David Alexander de Horne Rowntree v Performing Right Society Limited & Anor [2026] EWCA Civ 814

[2026] EWCA Civ 814Case No CA-2025-002741
IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE COMPETITION APPEAL TRIBUNAL
(Justin Turner KC, Professor David Ulph CBE and Hugh Kelly)
[2025] CAT 49
Venue Royal Courts of Justice, Strand, London, WC2A 2LLDate 29/06/2026
Before
LORD JUSTICE NUGEELORD JUSTICE ZACAROLILORD JUSTICE MILES
Between
DAVID ALEXANDER DE HORNE ROWNTREEAppellant(1) PERFORMING RIGHT SOCIETY LIMITEDRespondents(2) PRS FOR MUSIC LIMITEDRespondent
Tim Ward KC and Jack Williams (instructed by Willkie Farr & Gallagher (UK) LLP) for AppellantMarie Demetriou KC and Charlotte Thomas (instructed by Macfarlanes LLP) for RespondentsHearing Hearing date : 11 June 2026
Approved JudgmentThis judgment was handed down remotely at 10.30am on 29 June 2026 by circulation to the parties or their representatives by e-mail and by release to the National Archives..............................

Introduction

[1]This is an appeal from the decision of the Competition Appeal Tribunal (the Tribunal) dated 27 August 2025(a) granting reverse summary judgment or striking out the claims, and(b) refusing to certify them as eligible for inclusion in collective proceedings.[2]The appeal concerns the application of the rules of the first respondent (PRSL) which is a collective management organisation (CMO) in the United Kingdom which, through its wholly owned subsidiary, the second respondent (PRSFM), collects and distributes performing rights royalties for music copyright works to its members. PRSL is a company limited by guarantee with about 175,000 registered members. About 165,000 are songwriter members and about 10,000 are publishers. I shall refer to the respondents together as the PRS.[3]The PRS is non-profit making. Its purpose is to licence the musical works, collect royalties and distribute its net income to its members according to its rules.[4]The appellant (the PCR) is a songwriter member of the PRS. He is the drummer of Blur. He sought a collective proceedings order (CPO) pursuant to section 47B of the Competition Act 1998 (the Act) proposing himself as the class representative. The proposed proceedings are opt-out proceedings. The proposed class consists of any person who between 9 March 2017 and the date of issue of the proceedings, was a songwriter member of PRSL (i.e. a creator of music or lyrics of a work to which the copyright has been assigned to the PRS).[5]Members of the PRS assign the copyrights in musical works to the PRS and the PRS collects royalties for the exploitation of those works. The PRS has a series of rules or regulations including a Distribution Policy.[6]In broad terms, under these rules members are entitled to distributions from the PRS in amounts calculated by reference to the numbers of uses of the individual musical works during a relevant period.[7]The PRS uses several methods for matching receipts with musical works and the share profiles of individual members (whether writers and publishers) in respect of those works. It seeks wherever possible to carry out that matching process.[8]The PCR’s claim concerns royalties received by the PRS which it cannot match to the correct songwriter or publisher member owing to inaccurate or insufficient information. These are sometimes referred to colloquially as black box royalties, and the PCR’s pleading adopts that term. The PRS prefers the more monotone “unidentified royalties”, pointing out that there is no suggestion that the problem arises from a failure of transparency; rather it arises from the absence of information which would enable the sums to be matched. With this caveat in mind, it is convenient to use the black box royalties label.[9]Under the PRS’s existing rules black box royalties have been, and are, distributed to writer and publisher members in the same proportions as distributions of matched royalties. This has been described as pro rata distribution.[10]The essential complaint is that a pro rata distribution is unfair to writer members because greater proportions of the sums making up the black box royalties are likely in the aggregate to be attributable to writer members rather than to publisher members. The PCR says that this amounts to an unfair abuse of dominance under the Chapter II prohibition contained in section 18 of the Act. The claim essentially seeks a redistribution of the unidentified royalties by way of damages and an injunction requiring a non-infringing distribution in the future.

Further factual background

[11]The relevant facts may be taken from the Tribunal’s judgment ([2025] CAT 49, references below to its paragraphs are in the form “[Jx]”). The PRS[12]PRSL is a company limited by guarantee and exists only for the benefit of its members. The statutory board of directors of the PRSL is known as the Members’ Council. This currently comprises four independent nonexecutive Council Members, ten publisher Council Members, ten writer Council Members and the CEO. In previous years covering the period of the proposed claim there has been a similar, although not identical, constitution. PRSFM has a board comprised of a mixture of publisher and writer members: the current composition being, one executive board member, four writer Council Members, four publisher Council Members and four non-executive board members.[13]The PRS has a Distribution Committee (the DC). This was explained in the evidence of Mr Arber, the Director of Operational Improvement, which was accepted by the Tribunal. The DC’s specific functions include considering and approving PRS management’s proposals on distribution policy, which involves considering the financial impact that policy decisions may have on members’ earnings, and considering the balance of cost-efficiency and accuracy. When carrying out their duties, the members of the DC must act in the interests of all the members of PRS as a whole. The DC includes an equal number of writer members and publisher members as well as independent non-executive directors. Assignments of copyrights to

The PRS

[14]Members assign their performing rights in musical works to the PRS pursuant to standard form terms. The PRS agrees to use its best endeavours to pay its members amounts calculated by reference to the royalties received by it in respect of the relevant musical works.

