s stance being that ATR has wrongfully repudiated it. On 6 May 2004, ATR claimed the payment of rebates exceeding £58m, relying on provisions in Schedule 7 to the MRA. It is currently suing 31 of the 49 signatory Courses (those that have not since signed up to a new ATR service) and it has reserved the right to bring a claim for competition law damages.[89]The RCA appellants urged upon us that the present appeals are not just about an agreement which has been actually or purportedly terminated. They face the threat of a competition damages claim, which could (under section 47A of the Act) be brought before this Tribunal; and any decision of this Tribunal upholding the OFT’s decision that the MRA infringed the Chapter I prohibition would bind the parties in any such claim. We recognise that the outcome of these appeals is of considerable commercial importance. XI THE DECISION OF THE OFT[90]ATR has purported to terminate the MRA, but the OFT described ATR’s activities in the present tense. In summarising their Decision, we will do the same. The OFT found that one aspect of the notified arrangement infringed the Chapter I prohibition and did not qualify for individual exemption. That is the collective selling by the 49 Courses of the media rights necessary for the production of the ATR interactive digital TV channel and website. The OFT approached the matter on the basis that the principal concern in relation to a suspected infringement of the Chapter I prohibition is whether any agreement or concerted practice has an appreciable effect 42 on competition. This required the identification of the products relevant to the notified arrangement and of the product market in which they lay, the OFT recognising that the relevant market may be wider. It recognised that the identification of the relevant product in complex service industries may be difficult. Relevant products[91]The OFT identified five relevant products: two “upstream” sets of rights licensed by the Courses, being necessary inputs for three “downstream” products supplied by ATR.

(a) The two upstream rights

[92]The OFT’s view was that, as the Courses could license each of these independently, and as the OFT considered them likely to have separate demand substitutes, they comprised separate relevant products. They are:(i) The viewing rights These are the rights licensed by the Courses necessary to permit ATR to supply programming covering races to TV distributors for viewing.(ii) The Non-LBO bookmaking rights These are the interactive rights licensed by the Courses necessary to permit ATR to supply programming covering races to UK bookmakers other than LBOs, for distribution in combination with betting services. We will call them either “the interactive rights” or “the Non-LBO rights”.

(b) The three downstream rights

[93]These are: 43(i) British horseracing programming supplied to TV channel distribution for viewing ATR has used the viewing rights to create the television channel (“the Channel”), supplied to TV distributors (via the DSat platform and cable). It also supplies smaller pieces of footage to TV distributors, including both free-to-air broadcasters and pay TV operators. It allows a significant level of analogue terrestrial coverage of live racing, through the sub-licensing of television rights to the BBC and Channel 4.(ii) Access to in-vision betting services via interactive digital TV (“iDTV”) The OFT found that, using the interactive rights, ATR supplies punters with access to in-vision fixed odds and pool betting on British horseraces via iDTV. The punter can press a button on the digital TV remote control while watching the Channel. He will continue to see the pictures, which will fill ¼ of the screen, and the sound will be uninterrupted. The remaining ¾ of the screen will be filled with text and he can place bets by navigating through a series of text menus, with the available bets and odds being displayed on the screen. The Channel also provides services with supplementary information about race cards, odds, form and results.(iii) Internet access to betting services linked with live pictures of British horseracing Using the interactive rights, ATR has created its website (“the Website”). This provides information relating to horseracing, access to several fixed-odds bookmakers’ services and the pool betting service. It incorporates live streamed video coverage of horseracing, allowing punters to watch races on which they have placed bets. When the race starts, a second window opens on the PC screen, displaying sound and pictures of the race. Users can access the Website via PCs, WAP mobile phones and PDA devices. Relevant markets[94]In identifying the markets relevant to these five products, the OFT used the hypothetical monopolist test, which it summarised as follows: namely, whether a hypothetical monopolist of a certain set of products could maximise profits by 44 consistently charging higher prices than it would if it faced competition. If the hypothetical monopolist would be prevented from profitably setting prices above the competitive level by customers switching to alternative products, such substitutes should be included in the relevant market.

(a) The market relevant to the supply of viewing rights

[95]The OFT found the market relevant to these rights to be sufficiently broad for the MRA not to have an appreciable effect on competition: it considered that other television programming can attract a similar viewing audience to horseracing.

(b) The market relevant to the supply of the Non-LBO rights

[96]The OFT found that internet and in-vision iDTV bookmaking services are provided to punters by Non-LBO bookmakers; and that the suppliers of these services are the two categories of Non-LBO bookmaker who might choose to purchase sound and pictures of British horseraces. It concluded that the position in the UK is that the market is limited to the supply of live sound and pictures of British horseraces, as opposed to those of foreign horseraces, the latter accounting for less than 5% of off-course betting turnover. It found that foreign horseracing is generally used in UK LBOs to fill gaps between UK races rather than as a stand-alone product. The courses’ market share, measured by betting turnover, would not be materially affected by the inclusion of foreign horseracing in the relevant market.[97]The OFT found that, in licensing their rights, racecourses can set terms and conditions which determine how the broadcasting of live sound and pictures can be used. They can charge different prices for:(a) the supply of viewing rights for programming for subsequent supply to TV distributors; and(b) for programming that is to be combined with bookmaking services. They can charge different prices to different bookmaking outlets (LBOs and Non-LBOs) and to bookmakers in different countries. It follows that switching behaviour by TV distributors, foreign bookmakers or LBOs will not constrain the price of rights to Non-LBO bookmakers. The demand by these bookmakers is a derived demand. The OFT accordingly analysed the impact of an increase in the price of live sound and pictures of British horseracing to UK 45 Non-LBO bookmakers: internet and iDTV bookmakers are such bookmakers who might choose to purchase such sound and pictures; and ATR is a UK non-LBO bookmaker which supplies access to other Non-LBO bookmakers’ services.[98]The OFT considered that an increase in the price of pictures should not increase the unit price of bets: the Non-LBO bookmaker covers the cost of pictures from the profits generated from the additional demand the pictures stimulate. Its analysis was that if a monopolist licenses rights to a distributor for a fixed fee (independent of turnover), an increase in the fee does not increase the distributor’s marginal cost of production and (if the distributor’s marginal costs do not change) there are unlikely to be economic reasons for the distributor to change the price it charges the downstream customers. The OFT concluded that it followed that an increase in the monopolist’s licence fee “may not” affect the price at which the product is offered to downstream consumers. It noted that (paragraph 93, footnote 57):
“If the higher fee results in some distributors exiting the market, because their fixed costs are higher, then such exit may lead to an increase in the price paid by final consumers (punters). However, the OFT notes that, in any market, non-marginal distributors may make supra-normal profits (although if entry barriers are low then entrants will not be able to make supra-normal profits). An increase in the licence fee that only affects these non-marginal distributors merely captures some of those profits and transfers them upstream. Accordingly, such a non-marginal distributor’s behaviour will be unaltered (assuming it continues to purchase the licence).”
[99]The OFT’s overall conclusion was that a Non-LBO bookmaker will continue to purchase pictures of British horseracing following a small, significant increase in their price. In so concluding, it explained first the impact of pictures on betting turnover.

(i) The impact of pictures on betting turnover

[100]The OFT found that the evidence (including the inferences justified by the increase in LBO betting turnover when audio coverage is provided, or when both sound and pictures are provided) is that, by providing live British horseracing 46 coverage, a Non-LBO bookmaker will significantly increase demand for British horseracing bets.[101]It found that the provision on a website of live pictures attracts extra custom which would be lost if the website ceased to provide live pictures in response to a picture monopolist’s price increase. Combining live pictures and betting enhances both the gambling and the viewing experience: part of the excitement of placing a bet on a sporting event is in watching the event unfold. Punters are therefore more attracted to bookmakers who provide pictures. Similarly, in-vision bookmakers located on a channel devoted to British horseracing have an advantage over Non-LBO bookmakers unable also to offer pictures. The Channel is likely to attract viewers interested in betting on British horseraces and may prompt viewers to place impulse bets. Channel viewers would have to change channels to bet with another iDTV bookmaker. One attraction of iDTV betting is the convenience of betting on an event merely by changing to a different iDTV screen; another is that the punters can watch the event on which they are betting. The evidence led the OFT to conclude that iDTV bookmakers will attract substantial extra custom by providing services primarily dedicated to British horseracing, an advantage they would lose if they ceased taking live pictures in response to a picture monopolist’s price increase. As regards fixed-odds bookmakers, ATR provides punters with access to them in conjunction with live pictures of races. The OFT’s conclusion was that the price that such bookmakers are willing to pay for such access indicates their assessment of the additional business they will attract from the simultaneous provision of live pictures. It also found that the evidence from LBOs was that supplying live pictures in conjunction with betting opportunities encourages punters to bet substantially more. (ii) Possible substitutes for betting on British horseracing[102]In identifying the relevant product market, the OFT also focused on the punters for whom the provision of pictures of British horseracing are especially important, and who will not be attracted by pictures of alternative events, for example greyhound racing. These punters would be lost by the non-LBO bookmaker if it ceased to purchase pictures of British horseracing. 47[103]The OFT’s conclusion was that most punters attracted to bet by pictures of British horseracing will not be attracted to bet by pictures of foreign horseracing or other competing events. This implied that pictures of such alternative events are not close substitutes for pictures of British horseracing. This was supported by the OFT’s conclusion that most punters who bet on British horseracing have built up a reservoir of knowledge about such racing, which they will not have about other competing events: and that they will not wish to incur the costs of acquiring an equivalent knowledge about such events. The OFT found that most punters attracted by pictures of British horseracing will not be attracted to bet by pictures of numbers games or other forms of gaming. Their explanation of this is that it is important to punters to feel they are exercising judgment when placing bets, whereas numbers games do not allow them to exercise judgment in the same way that betting on horseracing does. The one exception is the National Lottery, which many horseracing punters also play.[104]The OFT rejected the suggestion that there might be substitution between fixed odds betting and the National Lottery. Bookmakers are prevented from taking bets on the Lottery. Even if punters place additional wagers on the Lottery when supplied with pictures of the Lottery, bookmakers cannot boost their income by screening those pictures. The OFT concluded that UK non-LBO bookmakers would not regard sound and pictures of the National Lottery as a substitute for the sound and pictures of British horseracing. The OFT’s conclusion on relevant product market definition[105]The OFT’s overall conclusion was that sound and pictures substantially boost demand for bets. Most punters attracted to place bets by pictures of British horseracing are not attracted by pictures of other sports and betting opportunities. The implication was that the value of the additional demand that can only be attracted by pictures of British horseracing is large. The demand will be lost if a bookmaker ceases to purchase the pictures. Accordingly, a Non-LBO bookmaker would be likely to continue to purchase pictures of British horseracing following a small, significant price increase above competitive levels. 48[106]The OFT therefore considered that a hypothetical monopolist could increase the price of live pictures of British horseracing to UK Non-LBO bookmakers above the competitive level; and that a hypothetical racecourse monopolist could increase the price of the rights used to produce programming of British horseraces for combination with UK Non-LBO bookmaking services. The OFT found, therefore, that the supply of Non-LBO bookmaking rights by the Courses to ATR is a relevant product market in which the notified arrangement may have an appreciable effect. It defined the relevant geographic market as the UK (on which there is no dispute). The OFT’s findings on the Courses’ and ATR’s position in that market[107]An OFT guideline on the Chapter I prohibition is that “an agreement will generally have no appreciable effect on competition if the parties’ combined share of the relevant market does not exceed 25 per cent.” The OFT found the notified arrangement to be capable of having an appreciable adverse effect on the market for the supply of Non-LBO bookmaking rights. The Courses’ combined share of the relevant market was very high: it accounted for 90% of Ladbrokes’ betting turnover on British horseraces in 1999 and 91% in 2000. If market share is measured by reference to their share of British horseracing fixtures, they had a share of about 90%. The ten courses that sold their rights to GG Media accounted for the remaining share.

