Lifestyle Equities CV & Anor v Royal County of Berkshire Polo Club Ltd & Ors (Rev1) [2023] EWHC 1839 (Ch) (19th July 2023)

Neutral Citation Number: [2023] EWHC 1839 (Ch) Case No: IL-2018-000115 IN THE HIGH COURT OF JUSTICE BUSINESS AND PROPERTY COURTS OF ENGLAND AND WALES INTELLECTUAL PROPERTY LIST (ChD) INTELLECTUAL PROPERTY EUROPEAN UNION TRADE MARK COURT Rolls Building Fetter Lane London, EC4A 1NL th July 2023 Before : THE HONOURABLE MR JUSTICE MELLOR - - - - - - - - - - - - - - - - - - - - - Between : (1) LIFESTYLE EQUITIES C. V. (2) LIFESTYLE LICENSING B. V. (both companies incorporated under the laws of the Netherlands) Claimants - and - (1) ROYAL COUNTY OF BERKSHIRE POLO CLUB LIMITED (2) MR DAVID BAXTER GENTLE AS THE PERSONAL REPRESENTATIVE OF THE ESTATE OF GRETA MAE MORRISON (3) JAMES TARA MORRISON (4) THE PARTNERSHIP (LICENSING) LIMITED (5) JONATHAN ERIC BOWER TOWNSEND (6) MAYS ZONA LIBRE S. A. (a company incorporated in Panama) (7) EMPRESAS POLAR S. A. (a company incorporated in Chile) (8) EMPRESAS HITES S. A. (a company incorporated in Chile) (9) TIENDAS PERUANAS S. A. (a company incorporated in Peru) (10) SEARS OPERADORA MEXICO, SA DE CV (a company incorporated in Mexico) (11) ABDUL GHANI MAMOUN TR LLC (a company incorporated in the UAE) Defendants - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Michael Edenborough KC and Thomas St Quintin (instructed by Brandsmiths ) for the Claimants Michael Silverleaf KC (instructed by Maitland Walker LLP ) for the Defendants Hearing dates: 10 th , 13 th -15 th , 20 th & 21 st June 2022 - - - - - - - - - - - - - - - - - - - - - APPROVED JUDGMENT I direct that pursuant to CPR PD 39A para 6.1 no official shorthand note shall be taken of this Judgment and that copies of this version as handed down may be treated as authentic. This judgment was handed down remotely by circulation to the parties’ representatives by email. It will also be released for publication on the National Archives and other websites. The date and time for hand-down is deemed to be Wednesday 19 th July 2023 at 2pm. ............................. THE HON MR JUSTICE MELLOR Mr Justice Mellor: Introduction .. The claims and counterclaims in more detail . The date for assessment . The Claimants’ case in more detail . The witnesses . Applicable law. Section 10(2) Assessment of the Likelihood of Confusion . Direct/indirect confusion . The Average Consumer Context and the ‘crowded’ market . Relevance of Co-existence Agreements . Section 10(3) and equivalents. The history of the BHPC brand . the Claimants’ “evidence” . Marginal Annotations . Bundle F . BHPC Sales Figures . Bundle F in more detail . Dream Project srl Catalogues & Look Books from other licensees . ‘Industry Awards’ Promotion of the BHPC brand in the UK and EU .. Interim conclusions as to the status of the BHPC brand . The history of RCBPC . Dates of first use of the RCBPC brand in various countries . Overall use . The market in which the two brands operate . The relevance of polo as a sport The growth and popularity of polo-themed brands . The evidence of Señor Garcia . The relevance of other polo-themed brands . The Cs’ case on the words ‘POLO CLUB’ Apparent Conflicts between the brands . Complaints from Licensees . F/9/160/pp2401-2408 . Ms Borycz’s evidence of confusion . The section 10(2) case . Overview - the Cs’ principal contentions . Average consumers . UK - analysis of similarity . EU - similarity analysis . Panama - similarity analysis . Mexico, Chile, Peru . UAE . Assessment of distinctive character Global assessment of the likelihood of confusion . Indirect Infringement Direct Infringement . Panama . Mexico, Chile, Peru . UK . Infringement of the EUTMs . Infringement under section 10(3) and/or art 9(1)(c). Passing off . What I have not relied on .. Joint and several liability . Applicable Principles . D1 as liable for the acts of its licensees . The other allegations of joint liability - summarised at [11 iv)] above. validity of marks . Conspiracy to injure by unlawful means . Conclusion .. Postscript regarding CPR 31.22(2) Applicable Principles . Analysis . Introduction 1. In essence this action is concerned with whether the second sign shown below (which belongs to the First Defendant (D1), ‘Sign 3’) conflicts with the first mark (which belongs to the Claimants or the Cs) in a number of different territories, namely, the UK, the EU, Chile, Panama, Peru, Mexico and the UAE: 2. In a number of trade mark oppositions (and cancellation actions) around the world, C1 has persuaded trade mark tribunals that there is a likelihood of confusion between the first mark and the second sign. In most (if not all) of these cases, the relevant tribunal either did not receive evidence of the situation in the market or did not take account of it. The Cs rely on these decisions as supporting a finding of a likelihood of confusion between the second sign and the first mark. 3. In this trial, the Defendants put in a considerable amount of evidence to establish how (in their submission) the Cs’ mark, the Defendants’ Sign and other ‘polo’ brands have continued to exist alongside each other in many countries around the world. Via this evidence, the Defendants pose what they characterise as a simple dilemma: i) On the one hand, if there really was a likelihood of confusion between C1’s mark and the Defendants’ Sign, confusion must have been occurring for years on a very widespread scale between all the ‘polo’ brands; ii) On the other hand, the existence of all the ‘polo’ brands in the market indicates that consumers are able to distinguish between them. If so, consumers also distinguish between the Cs’ mark and the Defendants’ Sign. So, they say, there is no likelihood of confusion. 4. For their part, the Cs take the following position: i) The Cs accept that both they and D1 have co-existence agreements with Ralph Lauren (‘RL’), the owner of the well-known Polo Ralph Lauren brand (‘RL Polo’), in which RL imposed certain restrictions on their respective branding but which permitted their current respective branding; ii) The Cs accept that, of the ‘polo-themed’ clothing brands in the market, the RL Polo brand is No.1 in this market and US Polo Association is No.2; iii) They assert that their brand is ranked No.3 in the market; iv) They assert that those three brands are able peacefully to co-exist; v) Despite that, they assert there exists a likelihood of confusion between C1’s mark and the Ds’ Sign 3. 5. The situation is not as simple as either side suggests and their positions beg the main question I have to decide, which is whether there exists a likelihood of confusion between the Cs mark and the Ds sign(s) in various countries. So far as is possible on the evidence led in this trial, it is necessary to assess the situation in each country in issue. However, as will appear, it is only possible or feasible to make the assessment in relation to certain groups of countries. 6. Although I have highlighted the key issue for determination, as is regrettably common, the parties have managed to raise a significant number of issues for determination, some of which descend into minute details found in the evidence and documents. Furthermore, the issues in these trade mark cases often require a significant citation of caselaw (or summaries thereof) so that the Court is correctly oriented to decide each issue. 7. Although this case can be decided with an application of established principles, it does raise in a stark form the scope of protection which is or should be afforded to a registered trade mark which is used in a crowded market i.e. a market in which there are several trade marks using similar motifs and which incorporate the same or similar words. The overall question in such circumstances of ‘How far does the exclusive right which is conferred by registration extend?’ can only be answered by addressing the conventional approach laid down in the CJEU caselaw. The claims and counterclaims in more detail 8. C1 is a Dutch company and the registered proprietor of all the registrations of the mark shown above in all countries relevant to this claim, although it does not own the mark worldwide (specifically not in the USA). C2 is also a Dutch company and the exclusive licensee of all the marks in issue. Mr Edenborough KC leading Mr St Quintin argued the case for the Cs. 9. All claims against the Sixth to Eleventh Defendants have been stayed, so all references to ‘the Defendants’ (‘the Ds’) are to the First to Fifth Defendants (hereafter D1-D5, as appropriate). D1 is the company which operates the Royal County of Berkshire Polo Club. The Club was founded in 1985 by the late Bryan Morrison (then the manager of Pink Floyd). His widow (now deceased) was the Second Defendant, now represented by her personal representative, Mr Gentle, who is a solicitor. The Third Defendant is the son of Mr and Mrs Morrison and the current chairman of the Club. 10. Until around November 2018, D4 was D1’s licensing agent, with D5 its sole director. D4 and D5 have no ongoing connection with the matters in dispute and have joined forces with D1-3 to defend the claim. Hence, D1-5 were represented by Mr Silverleaf KC, instructed by Maitland Walker LLP. 11. The principal claims are for infringement of the Cs’ registered trade marks, registered in the form shown above. The Cs allege: i) Infringement in the UK and EU due to the existence of a likelihood of confusion, also in Chile, Panama, Peru, Mexico and the UAE (‘the Overseas Territories’) by acts carried out in those territories. ii) Section 10(3)/Article 9(1)(c) infringement in the UK and EU. iii) Passing off in the UK by D1 and D4. iv) A variety of allegations of joint liability which I consider below, but the principal claims are that: a) D2 & D3 are jointly liable with D1; b) D5 is jointly liable with his company D4; c) D1-3 are jointly liable for the acts of D4-5 and of D6-D11; d) D4-5 are jointly liable for the acts of D1, D2-3, and D6-11 v) Unlawful means conspiracy carried out by D1-D5; and vi) That certain UK registered trade marks owned by D1 are invalid or should be revoked for lack of genuine use. Although bad faith was originally a ground of invalidity asserted by the Cs, this ground was dropped during trial, leaving the relative grounds of invalidity. 12. The Ds counterclaim for orders that the Cs’ UK and EU registered trade marks are invalid. At the start of trial, the Ds sought relief for unjustified threats of trade mark infringement and alleged abuse of process, but those claims were also dropped during the trial. 13. Although the principal claims concern the Ds’ Sign 3, set out above, other signs used by D1 are in issue. By the time of closing arguments, the Cs indicated I could concentrate on Sign 3 (shown above), a variation thereof used in Germany which featured a straight mallet, and Sign 1 which looks like this: 14. D1 admits: i) Having used Signs 1 and 3 in the UK; ii) The use of Sign 3 (with a straight mallet) by its licensees in Germany; iii) The use of Sign 1 by its licensees in Latin America; iv) The use of Sign 3 by its licensees in Latin America outside Mexico. 15. The Ds have abandoned the use of Sign 1, except in Mexico, due to the terms of their co-existence agreement with RL. Sign 2 is the picture from Sign 1 with no words, although the Ds say it has never been used. The Ds deny use of Signs 4 and 5, set out below. The Cs point out that D1 obtained a UK registration featuring Sign 4, but that does not prove it has been used: 16. Realising that the resolution of all the various claims of trade mark infringement would be extremely complicated (and costly), fortunately the parties agreed to simplify matters in certain respects: i) First, that all allegations of trade mark infringement (including in respect of the Overseas Territories) are to be determined by the application of UK/EU law, save in respect of certain specific defences to infringement under the national law of one of the Overseas Territories - those defences to be considered after judgment if they remain relevant. Those defences apart, all issues of liability are to be determined by this trial. ii) Second, that questions of infringement are to be assessed using a single relevant date, the date of the claim form which was 25 th June 2018. This apparent agreement (made in correspondence in June 2021) gives rise to the first issue I have to resolve. The date for assessment 17. Both sides noted this agreement in their Opening Skeletons, but it was only in their Closing Skeleton Argument that the Cs spelt out their contention that the agreement meant that infringement is to be assessed as if it commenced on the date of the claim form i.e. that the use of the allegedly infringing signs only began in June 2018. By contrast, the Ds said the agreement simply meant that infringement was to be assessed at the date of the claim form, in effect on the basis of the situation which then existed. 18. There is a very significant difference between these two positions. If the Cs’ contention was correct, this would be a very different case and one in which the previous opposition and cancellation decisions between the parties would be highly influential. It would also be a case in which the Ds had effectively relinquished their main argument. For the reasons which follow, I do not believe that the Ds did relinquish their main argument by making this agreement. Furthermore, there was no sign, in the way the Cs conducted their case between June 2021 down to the service of their Closing Skeleton Argument, that the Cs considered the Ds had done so either. Certainly, if this was the Cs’ case, it should have been made front and centre in their opening of the trial. As I said, it emerged only in closing. 19. In order to resolve this issue, it helps to understand the background against which the alleged agreement was made. A short chronology provides the relevant background. 20. As formulated in the original Particulars of Claim against D1-D11, this action involved a very complex series of claims, in that the alleged infringements had to be assessed in each country in issue which would require (at least) evidence of foreign law, the date when the alleged infringement commenced in each country and the relevant circumstances. After D1-3 had served their defence and counterclaim, Morgan J. had to deal with some interim skirmishing in which both sides sought to simplify this action. The details are set out in his judgment of 14 th December 2018, the neutral citation of which is [2018] EWHC 3552 (Ch) , resulting in his Order stamped on 11 th March 2019. In very brief summary, D1-3 sought to strike out various aspects of the claims relating to D6-D11 and the claim for conspiracy to injure by unlawful means. Their applications failed. For their part, the Cs sought summary judgment or strike out of various counterclaims brought by D1-3. The Cs’ application succeeded in part, leading to amendments to the Claim Form and to the Particulars of Claim, and certain parts of the Defence being struck out. 21. The Cs brought a further application to strike out or for summary judgment which was refused by Pat Treacy (sitting as Deputy Judge of the Chancery Division) on 17 th July 2019. As her judgment on consequential matters indicates [2019] EWHC 2413 (Ch) , her holding Order of 17 th July 2019 required D4-5 to make various amendments to their pleadings as agreed between the parties at the start of the hearing of the application and required the Cs to identify portions of the Defence and Counterclaim of D4-5 in respect of which they sought further particulars or amendment. 