Royalty revenues

[15]There are four main categories of royalty revenue:(i) broadcasting;(ii) public performance;(iii) online; and(iv) international.[16]The broadcasting category relates to the licensing of PRS members’ works broadcast on TV and radio, including BBC, ITV and Sky.[17]The public performance category relates to the licensing of the performance of PRS members’ works in public, whether as recorded music or live. Examples include performing live music in concerts and playing recorded music at public venues.[18]The online category relates to the licensing of PRS members’ works for digital services of all types, including music downloads and streaming, podcasts, and video on demand services. This category can broadly be split into two sub-categories: multi-territorial online licensing (MTOL) and non-MTOLonline. Digital music service providers which operate in multiple territories, such as Spotify or Apple Music, generally fall into MTOL, whereas non-MTOL online generally covers online audio-visual services such as Netflix, video games and audio services which predominantly operate only in the UK.[19]The international category relates to the licensing of PRS members’ works in territories outside the UK. This typically involves and relies upon licensing and distribution activity undertaken by Collective Management Organisations (CMOs) based overseas in the relevant territory, with whom PRS has entered into reciprocal agreements. PRS relies on information received from those affiliate CMOs to make onward distributions to its members.[20]There are two main types of licensing: repertoire-specific licensing (which generally applies only to MTOL); and blanket repertoire licensing (which generally, applies to broadcast, non-MTOL online, international and public performance).[21]Blanket repertoire licences authorise the use of any and all works in the PRS repertoire. The amount received by PRS pursuant to such licences does not tend to vary based on the number of times particular underlying works are exploited. It is typically a fixed amount for a given period (which can vary from period to period) covering the term of the relevant licence for the use of the whole repertoire.[22]Under a blanket repertoire licence there is no inherent distribution value for any particular usage on any given source or platform. The usage is not valued until the late stages of PRS’s distribution process are completed and would depend, among other things, on the number of times a particular work is performed as a proportion of the total usage of works performed.[23]In the case of repertoire-specific licences, PRS receives revenue collected from repertoire-specific licensing on a transactional basis by reference to individual usage, which has an associated value ascribed to it prior to the distribution stage.

Sources of use information

[24]The PRS uses various methods to identify which works have been performed. One is the use of what is called census information provided by the relevant licensee. This applies to most online streaming and many radio stations. A second method is the use of sampling, which involves collecting a representative sample of actual performances over a period and using it as a basis for distributing royalties more generally. This tends to be used where usage is repetitive. A third method is analogy-based distribution which is used where there is insufficient data available to enable sampling such as in the case of usages in pubs and restaurants. The appropriateness of these various methods is not the subject of this dispute. They are therefore proper methods for matching royalties received by the PRS with the use by licensees of musical works.

Recorded shares of members in a work

[25]Royalties are, where possible, distributed by reference to the number of uses of specific works. It is necessary for the PRS to be able to identify the writers and publishers entitled to a share of royalties in respect of each work used. The PRS relies upon its members to register their works on its database and to provide accurate information about the writers and publishers entitled to a share of the royalties. Where the writer of a work has a publisher (as is usual), the publisher is responsible for registering the work. The publisher is also required to register the interest shares of the writers and must register the relevant publishing agreement. The standard arrangement is that the publisher receives 50% of the royalties and the writers (there may be one or more) the other 50%. PRS does not accept works where it is sought to allocate more than 50% of royalties to the publisher. In some cases the writer may receive more than 50% of the royalties. Where the work is unpublished the writer is responsible for registering their interest in the work.[26]When a work is registered with the PRS it is issued with a tune-code which is used to identify that work. Subsequently, it should be issued with an International Standard Musical Work Code or ISMW and may receive one or more international codes.

Black box royalties

[27]In order to calculate the amounts payable to members for the royalties which it receives in respect of works registered with it, the PRS needs to be able(a) to identify the relevant work based on the information provided to it by the licensee and match it to the work registered in its database,(b) to identify the “share picture” for the work, i.e. the interested parties and their respective shares in the work, and(c) identify the contact and payment details of the person(s) to whom the royalty distribution should be paid. At any of these stages, there may be a data problem which complicates the process. This potentially gives rise to black box royalties.

The claims made by the PCR in the CPO Claim Form

(c) The claims made by the PCR in the CPO Claim Form

[28]In the Claim Form (the CF) the PCR alleges that the PRS is dominant in the market for the administration of performing rights royalties for songwriters and/or writers and publishers in the United Kingdom. He alleges that the PRS has abused and continues to abuse its dominant position contrary to the Chapter II Prohibition (and before 31 December 2020 contrary to Article 102 of TFEU). It is also alleged that the PRS has breached and is continuing to breach the Chapter I Prohibition (and before 31 December 2020 Article 101 TFEU) by its agreements with its members and/or its royalty distribution decisions. As the PCR did not rely on the Chapter I Prohibition on the appeal, I shall not refer to it further below.[29]The CF alleges at [8] to [10] that:
“8. The unfairness arises because PRS pays royalties to publishers which likely belong to songwriters and should be paid to them. More specifically the claim relates to the portion of performing rights royalties collected by PRS which cannot be matched with the correct songwriter or publisher because of inter alia incomplete reporting by licensees or data issues, including on the PRS database. While the musical work may be identified, failings frequently arise as to the identity of a songwriter, and much less frequently as to the publisher. For example, the songwriter’s name may have been misspelt. These unmatched royalties are described colloquially in the industry as “Black Box”. 9. The majority of Black Box royalties held by PRS belong to writers because (a) the vast majority of the 165,000 PRS members are songwriters and there will therefore be more reporting and data issues applicable to songwriters; (b) PRS collects royalties on behalf of songwriters in respect of use worldwide, but, by contrast, tends to collect royalties on behalf of publishers only in respect of use in the United Kingdom. Entities abroad (such as foreign CMOs) tend to pay publishers direct whereas they account to PRS for the songwriter share of the royalties collected abroad. 10. Thus reporting and data issues are more likely to apply to the songwriters and PRS administers a large proportion of royalties belonging to songwriters. Its likely therefore that the majority of black box royalties are songwriter royalties.”
[30]The PCR alleges that most if not all black box royalties “relate to songwriters”:
“45.1. The number of songwriter members is far greater than the number of publishers. Of the 165,000 PRS members, well in excess of 160,000 members are likely to be songwriters. Accordingly, the reporting and data defects will more likely apply to a songwriter than to a publisher. There are only several large publishers in the United Kingdom, and they conclude many hundreds or thousands of agreements with individual songwriters. 45.2. Most of the royalties processed by PRS relate to songwriters. As explained above in the case of public performances outside the United Kingdom, publisher members typically choose to be paid their share of the royalty outside of the United Kingdom. Publishers also tend to be paid direct in respect of multi-territorial online licensing. 45.3. When staff at PRS input musical works and interested party data into PRS’s databases, they are more likely to transcribe publisher details correctly as there is more familiarity with the names of publishers. 45.4. Publishers are more likely to verify the accuracy of a royalty payment than a songwriter. Publishers have significant resources to check whether their royalty payments are correct and to conduct royalty payment audits and, where necessary, question the amount of a royalty payment with PRS.”
[31]The primary claim advanced before the Tribunal was for contravention of the Chapter II Prohibition. PCR alleged that the PRS had a dominant position in the market for the administration of performing rights royalties for songwriters and/or for songwriters and publishers in the UK. The abuse of dominance was pleaded in [61] to [63]:
“61. Songwriters are required to assign their performing rights to PRS which has special responsibility to ensure that songwriter royalties are properly allocated and the right balance is struck between the freedom of writers to dispose of their works and the effective management of their rights by PRS. In addition, although PRS is not responsible for actually distributing the entirety of Black Box preforming rights royalties from the exploitation of PRS repertoire (because, for example, such royalties are paid direct to publishers by foreign CMOs and other entities), PRS is nevertheless required to ensure that publisher and songwriter members receiver their equitable allocation of the overall Black Box royalties. 62. In breach of this special responsibility PRS is imposing unfair trading terms on its songwriter members through its policies that permit the distribution to publishers of performing rights royalties owed to PRS songwriter members. Moreover, the distribution policies of PRS and their implementation are not equitable since they require songwriters to dispose of their works without entitling them to obtain the full royalties owing to them. 63. To the extent that songwriter royalties cannot be matched to the correct songwriter – hence the Black Box – there is no objective reason why the unmatched royalties should not be allocated to the songwriters as a class. There is no objective reason why any portion of unmatched songwriter royalties should be paid to publishers as a class.”
[32]It should be noted that the CF does not make any complaint about the efforts which the PRS has taken to identify works and/or writers.[33]The PRS has not disputed that royalties associated with unidentified works are distributed pro rata between publishers and writers by reference to their matched royalty entitlements. Moreover, for the purposes of the strike out application (but not otherwise) the PRS has accepted that the Tribunal should assume the truth of the PCR’s allegation that it was likely that a larger proportion of the unidentified royalties would (assuming that there had been no data failures) have been payable to writer members as a class than was the case under PRS’s distribution rules.