(i) Potential competitors in the market

[108]The OFT considered the extent to which the entry of potential competitors into the market may constrain the competitive behaviour of the Courses. The Jockey Club and BHB regulate British horseracing. Without a licence from the Jockey Club, a new racecourse can only hold unrecognised meetings (known as “flapping”). This is perceived as having poor integrity in comparison with “recognised meetings”, a feature which will reduce an unlicensed racecourse’s income, particularly from bookmakers. In addition, owners, trainers, riders and officials involved with an unrecognised race meeting can be disqualified by the Jockey Club for up to twelve months, during which they will be unable to have any connection with any recognised race meeting. The result is that unlicensed racecourses will face difficulty in attracting participants. 49[109]The OFT concluded, therefore, that a new racecourse would wish to be licensed by the Jockey Club. A course so licensed can then, however, only hold races on days when it has been allocated fixtures by the BHB. It will also require a substantial number of fixtures in order to justify its investment. The OFT found that of the eight applications made to the BHB in 1998, none had yet entered the market. It said that this reflected difficulties in obtaining planning permission and BHB fixtures: there had been no new racecourse in the UK since 1927. The OFT concluded that the threat of new entrants would not constrain the behaviour of the Courses. (ii) Barriers to output expansion[110]The OFT considered whether courses might expand their output by holding more fixtures and/or more races per fixture. The BHB’s permission is required for the holding of additional fixtures; and that of the BHB and the Jockey Club is needed before the number of races per fixture can be increased. There are also constraints on the timing of races, which generally prevent racecourses from running them simultaneously. The OFT’s conclusion was that there were significant barriers in the way of courses expanding their output of Non-LBO bookmaking rights. (iii) Buyer power for the Non-LBO rights[111]Buyer power may offset the potential market power of the seller; but requires that the buyer should be large in relation to the market. The OFT referred to the applicants’ case that there was effective actual competition from Carlton and GG Media; SIS also made a bid in the summer of 2000; and there was potential competition from other broadcasters, rights brokers and others who could have bid for the rights: TVG, an American company providing interactive TV betting on horseraces in the USA, said that it was thinking about bidding, but in the event did not. The OFT’s conclusion was that, as there were several credible potential buyers of the Non-LBO rights, any of them was in a relatively weak position in relation to the Courses. Its conclusion was that there was insufficient buyer power to offset the market power possessed by the Courses. 50 Market power: the overall conclusion[112]The OFT concluded that collectively the Courses held market power in the market for the supply of the Non-LBO rights, and that the notified arrangement was capable of having an appreciable effect on competition in that market. It turned to consider whether such effect exists. Does the arrangement have an appreciable effect on competition within the UK?[113]The OFT referred to section 2 of the Act. It found the Courses to be “undertakings” within section 2(1) (there is no challenge to that). It found they had participated in an “agreement and/or concerted practice” within section 2 to “collectively sell” their Non-LBO rights, a sale effected by the MRA. It disagreed that such selling fell outside section 2 as being:(i) a sale which, because it involved rights to sporting events, promoted a legitimate objective; or(ii) a sale necessary to form a new product or service. It reached no conclusion on whether the agreement or concerted practice had the object of preventing, restricting or distorting competition; but held that it had the effect of doing so and that it may affect trade within the UK. We summarise the OFT’s reasons for these conclusions. (a) Was there an agreement or a concerted practice?[114]The OFT was satisfied that the 49 Courses’ conclusion of the MRA, negotiated for them by the RCA, amounted to an agreement and/or a concerted practice. It relied on:(i) the fact that when each Course signed the MRA it would have seen the names of all the other Courses listed on it and knew that each would be committing itself to adopt identical conduct with regard to the disposition of its rights;(ii) the provisions of schedule 14 to the MRA, providing for the distribution between the Courses of the money payable to the RCA by ATR, a formula arrived at after discussion between a Distribution Group of the Courses;(iii) the fact that each Course knew that the RCA was acting for the Courses in negotiating the MRA, even though it was not bound to accept the outcome: the OFT inferred that there was material contact between the Courses;(iv) that the applicants to the notification had stated in it that one reason for the centrally negotiated agreement was to realise the “full value” of the 51 Courses’ rights; and(v) the applicants’ own assertion of the “principle of solidarity”, by which financially stronger sporting participants support weaker ones. (b) Did the collective sale promote a legitimate sporting objective?[115]The OFT held there was no broad category of agreements between undertakings in the sporting sector that is in principle not subject to the Act. All such agreements must be assessed by reference to their effects on competition.[116]The OFT accepted that, in principle, an agreement effecting negative effects on competition may not offend the Chapter I prohibition if it was necessary in order to promote a legitimate objective and they referred to decisions of the European Court of Justice (“the ECJ”) supporting this approach (one of them, Case C-250/92 Gøttrup Klim e.a. Grovvareforeninger v. Dansk Landbrugs Grovvareselskab AmbA [1994] ECR I-5641 featured prominently in the argument to us). But it held that the Courses would have sold their rights anyway (without collective selling); that the notified arrangements reduced their incentive to improve their output; that there was no basis for the claim that it enhanced the distinctive characteristics of horseracing; that collective selling was not necessary to counterbalance the buyer power; that the ECJ case law showed that an agreement would infringe Article 81(1) of the EC Treaty if it was inconsistent with “workable competition”, rendered the relevant market excessively rigid or significantly increased barriers to entry or expansion. The OFT said the Courses had a 90% market share, were shielded by barriers to entry and output expansion and, by eliminating all competition between themselves, achieved a collective sale which was incompatible with the concept of “workable competition.” (c) Was the collective sale necessary in order to create a new product?[117]The RCA and BHB cases were that ATR needed a minimum amount of racing coverage (a “critical mass”) to make its channel and website commercially viable and that collective negotiation was the only way to meet this need. The OFT rejected this. It said the cases cited by the RCA in support of the argument were ones where the value created by undertakings by their co-operation was greater than if each had 52 contributed its output independently. This did not apply to the collective sale by the Courses. The OFT said that it:
“262. … accepts [ATR’s] need for a ‘critical mass’ of rights (although this does not imply that the value to [ATR] of a volume of rights just falling short of this critical mass is (almost) zero. However, it finds that the collective selling by all the Courses together was not necessary to achieve this aim. See paragraphs 397-404. For example, buyers could assemble the necessary critical mass. Therefore the collective selling in this case cannot be excluded from the scope of section 2 of the Act on this ground. The OFT considers that the precedents and extracts of the Notice on Horizontal Agreements cited by the RCA do not apply on the facts of this case, and that the collective selling falls within the Chapter I prohibition.”
[118]The OFT explained that conclusion further in subsequent paragraphs (281, 397 to 404). It found, first, that no single racecourse or racecourse group had a veto: they considered that only groups of racecourses acting collectively could have vetoed the agreement. As there were no veto holders, the Non-LBO rights of individual Courses were not “complements” in the economic sense: the Courses’ respective rights were substitutes. The constraint on their pricing was eliminated by the collective selling, which allowed them to act as a bloc and so substitute co-operation for competition.[119]In paragraph 397, it found that “by implication, buyers should be able to compile the necessary critical mass without collective selling by racecourses i.e., buyers can handle the practical and logistical aspects of assembling the necessary portfolio of rights.” It rejected the argument that central negotiation by the RCA was necessary for such assembly and preferred ATR’s argument that it was not. It found (in paragraph 401) that buyers could have assembled the necessary critical mass by making “any contracts conditional on obtaining rights from sufficient courses (as itself occurred in the Notified Arrangement, itself a conditional agreement). Such ‘conditional contracts’ could also have allowed [ATR] to assemble the necessary portfolio of rights prior to launch.” It found that: “Both BSkyB and Channel 4 produce TV channels and therefore have considerable experience of assembling the packages of rights necessary to launch channels by negotiating with many suppliers. Indeed [ATR] has stated that ‘it would be highly unusual and unrealistic for a television channel not to obtain programming rights from multiple rights sources, were it free to do so.’ [Fax from ATR to the OFT dated 15 August 2003]” 53 (d) Did the collective selling have an anti-competitive effect?[120]The OFT concluded that the collective sale of the Non-LBO bookmaking rights had an appreciable effect in preventing, restricting or distorting competition: it had two anticompetitive effects, namely it:(i) increased prices (due to a restriction of price competition between the Courses); and(ii) restricted incentives for non-price competition between the Courses. (i) Did the collective sale increase prices?[121]The RCA and BHB case was that certain courses had a veto on any agreement and so possessed the same market power as a monopolist. Collective selling therefore could not have had a relevant effect on the overall price. The OFT accepted that ATR required a critical mass of rights but considered that no course or group of courses had a veto. It referred to ATR’s evidence that its business plan was formulated on the basis that 70% of betting turnover was sufficient. Not even RHT, the largest course-owning group, had more than 26.2%. Thus any course, or group, was at risk of being excluded. It referred to condition 2.2.2(i) of the MRA and said that ATR’s evidence (in a letter of 7 August 2003) was that it was not essential for it to sign up with all the courses there referred to and that the venture could have been viable without some of them. The OFT referred to Ayr (a TRG course with only 2.1% of betting turnover) and did not accept that in practice it would have had a veto. Its overall conclusion was that only groups of courses acting together could have vetoed the notified arrangement. 121. The OFT rejected the argument that the courses’ rights were complementary, that the courses were not therefore in competition with each other and that the individual selling of complementary products can result in higher prices than collective selling. It accepted the principle of the last point, but did not accept that the courses’ rights were true complements. ATR’s need for a critical mass of rights did not prove otherwise, any more than that a bookshop needs a critical mass of books to be viable. Since no racecourse had a veto, the respective rights of each were true substitutes in the economic sense. The maximum a buyer will pay for any course’s 54 rights is the incremental value they generate: the incremental value of the 59 th course, once the rights of the other 58 have been acquired, will be low: and each of the 59 risked being at the back of the queue. The consequence is that, absent collective selling, each course’s pricing is constrained by rivalry. Collective selling prevented the buyers from exercising choice and striking deals with individual courses or groups. The OFT relied on ATR’s fax of 15 August 2003 that it “would not have been able to bid for the [R]ights individually, given that the RCA was only interested in offering the [R]ights in a single package.” The OFT concluded that the collective selling raised the price of the Non-LBO rights.[122]The OFT dealt next with the points that there was no evidence that the price was higher as a result of collective selling; that it had not shown what the counterfactual price would have been had there been individual negotiation; that neither ATR nor Carlton increased their bids during the negotiations; that neither complained that the manner of sale had increased the price; and that buyers were only interested in negotiating with a single body and were not interested in playing the courses off against each other. The OFT said these points were “not persuasive.” The collectivity between the courses gave them a power they did not hold individually; and even if ATR and Carlton did not increase their bids, that did not mean they were not anyway above the competitive level. The OFT accepted ATR’s denial (in its fax of 15 August 2003) of the assertion that the buyers were not interested in playing off courses against one another: it regarded the contrary case as unsupported by evidence and as counter to the buyers’ interests. It said the conclusion that collective selling increased prices was supported by contemporaneous statements from the RCA that if the 42 stuck together a “remunerative conclusion will be reached”, the statement to like effect by Mr Hutchinson of Ripon course on 6 March 2000 (earlier quoted) and certain statements to like effect made by the BHB in its representations to the OFT.[123]Finally, in this context the OFT dealt with the argument that, even if the price was increased, any increase was a competitively-neutral transfer from ATR to the Courses which did not affect the ultimate consumer: punters were not affected and output was not restricted. The OFT held that a finding of direct detriment to the final consumer was not a condition to an infringement of the Chapter I prohibition. It said the key legal question was whether an agreement prevents, restricts or distorts 55 competition on a relevant market within the UK and that such market need not be a retail market.[124]The OFT’s overall conclusion was that the effect of the Courses’ collective selling of the Non-LBO rights gave them a market power they did not hold individually and that they were likely to exercise such power to raise prices. They concluded that the effect of the collective selling was to increase the price of the Non-LBO rights above the competitive level. (ii) Did such selling restrict incentives for non-price competition between Courses?[125]The OFT’s conclusion was that it did, in particular that it restricted incentives for the Courses to improve the quality and nature of their output. It held that the effect was exacerbated by the ten-year duration of the MRA. It said that (absent collective selling) such incentives would exist. It said the relevant counterfactual in considering the question of incentives was agreements for the acquisition of the rights achieved by the Courses selling individually or in small groups rather than collectively. Under the MRA, over 94% of the expected ATR payments were guaranteed and fixed and divisible between the Courses under a pre-set formula. This reduced the incentive for a Course to improve the attractiveness of its output (for example, with regard to the nature, timing or quality of its races) relative to that of rival Courses, because such improvements would not significantly affect its ATR income. Whilst the Course will share in any top-up revenue payable by ATR if its revenue is sufficiently high, it will only receive a small fraction – too small to be an incentive to improve its output. The RCA’s response to the last point was that it had a power to allocate top-up revenue to the particular Course or Courses that had created it, a power which preserved all Courses’ incentives. The OFT accepted the theory but regarded it as unlikely to work in practice. First, how could it be shown which Course or Courses had pushed ATR’s revenue over the relevant threshold? Second, as a trade association, the RCA was poorly placed to allocate top-up money to one Course rather than another. A further argument was that the Orders and Rules of Racing anyway limit the scope for competition between Courses, but the OFT preferred the view that, even so, there was still scope for some freedom to vary their output under the Orders and Rules, including the characteristics of the racing they stage. 