22. Eventually, the Costs and Case Management Conference came before Master Teverson, resulting in his Order dated 31 st March 2021. For present purposes, the salient points in his Order are as follows: i) First, he ordered a split trial of liability and quantum. ii) Second, the fifth recital in his Order recorded that the Cs and D1-5 had already agreed that questions of infringement of non-UK or EU trade marks were to be assessed on the basis of UK law. iii) Third (as also recorded in that fifth recital), provision was made in the Order for the Cs to serve written particulars of ‘the ways and the extent to which the Claimants allege that the characteristics of the average consumer in each of the non-UK or EU territories in which infringement is alleged differs from the UK or EU average consumer’, and for D1-5 to respond, with permission to apply for further directions. iv) Fourth, provision was also made for the Cs to supply to D1-5 ‘a specific request for information designed (a) to reformulate the issues for disclosure in respect of the alleged joint and several liability of [D1-D5]… and /or (b) to add a further issue or further issues in respect of the dates on which the Signs complained of in the Amended Particulars of Claim were used in each relevant territory’, with consequential provisions allowing the Cs to apply for amendment to the Disclosure Review Document. 23. From the correspondence made available to me, it is apparent that the parties continued to discuss the three outstanding issues which concerned (a) the average consumer in foreign territories (b) disclosure regarding joint liability and (c) disclosure regarding ‘dates of usage’. 24. The discussion over ‘dates of usage’ culminated in the solicitors for D1-5 writing on 16 June 2021 and agreeing the proposal set out in the letter of 14 June 2021 from the solicitors for the Cs that ‘alleged infringements can be assessment [sic] as at the date of the claim form’ with the result that the related item could be removed from the DRD. 25. Unfortunately, it is clear to me that when this point was agreed in correspondence, the parties were not ad idem as to what was meant by ‘assessed’, as is apparent from what I have set out above. 26. The Cs appear to have realised that there might be difficulties with this supposed agreement, so in cross-examination of Mr Amoore, Counsel asked about the dates of first use, mostly by reference to the dates of licence agreements. This was a proportionate way to proceed. It would also have been a proportionate and sensible way to resolve the issue over disclosure back in June 2021. 27. I have summarised this part of Mr Amoore’s evidence below, in the context of my findings on the Ds’ business under the RCBPC brand. The Claimants’ case in more detail 28. The Cs suggest this is a case for a straightforward application of established principles which, they say, establish a likelihood of confusion between their Mark and the Ds’ Signs. The Cs also suggest that their trade under the Mark has been extensive, giving rise to a reputation, that use of the Ds’ Signs would plainly give rise to a link being made with the Mark, and they claim damage in the form of dilution of the distinctiveness of their Mark. 29. The Cs (indeed both sides) considered that passing off stands or falls with the allegation of likelihood of confusion. 30. The Cs say their allegations of joint liability are made good on the evidence. 31. Although the Cs continued to press their claim based on the alleged unlawful means conspiracy, I found it difficult to understand how that claim could add anything since the unlawful means relied upon are the infringements of registered trade mark and/or passing off and the facts said to give rise to joint liability are those relied upon in support of the alleged conspiracy. Those points notwithstanding, I will assess this claim later, along with various other subsidiary issues which remain live. 32. In their attempts to stick to their position, the Cs sought to deflect the Ds’ case in several different ways. The common theme however was that the Cs took various extreme positions which did not represent the way the Ds put their case. For example, in their Opening, the Cs characterised the Ds as arguing that horse-and-player devices have no distinctive character at all and submitted that if the Ds failed in that argument, the horse-and-player element of the Marks should be held to be distinctive. A much more nuanced approach is required, as I shall explain. The witnesses 33. The Cs had two witnesses: Ms Aleksandra Borycz and Mr Eli Haddad. 34. Mr Haddad is the Managing Director of C2, and was until the end of 2018 the CEO of C1, after which point his son, Daniel, took on that role. As Mr Edenborough submitted, it certainly appears that Mr Haddad remains heavily involved in C1’s business. 35. In his first witness statement, Mr Haddad gave evidence about the history of BHPC, some details about the Cs’ trade, evidence about the Cs’ sales and promotion in the UK and the EU, complaints the Cs received from licensees about sales of Ds’ products (the Cs say these evidence examples of actual confusion), and his dealings with Ds. In his second witness statement, Mr Haddad responded to some evidence from Mr Amoore about a conversation between Mr Haddad and D5. 36. I assess the content of Mr Haddad’s evidence later. It is fair to say that he is a very passionate advocate for the BHPC brand. The extent of his passion means I must treat his evidence with some caution. Indeed, in closing Mr Edenborough KC accepted that Mr Haddad was ‘not a witness without shortcomings in the way he gave his evidence. He was often verbose…’. Indeed he was, giving quite long speeches at times, which were not answers to the question put. The cross-examination revealed a number of points on which Mr Haddad had no personal knowledge but he was nonetheless prepared to assume and convey as fact whatever he perceived was in the best interests of the brand. 37. Ms Borycz is the global brand manager for the BHPC brand and has been a consultant to the owners of the brand for 15 years. In her first witness statement she gave evidence about her role and about certain incidents which the Cs allege amount to instances of actual confusion which occurred at the Albrook Mall in Panama in 2014 and 2015 and at a separate department store (Stevens) also in Panama in 2018. In her second witness statement, she corrected an error in her first regarding the status of the Cs’ licensee’s store at the Albrook Mall. 38. Ms Borycz resides in Italy I believe and gave her evidence via a rather inconsistent videolink. As her cross-examination continued, she grew somewhat agitated about the time it was taking because of her young baby. She was also giving evidence about events in 2014, 2015 and 2018 and the passage of time did not help matters. Again, I have to consider her evidence in detail below. Although nowhere near as passionate as Mr Haddad, I was struck by her evident loyalty to the BHPC brand. 39. The Ds relied on witness statements from six witnesses. 40. Señor Garcia is Vice-President of CAA Global Brand Management, the business which has acted as D1’s licensing agent since January 2018, effectively taking over from D4. In his witness statement, he gave evidence about CAA, about Latin American markets relevant to this claim and his awareness of the ability of the public in those markets to distinguish between different polo-themed brands. He was a ‘trade’ witness. 41. Señor Garcia gave his evidence in English, assisted with the meaning of the questions put to him by an interpreter. That, together with a videolink which was not of the best quality, presented some challenges. His English was reasonable but he was unsurprisingly not always able to understand some of the more precise questions which Counsel put to him. Notwithstanding these challenges, Señor Garcia was able to convey what he wanted to say. The Cs demonstrated that there were certain mistakes in Señor Garcia’s evidence, which he acknowledged. I assess his evidence further below. 42. Mr Morrison (D3) is the Chairman and CEO of D1. In his first witness statement (made for one of the interim applications), he gave evidence about his role within D1, some background about the Club, and about his role in D1’s overseas merchandising activity. In his second witness statement he gave evidence about the date on which D1’s relationship with D4 ended (which was after an extended run off period), and about a meeting that he and Mr Amoore had with Mr Haddad on 12 March 2012. 