Relevant provision of the Act and the Rules

[34]Sections 47A and 47B of the Act provide:
“47A Proceedings before the Tribunal: claims for damages etc. (1) A person may make a claim to which this section applies in proceedings before the Tribunal, subject to the provisions of this Act and Tribunal rules. (2) This section applies to a claim of a kind specified in subsection (3) which a person who has suffered loss or damage may make in civil proceedings brought in any part of the United Kingdom in respect of an infringement decision or an alleged infringement of— (a) the Chapter I prohibition, or (b) the Chapter II prohibition. (3) The claims are— (a) a claim for damages; (b) any other claim for a sum of money; (c) in proceedings in England and Wales or Northern Ireland, a claim for an injunction. …” “47B Collective proceedings before the Tribunal Subject to the provisions of this Act and Tribunal rules, proceedings may be brought before the Tribunal combining two or more claims to which section 47A applies (“collective proceedings”). …”
[35]The reference to the “Chapter II prohibition” is to section 18 of the Act, which relevantly provides that: “any conduct on the part of one or more undertakings which amounts to the abuse of a dominant position in a market is prohibited if it may affect trade within the United Kingdom”, and that conduct “may, in particular, constitute such an abuse if it consists in – (a) directly or indirectly imposing unfair purchase or selling prices or other unfair trading conditions…”.[36]The reference in section 47B(1) to “provisions of this Act and Tribunal rules” includes reference to the eligibility condition and the authorisation condition.[37]The claims must be eligible for inclusion in collective proceedings: s. 47B(5)(b) and rule 77(1)(b) of the Competition Appeal Tribunal Rules 2015 (the Rules). The eligibility condition comprises three cumulative requirements, set out in s. 47B(6) and r. 79 of the Rules, namely(a) the proposed claims must be brought on behalf of an identifiable class of persons: r. 79(1)(a);(b) the proposed claims must raise common issues i.e. the “same, similar or related issues of fact or law”: s. 47B(6) and r. 79(1)(b); and(c) the proposed claims must be suitable to be brought in collective proceedings: s. 47B(6) and r. 79(1)(c).[38]The Tribunal must also be satisfied that the PCR bringing the proceedings can be authorised as the class representative: s.47B(5)(a) and r. 77(1)(a). The authorisation condition is met if the Tribunal considers that it is “just and reasonable” for the PCR to act as a representative in the proceedings: s. 47B(8)(b) and r. 78(1)(b). The factors relevant to the determination of whether it is just and reasonable for the PCR to act as the class representative are set out in r. 78(2).[39]Rule 79(4) of the Rules provides that at the hearing of an application for a collective proceedings order the Tribunal may hear an application to strike out the claims under r. 41(1) or for summary judgment under r. 43(1). As recognised in [J34]-[J35], the same approach applies to these rules as in ordinary civil litigation. There is no need to rehearse the well-known principles in this judgment.