56 (e) Did the collective selling have an effect on trade within the UK?[126]The OFT concluded that the notified arrangement, including the Courses’ collective selling of their non-LBO bookmaking rights by the MRA, may have an influence, direct or indirect, actual or potential, on the pattern of trade within the UK due to its effect on the conduct of the applicants, the Courses, and their commercial rivals and customers within the UK. The OFT concluded that the Chapter I prohibition applied to the MRA. Did the agreement qualify for an individual exemption under section 4 of the Act?[127]The OFT referred to the section 9 criteria. It concluded that the notified arrangement as a whole improved coverage of British horseracing on television, distributed it more widely and made its access cheaper for consumers; but it considered that the collective selling did not contribute to these results: it raised the price payable by ATR for the rights. It did not consider that funding the Courses and British horseracing justified the collective selling that had these effects. It did not accept that the notified arrangement contributed to economic or technical progress by reducing transaction costs as compared with the potential costs of individual sales.[128]The OFT accepted that viewers and punters will receive a fair share of the benefits from the Channel, Website and terrestrial television coverage: they will receive the benefit of products not previously available at competitive prices. It did not accept that the collective selling made any contribution to these benefits in which the end-consumers may share. Further, it considered that collective selling was not indispensable to attaining the benefits resulting from the agreement. Collective selling afforded the possibility of eliminating competition with respect to a substantial part of the products in question, namely the supply of non-LBO rights. The OFT concluded that the agreement was not eligible for an individual exemption. 57 XII THE APPEALS TO THIS TRIBUNAL[129]The two sets of appellants did not make common cause on all points. Some points made by the RCA appellants were not made by the BHB and vice versa. To the extent that their grounds of appeal touched on common matters (for example, as to the relevant product market) they did not always advance overlapping arguments. The OFT sought to make some capital out of the differences of approach reflected in the two appeals, suggesting that they pointed to a lack of credibility for the appellants. There is nothing in that. Each ground of appeal must be judged on its merits. One good point may be enough to entitle an appellant to win. We consider the points arising on the appeals under the following numbered heads. We should say that we have been provided with pleadings and skeleton arguments running to hundreds of pages. To deal comprehensively with all points made in them would extend this judgment intolerably and we have therefore not attempted to do so, although we make clear that we thereby intend no discourtesy to counsel’s careful arguments. 1. Burden of proof[130]The principal legal basis on which the appellants founded their appeal is that the legal burden of proving that the MRA had as its effect an appreciable restriction of competition was on the OFT. Their stance is that they had to do no more than show how the OFT failed to prove that the MRA involved any infringement of the Chapter I prohibition.[131]Subject to one qualification, there was no issue that the legal burden of proof of the alleged infringement of that prohibition lay with the OFT. We were referred to Article 2 of Regulation 1/2003 in relation to Article 81(1) of the EC Treaty and it was not suggested that a different principle applies to section 2 of the 1998 Act. To like effect, we were referred to Joined Cases 29/83 and 30/83 CRAM and Rheinzink v. Commission [1984] ECR 1679, paragraph 16ff; and Napp Pharmaceuticals Holdings Limited v. Director General of Fair Trading [2002] CAT 5, paragraph 110. In Napp, and in JJB Sports plc and Allsports Limited v Office of Fair Trading [2004] CAT 17 this Tribunal confirmed that the standard of proof is the civil standard of balance of probabilities, although the seriousness of an infringement of the Act, involving as it 58 may the imposition of penalties, is a factor to be taken into account in considering the probabilities of an infringement having occurred (compare Re H and Others (Minors) [1996] AC 563 , at 586, per Lord Nicholls of Birkenhead; and Secretary of State for the Home Department v. Rehman [2003] 1 AC 153 , at paragraph 55, per Lord Hoffmann). Napp also confirmed that the OFT may only rely on inferences flowing naturally from a given set of facts in the absence of countervailing indications (paragraph 110).[132]The OFT submitted, however, that this position is qualified in cases in which the decision-maker has to decide whether what appears to be a restriction of competition is justified by the particular circumstances of the case. It submitted that, in such cases, whilst the legal burden of proving the infringement of the Chapter I prohibition remains with the decision-maker (here the OFT), the evidential burden of demonstrating that the apparent restriction on competition is justified falls upon the undertaking advancing such assertion: he who asserts must prove. The OFT submitted that, to the extent that the appellants defended the prima facie anti-competitive effect of the MRA as being “necessary” to achieve a pro-competitive outcome, the evidential burden of showing it lay on them.[133]We accept this. It cannot be for the OFT to set up and disprove a case founded on the “necessity” argument. If, as the appellants claimed, any apparently anti-competitive effect of the collective dealing between the Courses and ATR was justified by the necessity of such dealing, it was for them to demonstrate it by evidence. Once that evidence was before the OFT, the overall legal burden still remained on the OFT to prove the infringement of the Chapter I prohibition that it was asserting. But unless the appellants first made out a necessity case on the facts, no such case would arise for consideration.[134]The OFT also submitted that, to the extent that the appellants claimed an exemption under section 4, the legal burden of proof was on them. Regulation 1/2003 shows this to be so as regards claims to the benefit of Article 81(3), and we can see no reason why the same should not apply to claims under section 4. We do not understand how it could be otherwise. The BHB submitted that such claimants are only subject to an evidential burden, but we do not follow that. We conclude that the 59 appellants are subject to the legal burden, again one which can be discharged on the balance of probabilities. 2. The relevant market[135]The OFT’s Decision was that the relevant product market is the supply of Non-LBO rights. Its reasoning was:(i) that internet and iDTV bookmaking services are provided to punters by Non-LBO bookmakers (paragraph 28);(ii) there is a market for the supply of sound and pictures of British horseraces (paragraph 84);(iii) in licensing their rights, racecourses supply inputs used to produce and broadcast live sound and pictures of British horseracing (paragraph 90);(iv) the demand for Non-LBO rights is a derived demand requiring an analysis of the impact of an increase in the price of live sound and pictures of such horseracing to Non-LBO bookmakers (paragraph 91); and(v) such bookmakers would continue to purchase pictures of such horseracing following a small but significant price increase in their price (paragraphs 91 to 126). The OFT was led to its conclusion in (v) by evidence showing that the supply of pictures in conjunction with betting opportunities encourages punters to bet more and by the further evidence that there is no real substitute for British horseracing.[136]The RCA appellants submitted that the flaw in this reasoning was that the OFT failed to consider the effect of competition from interactive and internet betting and betting exchanges without pictures on interactive betting with pictures. ATR’s business model assumed that its one-stop shop of live pictures and interactive betting via its website would satisfy its viewers’ betting needs. But the advent of many internet betting sites and exchanges meant that many punters watched the ATR channel but placed their bets on other online sites or via their PC or their mobile phone – which is why the ATR venture failed. The punters watched its channel but used other sites for their betting, which offered either better odds or different bets. It is, therefore, said that the flaw in the OFT’s Decision is that it does not address the question of whether a website with pictures faces competition from interactive sites without pictures; and, therefore, that the OFT’s finding that a Non-LBO bookmaker would continue to buy pictures of British horseracing following a small, significant price increase is in consequence seriously undermined. The BHB also submitted that the OFT’s definition of the relevant market was unduly narrow and that the market 60 must at least also include the data rights thought to be controlled by the BHB, which are said to be complementary to the Non-LBO bookmaking rights.[137]We have concluded that the OFT’s conclusion that the Non-LBO rights constituted a distinct economic market was flawed. The OFT adopted the hypothetical monopolist test in arriving at their conclusion. Its reasoning depended crucially on the fact that, under the MRA, payments to the Courses were largely fixed. It recognised (in paragraphs 167 to 170) that in the downstream market there is a high degree of substitutability between alternative means of placing off-course bets (websites with pictures, websites without pictures, in-vision iDTV betting, telephone betting and LBOs). The evidence showed that the degree of substitutability is so high that bets placed through ATR were at the same odds as were generally available elsewhere.[138]We consider that, in normal circumstances, this degree of downstream substitutability would be reflected in a corresponding degree of upstream substitutability between the rights required to provide betting services in different ways (for example, the rights to provide betting with pictures in LBOs and through internet or iDTV services). An increase in the unit cost of the rights to one kind of supplier (the unit cost of the rights to other suppliers remaining unchanged) would either be reflected in the odds available from that supplier, causing punters to take their custom elsewhere; or would induce the supplier to cut other costs (for example, marketing costs), with similar consequences for customer demand. This would mean, in the present case, that the relevant market must at least be wide enough to include both the bookmaking rights to LBOs as well as the Non-LBO rights. The OFT’s position is, however, that this expected link between the downstream and upstream markets is absent.[139]The problem with the OFT’s adoption of the hypothetical monopolist test in their journey to the conclusion that the Non-LBO rights constituted a distinct economic market is that it required the OFT to ask itself whether, starting from the pricing in a competitive situation, it would be profitable for the monopolist to make a small but significant non-transitory increase in price (“the SSNIP test”). As, however, the provision by the Courses of the Non-LBO rights involved the provision of a novel service, there was no empirical evidence of a competitive price for it and so the 61 SSNIP test could only be applied to a hypothetical counterfactual of the competitive situation. The OFT’s analysis proceeded upon the basis that such counterfactual competitive price must be(a) fixed and(b) lower than that charged by the MRA.[140]As regards point (a), we have summarised the OFT’s point that an increase in a fixed fee does not increase the distributor’s marginal cost of production and so will be unlikely to generate economic reasons for it to change its price to consumers and so that, in turn, the consumers’ switching conduct may not constrain the monopolist’s pricing. The OFT contrasted this with the case of a manufacturer of car parts who increases the unit price of parts to car manufacturers, which will increase the latters’ marginal costs and lead to an increase in the price of cars. This may, in turn, lead to a fall in the demand for cars and (in consequence) a demonstration that the increase in the unit price of the parts was not profitable: final consumer price sensitivity may therefore constrain the price of car parts.[141]If, in the circumstances of the present case, therefore, the licence fee for the Non-LBO rights was not independent of the distributors’ turnover, the latters’ marginal cost would (on the OFT’s analysis) be affected by the consumers’ switching conduct in the downstream market and would be expected to constrain the hypothetical monopolist’s pricing of those rights. That would suggest that the market would be at least so wide as to comprise the licensing of both Non-LBO and LBO rights. We understand that, as regards market definition, it is therefore essential to the OFT’s analysis that the competitive counterfactual is one in which the Non-LBO rights are licensed for fees which are either fixed or predominantly fixed.[142]The problem with this is that it exposes an internal inconsistency in the OFT’s reasoning. The OFT’s Decision was that the collective selling by the 49 Courses had not only led to an increase in the price payable by ATR for their rights, it had also resulted in an agreement which restricted incentives within the relevant market for the Courses to improve the quality and nature of their output. The OFT found that, absent collective selling, such incentives would exist. By the time of the argument before us, the OFT’s preferred counterfactual was one in which (as the OFT’s skeleton argument put it) the Courses sold their rights to ATR “under agreements under which a high proportion of the payments for the [Non-LBO] rights was payable under a revenue 62 split formula.” This amounted to an assertion that the appropriate counterfactual is not one under which the price for the rights is predominantly represented by fixed fees. It follows that, for the purposes of their Decision, the OFT applied the SSNIP test to a counterfactual that, for the purposes of these appeals, it now disclaims.[143]The OFT was sensitive to this difficulty and sought to overcome it by submitting that the SSNIP test could and should take account of the effects not only of an increase in price but also of a change in the pricing structure from variable to fixed fees. It submitted that, unless this was done, there would be a risk of underestimating the market power of the Courses. We do not accept that this approach is correct. If the pricing structure in the appropriate counterfactual situation makes a conventional SSNIP test inappropriate, that test cannot properly be used as a primary indicator of the relevant economic market. The OFT’s argument amounted to an inappropriate attempt to use the alleged market power of the Courses in order to define the relevant market. That is to put the cart before the horse. The market is not defined by reference to the supposed power of the players in it. It is only after it has been defined that it is possible to assess the players’ market power.[144]On the basis of the OFT’s suggested counterfactual, we are not, therefore, satisfied that the OFT established that the Non-LBO rights constituted a distinct economic product market.[145]We consider later in this judgment the question of whether it would in practice have been possible (as the OFT has asserted) for the Courses to have sold their rights to ATR otherwise than by collective selling or negotiation. Assuming it was, we are also of the view that the most plausible competitive counterfactual would have been one in which the licence fee structure was predominantly fixed, as under the MRA. For completeness, we now consider the correctness of the OFT’s economic analysis on that assumption. Even on this assumption, we consider there are flaws in the reasoning that led the OFT to its conclusion on market definition.