43. Mr Amoore is the general manager of D1. In his first witness statement he gave evidence about D1’s background, his role within D1, his role in D1’s overseas merchandising activities, how he reports to its directors, D1’s use of its trade marks in the UK, the status of disputes in various jurisdictions between D1 and Cs, his views of the intentions of D1 and its directors, his lack of receipt of reports of confusion, his limited knowledge of other brands in the markets in issue in this case, and about co-existence agreements entered into by D1 and third parties, and by Cs and third parties. His second witness statement covered the date on which D1’s relationship with D4 ended, and a meeting that he and Mr Amoore had with Mr Haddad on 12 March 2012. He also gives evidence about the knowledge he had of Cs’ Brand prior to that meeting. 44. Ms Virgin is one of the solicitors acting for the Ds. Her witness statement refers to an exhibit prepared by D5 and gives evidence about her efforts to verify a sample of the contents of that exhibit. She was briefly cross-examined. 45. Mr Morrison, Mr Amoore and Ms Virgin were all straightforward witnesses, whose evidence I accept. 46. The evidence from the Ds’ two final witnesses was accepted without cross-examination: i) Ms Muller gave evidence about the creation of an Excel worksheet pivot table from data provided to her. That table shows the sales reported by Ds’ licensees. ii) Ms Tak gave evidence about the correct description of figures she provided to Mr Amoore and potential explanations for the differences between those figures and those provided by Ms Muller. Applicable law. 47. The applicable law was very largely agreed. In one sense, this case calls for an application of the well-established principles which underpin the causes of action for infringement under section 10(2)/article 9(1)(b) and under section 10(3)/article 9(1)(c). Although I was and am very familiar with these principles, it is useful nonetheless for me to remind myself of them. Section 10(2) Assessment of the Likelihood of Confusion 48. I can start with the following summary of the approach to be taken to the assessment of the likelihood of confusion. It is a succinct summary of a number of CJEU rulings and decisions on the provisions of EU TM law which are concerned with the likelihood of confusion. The summary was developed in the context of oppositions decided at the UK IPO but was endorsed for use in the infringement context by the Court of Appeal in Specsavers v Asda [2012] FSR 19 (‘ Specsavers’ ), in Maier v ASOS [2015] EWCA Civ 220 at [75], in Comic Enterprises v Twentieth Century Fox [2016] EWCA Civ 41 (‘ Comic Enterprises’ ) per Kitchin LJ at [31], and in Liverpool Gin Distillery v Sazerac Brands [2021] ETMR 57 (CA) per Arnold LJ at [9]. This citation is taken from Liverpool Gin (noting that this contains the correct sub-paragraph (k) - see Arnold LJ in Match v Muzmatch [2023] EWCA Civ 454 at [30]): ‘(a) the likelihood of confusion must be appreciated globally, taking account of all relevant factors; (b) the matter must be judged through the eyes of the average consumer of the goods or services in question, who is deemed to be reasonably well informed and reasonably circumspect and observant, but who rarely has the chance to make direct comparisons between marks and must instead rely upon the imperfect picture of them he has kept in his mind, and whose attention varies according to the category of goods or services in question; (c) the average consumer normally perceives a mark as a whole and does not proceed to analyse its various details; (d) the visual, aural and conceptual similarities of the marks must normally be assessed by reference to the overall impressions created by the marks bearing in mind their distinctive and dominant components, but it is only when all other components of a complex mark are negligible that it is permissible to make the comparison solely on the basis of the dominant elements; (e) nevertheless, the overall impression conveyed to the public by a composite trade mark may, in certain circumstances, be dominated by one or more of its components; (f) and beyond the usual case, where the overall impression created by a mark depends heavily on the dominant features of the mark, it is quite possible that in a particular case an element corresponding to an earlier trade mark may retain an independent distinctive role in a composite mark, without necessarily constituting a dominant element of that mark; (g) a lesser degree of similarity between the goods or services may be offset by a greater degree of similarity between the marks, and vice versa; (h) there is a greater likelihood of confusion where the earlier mark has a highly distinctive character, either per se or because of the use that has been made of it; (i) mere association, in the strict sense that the later mark brings the earlier mark to mind, is not sufficient; (j) the reputation of a mark does not give grounds for presuming a likelihood of confusion simply because of a likelihood of association in the strict sense; (k) if the association between the marks creates a risk that the public might believe that the respective goods or services come from the same or economically-linked undertakings, there is a likelihood of confusion.’ 49. The Cs submitted that paragraphs (b), (g) and (h) were particularly important in this case. 50. In any particular case, this summary often needs to be supplemented with caselaw dealing with particular points which arise. The Cs drew my attention to the following points. 51. First, the guidance from the General Court in Gérard Meric v OHIM , Case T-133/05, [2006] ECR II-2737 on the comparison of goods or services: ‘29. …goods can be considered identical when the goods designated by the earlier mark are included in a more general category, designated by the trade mark application or when the goods designated by the trade mark application are included in a more general category designated by the earlier mark.’ 52. Second, the significance of the presence or absence of actual confusion. The caselaw shows that this is a nuanced issue which depends on the circumstances. It is sufficient for me to refer to two passages. In Maier v ASOS [2015] ETMR 26 , at [80], Kitchin LJ said the following: ‘If the mark and the sign have both been used and there has been actual confusion between them, this may be powerful evidence that their similarity is such that there exists a likelihood of confusion. Conversely, the absence of actual confusion despite side by side use may be powerful evidence that they are not sufficiently similar to give rise to a likelihood of confusion. This may not always be so, however. The reason for the absence of confusion may be that the mark has only been used to a limited extent or in relation to only some of the goods or services for which it is registered, or in such a way that there has been no possibility of the one being taken for the other. So there may, in truth, have been limited opportunity for real confusion to occur.’ 53. On the same point, the Cs also relied on passages from Jack Wills v House of Fraser [2014] FSR 39 ; BMW v Technosport [2017] EWCA Civ 779 at [24]; and Fine & Country Ltd v Okotoks Ltd [2012] EWHC 2230 (Ch) , (a passing off case). It is sufficient to cite this passage from the judgment of Arnold J. in Jack Wills at [99]: ‘99 No evidence of actual confusion. Counsel for House of Fraser relied strongly on the absence of any evidence of actual confusion. As I have said in a number of judgments, absence of evidence of actual confusion is not necessarily fatal to a claim under art.5(1)(b). The longer the use complained of has gone on in parallel with use of the trade mark without such evidence emerging, however, the more significant it is. Other relevant factors are the scale of the use complained of and the likelihood of actual confusion being detected.’ Direct/indirect confusion 54. As Arnold LJ observed in Liverpool Gin at [10]: ‘It is well-established that there are two main kinds of confusion which trade mark law aims to protect a trade mark proprietor against (see in particular Sabel BV v Puma AG (C-251/95) [1997] ECR I-6191 at [16]). The first, often described as “direct confusion”, is where consumers mistake the sign complained of for the trade mark. The second, often described as “indirect confusion”, is where the consumers do not mistake the sign for the trade mark, but believe that goods or services denoted by the sign come from the same undertaking as goods or services denoted by the trade mark or from an undertaking which is economically linked to the undertaking responsible for goods or services denoted by the trade mark.’ 55. Arnold LJ went on to detail, by reference to L. A. Sugar and Cheeky Italian , some now well-known dicta concerning the concept of indirect confusion. I will not lengthen this judgment by reproducing his paragraphs [11]-[13] but I have those points well in mind. The Average Consumer 56. The characteristics of the average consumer are well-known. The average consumer is deemed to be a reasonably well-informed and reasonably observant and circumspect. 57. Based on Interflora Inc v Marks & Spencer [2015] FSR 10 (‘ Interflora ’), the Cs submitted that a likelihood of confusion exists if any significant number of people with the characteristics of the average consumers of the goods in issue is likely to be confused, citing this passage at [129]: ‘… in light of the foregoing discussion we do not accept that a finding of infringement is precluded by a finding that many consumers, of whom the average consumer is representative, would not be confused. To the contrary, if, having regard to the perceptions and expectations of the average consumer, the court concludes that a significant proportion of the relevant public is likely to be confused such as to warrant the intervention of the court then we believe it may properly find infringement.’ 58. I consider it is helpful to refer to a distillation of the points made by Kitchin LJ in the slightly later case of Comic Enterprises , where he drew together various points made in his earlier judgments in Specsavers and Interflora . The distillation is set out in Kerly (16 th Edition, 2018 and 1 st Supplement, 2020) at 3-006. I omit the first three points, because they correspond to paragraphs (a) to (c) cited above: ‘(4) In assessing the likelihood of confusion arising from the use of a sign the court must consider the matter from the perspective of the average consumer of the goods or services in question and must take into account all the circumstances of that use that are likely to operate in that average consumer’s mind in considering the sign and the impression it is likely to make on him. The sign is not to be considered stripped of its context. (5) The average consumer is a hypothetical person or, as he has been called, a legal construct; he is a person who has been created to strike the right balance between the various competing interests including, on the one hand, the need to protect consumers and, on the other hand, the promotion of free trade in an openly competitive market, and also to provide a standard, defined in EU law, which national courts may then apply. (6) The average consumer is not a statistical test. The national court must exercise its own judgment in accordance with the principle of proportionality and the principles explained by the Court of Justice to determine the perceptions of the average consumer in any given case in the light of all the circumstances. The test provides the court with a perspective from which to assess the particular question it has to decide. (7) In a case involving ordinary goods and services, the court may be able to put itself in the position of the average consumer without requiring evidence from consumers, still less expert evidence or a consumer survey. In such a case, the judge can make up its own mind about the particular issue it has to decide in the absence of evidence and using its own common sense and experience of the world. A judge may nevertheless decide that it is necessary to have recourse to an expert’s opinion or a survey for the purpose of assisting the court to come to a conclusion as to whether there is a likelihood of deception. (8) The issue of a trade mark’s distinctiveness is intimately tied to the scope of the protection to which it is entitled. So, in assessing an allegation of infringement under art.5(1)(b) of the Directive arising from the use of a similar sign, the court must take into account the distinctiveness of the trade mark, and there will be a greater likelihood of confusion where the trade mark has a highly distinctive character either per se or as a result of the use which has been made of it. It follows that the court must necessarily have regard to the impact of the accused sign on the proportion of consumers to whom the trade mark is particularly distinctive. (9) If, having regard to the perceptions and expectations of the average consumer, the court concludes that a significant proportion of the relevant public is likely to be confused such as to warrant the intervention of the court then it may properly find infringement.’ 59. As a footnote says, point (4) comes from Specsavers at [87] and points 5-9 are Kitchin LJ’s own summary of the more detailed consideration of the characteristics of the average consumer in Interflora from [107]-[130]. 60. The Cs also relied on a passage from the judgment of Fancourt J. in Sazerac Brands v Liverpool Gin Distillery [2020] EWHC 2424 (Ch) at [49], not challenged on the appeal which upheld his judgment - Liverpool Gin Distillery v Sazerac Brands [2021] EWCA Civ 1207 , [2021] ETMR 57 . Context and the ‘crowded’ market 61. The significance of the correct context was a major issue in Specsavers and is reflected in point (4) above. It is also relevant to note how Kitchin LJ deployed ‘context’ in Specsavers itself. The Court decided it was necessary to address Specsavers’ ‘cumulative effect’ argument in order to address the allegation of infringement of the Specsavers Shaded and Unshaded logo marks (set out in Appendix 1 to the Judgment). The CA did not consider at this point the allegation of infringement of the wordless logo mark, since that was stayed pending the reference to the CJEU. The cumulative effect argument brought together combinations of one or both of the straplines (“Be a real spec saver at Asda” and ‘Spec savings at Asda” - see at [4]) and one or both of the two versions of the Asda Logo (set out in Appendix 2). 62. At [161], Kitchin LJ said: ‘161. For the reasons I have given at [77] to [87] above, I believe that in assessing the likelihood of confusion under Article 9(1)(b), the court must take into account all the circumstances of the allegedly infringing use that are likely to operate in the average consumer’s mind in considering the sign and the impression it is likely to make on him. It follows that it is appropriate to consider the cumulative effect of the signs in issue, subject to the requirement explained by this court in L’Oréal that the test remains founded upon the mark as registered. In my judgment, the position under Article 9(1)(c) must be the same.’ 63. His conclusion came at [164]: ‘164. In assessing whether the use of the Asda logo has taken unfair advantage of the distinctive character or repute of the Specsavers Shaded and Unshaded logo marks it is of course necessary to carry out a global assessment. So I must also have regard to all relevant circumstances, including the significant reputation attaching to Specsavers’ marks, the fact that the goods are identical and the fact that it was Asda’s intention to target this campaign at Specsavers and to convey the message that Asda offered good, if not better, value. Taking all these matters into account I am satisfied that the use of the Asda logo (in both its forms) as part of the campaign including the straplines was such as to create a link with Specsavers Shaded and Unshaded logo marks in the mind of the average consumer; that this link did confer an advantage upon Asda; and that this advantage was unfair and without due cause. As in the case of the straplines, the use of the Asda logo permitted Asda to benefit from the power of attraction, reputation and the prestige attaching to Specsavers and its Shaded and Unshaded logo marks and to exploit without paying compensation the marketing efforts which Specsavers has made. I would therefore find infringement of the Shaded and Unshaded logo marks by the use of the Asda logo (in both its forms) as part of the composite advertising and promotional campaign.’ 