The Tribunal’s judgment

[40]The Tribunal concluded that the claim should be struck out or dismissed. It was also not satisfied that a CPO should be made. The Tribunal’s essential reasons may be summarised as follows.[41]Section 47A requires the members to have a claim. Section 47B does not itself provide a cause of action. The relevant individual claims of potential class members must be actual and not contingent or potential future claims which can start or crystallise only upon the making of a CRO [J42].[42]The case being advanced by the PCR is essentially that the rules of the PRS are unfair because people who are potentially “owed” unidentified royalties are those members whose works have been performed but who cannot be identified and/or paid. However, the Tribunal held that the class is not drawn to describe those people: it is drawn to describe all writer members of the PRS. No reasons have been advanced capable of supporting the contention that members of the class share the common feature of being “owed” black box royalties. Further no coherent reasons have been advanced to support the proposition that, notwithstanding that members of the class are not owed the black box royalties, it is unfair and abusive not to pay them these royalties, or a greater proportion of them [J61]-[J62].[43]There is no suggestion that it is wrong in principle for the PRS to redistribute black box royalties away from the people who might have a potential claim to those royalties to others. The complaint relates to the manner of this redistribution. The black box royalties could be redistributed in many different ways and some of these ways might be considered by some people to be preferable to others: they could be given to those members who are the neediest; or to a musical charity; or to every member in equal portion; or, as they are, distributed pro rata. Merely because one method of distribution might be considered preferable to another does not of itself mean that the PRS’s chosen method of is unfair or abusive [J63].[44]For these reasons the CF did not disclose facts which support the claim for infringement of Chapter II of the Act, nor did the collective action have a reasonable prospect of success. The applications for summary judgment and to strike out the claims therefore succeeded, and certification was refused [J64].[45]The Tribunal went on to consider the question whether the PCR had advanced a realistic methodology for assessing loss on a class wide basis (aggregate damages). In order to make an aggregate award of damages it is necessary to formulate a counterfactual which would enable the Tribunal to identify the revenues which would in the counterfactual have been paid to the class [J67].[46]The Tribunal was not satisfied that the expert evidence of Mr Savage, which was relied on in the CF, set out a realistic methodology for assessing aggregate damages. Mr Savage explained that he needed to identify the amounts of black box royalties due to writer members which have otherwise been distributed to publishers and that he will need to calculate(a) “the monetary amount of Black Box royalties that are due to PRS writer members over the class period”, and(b) “the monetary amount of Black Box royalties that PRS writer members actually received.”[47]The Tribunal noted that in cases where the data did not permit the PRS to identify the relevant work, it is prima facie unknowable, in respect of those works, what proportions of royalties are “owed” to publishers and to writers. Mr Savage’s report does not explain how his audit of the data could establish the extent to which royalties associated with unknown works were impacting writers more than publishers. He gave no explanation of how in such cases the “expected distribution” is to be ascertained [J72].[48]The Tribunal accepted that it was only wielding a broad axe at the certification stage and that due allowance should be made for any imperfections in the methodology which were capable of being addressed in the future, but it concluded that the PCR had not, even in general terms, identified a plausible approach to how it might estimate the sums that should be paid to writer members on the counterfactual. It followed that the PCR has not identified a sufficiently credible or plausible method of how damages fall to be assessed [J77]. For this additional reason the Tribunal would not have certified the case as suitable for collective proceedings.[49]The Tribunal went on to say that if the application for strike out and summary judgment had not succeeded it would have been receptive to the submission that in the light of further information that had been provided by the PRS during the course of this hearing between 13 February and 16 June 2025, the PCR should have had a further opportunity to formulate the counterfactual in the light of that information. The Tribunal said however that it is not clear that the defects were easily remediable [J78].[50]The Tribunal then considered the costs and benefits of collective proceedings. It noted the estimate of the PCR’s own costs of the proceedings of £13m. It also referred to a legal funding agreement (LFA) with funders, LCM, which (subject to the control of the Tribunal) would entitle the funders to a very substantial Funders Fee, depending on the date of conclusion of the proceedings. The tribunal considered the size of the claim. It recorded that the PRS had advanced an estimate of the unidentified royalties paid to publisher members over the relevant period at about £55m and that the claim would be some proportion of that amount. The Tribunal noted that this calculation did not include amounts paid to some publishers directly in the cases of the MTOL and foreign CMO arrangements. The PCR contended that additional amounts of up to £150m of unidentified royalties could have been paid to publishers in this way. The Tribunal considered that it could not resolve this issue and proceeded on the basis that the claim would be for some proportion of the claimed sum of £200m potentially paid to publishers. At [J99] the Tribunal observed that the size of the claim was not known because the PCR was unable to produce an estimate of the proportion of that £200m of black box royalties which should have been paid to writers in the counterfactual.[51]At [J103] the Tribunal said that it remained doubtful that the cost-benefit of these proceedings would favour certification. If the proceedings were to succeed, the class might in effect find itself contributing to the costs of the litigation, the Funders Fee, and any payment of damages. It was reasonably in contemplation that at the end of these proceedings, even if the claim was successful the principal beneficiaries might be the legal advisors and LCM and not the class. The agreement reached under the LFA was to pay a Funders Fee which might reasonably be expected to be as much as, or more than, sums awarded in damages.[52]At [J104] the Tribunal repeated that the PCR had failed to formulate the counterfactual and justify certification. It said that if the size of the claim could have been better understood it would have revisited proportionality, but that as matters currently stood it was not persuaded that the costs-benefit favours continuation of these proceedings. The central problem was that the class was, in a manner of speaking, suing itself.[53]The Tribunal also noted that under r. 79(2)(g) of the Rules one of the matters to be taken into account in determining whether the claims are suitable to be brought in collective proceedings is the availability of alternative dispute resolution and any other means of resolving the dispute. The Tribunal concluded that there had been no proper attempt to resolve the dispute through the constitutional means available to the PRS’s members, including by making representations to the Members’ Council. There was no evidence that this had been attempted by members of the class prior to this claim being put together by the solicitors and funders. Nor was it clear that those writer members on the governing bodies supported these proceedings. It was reasonable to contemplate that there was scope for formulating alternative ways of distributing the unidentified royalties which might benefit the class more than spending what it called eye-watering sums on lawyers and funders. If this claim had not been dismissed, the Tribunal would have encouraged the parties to look at alternative ways of resolving this dispute [J107].[54]The Tribunal finally considered the authorisation condition. It rejected the PRS’s argument that the PCR was not suitable to be authorised as a class representative. It concluded that Mr Rowntree was suitable to act as the class representative but, depending upon how the case progressed, if it progressed at all, this was a matter which might need to be revisited [J117].

Grounds of appeal and respondents’ notice

[55]There are two grounds of appeal. The first is that the Tribunal erred in law in concluding that the claim had no reasonable prospects of success on the basis that the class was not drawn to describe those who were ‘owed’ black box royalties, and that as a result the members of the class lacked a common feature of being ‘owed’. The Tribunal erred in law regarding the requirements for constituting a valid class, at least for the purpose of certification, and the meaning of the commonality requirement in s. 47B(6) of the Act.[56]The second is that the Tribunal erred in law in concluding that there were no coherent reasons supporting the allegation that it was unfair and abusive for the PRS not to pay the members of the class a greater proportion of the royalties. The Tribunal misdirected itself in law as to the scope of what may arguably constitute an unfair trading practice contrary to the Chapter II prohibition for the purposes of an application to strike out or for summary judgment.[57]The PRS has filed a respondents’ notice contending that the Tribunal’s judgment should be upheld for the additional reason that, even if the PCR’s claim had surmounted the summary judgment/strike-out threshold, had the Tribunal had available to it the judgment of the Supreme Court dated 18 December 2025 in Evans v Barclays Bank plc & ors [2025] UKSC 48, [2026] Bus LR 328, it should, and any properly directed reasonable tribunal would, have concluded that the claim should not be certified on an opt-out basis given that:(i) it is manifestly very weak;(ii) it involves the class “in a manner of speaking, suing itself” ([J105]); and(iii) there is an alternative means of resolving this (at most) intra-PRS dispute using PRS’s governance procedures which would obviate the need for (highly costly) collective proceedings, and which there is no evidence any member of the class has attempted ([J107]).