[146]First, in our summary of the OFT’s Decision we quoted footnote 57 to paragraph 93. A point there made is that if the higher fee results in “some distributors” exiting the market because their fixed costs are higher, such exit may 63 lead to an increase in the price paid by final consumers. We have concerns as to the relevance of this to the present case, in which the primary question was rather whether ATR would enter the relevant market and introduce an entirely new product. Moreover, in paragraph 93, and having expressed the view that an increase in the monopolist’s fixed fee to distributors does not increase the distributor’s marginal cost of production, the OFT continued by saying: “Thus an increase in the monopolist’s (fixed) licence fee may not affect the final price at which the product is offered to consumers [there is then the reference to footnote 57]. Accordingly, ultimate consumers’ price sensitivity (i.e. their switching conduct in response to a price rise) may not constrain the monopolist’s pricing. Rather, it is the licensees’ ability to recover the licence fee that constrains the monopolist’s prices. A bookmaker covers the cost of the licence fee it pays for pictures from the profits generated from the additional demand those pictures stimulate.” (Our italics)[147]This is the assessment upon which the OFT relied in order to satisfy themselves on the central issue of whether or not the otherwise expected link between substitutability on the downstream and upstream markets is broken. In our view, its negative conclusion on this cannot be justified by an assessment expressed in terms of what “may not” have been the economic consequences of the matters to which they here referred.[148]For this part of its analysis the OFT assumed that the price payable by ATR was such that it could expect to earn supra-normal profits (see paragraph 98 above). Later in this judgment (in the section headed “Was the OFT entitled to find that the MRA had the effect of increasing prices?”) we consider whether the OFT has sufficiently demonstrated that the price payable was above the competitive price and we conclude that it has failed to do so. Accordingly, the OFT’s findings as to these matters were, in our view, insufficiently solidly based to justify the conclusion that it built upon them.[149]Secondly, the OFT’s starting point for its analysis of the relevant product in this case was that the Non-LBO rights were a separate product. It gave no consideration to whether they were complementary with other products that ATR was acquiring simultaneously, in particular the BHB data rights. According to paragraph 64 5.4 of the OFT guideline on market definition applicable at the time of the Decision (the March 1999 edition) “Complements are included in the same market when competition to supply one product constrains the price for the other.” The OFT did not consider the point but we consider that the BHB data rights and the Non-LBO rights must be complementary, since the principal buyers of the relevant rights – the LBOs, as well as the providers of internet and iDTV betting services – required licences for both. If so, any increase in the amount a licensee has to pay for the Non-LBO rights is likely to be matched by a decrease in the amount it is willing to pay for the data rights. This was recognised by both Carlton and ATR in their bids for the Non-LBO rights. Carlton budgeted for a maximum charge of £1.5m p.a. for the BHB rights. ATR stipulated that the courses and/or the RCA would ensure that the BHB data rights were made available to ATR “free of charge”. Moreover, it is not possible to identify the price paid by ATR for the Non-LBO rights alone because the component of the total price referable to the BHB data rights cannot be isolated. The OFT should have considered this in coming to their Decision but did not.[150]We conclude, therefore, that the OFT was in error in identifying the narrow product market that it did for the Non-LBO rights. Our rejection of that market as the relevant one would not, by itself, be fatal to the Decision if (assuming all else against the appellants) we were satisfied that the collective selling or negotiation by the 49 Courses had an appreciable anti-competitive effect within a broader relevant market. The OFT did not, however, consider in its Decision or in its submissions to us what the position would have been on the assumption that the relevant product market was wider than the market merely for Non-LBO rights. We are, therefore, in no position to form any view on that subject. Our conclusion on this aspect of the appeals means that they must be allowed and the OFT’s Decision set aside. In case, however, we are wrong in this conclusion, we deal also with some at least of the further issues which arose. We do so on the assumption, contrary to our holding, that the OFT identified the correct relevant product market. 3. Did the MRA infringe the Chapter I prohibition: introductory[151]We first outline certain principles that we did not understand to be in dispute. First, the Chapter I prohibition applies to agreements or concerted practices having as 65 their “object or effect” the prevention, restriction or distortion of competition. These are alternative, not cumulative, requirements (Case 56/65 , Société Technique Minière v. Maschinenbau Ulm GmbH [1996] ECR 235). The OFT made no finding that the MRA had as its “object” any anti-competitive restrictions. It found only that it had an anti-competitive “effect.”[152]Second, there is no presumption that an agreement has any actual or potential anti-competitive effect so as to infringe the Chapter I prohibition. What must be shown, with a reasonable degree of probability, is that it has an appreciable effect, whether actual or potential, on competition within the United Kingdom. There is, for example, unlikely to be such an effect where the parties have a small market share; or where, by way of further example, two competitors, each with a large market share, form a joint venture to develop a new product which neither has the resources to develop on its own.[153]Third, in considering whether an agreement has an appreciable effect on competition, the ECJ said in Société Technique Minière:
“The competition in question must be understood within the actual context in which it would occur in the absence of the agreement in dispute. In particular it may be doubted whether there is an interference with competition if the said agreement seems really necessary for the penetration of a new area by an undertaking.”
This principle has been incorporated as follows into the Commission’s Guidelines on the application of Article 81(3):
“The assessment of whether an agreement is restrictive of competition must be made within the actual context in which competition would occur in the absence of the agreement with the alleged restrictions.”
The practical effect of this guideline is that, in the present context, the effect of the MRA has to be compared with that which would have prevailed had it not been entered into, an exercise requiring an assessment of the competitive landscape that would exist in its absence (“the counterfactual”), but within the context of the market as it was at the time of the MRA. 66[154]Fourth, collective selling by undertakings not in competition with each other does not restrict competition. This is recognised by paragraphs 24 and 143 of the Commission’s Horizontal Guidelines as to agreements which will fall outside the scope of Article 81(1): “24. Some categories of agreements do not fall under Article 81(1) because of their very nature. This is normally true for cooperation that does not imply a coordination of the parties’ competitive behaviour in the market such as … cooperation between non-competitors … [and/or] … cooperation between competing companies that cannot independently carry out the project or activity. … 143. If the parties clearly do not compete with regard to the products or services covered by the agreement the agreement cannot create competition problems of a horizontal nature … This also applies if a cooperation in commercialisation is objectively necessary to allow one party to enter a market it could not have entered individually, for example because of costs involved. Observations to like effect are to be found in European Community Law of Competition , Bellamy & Child, 5 th Ed, paragraph 4-097.[155]Fifth, an agreement is not restrictive of competition if it is necessary to create and/or operate a new service. Paragraph 87 of the same Horizontal Guidelines provides:
“… cooperation between firms which compete on markets closely related to the market directly concerned by the cooperation, cannot be defined as restricting competition, if the cooperation is the only commercially justifiable possible way to enter a new market, to launch a new product or service or to carry out a specific project.”
[156]The RCA appellants extract from all this three statements of principle which they say are directly applicable to the present appeal, being principles we do not understand to be in dispute. They are:(a) collective selling by undertakings not in competition with each other does not infringe section 2 of the Act; 67(b) an agreement is not restrictive of competition if it is necessary to create a new product; and(c) in the case of restriction of competition by effect, there is no presumption of anti-competitive effect. We now consider whether and, if so, how each of those principles applies to the present case. 4. Did the MRA amount to “collective selling”?[157]The OFT’s Decision was that the conclusion of the MRA, negotiated for the Courses by the RCA, amounted to an agreement or concerted practice within section 2, which the OFT characterised as “collective selling”. Its reasons were that:(i) when each Course signed, it knew that each other Course that signed would be doing so on the same terms;(ii) the licence fee distribution formula in Schedule 14 to the MRA had been arrived at by the agency of a Distribution Group of Courses, inviting the inference that there had been discussion between the Courses about pricing and fund distribution;(iii) each Course knew that the RCA was negotiating the MRA on its behalf, although it was not committed to signing it, which also invited the inference that there was contact between the Courses;(iv) the applicants to the notification had said that one reason for the centrally negotiated agreement was to realise the “full value” of the Courses’ rights; and(v) the applicants had also referred to the “principle of solidarity”, which again implied collective action.[158]Whilst the RCA appellants accept that the MRA was centrally negotiated for all 59 courses by the RCA, they dispute that the Courses sold their rights collectively pursuant to an agreement or a concerted practice. They submitted that none of the five points relied upon by the OFT proved, either directly or by inference, that the 59 courses (or any of them) had agreed between themselves (even informally) that they would only sell their rights collectively or that they had adopted a “concerted practice” to that end. All that the evidence shows is that the 59 courses authorised the RCA to negotiate the terms of a collective sale of their rights to the various bidders. They did not, however, authorise the RCA to commit them to any sale so negotiated; 68 there is no evidence that any of the courses was otherwise committed to signing up to any such sale; and the evidence shows they were not so committed. Ten courses decided to sell their rights to GG Media and there is no evidence supporting a finding that any of the other 49 had committed itself to signing the MRA. Any of them could have decided not to sign at all, albeit that that was improbable.[159]The BHB did not support this aspect of the RCA appellants’ case and we do not accept the submission. Section 2(1) refers to “agreements between undertakings, decisions by associations of undertakings or concerted practices …”. We accept the OFT’s submission that these three concepts are not clearly distinct, that there is an element of overlap between them and that the ECJ has adopted a broad interpretation of them: in particular, it is not necessary for arrangements (whether agreements, decisions or concerted practices) to be contractually binding for them to infringe the competition rules. In his oral submissions, counsel for the OFT placed his main focus on the argument that there was a concerted practice between the 49 signatory Courses who were parties to the MRA and he pointed in particular to the arrangement between the courses arranged by the RCA in November 2000 under which exclusivity was given to ATR between then and May 2001, when the MRA was concluded. We agree that the MRA was the fruit of a collective negotiation conducted by the RCA on behalf of all the racecourses by the RCA and we agree also that it could properly be characterised as the result of a concerted action by the 49 signatory Courses. We therefore at least agree with the OFT’s conclusion in paragraph 215 of its Decision that “the 49 Courses’ conclusion of the [MRA], negotiated on their behalf by the RCA, amounts to … [a] concerted practice between those Courses within the meaning of the Chapter I prohibition ….” 5. Was the collective selling, or negotiation, “necessary” for the creation of the new product the rights for which the Courses were selling?[160]The RCA appellants submitted that collective selling, or negotiation, was anyway necessary for the creation of the new product that ATR was proposing to establish, its creation being a legitimate objective in which they and ATR were jointly engaging for their respective benefit and for that of British racing generally. If so, they submitted that there was no infringement of the Chapter I prohibition. They said 69 the OFT was wrong to reject this argument in its Decision. The BHB agreed with and supported the RCA appellants’ argument in this respect, to which we now refer more fully.[161]Counsel for the RCA appellants referred to three authorities. First, Gøttrup-Klim (cited above), a decision of the ECJ made on a reference by a Danish court. It concerned two co-operatives, DLG and LAG. DLG had existed since 1969 and its object was to provide its members with pesticides at the lowest prices. Its members included “B” members, including the plaintiffs, who were entitled to take part in its management. In 1975, the plaintiffs formed a separate co-operative association, LAG, which specialised in the distribution of farm supplies. This led to competition between LAG and DLG, which led in turn to an amendment by DLG of its statutes so as to exclude from its membership anyone who was a member of LAG. The plaintiffs challenged the validity of the rule change. One question raised was whether DLG’s rule change, whose effect was to prevent its members from co-operating in other forms of organised co-operation which were in direct competition with it, was outlawed by Article 85(1) (now Article 81(1)) of the EC Treaty. It was held that it was not. The ECJ said, at paragraph 28, that the question could not be answered “in the abstract” but depended on “the particular clauses in the statutes and the economic conditions prevailing on the markets concerned.” The court said:
“33. Where some members of two competing cooperative purchasing associations belong to both at the same time, the result is to make each association less capable of pursuing its objectives for the benefit of the rest of its members, especially where the members concerned, as in the case in point, are themselves cooperative associations with a large number of individual members. 34. It follows that such dual membership would jeopardize both the proper functioning of the cooperative and its contractual power in relation to producers. Prohibition of dual membership does not, therefore, necessarily constitute a restriction of competition within the meaning of Article 85(1) of the Treaty and may even have beneficial effects on competition. 35. Nevertheless, a provision in the statutes of a cooperative purchasing association, restricting the opportunity for members to join other types of competing cooperatives and thus discouraging them from obtaining supplies elsewhere, may have adverse effects on competition. So, in order to escape the prohibition laid down in Article 85(1) of the Treaty, the restrictions imposed on members by the statutes of cooperative purchasing associations must be 70 limited to what is necessary to ensure that the cooperative functions properly and maintains its contractual power in relation to producers.”