64. The Cs relied on two passages from the Judgment of Daniel Alexander QC (sitting as a Deputy Judge of the Chancery Division) in Planetart LLC v Photobox Ltd [2020] ETMR 35 . As the Cs submitted, the Deputy Judge considered context at [23]-[25] and [164]. Having cited from Specsavers at [85]-[87], which included an extract from the judgment of Arnold J. in Och-Ziff Management Europe Ltd v Och Capital LLP [2010] EWHC 2599 (Ch) at [87]-[88], he continued at [24] & [25], returning to context at [164]: ‘24. In my view, Arnold J, as he then was, in Och-Ziff was saying that the CJEU took the view that, in considering infringement of a registered trade mark, it was not appropriate to look so broadly at the context that use which was prima facie infringing was nonetheless to be regarded as non-infringing because other, separate, acts of the defendant had countered actual deception. An extreme example is where a defendant uses a well-known brand for counterfeit goods but nonetheless makes it very clear that the goods are in fact counterfeit so that no actual purchaser is confused. There may be no actual confusion as a result of the use of the sign but there is nonetheless trade mark infringement because the court must focus on the use of the sign in question not the other statements by the defendant as to the trade origin of the goods. 25. Accordingly, while it is right to take the context in which the given sign will be seen into account, I am not persuaded that it would be right to expand the view so broadly as to take account of the fact that a given sign only appears in this case after a different sign has been used. To that extent, each use of the signs must be examined separately in what might be described as its “local” context.’ ‘164. I have also considered, in accordance with the guidance in Specsavers , and generally whether there is anything about the context of presentation of the marks which negates that result. In my view there is not. It is also necessary for the court to be cautious in adopting an overly expansive approach to taking account of context in a trade mark claim. One purpose of registered trade mark protection (in which it is distinguished from passing off) is to provide an element of exclusivity in the use of a registered mark, regardless of the wider context in which it is used, so long as the conditions for protection are fulfilled.’ 65. Based on these passages, the Cs submitted as follows: i) First, that it is not appropriate to expand the context that is to be taken into account to include other, separate acts of the alleged infringer. ii) A fortiori , it is not appropriate to expand the context to include separate acts of people other than the alleged infringer. iii) That means, therefore, that the separate acts of other polo brands is, as a matter of law, irrelevant. iv) That negates completely Ds’ position on the relevance of other polo brands purportedly existing in the marketplace. 66. As far as I could tell, the Ds did not engage with the Cs’ arguments as to context but it seems to me that these submissions confuse two separate but related concepts and involve a non-sequitur between the first two and the last two points. The concepts are related in the sense that they both involve things that impinge on the mind of the average consumer and (may) influence the result. 67. There are sound policy reasons for not taking an over-expansive view of the context of the allegedly infringing use. These can be readily understood in the examples which the Deputy Judge had in mind in [24]. The use of ‘Fake Rolex’ or ‘Imitation Louis Vuitton’ does not escape infringement of the famous marks. 68. Although, as I have indicated, the Cs suggested this case was all about context, the Ds put their case differently. Instead, the Ds submitted the key here was to focus on the nature of the mark, the message it conveys to the consumer and hence on its distinctive character. Implicitly, the Ds therefore agreed with the notion that the relevant context was ‘local’, and so do I. 69. Taking a step back from the detail, if I assume for a moment that RL Polo, USPA and all the other third party ‘polo’ brands had never existed and BHPC was the first ‘polo’ brand which created the market and Sign 3 was freshly launched onto the market (for this purpose, assume UK), the infringement action would look very different - in short, it would be far more likely to succeed. 70. Instinctively, the long-standing presence of RL Polo, USPA and possibly other third party ‘polo’ brands must create a different situation. I agree with the Ds that it is necessary to assess the nature of the Cs mark, what it conveys to the average consumer and its distinctive character in this market which can be characterised as somewhat crowded with ‘polo’ themed brands. This is not a ‘context’ issue, and the distinction is clear: context is concerned with an examination of the use complained of, whereas the Ds are saying that it is the Cs mark which brings to that examination the relevance of other ‘polo’ brands in the market, provided they impinge on the way in which the average consumer views and recalls the Cs’ mark. 71. The issue of a crowded market has been touched upon before but only in circumstances where it was found that the registered trade mark in question was distinct from other similar figurative elements used in the market. For example: i) In Specsavers itself, on the return to the Court of Appeal [2014] EWCA Civ 1294 , following the ruling of the CJEU (Case C-252/12), a settlement with Asda had been reached (which meant that the Court did not need to decide the issue of infringement of the Wordless Logo mark) and the only issue was whether Specsavers had made genuine use of the Wordless Logo mark. In the unusual circumstances of that case, the Court of Appeal found that genuine use of the Wordless Logo mark had been made through use of the Shaded logo mark (which included the word Specsavers). The second reason for their conclusion was because the Specsavers overlapping ellipses were themselves distinctive, as against the various logos of other opticians which used depictions of spectacles or two lenses: ‘25. Second, it is a particular feature of this business that none of the major competitors of Specsavers has a logo which is remotely similar to the Shaded logo mark or the Wordless logo mark. The judge had before him in the evidence of Mr Richard Holmes the marketing director of Specsavers Optical Group, the third claimant, reproductions of the logos of Boots Opticians, Optical Express, Dollond and Aitchison and Vision Express, and each of them is quite different from the Shaded logo mark and the Wordless logo mark. This is not a case in which it can be said that Specsavers are seeking to secure for themselves a monopoly in a relatively banal or commonly used background. Indeed Dame Mary Perkins, one of the founders of Specsavers, explained in her evidence that she chose the Wordless logomark, comprising as it does overlapping elipses, precisely because she felt it was more abstract than either a pair of glasses or an image of two eyes, and so would be immediately recognisable by the public as denoting Specsavers.’ ii) In Jack Wills Ltd v House of Fraser (Stores) Ltd [2014] EWHC 110 , Jack Wills’ mark comprised the simplest version of the Jack Wills logo, which Arnold J. described as consisting of a silhouette of a pheasant with a top hat and cane. The Ds’ alleged infringement was ‘the Pigeon Logo’. It is apparent that the Judge received evidence of use of other ‘bird logos’ in the UK clothing market and he dealt with their relevance in this passage: ‘87 Distinctiveness of the Trade Marks. Counsel for Jack Wills submitted that the Trade Marks were inherently very distinctive. Counsel for House of Fraser did not suggest that the Trade Marks were devoid of distinctive character. His primary submission was that the Trade Marks and the Pigeon Logo were distinctively different from each other. I shall consider that question below. In the alternative, he submitted that, if the differences between the Trade Marks and the Pigeon Logo did not suffice to enable the average consumer to distinguish between them, then it followed that the Trade Marks were insufficiently different from other bird logos to possess distinctive character. I do not accept this argument. Leaving aside the fact that it ignores the difference between the dates at which the validity of the Trade Marks and the issue of infringement fall to be assessed, the Pigeon Logo is closer to the Trade Marks than the Trade Marks are to any other bird logo of which there is evidence of use in the United Kingdom. Taking account of the other bird logos of which there is evidence of use, I consider that the Trade Marks have a substantial degree of inherent distinctive character. An important factor in the distinctive character of the Trade Marks is their anthropomorphic aspect, and in particular the fact that the bird is equipped with accessories associated with an English gentleman.’ 72. I take comfort and reassurance that both Judges (both with very extensive experience in the law of trade marks) dealt with what I have termed the ‘crowded market’ as impacting (or not, as the case may be) on the distinctiveness of the claimant’s registered trade mark. 73. Finally, I also note that in Lifestyle Equities CV v The Copyrights Group Ltd [2021] EWHC 1212 (‘ Greenwich Polo ’), Marcus Smith J. had to deal with an argument which is similar to the point which the Ds run here. When holding that he could not regard the marks as similar, he started by considering the significance of the polo and horse rider motif generally (see his footnote 47) (he did so because of one of the Cs’ arguments): ‘The figurative elements are different. I should say, at the outset, that I regard the figurative part of both the Logo and Greenwich Sign 3 as the least significant element of these marks. The fact is that the rider (or riders) on a polo horse (or horses) with polo mallet (or mallets) does not do anything more than evoke the sport of polo. This is not a case of a Nike “swoosh” or some other purely figurative sign that links to or evokes a particular brand or producer. As Mr Durbridge makes clear in his evidence, many brands, clubs and entities deploy the polo horse and rider motif in their branding, but always with an additional word element to designate club, origin or brand. The Claimants objected to this evidence on the ground that all that was in issue was the similarity (or otherwise) between the Logo and Greenwich Sign 3, and the resultant risk of confusion to consumers. That of course is right, but irrelevant. The point being made by the Defendants was that the polo horse and rider motif was generic as a reference to the sport in general, and not to any particular brand. The Defendants put the point as follows in their written opening submissions: “11. [The Claimants] object to all the Signs. Their position appears to be that any branding with a polo player on a horse infringes, however many horses and whichever way they are facing. [The Claimants] consider their exclusive monopoly to extend to parties who “imitate the appearance, the spirit and the style of the Brand”… 12. [The Defendants’] position is that the respective marks are not similar enough, but there is also no likelihood of confusion because there is a crowded and well developed market for consumer items, in particular clothing, using logos of polo players on horses. Consumers will be familiar with there being various brands on the market and appreciate that not all of them come from the same brand owner. Polo Ralph Lauren being a well-known example. 13. [In Greenwich Polo Club’s] case, it is a polo club and making a connection with that polo club in its branding does not deceive or confuse the public. It is not as though Greenwich Polo Club is trying to copy Beverly Hills. That would be absurd…” The point is that the polo horse and rider motif, on the evidence I have seen, adds nothing to the distinctiveness of the Logo, save to make clear that there is a generic link to polo and to sport.’ 74. Once again, it is apparent that Marcus Smith J. considered that the relevance of other ‘polo’ brands went to distinctiveness or the lack of it. I should make it clear that the point I derive from Greenwich Polo is additional support for the point of law that the relevance of a ‘crowded market’ is to distinctiveness of the registered trade mark in issue. I should also emphasise that, whilst that case involved the same Cs’ registered trade marks as in this case, the Greenwich Signs were very different from those I have to consider. Furthermore and most importantly, I have received different evidence in this case - I have not received any evidence from Mr Durbridge. 75. For completeness, I mention that the Ds drew my attention to the terms of the Order made by Birss LJ when refusing the Cs permission to appeal in that case. Relevance of Co-existence Agreements 76. In response to the Ds’ reliance on various co-existence agreements, the Cs made two points: i) First, they said their relevance was not clear, and the Cs assumed the Ds were saying the co-existence agreements evidence a lack of confusion between the brands which were the subject of those agreements. The Cs asserted that was wrong because they said the need for a co-existence agreement demonstrates the brands are in conflict and a co-existence agreement records the basis on which the rival brands will tolerate the use by a rival of a similar mark. ii) Second, the Cs submitted that co-existence agreements are not relevant to the assessment of the likelihood of confusion as a matter of law , citing Case T-90/05 Omega SA v OHIM at [49] in support of this proposition. 77. I will discuss the proposition sought to be established from Omega in this section of the judgment. It is appropriate to discuss the relevance of the co-existence agreements relied on here in the context of my analysis of the market, see below, but the Cs’ point, as a matter of generality, is incorrect. Co-existence agreements are a prudent and often very cost-effective way of avoiding unnecessary oppositions and litigation which involve a measure of compromise on both sides. On the senior user’s side, he may forego the opportunity to make arguments as to conflict between the marks which may or may not succeed (and similarly for the junior user) in return for the junior user either making some (usually relatively slight) change in his branding or agreeing to restrictions on use and/or registration. Argument over unnecessary, perceived or essentially imaginary conflict is therefore avoided. Co-existence agreements can go wrong (cf the rounds of litigation over the Apple mark). However, the general point is that the mere existence of a co-existence agreement does not imply a conflict between the marks which would have resulted in a finding of conflict by a court or tribunal - it depends on the circumstances. Certainly, a co-existence agreement implies that the parties are prepared to tolerate the use agreed to, which also implies any damage is unlikely to be serious or can be remedied. 