Summary of the arguments on the appeal

[58]In brief outline the PCR submitted as follows. As to ground 1, on the Tribunal’s own findings at this stage, it is at least arguable that the class as a whole has suffered an aggregate loss in an amount of up to £200 million. While actual quantification of that loss will be a matter for trial, the facts of the case are such that the possibility (which is unknown at this stage) that the class might contain some members who have individually suffered no loss does not affect the overall quantum in this case, and could not therefore result in any overcompensation or prejudice to the PRS. It is not necessary at this stage of the proceedings for a proposed class representative to establish that every member of a proposed class has suffered a loss. These are matters that will be resolved at trial on the evidence of the issues. The collective proceedings regime permits that to be dealt with at a later stage and by means of a “top down” analysis, with scope for winnowing out any class members who have not suffered a loss later, should that even prove necessary. This is not a case in which it is already known that there are proposed class members who have not suffered any loss. It is also not a case where the presence of any potential no-loss class members inflates the aggregate damages. Thus there is no potential for prejudice to the PRS. There is also sufficient commonality amongst the proposed class members, contrary to the Tribunal’s remarks.[59]As to ground 2, the Tribunal should have concluded that this was at least sufficiently arguable to permit the claim to proceed to certification in order that the allegation could be tested in the light of evidence and disclosure, and in light of a revised CPO application which the PCR wishes to make. Applying domestic and European authorities, it is at least arguable on the facts that the PRS has created an unfair and unlawful system that routinely underpays songwriters as a class due to a systemic bias against them in how black box royalties are distributed. Under s. 18 of the Act, all songwriters are entitled to a fair distribution system (imposed on them by a near monopoly exclusive and unavoidable trading partner) rather than one which, at least arguably, is systematically and disproportionately biased against them and in favour of publishers; and is not necessary or proportionate. It is not generally appropriate to strike out a claim on assumed facts in an area of developing jurisprudence, novel points of law should be based on actual findings of fact, and difficult issues of law are best resolved once all the facts have been ascertained at a trial. The law on what constitutes unfair trading conditions is in a state of development.[60]In brief outline, the PRS submitted in answer to ground 1 that there is an obvious mismatch between the PCR’s proposed class – songwriters – and those who might do better under an alternative approach to allocating black box royalties than in proportion to identified royalties. The only possible distribution strategy the PCR has even hinted at is a system in which each member of the PRS receives exactly the same share of black box royalties, whether they are likely in reality to have contributed to a million pounds of the unidentified royalties or none at all. But if the PRS adopted this rule any songwriter who currently receives above average identified royalties (and therefore also above average unidentified royalties, on the pro rata basis) would lose out by having their share of black box royalties reduced to the same average share as every member of the PRS. Hence those members have suffered no financial detriment capable of constituting actionable “loss” on the PCR’s case (quite the contrary). The PCR’s only means of avoiding this obvious problem is to refuse to commit to identifying what he says would actually constitute a lawful distribution by PRS, against which PRS’s current system could be measured. But he cannot in this regard rely on his own failure. He needs to show that there is a coherent case that, at least in principle, the class as a whole has suffered loss, and he has failed to do so.[61]Ground 2 fails, in essence, because the PCR has failed to explain how PRS’s approach to the allocation of black box royalties is an abusive imposition of unfair terms. PRS allocates them in proportion to identified royalties. To show unfairness constituting an abuse according to the case-law on which the PCR relies, he would need to demonstrate that the distribution system either pursues an illegitimate objective or is disproportionate to that objective. He does neither. Part of his problem in this regard is that he does not even commit to a principle that would govern his own alternative system of allocation, let alone cogently explain how PRS’s actual approach is disproportionate. The only suggested rule of distribution (i.e. per capita sharing) is demonstrably unrealistic (see above).