[162]The ECJ then referred to certain features of DLG’s statutes and continued:
“40. Taking all those factors into account, it would not seem that restrictions laid down in the statutes, of the kind imposed on DLG members, go beyond what is necessary to ensure that the cooperative functions properly and maintains its contractual power in relation to products. 41. As regards the penalties imposed on the plaintiffs as a result of their exclusion for infringing DLG’s rules, these would not appear to be disproportionate, since DLG has treated the plaintiffs as if they were members exercising their right to withdraw. 42. So far as concerns the membership period, this has been reduced from ten to five years, which does not seem unreasonable. 43. It is significant, in the last analysis, that after their exclusion, the plaintiffs succeeded, through LAG, in competing vigorously with DLG, with the result that in 1990 their market share was similar to DLG’s. … 45. The answer to the second set of questions referred by the national court must therefore be that a provision in the statutes of a cooperative purchasing association, forbidding its members to participate in other forms of organized cooperation which are in direct competition with it, is not caught by the prohibition in Article 85(1) of the Treaty, so long as the abovementioned provision is restricted to what is necessary to ensure that the cooperative functions properly and maintains its contractual power in relation to producers.”
[163]The next case in time to which we were referred was Métropole télévision v. Commission [2001] ECR II-2459 , a decision of the Court of First Instance. The applicants there challenged the Commission’s finding of an infringement of the former Article 85(1), arguing that it was necessary to weigh up the pro- and anti-competitive effects of an agreement in order to determine whether it was caught by the prohibition of the article - an application of the “rule of reason”. In paragraph 72, the court said that both the CFI and the ECJ had “been at pains to indicate that the existence of a rule of reason in Community competition law is doubtful.” In paragraph 74, the court explained that Article 85(3) provided the possibility for an exemption for an agreement that restricted competition and that “it is only in the precise framework of that provision that the pro- and anti-competitive aspects of a restriction may be weighed.” The court referred to various authorities said to support that, adding that 71 “Article 85(3) of the Treaty would lose much of its effectiveness if such an examination had to be carried out already under Article 85(1) of the Treaty.”[164]That approach could not, however, obviously be reconciled with, for example, the Gøttrup-Klim case. The court recognised that and sought to explain the position as follows:
“75. It is true that in a number of judgments the Court of Justice and the Court of First Instance have favoured a more flexible interpretation of the prohibition laid down in Article 85(1) of the Treaty [and it referred to various authorities, including Gøttrup-Klim]. 76. Those judgments cannot, however, be interpreted as establishing the existence of a rule of reason in Community competition law. They are, rather, part of a broader trend in the case-law according to which it is not necessary to hold, wholly abstractly and without drawing any distinction, that any agreement restricting the freedom of action of one or more of the parties is necessarily caught by the prohibition laid down in Article 85(1) of the Treaty. In assessing the applicability of Article 85(1) to an agreement, account should be taken of the actual conditions in which it functions, in particular the economic context in which the undertakings operate, the products or services covered by the agreement and actual structure of the market concerned [it then referred to various authorities, not including Gøttrup-Klim]. 77. That interpretation, while observing the substantive scheme of Article 85 of the Treaty and, in particular, preserving the effectiveness of Article 85(3), makes it possible to prevent the prohibition in Article 85(1) from extending wholly abstractly and without distinction to all agreements whose effect is to restrict the freedom of action of one or more of the parties. It must, however, be emphasised that such an approach does not mean that it is necessary to weigh the pro and anti-competitive effects of an agreement when determining whether the prohibition laid down in Article 85(1) of the Treaty applies.”
[165]Finally, we were referred to the later case of Wouters v. Algemene Raad van de Nederlandse Orde van Advocaten [2002] ECR I-1577 , a decision of the ECJ. The issue was whether a regulation of the Dutch Bar preventing its members from forming a partnership with an accountant infringed the former Article 85(1) as being anti-competitive. The appellants’ case was that multi-disciplinary partnerships of members of the Bar and accountants enabled a better response to the needs of clients. The court’s conclusion, at paragraph 86, was that the national legislation imposing the prohibition “has adverse effect on competition and may affect trade between Member States.” They said much the same in paragraphs 93 and 94. However, as in Gøttrup 72 Klim , they concluded that the restriction in the regulation did not in fact infringe the Article 85(1) prohibition, saying:
“97. However, not every agreement between undertakings or every decision of an association of undertakings which restricts the freedom of action of the parties or of one of them necessarily falls within the prohibition laid down in Article 85(1) of the Treaty. For the purposes of application of that provision to a particular case, account must first of all be taken of the overall context in which the decision of the association of undertakings was taken or produces its effects. More particularly, account must be taken of its objectives, which are here connected with the need to make rules relating to organisation, qualifications, professional ethics, supervision and liability, in order to ensure that the ultimate consumers of legal services and the sound administration of justice are provided with the necessary guarantees in relation to integrity and experience (see, to that effect, Case C-3/95 Reiseburo Broede [1996] ECR I-6511 , paragraph 38). It has then to be considered whether the consequential effects restrictive of competition are inherent in the pursuit of those objectives.”
[166]The court then focused more specifically on the relevant regulation, and included the following observations in its conclusions: “105. The aim of the 1993 Regulation is therefore to ensure that, in the Member State concerned, the rules of professional conduct for members of the Bar are complied with, having regard to the prevailing perceptions of the profession in that State. The Bar of the Netherlands was entitled to consider that members of the Bar might no longer be in a position to advise and represent their clients independently and in the observance of strict professional secrecy if they belonged to an organisation which is also responsible for producing an account of the financial results of the transactions in respect of which their services were called upon and for certifying those accounts. …107. A regulation such as the 1993 Regulation could therefore reasonably be considered to be necessary in order to ensure the proper practice of the legal profession, as it is organised in the Member State concerned.108. Furthermore, the fact that different rules may be applicable in another Member State does not mean that the rules in force in the former State are incompatible with Community Law …109. In light of those considerations, it does not appear that the effects restrictive of competition such as those resulting for members of the Bar practising in the Netherlands from a regulation such as the 1993 Regulation go beyond what is necessary in order to ensure the proper practice of the legal profession (see, to that effect, Case C-250/92 DLG [1994] ECR I-5641 , paragraph 35 [the Gøttrup Klim case] ).” 73[167]We confess to some difficulty in reconciling the approach of the ECJ in Gøttrup-Klim and Wouters with that of the CFI in Métropole, but find it unnecessary to dwell on the explanation in Métropole as to the rationale that the CFI perceived as underlying cases such as Gøttrup-Klim and Wouters (the latter of course being decided after Métropole). We consider that these two decisions of the ECJ show that the assessment of whether or not a particular arrangement constitutes an infringement of Article 85(1) (now Article 81(1)), or therefore of the Chapter I prohibition, is a rather more flexible exercise than the CFI was perhaps willing to appreciate. It is not enough that the arrangement is apparently anti-competitive, as in Gøttrup-Klim and Wouters. What those cases show is that ostensibly restrictive arrangements which are necessary to achieve a proper commercial objective will not, or may not, constitute an anti-competitive infringement at all. Whether or not they will do so requires an objective analysis of the particular arrangement entered into by the parties, assessed by reference to their subjective “wants” and against the evidence of the particular market in which they made their arrangement. The task then is to consider whether the restrictive arrangement of which complaint is made is “necessary” to achieve the objective. The RCA appellants also submitted that the concept of “necessity” in this context is not an absolute one, but has an element of flexibility about it, for which they referred us to paragraph 109 in the Métropole case in which the court observed that “If, without the restriction, the main operation is difficult or even impossible to implement, the restriction may be regarded as objectively necessary for its implementation.” We also accept this last submission: competition law is not an area of law in which there is much scope for absolute concepts or sharp edges.[168]The RCA appellants said that the present case falls squarely within this concept of “necessity”. ATR entered into the MRA in order to create a wholly untried and innovative product - the channel and linked interactive website. The Courses entered into the MRA in order to sell ATR the rights which would enable ATR to create and exploit that new product and so in turn to generate a new source of much needed income for the Courses with a view to replacing the income drought with which they were faced by the proposed abolition of the levy. The creation of that new product was a legitimate commercial objective from everyone’s point of view: 74 horseracing generally, the punters and the parties to the MRA. The new channel was to be financed principally from betting income and so ATR needed, so far as possible, to have uninterrupted race coverage on each day of each week of each year in order to maximise its betting income. That need required it in turn to seek to acquire the interactive rights from as many racecourses as possible. It did not need (albeit that it would have liked) to acquire the interactive rights of all 59 courses, but it did at least need to acquire a “critical mass” of such rights in order for the new venture to be viable. There is no agreement between the parties as to what ATR’s critical mass was, but the OFT now accepts that ATR (or any other bidder) required at least a majority of the rights of the 59 courses by reference to betting turnover and that therefore there could only be one successful bidder for the rights.[169]Save for the quantum of the necessary critical mass, all this was common ground. The point of difference was the next step. Whereas the RCA and the Courses claimed that, from a practical point of view, collective selling, or negotiation, was the only realistic way to achieve a sale and purchase of the rights (being rights which had never been dealt with in the market before), the OFT found that the buyers could have assembled the rights themselves by individual negotiation with the course owners or with small groups of courses. In practice, they would have had to negotiate with up to about 37 separate course owners. It made the point that any such separate contracts so negotiated could have been made conditional on obtaining sufficient rights from other courses “(as in fact occurred in the Notified Arrangement, itself a conditional agreement)”. It also said that BSkyB and Channel 4 had considerable experience of assembling packages of rights necessary to launch channels.[170]The suggestion that the acquisition of the necessary critical mass by individual negotiation with up to 37 course owners either could have been done, might have been done, or was ever even contemplated as something which could or might have been done, appears to us to represent a triumph of theory over commercial reality and to ignore the evidence of the events leading up to the MRA. We have summarised the course of the negotiations fairly fully and do not refer to those details again here. We regard it as apparent from the totality of the negotiations that Channel 4, Arena and ATR (whom we lump together for this purpose, since ATR simply continued the negotiations earlier started by Channel 4 and Arena) - and also Carlton - had one 75 major objective in mind. That was to acquire the rights of as many courses as they could and, with a view to achieving that end, to deal with the courses as a whole. There is no evidence that, at the time of the negotiations, any of the bidders ever had it in mind to pick off the courses individually or in small groups and the notion that they might have tried to do so appears to be unrealistic. We do not overlook the staged acquisition operation that Arena proposed in its discussion document in January 2000, but record that it then took no steps towards carrying it out. Since there could and would be no deal with any courses until the necessary critical mass had been tied up, each individual deal would have had to be made conditional upon the signing of other courses until such time as the critical mass had been achieved. This could have been done in theory. In practice, the bidders never sought to do it, for the obvious reason that so to have proceeded would, in practice, have been extremely difficult if not impossible. The OFT’s point that the availability of this route was in part illustrated by the fact that the MRA was itself a conditional agreement was, in our view, a false one. The MRA reflected the fruit of the centrally negotiated agreement and its stated conditions were in substance no more than a public statement by ATR to all 59 courses to the effect that the MRA would only bite as a matter of contract if enough courses signed up to it. A conditional agreement of that sort was quite different from that which would have been required if ATR had sought to negotiate separate deals with up to 37 different owners. We do not know what conditions the OFT had in mind as those to which the contracts might have been subject: but if they were, for example, equivalent to those in clause 2.2 in the MRA, then that would appear to present a good many courses further down the queue with an obvious veto on the self-assembly exercise ever being consummated except at prices dictated by them. We also consider that the OFT’s point that the ATR consortium had experience in assembling rights from different suppliers ignored the fact that the courses’ interactive rights were being sought for the creation of a wholly new product: we are unaware that there was any evidence that such earlier self-assembly experience involved the self-assembly of a particular critical mass of rights.[171]In our view, therefore, an acquisition via a central negotiation was the only realistic way forward both from the viewpoint of both bidder and sellers and we regard it as probable that any initial attempt at a self-assembly exercise via individual negotiations would have led quickly to a centrally negotiated one. This is illustrated 76 by the failure of Channel 4’s attempt first to sign up with the Super 12 and then with the 47. Put another way, we consider that the weight of the evidence before the OFT, being evidence as to what actually happened, showed that collective negotiation was the necessary way forward, and that is the way that was in fact adopted.[172]In so concluding, we are saying nothing new. That always was, and remains, the view of the Courses, the RCA and the BHB. Once upon a time it was also the view of ATR, which expressly agreed in the notification that:
“It was necessary for the Rights to be sold pursuant to a centrally negotiated agreement in order to put together a package of rights sufficient to be attractive to purchasers and to allow the radical move away from the current method of funding British Racing (i.e. the Levy plus limited commercial revenues). The involvement of a significant number of courses was necessary in order to achieve an efficient sale of the Rights and the necessary ‘critical mass to make [ATR’s] product offering feasible.”