Analysis

[62]It is logical to start with the second ground of appeal as it raises the question whether there is a properly sustainable claim at all.[63]As already explained, the Chapter II prohibition provides that “any conduct on the part of one or more undertakings which amounts to the abuse of a dominant position in a market is prohibited if it may affect trade within the United Kingdom”, and that conduct “may, in particular, constitute such an abuse if it consists in – (a) directly or indirectly imposing unfair purchase or selling prices or other unfair trading conditions…”.[64]There is no dispute at this stage that the PRS is in a dominant position in the relevant market or that its distribution rules may affect trade within the UK. The issue is whether its decision to make pro rata distributions is capable of constituting an abuse of its dominance by directly or indirectly imposing unfair trading conditions on the writer members.[65]We were taken to summaries of the law of abuse by the imposition of unfair trading conditions in the decision of this court in London & South Eastern Railway Ltd v Gutmann [2022] EWCA Civ 1077, [2022] ECC 26, at [93] to [102]; and the decision of the Tribunal in Gutmann v First MTR South Western Trains Ltd [2025] CAT 64, [2025] Bus LR at [48] to [88]. It was common ground that these passages were a sufficient statement of the law for present purposes. As the arguments of the parties did not turn on any analysis of particular passages within those summaries, or indeed on the details of the cases summarised in them, it suffices to record that both cases contain helpful general surveys of the relevant principles, and that nothing would be gained from repeating them at length here. I would however refer to some important general observations in [79] and [80] from the Tribunal’s summary in Guttmann v First MTR South Western Trains Ltd (with which I agree):
“79. We have no doubt that abuse is a broad concept, and that the concept of exploitative abuse by “unfair” conduct should develop to reflect new patterns of commerce. However, that concept is not unlimited. Competition law is not a general law of consumer protection. And where the allegations concern systemic conduct, the fact that the dominant company could have carried out a particular aspect of its business better, or in a different way that would have benefited consumers, does not mean that this conduct crosses the line to constitute abuse …” 80. It must be emphasised that abuse of dominance is prohibited and therefore unlawful. Such conduct renders a dominant company liable to potentially very significant fines, and is classified as quasi-criminal for the purpose of Art 6 of the European Convention on Human Rights. That is why the Tribunal has held that “strong and compelling evidence” is required to establish abuse: Napp Pharmaceutical Holdings Ltd v DGFT [2002] CAT 1 at [109]. The competition law prohibition of abuse does not create an obligation on the dominant company to organise or conduct its business so as to achieve the best outcome for its customers, or a fortiori for a sub-group of its customers …”
[66]The PCR’s claim has been summarised above. He alleges that the PRS, a dominant undertaking, has imposed unfair terms on its members. He alleges that the majority of black box royalties held by the PRS “belong” to writers because(a) the vast majority of members are songwriters and there will therefore be more reporting and data issues applicable to songwriters; and(b) the PRS collects royalties on behalf of songwriters in respect of use worldwide, but, by contrast, tends to collect royalties on behalf of publishers only in respect of use in the UK. Entities abroad (such as foreign CMOs) tend to pay publishers direct whereas they account to PRS for the songwriter share of the royalties collected abroad. On this basis it is alleged that reporting and data issues are more likely to apply to the songwriters, and the PRS administers a larger proportion of royalties “belonging” to songwriters. Thus the majority of unidentified royalties are “songwriter royalties”. The PCR’s case is that the actual distribution of unidentified royalties is unfair because, statistically, it is likely to favour publishers in aggregate over writers in aggregate.[67]The case is framed in terms of the black box royalties being “songwriter royalties”, with such royalties being “owned by” or “belonging to” songwriters as a class. This way of putting the case bears two comments at the outset. The first is that none of the unidentified royalties can be said to belong to or to be owned by (or indeed owed to) any specific members (or groups of them). This is not a cavil about strict legal concepts of belonging or ownership or obligation. It is more basic than that. The black box royalty problem arises from data failures; in other words from conditions of ignorance. In a perfect world, with complete data, the PRS would simply distribute royalties to individual members according to the numbers of uses of individual musical works and the members’ shares in those works. But in the real world, where there are data or reporting failures, the PRS cannot do this.[68]In theory it is possible to postulate the notional “true distribution” that members would receive in the perfect world of complete data, and to say (theoretically) that the actual distribution is unfair to the extent that it fails to reflect the “true” one. But that is a theoretical exercise that tells one nothing about the real world. The “true distribution” is by definition unknown and unknowable. It follows that while all members can say that they might have been entitled to payments under “true” (or perfectly accurate) distribution, none of them is able to say that they were actually entitled to any given amount. Hence none of the members can properly claim that any given amount of the black box royalties “belonged” to or was “owned” by them. Nor can they do so by being lumped into a class. I shall return to this point below.[69]The second observation on the way the claim is framed is that under the PRS’s distribution policies there is no rule of sharing by reference to membership of a class of members (e.g. writers or publishers). The PRS’s aim is to distribute royalties on a work-by-work basis - that is indeed how matched royalties are divided. They are shared according to the share-profile of individual members in the identified works used by licensees. The underlying principle is that royalties are to be distributed as far as possible to the members who have an interest in the works which have generated the royalties. These members may be writers, or they may be publishers. The owners of shares in works which are played more than others receive a greater amount. There is nothing in the rules which suggests or requires a class-based approach.[70]These are both important points, on which I shall expand below. However, they show that the case as pleaded, based as it is on “songwriter royalties”, is a manifest oversimplification.[71]As the Tribunal noted, the PCR does not complain in the CF about the PRS’s procedures for seeking to address data failures.[72]There was indeed evidence before the Tribunal of the extensive lengths to which the PRS went to achieve matching. These include automatic and manual matching and procedures for seeking to improve the information provided to and recorded by the PRS in respect of members’ interest-profiles in specific works. The evidence shows that the PRS has sought to improve its methods for achieving matching. The PRS uses a threshold floor of £3 for seeking to match revenues manually (which has been lowered from the earlier floor of £5). As counsel for the PRS pointed out, there is always a balance to be struck between the costs and the benefits of seeking to achieve perfect matching. The PRS could in theory continue spending black box revenues in trying to achieve matching. But at some point the costs would become disproportionate. As recorded in [‎13] above the Tribunal recorded that the DC’s specific functions include considering and approving PRS management’s proposals on distribution policy, which involves considering the financial impact that policy decisions may have on members’ earnings, and considering the balance of cost-efficiency and accuracy.[73]The PCR accepted that there was no challenge to the PRS’s processes for seeking to match royalties to members. Counsel submitted however that it did not follow that it would be impossible for more accurate matching to take place. But he accepted that there was a trade-off between greater accuracy and the costs of achieving it. There was no allegation that in approving pro rata distributions the PRS had assessed the trade-off in the wrong place. In my judgment it follows that the PCR’s case must be considered on the footing that the PRS has taken reasonable and proportionate steps to achieve accuracy and that the data failures which have generated the black box royalties are an inherent feature of the case. There is to my mind no realistic basis for supposing that greater accuracy (or approximation to “the true distribution”) will be achieved through the disclosure of documents by the PRS.[74]The PCR also accepts that the PRS has to adopt some rule for sharing black box royalties. It is a non-profit making body, whose purpose is to collect and distribute royalties. It needs to select some way or another of sharing the black box royalties and cannot simply accumulate them.[75]So what is the basis of the claim that in choosing the pro rata approach the PRS has imposed unfair conditions on its members?[76]It cannot be enough that there is some variance between the actual distribution and the “true distribution”. This is because any distribution rule which the PRS chooses to adopt for sharing the black box royalties will inevitably vary (and vary materially) from the “true distribution”.[77]Counsel for the PCR realistically accepted that the case was not merely that there was some variance between the “true distribution” and the actual distribution. He accepted that in order to determine whether there was abusive unfairness it was necessary to compare the actual distribution with another, counterfactual, non-infringing distribution.[78]I agree with the PRS’s submission that any assessment of the fairness of an actual distribution rule must be a relative exercise. It is necessary to compare the actual distribution with another proposed rule. Once one accepts that the PRS has to adopt some rule of distribution for the black box royalties, the fairness of its chosen rule can only be assessed by comparison with some other rule. It makes no sense to speak of any given rule as “fair” or not in an abstract sense. In other words the claimant must identify a realistic non-infringing counterfactual rule of distribution. The PCR has not however pleaded or advanced a counterfactual distribution. The only comparison it has made in its pleading is between the actual rule and the “true distribution” (i.e. that which would have happened in a perfect, fully informed, world).[79]For the reasons already given that is not however the relevant comparison, since if the PRS had the information which would enable there to be a “true distribution”, there would be no black box royalties. Armed with the missing data the PRS would simply distribute the royalties to those actually entitled. The circumstance that has generated the black box royalties is precisely that the true distribution is not and cannot be known. Nobody can ever know how much would have been payable to any member in that world. Because it cannot be known, it cannot be used as a realistic comparator in determining whether the actual distribution is unfair.[80]The comparator is crucial. Suppose that instead of distributing the black box revenues to members the PRS had decided to distribute them per capita to members. This was indeed floated as a possibility in the PCR’s skeleton argument, but not pursued by counsel at the hearing. But it is highly illustrative. As counsel for the PRS explained, such a rule would be far worse for many songwriter members than the pro rata rule. There are some superstar writers, say Ed Sheeran, whose songs are known (from the matching data) to have been played billions of times. There are others (a large number on the evidence) whose registered songs have only very rarely, if ever, been played. To distribute the black box royalties per capita would demonstrably favour the latter over the former. This helps to illustrate the need to frame the counterfactual to enable proper scrutiny.[81]Counsel for the PCR frankly accepted that no counterfactual had been advanced in the CF and that no counterfactual was being advanced now. He argued nevertheless that the case should be certified and that the PCR should be allowed an opportunity to reformulate its case, including by identifying a counterfactual rule of distribution. He noted [J78], where the Tribunal said that if it had not struck out the claim or given summary judgment it would have been receptive to giving the PCR a further opportunity to formulate its counterfactual in the light of evidence that had been provided by the PRS.[82]When asked to explain, even in general terms, what an expert could do with more information, Counsel for the PCR explained that there could potentially be sampling of the data about black box royalties in order to seek to determine how much was attributable to the absence of information about the identity of the works, or from failures or errors in the ways members’ interests were recorded on PRS’s database. However, as Zacaroli LJ noted, even if this proved possible it would only mean that the PCR had established as a fact the assumption being made at this stage in the proceedings: i.e. that, on average, data failures are more likely to affect writer members than publisher members. But that exercise would not lead to the identification of a counterfactual rule of distribution. This is because the assumed state of facts is no more than an assumption at a high level of abstraction about an aggregate variance between the “true distribution” and the pro rata one; and counsel accepted that the true distribution was unknowable.[83]In my judgment the PCR has had every reasonable opportunity to identify (at least in general terms) the proposed counterfactual. The only candidate even floated so far is the per capita division: but the analysis set out above shows that it is not a plausible counterfactual. The Tribunal found that the PCR had been unable to advance a plausible case at the hearing. In the period since the judgment was given in August 2025 the PCR has not advanced a plausible counterfactual distribution. I would accept that there may well be cases where at the outset the proposed counterfactual terms can only be formulated in general, descriptive, terms, recognising that it will need to be refined as disclosure is given.[84]The problem here is that the PCR has not offered even a broad description of the proposed counterfactual distribution. I do not consider in these circumstances that it would further the overriding objective to allow the case to go forward on the basis of the PCR’s unsupported hope that his team might be able to do better with more time. In my judgment the reason why no counterfactual has been advanced, even in general terms, is manifest: it is that the very data failure problem which has given rise to the black box royalties means that there is no plausible basis for suggesting a more accurate (let alone fairer) distribution.[85]In conducting its ordinary operations the PRS has done what can be done to achieve an accurate or true distribution. Indeed it inevitably has to do this to carry out its functions. It is not criticised for the steps it has taken. Where it has been able to obtain the data or correct errors in the records it makes matched distributions. But where it cannot do that there is a black box problem (i.e. an information deficit). There is no realistic basis for thinking that the processes of litigation will unearth more information, so the deficit will remain. One consequence is that there is no reason to suppose that the proceedings would lead to the development of a more accurate (and therefore fairer) counterfactual rule. I return to a further elaboration of this point in [‎94] below.[86]In my judgment it is also not sufficient for the PCR to say that the counterfactual distribution is one that would avoid any systemic and unlawful bias against the writer members as a class. That is question-begging. It presupposes that the chosen rule of distribution is unfair and unlawful. But that question cannot be tested in the absence of a concrete counterfactual proposal, expressed even in general terms.[87]I turn to examine more closely the PCR’s argument based on the statistical observation that data failures were more likely to affect writers as a class than publishers as a class. This is at the heart of the case. The PCR contends that this demonstrates that the pro rata distribution of black box revenues is inherently biased against writers as a class. As explained above, the PRS has accepted that the premise of this argument must be assumed to be correct.[88]In my judgment even accepting or assuming the premise that, on average, writers are more likely to be affected by data issues than publishers, this does not lead to the conclusion that the pro rata distribution rule is inherently biased against writers as a class or otherwise unfair. I say this for four overlapping reasons.[89]The first arises from points already addressed: the argument is based on a comparison between actual distributions and the “true distribution” that would have been made in a world of perfect information. For the reasons already given, that is not the proper comparison. The “true” or accurate distribution is not and cannot be known. This is because the root of the black box royalties problem is the absence of accurate information; and, conversely, if accurate information were available, the relevant royalties would be matched.[90]Moreover, statistical probabilities are not the same as entitlements. Suppose that writers as a class are more likely to be affected by data errors, it may be said of any randomly chosen, writer (W) that he or she is more likely than a random publisher (P) to be affected by data issues. But this does not lead to the conclusion that W is unfairly affected by the pro rata distribution rule or that P is favoured by it. If W is a superstar he or she may do well under that distribution, and may do far better than P or, indeed, than the average publisher. There is nothing inherently unfair about the pro rata distribution simply because, in aggregate, writers are more affected likely to be affected by data errors.[91]The second reason is that writer members do not make up a monolithic class (and nor do publisher members) when it comes to distributions. There are manifest differences between the positions of writer members. Superstar writers have had their works played billions of times, whereas other writers have had very few plays. The fact that, on average, writer members may face more data issues than publishers does not lead to the conclusion that any given writer member has been adversely affected or impacted by the pro rata distribution rule even within the class of writer members. Indeed any distribution mechanism will almost certainly produce winners and losers relative to every other (though none can be a proper match).