The point that ATR was there acknowledging – and asserting - was that it needed a critical mass of rights and therefore engaged in the one commercially obvious – and necessary – way to achieve it, which was by a central negotiation. Of course, ATR much later decided that its own commercial interests lay in a repudiation of the MRA and the OFT chose to prefer ATR’s later evidence produced in August 2003 to the effect that what they had there said in the notification was (in effect) a mistake and that in reality, contrary to that assertion, central negotiation had not been necessary at all. That self-serving evidence was a self-contradiction and was unsupported by ATR’s actions at the time of the negotiations. It is also contrary to what was said in a briefing paper submitted by the RCA to the OFT in March 2003, one which was first endorsed by ATR and paragraph 3.9 of which confirmed that ATR was only prepared to negotiate through a central entity and would not have been willing to deal separately with up to 38 sellers: had ATR been faced with a need to do so, “it is likely that they would not have proceeded”. ATR made an express reference to paragraph 3.9, with apparent approval of its contents, in its letter of 25 July 2003 to the OFT. It is also of note that Arena (an ATR company), in response to section 26 notices served by the OFT, made the point that whilst certain of the larger courses would have had the resources to sell their interactive rights individually, the typical course would not, since as it would hold few, if any, fixtures of any intrinsic value to a broadcaster, it 77 could not make a commercially attractive offer to a broadcaster. Arena’s view was that (the largest courses apart), a typical course might not therefore be able to sell any interactive rights at all unless it were to do so as part of some wider negotiating process under which the buyer requires a larger package of rights of interest to both parties. Its evidence was that for most courses the sale of rights individually was simply not a viable option for either buyer or seller. The OFT appears to have ignored this evidence in its Decision.[173]No bidder went down the self-assembly route in this case, although there was, in practice, nothing to stop them doing so. They were free at any time to approach any racecourse and make it an offer. We presume that the reason for the adopted method of acquisition is because no bidder regarded anything other than central negotiation as offering a practicable way forward.[174]We record that the OFT placed reliance upon the decision of the Commission in the UEFA Champions League OJ [2003] L291/25 as providing a pointer against our conclusion on the necessity argument. We consider that counsel for the RCA appellants was correct in his response that in that case the argument failed on the facts identified in paragraph 131 of the decision.[175]We conclude, therefore, that the central negotiation in which the Courses engaged was necessary for the achievement both by them and by ATR of the legitimate commercial objective of creating the new product that ATR proposed to exploit for the benefit of itself, the punters, the racecourses and racing generally. In our view, the evidence pointed to that conclusion and there was no reliable evidence supporting the different view that the OFT preferred, which appears to us to have been founded in theory rather than reality. We conclude that the MRA involved no infringement of the Chapter I prohibition.[176]In coming to its different conclusion on this, the OFT also relied upon its assessment that no course, or group of courses, had veto rights on the acquisition by ATR of the necessary critical mass. For reasons we give when dealing below with the issue as to whether the negotiated sale resulted in an increase of prices, we also disagree with the OFT on the question of veto rights. 78 6. Did the MRA have an anti-competitive effect?[177]Given our previous conclusions, we do not regard this question as requiring decision, but will express our views on it. We start by reiterating that in the case of restriction of competition by effect, which is all the OFT contends for, there is no presumption of anti-competitive effect. Was the OFT entitled to find the MRA had the effect of increasing prices?[178]The RCA appellants’ case to the OFT was reflected in paragraphs 290 and 292 of the Decision, as follows:
“290. The RCA stated that there is no evidence that prices are higher as a result of collective selling. It stated that the OFT has not explained or demonstrated what the price would have been if there had been individual negotiation. Further, the RCA considered that the net cost of the Non-LBO Bookmaking Rights is not high compared to [ATR’s] projected revenue or costs. Neither [ATR] nor rival bidder Carlton increased their bids during negotiations, and neither of these companies complained that the price of the Courses’ rights was increased as a consequence of the way they were sold. The RCA stated that buyers were only interested in negotiating with a single body and were not interested in playing off the Courses against each other … 292. Racecourse group RHT stated that the manner in which the Courses sold their rights did not increase the price paid by [ATR]. First, the price paid by [ATR] was lower than an estimate of the value of the Rights made by Arena in July 2000. Second, 12 major racecourses (the ‘Super 12 Courses’) received a similar amount under the Rights Agreement as a previous offer they had received for their Rights.”
[179]The OFT rejected this as “not persuasive.” Its counter-assertion was that “the Courses collectively could exercise the market power that they did not hold individually.” The RCA appellants submitted that this was no sufficient answer to their evidential case and ignored the veto point: namely, that certain course owners, or groups of courses, held the key to the critical mass that ATR needed to assemble, with the commercial consequence that collective negotiation could not result in a higher price than individual negotiation. The economic theory underlying the veto point is not in dispute, but the OFT rejected the existence of any veto on the facts. It also 79 rejected as irrelevant the point that neither ATR nor Carlton increased the price of their bids. The RCA appellants submitted that a fundamental error in the OFT’s case was that the OFT was unable to provide their own estimate of what a “competitive price” for the Non-LBO bookmaking rights would be. That inability is admitted, the difficulty being that this was the first time that such rights had been sold by any racecourses, so posing an obvious problem to the OFT in proving that they were sold to ATR at an inflated price. But, the RCA appellants claimed, if the OFT’s case was that the collective selling had the “effect” of increasing prices, it had to prove it.[180]One point the OFT relied upon was the fact that the Arena offer in July 2000 for the non-terrestrial rights was £178m compared with ATR’s later offer of £195m. But the RCA appellants pointed out that the OFT failed to recognise that this apparent increase was simply a reallocation of £15m from the terrestrial payments to the non-terrestrial payment: the reduction of one broadly matched the increase in the other, a stipulation introduced by ATR (not the courses) so as to raise the threshold at which the courses would start to share in top up revenue. After the reallocation, the global amount of the ATR offer remained the same and was less than that offered by Carlton. Another point on pricing made by the OFT was that the (later) ATR offer included a commitment on marketing which the (earlier) Arena offer did not. But the RCA appellants pointed out that this was not by way of a payment to the courses and cannot be regarded as a payment for the Non-LBO rights. The marketing “spend” was for marketing and sponsorship opportunities that would otherwise have been sold by the courses to third parties, it was therefore spent for value received by the courses and cannot be brought into a price comparison with the Arena offer.[181]The OFT also rejected the point that the bidders were not interested in playing one course off against another, relying primarily on ATR’s August 2003 assertion to the contrary effect. By then, however, ATR had decided to depart from its different stance expressed in the notification, to which we have referred. The RCA appellants said that the OFT should have regarded ATR’s change of attitude with considerable circumspection, and contrasted its August 2003 stance with its earlier endorsement of the March 2003 submission to the OFT that (in paragraph 3.7) “It is the parties’ view that alternative methods of selling rights (in particular individual negotiations between Courses and ATR), in the particular circumstances of the case, would not have been 80 able to deliver the benefits to which the Notified Arrangement gives rise.” They also say that it was for the OFT to prove its assertion that the buyers were interested in playing one course off against another; and they said that, far from there being any such evidence, the evidence was to the effect that all the buyers, in particular ATR, wanted a one-package deal that included all the courses.[182]In paragraphs 299 and 300 of the Decision, the OFT referred to the Channel 4 offer to the Super 12 in February 2000 and to the letter of 6 March 2000 from Mr Crichton-Miller (the then chief executive of the RCA) to the chairman of the independent courses (i.e. courses other than the Super 12, or courses owned otherwise than a by Super 12 course owner) and a letter from the managing director of Ripon racecourse to Mr Savill (the BHB chairman) of the same date. The OFT asserted that the comments in those documents implied that the parties believed that collective selling by the courses would achieve a higher price. The OFT also relied on a file note dated 27 April 2000 of an RCA meeting produced by Cartmel racecourse which made mention of the “strength in the 42 courses remaining as one group or better still 59 courses” and the notifying parties’ comment in their Form N that central negotiations allowed the courses to “[realise] the full value of their rights.” Counsel for the RCA appellants submitted that these statements needed to be read in the context of the importance to the courses at that stage of achieving a deal: they were not directed to increasing the price paid by any successful bidder, but reflected an aspiration to ensure that a deal took place at all. Further, he said that, if the documents carry the meaning the OFT attributed to them, they proved at best that the collective selling had the object of increasing prices. But the OFT made no finding that the courses had any such object , its Decision focused solely on the OFT’s claim that the MRA had the effect of increasing prices. The OFT therefore had to produce evidence of effect. Counsel for the RCA also pointed out that in paragraph 387 the OFT recognised that the ten smaller courses which signed up with GG Media did so because they felt they could get a better deal, suggesting that the ATR price was not an inflated one.[183]We consider we have to approach this particular issue on the assumption (contrary to our earlier conclusion) that collective negotiation was not necessary to the signing of the MRA. Even so, we agree with the appellants that it was still for the OFT to prove a counterfactual situation in which ATR could, on the probabilities, 81 have acquired the Courses’ rights at prices appreciably lower than those reserved by the MRA. The OFT’s counterfactual was that ATR could have assembled its required critical mass of rights by separate negotiation with up to 37 course owners, or with small groups of them. The OFT’s stance is that, by acting collectively, the Courses were able to exercise market power they would not have had if they had acted individually; and that, in the absence of collective selling, it would have been in ATR’s interests to play the Courses off against each other. It asserted in the Rule 14 notice that:
“By negotiating as a bloc, via the RCA, the Courses hindered potential buyers striking deals with the individual courses or playing off the Courses against one another. For example, provided [ATR] acquired sufficient Rights, it could have credibly threatened not to purchase an individual Course’s rights. Such a threat would constrain the price that any course can charge. This threat is absent when courses negotiate en bloc. As a result, the total price of the rights will be higher.”
[184]The OFT further asserted that:
“If buyers assembled the necessary rights themselves, they could have made any contracts conditional on obtaining Rights from sufficient courses (as in fact occurred in the Notified Arrangement). Such conditional contracts would also have allowed [ATR] to assemble the necessary portfolio of rights prior to launch. Both BSkyB and Channel 4 produce TV channels. They therefore have considerable experience of assembling the packages of rights necessary to launch channels by negotiating with many suppliers.”
[185]The RCA appellants recognised that had ATR been faced with a situation in which it was faced with, say, ten equally sized groups of courses each of similar quality, collective negotiation and selling would be likely to lead to a higher price and to prevent ATR from playing the courses, or groups, off against each other. This, however, assumes that no one course or group was essential to the creation of the new ATR product and that each was perfectly substitutable by another. The RCA appellants asserted that the realities of the actual deal that ATR wanted were very different. Their case in the notification proceedings was that certain courses’ rights were essential to the viability of the ATR business. This meant that those courses had a veto on any deal with ATR. As the OFT accepted, ATR needed to obtain its required critical mass. To that end, the MRA was subject, by clause 2.2.2, to two key 82 conditions:(i) acceptance from all TRG courses (and any other courses which were part of groups including members of the TRG); and(ii) acceptance from courses accounting for 70% of UK off-course betting turnover. Their case was that, in practice, RHT had a veto on the ATR deal, as also did any TRG course.[186]The OFT accepted that ATR needed a critical mass, but disputed that any course or group had a veto. It dismissed RHT as a veto holder, because it held only 26.2% of off-course betting turnover, whereas 70% of such turnover was the condition in the MRA. It found that any one course, or group of courses, was at risk of not being included in the completed transaction. It disputed that the TRG courses had a veto on two grounds:(i) ATR had said in a letter of 7 August 2003 that it was not necessary for it to sign up all TRG courses and that the new channel would have been viable without some of them; and(ii) that, even though the TRG courses held many of the most popular races, it did not accept that (despite the terms of clause 2.2.2(i) of the MRA) each of them had a veto: for example, Ayr, only accounted for 2.1% of betting turnover. It also rejected the argument that the major courses had an effective veto because terrestrial coverage of British racing advertises and promotes the sport: it said these courses could sell their terrestrial rights to another broadcaster. It concluded that only groups of courses acting collectively could have vetoed the arrangement. The OFT’s conclusion was that the TRG courses could not have had a veto because certain of the TRG courses were not significant enough. It said in paragraph 279 of the Decision:
“While the OFT accepts that the TRG courses hold many of the best known and most popular races (and are the leading ‘brands’ in British racing), it does not accept that without the agreement of each of these courses no venture equivalent to “Attheraces” could be launched. For example, Ayr racecourse accounts for only 2.1% of betting turnover and only 2.2-2.4% of the viewers of British horseracing. Given the magnitude of these figures, the OFT does not accept that Ayr had a veto. Thus notwithstanding Clause 2.2.2(i) of the Rights Agreement, the OFT does not accept that each of the TRG courses (which include Ayr) held a veto.”
[187]The RCA appellants submitted there was no basis for that conclusion. It involved speculation as to what would have happened if Ayr had not signed; and even if ATR could and would have waived any non-signature by Ayr, no inference can be drawn that it would have waived any refusal by any other TRG course to sign. The 83 RCA appellants submitted that the OFT had failed to understand the importance of the TRG condition. Overall, they submitted that the OFT was wrong to conclude that there were no veto holding courses, a further reason for their conclusion on increased prices being wrong.[188]As for the 70% condition, whilst in theory the 70% threshold could have been satisfied without RHT, the RCA appellants said that RHT’s participation was in practice essential if ATR was to launch a viable product. Given that the GG Media courses accounted for 9.6% of turnover, the 70% threshold could not have been achieved without RHT. Even if the GG Media courses are regarded as having been in play, it would have been necessary to have achieved acceptances from almost all course owners other than RHT in order to achieve the 70% threshold. In addition, the larger courses had a preference for the ATR deal because it offered terrestrial television rights on Channel 4 in a way that the Carlton offer did not. RHT controlled seven of the Super 12, all TRG courses. Even if ATR had been prepared to drop TRG courses such as Ayr, it is inconceivable that ATR would allow the deal to proceed without the five Super 12 TRG courses that RHT did not control. The RCA appellants submitted that, in all these circumstances, the practical need to have RHT on side in order to achieve the 70% condition meant RHT had a veto. It is said that the OFT also failed to consider whether any course would have licensed its rights in circumstances in which RHT failed to do so. In addition, it is said that the OFT failed to take account of the importance of RHT in putting together sufficient terrestrial rights to offer packages for sub-licensing to Channel 4 and the BBC. Channel 4 had made it clear that if it did not get the rights to Cheltenham (an RHT course), it would pull out. As for the TRG condition, its effect was that ATR could refuse to complete if any of the TRG courses did not sign up. This gave each TRG course a veto, which could prevent the MRA from being concluded. Even if, in theory, ATR could have waived the non-satisfaction of these conditions, the RCA appellants said there was no evidence that they did not reflect the genuine preferences of ATR at the time of the MRA. There is no evidence, it is said, that the conditions would have been waived.[189]We come to our conclusion on whether the OFT has proved that the MRA had the effect of increasing prices. The OFT’s case is that the MRA has achieved an anti-competitive effect, because the price of the rights to ATR under the MRA is said to 84 have been higher than it would have been if there had been individual negotiation between the 49 Courses (or small groups of them) and ATR for the sale and purchase of the rights. It is said that the MRA price was higher than the competitive price. That being the OFT’s case, we consider that the OFT must show that it has proved it.[190]The first difficulty we have is that we are not satisfied that it is possible in the circumstances of this case to identify a price actually paid for the Non-LBO rights separately from the price also paid by ATR for the complementary rights necessary for the valuable exploitation of the non-LBO rights, in particular the BHB data rights: unless this can be done, then a comparison with the suggested counterfactual price necessarily becomes an inherently uncertain exercise. The second difficulty we regard the OFT as facing is, as the RCA appellants submitted, that it is incumbent on the OFT to identify the counterfactual situation by reference to which the competitive price is to be identified. This may not be difficult in a case in which the relevant product is already in existence and for which the competitive price can readily be identified. But that is not this case, a critical feature of which is that the rights had never previously been sold and were to be used by ATR to create an entirely new product. In addition, the OFT accepted by the time of the hearing before us that, at the material time, any bidder for the rights would require at least 50% of the betting turnover generated by the racecourses so that in practice there was likely to be only one successful bidder.[191]In these unusual circumstances, in order to prove the alleged anti-competitive effect of the MRA on the sale price of the rights, the OFT had to adduce convincing evidence establishing both:(i) that there was an alternative, less restrictive, way in which ATR could have concluded the purchase of the rights they needed to acquire, and;(ii) that that alternative process would have resulted in an appreciably overall lower price for the rights. Their case was that buyers would negotiate with individual courses (or with companies owning more than one course) either individually or in groups. It also considered that potential buyers might seek to play one course off against another.[192]As we have already indicated, we consider there was no satisfactory evidence that any of the bidders was ever interested in dealing with the 59 courses in this 85 manner. The bids were all made either to significant groups of courses (the Super 12 or the 42) or to all 59, which is unsurprising bearing in mind that the bidders’ ambitions were to acquire the rights of all 59 courses, or at least the necessary critical mass of such rights. No bidder ever appears to have had it in mind to deal with the course owners individually and the RCA appellants rightly criticised this suggested approach as fanciful. If the bidders would have approached each course, or small groups of courses, separately, what would they have offered them? How would they have allocated their intended overall investment between the courses as a whole? Would the offered price have been predominantly by way of minimum guarantees (as in the Channel 4 proposal to the Super 12)? Or would it have been substantially represented by a revenue sharing arrangement between bidder and seller? To what conditions would each separate deal have been made subject? Would this immediately have given other courses a veto? The difficulty we have with the OFT’s case on this aspect of its Decision is that the counterfactual by reference to which the competitive price is said to be identifiable is so imprecise. At the hearing, counsel for the OFT suggested an alternative, and perhaps rather more realistic, way in which the bidders might have acquired the courses’ rights. This was based on the Arena document of January 2000. Under that plan, Arena proposed that it would sign up with the Super 12, Arena (it had six courses of its own) and Northern on 1 February 2000, and would then negotiate and sign up with remaining courses on 1 March 2000, those courses negotiating via the RCA. The Super 12 had a betting revenue share of 34.5%, Arena 16.2%, Northern 10.6% and all other courses 38.7% (despite the size of its share, the OFT has not suggested that the co-ordinated behaviour of the Super 12 infringed section 2 of the Act). That submission amounted to the adoption of a counterfactual derived from a statement of intention by Arena, which Arena never implemented and upon which the OFT placed no reliance in its Decision. The OFT provided no analysis of the effect on competition of the concerted action in comparison with this counterfactual. That appears to us to be an unsound basis on which to found the OFT’s case as to what would, on the probabilities, have happened but for the subsequent collective negotiation conducted by the RCA.[193]A notional competitive price based on a hypothetical counterfactual which could, we consider, have no practical foundation in reality is no basis on which to found a conclusion that the MRA had an appreciably anti-competitive effect on the 86 price obtained for the Non-LBO rights. Moreover, whatever alternative purchasing strategy is hypothesised, the RCA appellants’ case was that if one or more racecourses, or groups of courses acting together, held a “veto” on a sale of the non-LBO rights (i.e. their participation in it was necessary to a purchaser) then separate negotiation of the purchase of the rights would, in theory, be expected to lead to the same overall price as collective selling. We agree with the principle of the theory and the OFT does not challenge it. The question of whether any course, or group of courses, had a relevant veto cannot, however, be considered independently of the question of what “critical mass” of rights a purchaser would need in order to make the project viable. For example, if a potential purchaser required rights of no less than 95% of betting turnover, any course with more than 5% of the rights could veto the purchase. Similarly, if any particular course was a “must have”, that course could also veto the purchase.[194]As to what “critical mass” was requisite for ATR’s purposes, ATR identified what it required by the preconditions it imposed in clause 2.2 of the MRA. Those conditions required the participation of:(i) all the TRG courses;(ii) all other courses owned by a TRG course owner; and(iii) courses which, overall, had between them at least 70% of the betting revenue. We accept that those conditions do not provide evidence of what might be regarded as the minimum rights which might objectively be needed for any interactive channel to be viable: that is likely to vary from bidder to bidder and will depend on the particular bidder’s business model. But we do not understand why they do not provide the best evidence of what ATR regarded as its minimum requisite critical mass if it was to be the acquirer of the rights: they were, after all, the contractual conditions that ATR required to be enshrined in the MRA.[195]The OFT’s final stance on this particular topic was that a critical mass of at least 50% of betting turnover was objectively necessary and that it was improbable that more than one commercially sustainable service could have been launched at the time of the negotiations with the courses. The OFT recognises, therefore, that there would have been only one successful bidder. The importance to the OFT of demonstrating such a relatively low critical mass is that it provides a relatively easy answer to the veto point. Whilst the OFT did not make any such point in its Decision, it now seeks to make good that 50% case by relying on the Arena “subject to 87 contract” offer of June 2000 to the 42, one condition of which was acceptance by courses with at least 50% of betting revenue. It also relies on the like condition in the subsequent offer, also to the 42, made on 14 July 2000.[196]The RCA appellants were, in our view, justified in regarding this as poor support for the view that 50% coverage represented the critical mass that Arena (or any other bidder) then and thereafter needed. The OFT did not seek the production of Arena’s business plan; and precisely what its negotiating tactics were at June and July 2000 when it made these two subject to contract offers is a matter of uncertainty. It can be said that, had Arena achieved a signing with the 42, and then (as it must have hoped) also signed up with the Super 12, it would have obtained 85% of betting turnover. What we do know is that when it made its offer to all 59 courses on 31 July 2000, its minimum requirements were a critical mass combining at least 80% of betting revenue and not less than 75% of terrestrial fixtures. In addition, no other bid set a critical mass condition as low as 50%: the next lowest was the 65% condition which Carlton identified in its letter to the courses of 18 November 2000.[197]The OFT relied next, in support of its case that the 70% condition in the MRA should be discounted, upon the fact that ATR had in 2003 itself made statements to the OFT that it would not necessarily have insisted on full compliance with the clause 2.2 conditions but would or might have waived them and settled for acceptances falling short of the 70% figure; or, for example, for acceptances not including all the TRG courses, Ayr (with betting turnover of just 2.1%) was regarded by the OFT as being readily dispensable. We heard no argument on whether, as a matter of contract, it would in principle have been open to ATR to waive full compliance with condition 2.2. Assuming that it would, we would again regard any such self-serving assertion by ATR as one calling for circumspection by the OFT, not least because in the Notification’s reference to the 70% condition ATR had joined in saying that: “The participation of a minimum number of courses (measured by betting revenues rather than by number of meetings or races) is necessary for the Model to work …”.