[92]The third reason is that, as explained above, the PRS does not make distributions (whether of matched or black box royalties) by reference to the membership of a class. It does not create a separate pot for writers and another for publishers. Where it has the data, it distributes to individual members (be they writers or publishers) work-by-work. Where it does not (and is therefore sharing out the black box) it replicates that distribution. The class as a whole cannot say that it is entitled to any part of the black box revenues. As already explained the framing of the case in terms of songwriter revenues is apt to mislead.[93]The fourth reason is that one could postulate any number of other statistical distributions showing members of a chosen putative class being on average more affected by data failures than non-members of the class. The point may be tested this way. Suppose it could be shown statistically that writers of folk songs tended to suffer on average from a higher (or indeed lower) incidence of data failure than others. Or that there was a higher (or lower) incidence of data failure for writers of songs written before a given date, say 2000, than afterwards. Applying these classifications would show that members of those classes are likely to suffer more (or less) from data errors than the average. But it could not realistically be contended that the PRS’s rule of distribution should somehow reflect these classifications or that a pro rata distribution was unfairly imposed on them. That could lead to any number of different distribution principles. The PCR has chosen for the purposes of the claim to isolate the class of writers; but since under the usual distribution processes membership of the class of writers is not a relevant qualifying criterion there is no proper basis for identifying that class rather than another when testing the fairness of the distribution of black box royalties.[94]I do not therefore consider that the case based on the statistical assumption concerning the position of the writers as a class comes close to grounding a case that the PRS has unfairly imposed conditions on them. In short, the assumed fact that the average writer was more prone to data errors than the average publisher does not support the case that pro rata division is inherently unfair.[95]It is also helpful here to return to a point addressed earlier: what is needed to assess the alleged breach of competition law is an appropriate yardstick for comparison, i.e. a counterfactual. The discussion about the “class” based nature of the claim further illuminates the nature of any proposed counterfactual and enables something more to be said about it. The ultimate basis of the pleaded complaint is that the pro rata rule departs from the “true distribution” and is therefore unfair. In my judgment, it must follow that in order to be a relevant (“fairer”) counterfactual, the proposed distribution would have to represent a closer approximation to the “true distribution” or, in other words be more accurate. However, as already explained, the very problem that has given rise to this claim is the PRS’s inability to match, or in other words, make accurate distributions. I have already explained that no realistic case has been advanced that, given the absence of the necessary information, further litigation processes will lead to a more accurate counterfactual distribution rule. In my judgment, the failure of the PCR to formulate such a yardstick, and the absence of a realistic prospect of identifying one hereafter, fatally undermines the claim. As explained in [‎93] above, the black box royalties problem arises precisely from the absence of accurate information; if accurate information were available, the relevant royalties would be matched.[96]I also consider that there is force in the submission of counsel for the PRS that the PCR’s submissions in the appeal that there was a systemic bias against writer members as a class amount in substance to a case of discriminatory or differential treatment of different categories of member. Counsel for the PCR referred repeatedly to unequal treatment and systemic bias of the class of writers. However, the pleading does not set out a case of discrimination under s. 18(2)(c) (“applying dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage”, with equivalent provisions in Art. 102(c) of the Treaty on the Functioning of the European Union). I do not consider that a properly pleaded case has been advanced under s.18(2)(c).[97]In assessing the case that the PRS has arguably imposed unfair trading conditions on its writer members, I also consider that there is substantial force in two further points relied on by counsel for the PRS. The first is that the evidence shows that the governing principle applied by the PRS is to distribute royalties wherever possible on a work-by-work basis. The PRS makes reasonable and proportionate efforts and adopts processes to achieve this by chasing down matching data. There is no criticism of these efforts or processes. Far from supporting a case of an abuse of dominance, this shows that the PRS is seeking to do its best to achieve a fair distribution. The pro rata distribution of the black box royalties is its best attempt to approximate a fair outcome in cases of data failure. The second is that black box revenues have to be distributed somehow. At this stage the DC considers and approves the proposals of management. It has to act in the interests of all the members. The DC includes an equal representation of writers and publisher members and further non-executive directors. In carrying out its functions the DC considers the trade-off between accuracy and the cost of taking further steps to achieve it: see [‎‎13] above.[98]Neither of these points can be conclusive. The fact that the PRS does not act unlawfully in relation to the matched royalties pot does not entail that it has not imposed unfair terms in relation to the black box royalties pot. And the fact that the DC has approved a particular distribution cannot, of itself, mean that the decision to choose that rule is incapable of being an abusively unfair condition for the purposes of s. 18 of the Act. However, I agree with counsel for the PRS that the complaint in this case is far removed from cases where a dominant undertaking, freed from the shackles of competition, has been found to abuse its position by imposing conditions on trading parties that could not be justified in a free market.[99]It is also worth standing back and considering the substance of the PCR’s complaint. The PRS seeks to distribute on a work-by-work basis and does so where it can match royalties to specific works and members’ interests in respect of them. Sometimes it receives royalties but, after making proportionate efforts, cannot achieve a match because of data or recording failures. No member can claim an entitlement to a share of the funds, because the information required for matching is missing. The PRS cannot simply accumulate these funds, but has to share them in some way. Any chosen rule is bound to depart from the “true” – but unknowable – shares that members would receive in a perfect world. The PRS therefore has to do the best it can and its DC approves the distribution. As a result, members who share in the matched revenues get something more than they otherwise would get (i.e. a pro rata share from the black box). I can for instance see no reason at all why it would have been a contravention of Chapter II if the PRS had chosen instead to donate the black box royalties revenues to an appropriate charity. Under the pro rata rule members as a whole do better than that. They receive something that they cannot say they are otherwise strictly entitled to under the rules of the PRS. The Chapter II prohibition does not create a general law of consumer protection and it is not enough for a group of consumers (here members) to contend that things might have been arranged more favourably for them. The PCR needs to show that he has a realistically arguable case that the PRS has engaged in exploitative abuse by unfair conduct. I do not consider that the PCR has come close to doing so.[100]Counsel for the PCR referred to statements in well-known cases for the proposition that it is not generally appropriate to strike out or summarily dismiss a claim on assumed facts in a novel or developing area of law (see, e.g., Begum v Maran (UK) Ltd [2021] EWCA Civ 326, [2022] 1 ALL ER (Comm) 940 at [23]-[24]).[101]There are, on the other hand, many cases stating that it does the parties no favours to allow a case that lacks reasonable prospects of success to continue. If these proceedings were to go ahead they would be very expensive and time-consuming for the parties and tribunals. Counsel for the PCR did not to my mind persuasively identify any aspect of the law governing the Chapter II prohibition which is either novel or developing.[102]For the above reasons, in my judgment there is no reasonably sustainable case that the PRS’s decision to share the black box royalties pro rata involves the abusive imposition of any unfair trading conditions on its members within the meaning of the Chapter II prohibition.[103]I would therefore dismiss the second ground of appeal. The proceedings must therefore be dismissed and it is unnecessary to address the other issues, which concern the certification of the proceedings as collective ones.[104]I agree.[105]I also agree.