[198]We take the view that a good working inference is that 70% such participation was so necessary. We do not consider that the OFT was in any position, on the basis 88 of the evidence before it, to make a finding that ATR’s required critical mass was anything less than that specified in clause 2.2. For reasons briefly given, we regard as somewhat flimsy its reliance upon Arena’s “subject to contract” offers in June and July 2000 to the 42 (which came to nothing) as supporting a conclusion that a critical mass of something materially less than 70% of betting turnover was required by ATR . In our view the best evidence as to the requisite critical mass that ATR required is to be found in the contract it actually made, namely the MRA. We can see no reason to assume other than that the specific contractual conditions incorporated into the MRA provide the best guidance as to the critical mass perceived by ATR to be requisite for the provision by it of a viable service. As it seems to us, those conditions must be viewed as the conditions that ATR would have sought to meet in its bid to acquire the necessary rights via the counterfactual purchasing process hypothesised by the OFT.[199]Those conditions (introduced at the behest of ATR) required acceptances by(i) all 17 TRG courses,(ii) the five non-TRG courses owned by RHT,(iii) the three non-TRG courses owned by Northern, and(iv) courses as a whole enjoying between them at least 70% of betting turnover. For practical purposes, it followed that each TRG course, the TRG courses as a group and each of Northern and RHT had a veto in respect of compliance with the first three conditions; that the Super 12 (with betting turnover of 34.5%) had a veto as regards condition (iv); and that, for practical purposes, so did RHT. If, as we consider, clause 2.2 provides the best evidence of the components of the minimum critical mass that ATR needed, we do not follow how the OFT could properly find that none of these courses, owners or groups each had a relevant veto on the conclusion of an agreement dependent on tying up the courses and groups identified in clause 2.2. The OFT advanced no case that for any such group to exercise such a veto would itself be impermissible as involving an infringement of section 2 of the Act. In addition, it is relevant to note that RHT, the TRG, Arena and Northern sold their rights to ATR for proportionately more than the independent courses, and therefore at a premium, so supporting their claim to a veto.[200]Ultimately, however, the whole veto question appears to us to be anyway probably too theoretical to be of real practical utility. The OFT’s case is founded on the assertion that ATR could have gone about the purchasing exercise in a fundamentally different way from that which ATR originally acknowledged was the 89 only practical way; and that, had it done so, it could have picked up the requisite rights at an appreciably lower price. Any such case has to be made out as a matter of fact, albeit only on the balance of probabilities. The task that the OFT had to assume in seeking to prove that case required it to engage in a speculation as to the outcome of events which never happened in a world that never was. That is not intended as a critical comment, it is merely intended to highlight what we regard as the particular difficulties of proof that the OFT faced in this case. For reasons we have briefly given earlier, we consider it more than likely that even if ATR had started out on the road of individual negotiation, it would in fact have finished up with a central negotiation. To make it all yet more speculative, there has also to be factored into the exercise that, in going down this alternative route, ATR would also have been faced with competing approaches to the courses from the other bidders, including in particular Carlton, which would have served to push the price up.[201]As to this last point, the history of the negotiations shows that there were in fact a number of credible bidders, who appeared to be serious about acquiring the rights. At the very least, ATR and Carlton were competing bidders, willing to pay substantial amounts to acquire them. In a situation such as this, where there was genuine competition for the rights between two serious bidders, it is in our view plausible that such competition would have pushed up the price for the rights even in the absence of any co-ordinated actions on the part of the Courses. In addition, ATR’s bid was pitched at a level that was consistent with a coherent business plan that it had produced, which suggests that the price it eventually paid represented its genuine, contemporaneous estimate of what the rights were worth to it: a price which would, therefore, represent the competitive price. The OFT’s counter argument is that the level of the bids had been pushed to their autumn 2000 level as a result of earlier co-ordinated actions by the racecourses. This is a perhaps possible interpretation of the events, but it is not the only one; and in our judgment the evidence before the OFT did not entitle them to be confident as to the correctness of their interpretation of such events.[202]In all this speculative uncertainty, and having regard to the particular circumstances of the case – the sale of a novel product, which had never previously been sold and whose value, so far as it could be assessed, depended on future revenue 90 generated by a novel venture - we conclude that the OFT failed to prove that the MRA resulted in an appreciable increase in the different price that would have been paid by ATR in the imprecise, and somewhat shifting, hypothetical counterfactual situation which formed the basis of the OFT’s case. 7. Did any price increase affect competition?[203]Even if (contrary to our conclusion) the OFT did prove that the collective negotiation resulted in an appreciable price increase, the OFT accepted that the manner in which the Courses sold their rights could not affect the end-consumers of interactive betting, because the ATR product would be “available at prices likely to be at the competitive level” (paragraph 369). If, therefore, collective selling increased the price of the rights, the only effect was to reduce ATR’s revenues and to increase that of the Courses. The OFT’s decision on this was expressed as follows:
“303. While the OFT aims to use its powers to ensure that markets work well for consumers, a finding of direct detriment to final consumers is not a condition of finding an infringement of the Chapter I prohibition. The key legal question is whether an agreement prevents, restricts or distorts competition on a relevant market within the UK: that market need not be a retail market. For the reasons set out in this document, the OFT concludes that this has occurred in this case.”
[204]The RCA appellants and BHB submitted that the OFT failed to explain how the charging of an increased price by the 49 courses to ATR was anti-competitive when the price to end-users was competitive. They said it is settled law that the purpose of Article 81 of the EC Treaty:
“is not to provide a general escape route for those wishing to avoid complying with contractual obligations which turn out to be more onerous than expected.” ( European Community Law of Competition, Bellamy & Child, 5 th Ed, 2-115. The commercial position was that this was an arm’s length deal between the parties. ATR was a joint venture company comprised of three experienced operators. There is no basis for any inference other than that it was a willing purchaser at the price it paid. It was itself a party to the notification seeking negative clearance for the sale or else 91 an exemption. In the event, ATR’s commercial expectations have not been fulfilled and so it has now sought to reverse its attitude towards the acquisition so as to enable it to argue that the bargain it freely entered into is fatally tainted by an infringement of competition law. The RCA appellants referred us to the following observation of Advocate General Jacobs in Case C-7/97 Bronner v. Mediaprint [1998] ECR 1-779: “It is important not to lose sight of the fact that the primary purpose of Article [82] is to prevent distortion of competition – and in particular to safeguard the interests of consumers – rather than to protect the position of particular consumers.”
[205]The BHB emphasised that competition pays full respect to a freely negotiated commercial agreement unless only it has an appreciably foreclosing effect on third party suppliers or customers, which the OFT do not allege in their Decision. Competition law is not there to assist a party to unravel a commercial agreement that has turned sour, nor is its role to mend a bad bargain. Even on the OFT’s case, the MRA gave rise to significant consumer and other benefits. The BHB emphasised that context in competition law (as in most areas of law) is everything.[206]We have difficulty with these submissions, which were premised on the basis we outlined at the beginning of this section of the judgment. The proposition appeared to us to be that, assuming all else against appellants, only ATR was adversely affected by the MRA, which was therefore not relevantly anti-competitive at all. The submissions appeared to us to be of potentially wide-ranging importance but were not developed before us to a point that satisfied us that we can or should attempt to rule on them. We accordingly make no decision on this limb of the appellants’ cases. 8. Was the OFT entitled to find the price structure resulting from the collective negotiation restricted incentives for non-price competition between the Courses?[207]The OFT’s position was that payments by ATR to the Courses under the MRA were largely fixed. This is because, apart from a small top-up element expected to be payable in the last two years of the term, they consisted of guaranteed payments distributable between the Courses in accordance with the RCA distribution formula. It followed that the method of payment provided no incentive to the Courses to alter 92 their fixtures or races in ways that might generate more betting revenue. On the OFT’s counterfactual case, the sales of rights would have been individually negotiated and a high proportion of the payment structure so negotiated would have been represented by a revenue split formula in the nature of a revenue sharing arrangement. The payment received by each course would then have been related directly to the betting turnover it generated, thus providing an appropriate incentive. The OFT’s case was that it was obvious that the bidders would have wished to structure their payments to the Courses in this way. They referred to examples of agreements which included like formulae, including the BAGS agreement, the first Arena proposal in January 2000 and Channel 4’s approach to the 47 in March 2000.[208]The RCA’s response to the Rule 14 notice was to question whether any such incentives would have been provided for in the suggested counterfactual situation; to argue that the courses had anyway retained incentives to improve their output; and to argue that the MRA’s impact on their incentives was anyway too small to affect their conduct.[209]We recognise that the payment structure in the MRA loaded much of the downside risk with ATR, with the upside potential (if profits exceeded expectations) being shared with the Courses. This was, on its face, unattractive to ATR, whilst the avoidance of the downside risk for 10 years was attractive to the Courses. In the event, the ATR venture proved to be a business failure. That was not because of any deficiencies on the part of the Courses, but (so it appears) on shortcomings which can be laid at ATR’s door. It is no surprise that the Courses looked unfavourably on a proposal which required them to shoulder a material proportion of these risks.[210]The crucial question is, however, whether the OFT is right that a sale of the rights other than by collective negotiation would probably have yielded a return to the Courses in which a high proportion of the payments was under a revenue split. We consider that there was no sufficient evidential foundation entitling the OFT to reach the decision that, absent collective negotiation, the successful bidder would have achieved a purchase of the necessary critical mass of rights on terms involving the loading upon the Courses of a material element of the risks of the new venture. Again, the burden of proving the case in this respect was on the OFT, whereas again its 93 conclusion was founded on speculation as to what might have happened in a world that never was. It was guesswork based on what the bidders would have wanted but which paid insufficient regard to the fact that the evidence showed that the courses would probably not have wanted anything other than substantial fixed payments. The sale of their interactive rights was not an inevitably “win win” opportunity for the courses: as explained in the evidence of Mr Gundill and Mr Davies, the sale of their interactive rights could also have an effect on their other income, for example from race attendance (turnstile income) and could reduce their turnover at LBOs. The attraction from the courses’ viewpoint in a return represented predominantly by fixed guarantees is also underlined by the fact that the competitive GG Media deal offered fixed amounts and was therefore regarded as superior to the ATR deal by ten courses.[211]To the extent that the OFT relied on other agreements said to have been concluded in analogous circumstances, we were not persuaded that the claimed analogies were helpful. The question turns on the probabilities of a different reward structure having been negotiated in the particular market for these novel rights in 2000/2001. The reliance on the Arena discussion document of January 2000 was also unhelpful, that document proposing a structure in which the courses were to become shareholders in a joint venture, which was a quite different situation. In so far as there is anything in the way of concrete guidance as to what might have happened in hypothetical negotiations between ATR and the courses in the world that never was, it was within the knowledge of the courses that before the RCA became involved in the negotiations the Channel 4 consortium had made an offer to the Super 12 including minimum guarantees amounting to £221m. On the OFT’s counterfactual suggested to us in the course of argument, ATR would or might have engaged in separate negotiations with the Super 12, Arena, Northern and then the rest. It is improbable that the Super 12 would have settled instead for a split revenue consideration; and with such a deal on offer to the Super 12, we regard it as improbable that Arena, Northern and many (if any) of the other courses would have been prepared to accept terms that did not include substantial minimum guarantees either. In our view, there was no sufficient evidential basis for the OFT’s conclusion on this aspect of the case. The OFT did not, therefore, prove that the MRA resulted in an appreciable restriction of non-price competition that would otherwise have existed. 94 9. Was the OFT’s section 2 analysis affected by the fact that the rights related to the transmission of recordings of sporting competitions?[212]Having come to the conclusions we have on the issues that we have discussed, we find it unnecessary to express any views on this question. 10. Was the OFT wrong in refusing an exemption under section 4?[213]Having come to the conclusion that the OFT was wrong to find that there was any infringement of the Chapter I prohibition, we do not prolong this judgment by considering the alternative ground of appeal under this head. XII RESULT[214]We allow the appeals and set aside the OFT’s Decision that the sale of the Non-LBO rights in the MRA infringed the Chapter I prohibition. Colin Rimer Andrew Bain Sheila Hewitt Registrar 2 August 2005 95

Cited in 9 later judgments