78. Omega v OHIM was part of the wider battle between the watchmakers Omega SA and the US engineering company, Omega Engineering. Omega SA had applied for a EUTM comprising the Greek symbol above the word OMEGA in a wide range of goods and services. Omega Engineering’s opposition was based on their earlier French mark OMEGA registered for goods in class 9 and services in class 42. Omega Engineering also relied on an agreement with Omega SA. 79. The Opposition Division upheld the Opposition in its entirety on Art 8(1)(b) grounds. It stated that OHIM could not be bound or influenced by the provisions of a private agreement between the parties. The Second Board of Appeal dismissed Omega SA’s appeal. 80. On appeal to the General Court, OHIM submitted that the terms of the co-existence agreement were ambiguous and repeated the point that it could not be bound by the terms of a private agreement between the parties. OHIM also submitted it was significant that disputes between the parties in the UK were resolved without taking account of the agreement. 81. The most specific reference to the terms of the agreement came in [49] of the Judgment of the General Court. It stated that, by the agreement, Omega SA undertook not to seek registration of the trade mark OMEGA in respect of apparatus industrially or scientifically employed for measuring or controlling variable parameters such as temperature, pressure, force, load, vibration, electrical conductivity, liquid level, acidity, humidity, strain and flow. It then stated that it was not apparent that Omega Engineering undertook not to oppose registration of the trade mark OMEGA by Omega SA. It concluded: ‘In any event, without its being necessary to consider the consequences of the agreement for the parties, the fact remains that the agreement is irrelevant to the assessment of the likelihood of confusion in the present case.’ 82. In context, that finding is entirely understandable. What it does not do is establish the more general proposition which the Cs sought to establish i.e. that all co-existence agreements are irrelevant to the assessment of the likelihood of confusion. Section 10(3) and equivalents. 83. In their Opening Skeleton Argument, the Cs set out a full statement of the principles applicable to their case under section 10(3) of the Act and art 9(1)(c) of the EUTMR. These principles are well settled and not in dispute. I will not lengthen this judgment by setting them all out. In summary, a claimant must establish four requirements (see Red Bull v Sun Mark [2012] EWHC 1929 (Ch) at [90]-[98]). The Cs summarised them as follows: i) The trade mark has a reputation in the territory in which it is registered, that is to say that it is known by a significant part of the relevant public in a substantial part of that territory. (see PAGO International v Tirolmilch (C-301/07) at [21]-[30] and Arnold J. in Red Bull at [90]: ‘This is not a particularly onerous requirement: see General Motors Corp v Yplon SA (C-375/97)… at [24] .’ ii) The defendant’s use of the sign gives rise to a ‘link’ with the claimant’s trade mark (but not necessarily confusion) in the mind of the average consumer. See Intel Corp Inc v CPM United Kingdom Ltd (C-252/07) at [30]. Whether or not a link exists requires a global appreciation, taking into account all relevant factors including (a) the degree of similarity between the conflicting marks, (b) the nature of the goods or services in question and the relevant section of the public, (c) the strength of the earlier mark’s reputation, (d) the degree of the earlier mark’s distinctive character, inherent or acquired through use, (e) whether a likelihood of confusion exists on the part of the public. iii) The defendant’s use of the sign causes one or more of three kinds of injury, namely (i) detriment to the distinctive character of the mark, (b) detriment to its repute, or (c) unfair advantage taken of the distinctive character or repute of the mark. Unfair advantage is further explained in Red Bull at [94]-[97]; detriment to distinctive character in Intel at [29] and [76]-[77] plus Environmental Manufacturing v OHIM C-383/12 P, as discussed in Maier v ASOS at [127]. iv) The defendant’s use of the sign is without due cause. See Leidesplein Beheer BV v Red Bull GmbH C-65/12 at [60]. 84. With all those principles in mind but bearing in mind the centre of gravity of the Cs’ case on infringement was on direct infringement, I turn to assess the evidence which I received. I start by relating the history of the BHPC brand, which is largely uncontroversial, before moving to more controversial aspects of the evidence. The history of the BHPC brand 85. The BHPC brand was originally created in 1982 by two students. Mr Haddad first became involved with the brand in the 1990’s as a licensee, one of many licences his company held, but limited to the US. Prior to his involvement, there was litigation with Ralph Lauren which seems to have culminated in a settlement agreement dated 20 th February 1985 between BHPC, Inc. (and various associated individuals) and Polo Fashions Inc (i.e. Ralph Lauren). Subject to various transitional provisions which do not matter for present purposes, that agreement placed certain restrictions on what BHPC could use as trademarks. Exhibit B (‘the BHPC Symbol’) is recognisable as the Cs’ horse and rider motif, albeit with more definition than appears in the registered mark. Exhibit D shows an earlier version of the BHPC mark with a footnote ‘Typeface to be changed as per Paragraph 1(e)’, apparently a reference to the typeface in which the words ‘Polo Club’ appeared in that symbol. In essence, the BHPC parties agreed: i) First, so far as the wording was concerned, they would not use as a trade mark: a) The word ‘polo’ or the words ‘polo club’, apart from the composite Beverly Hills Polo Club. b) In that composite, to ensure the words Beverly Hills and Polo Club were of equal prominence and in close proximity, in the same typeface and colour, and without the word polo being in any way emphasised. c) Not to use the typefaces used by RL. d) Not to use dark blue as the background colour on any packaging label or trade dress containing the words Beverly Hills Polo Club and/or the BHPC Symbol. ii) Second, so far as the BHPC Symbol was concerned: e) They would not use the BHPC symbol is an overall size smaller than 5 ½” by 5 ½”, except: f) They were permitted to use the BHPC Symbol in a smaller size if (i) in combination with the words Beverly Hills Polo Club in the configuration shown in Exhibit D (or another triangular label form) or (ii) in a repetitive pattern covering substantially all of the front or back of the relevant products, provided BHPC appeared in close proximity to the symbol and somewhere the words Beverly Hills Polo Club were displayed. 86. Corresponding limitations were agreed regarding registered trade marks. The terms applied worldwide. These detailed provisions were evidently what RL required in order to agree to co-existence of the two brands. On this basis, two actions and various trade mark oppositions were settled. Indeed, Mr Haddad accepted that all those requirements were designed to distance the BHPC brand and logo from the RL Polo brand and logo in order to make sure they were distinguishable. 87. Mr Haddad pointed out, correctly, that neither he nor either of the Cs were party to or bound by the Settlement Agreement with RL. Nonetheless, I understood his evidence to be that his companies had continued to abide by those limitations, an obviously sensible move. Mr Silverleaf put to Mr Haddad a summary of RL’s Polo business from FGC-3 (E/11/p393) and he had no reason to say it was inaccurate. Mr Haddad also accepted, but in any event I so find: i) That the RL polo-player logo was well-known across the world. ii) That the RL polo business had been active in the UK and Europe since 1981, in the Middle East and UAE ‘for many years’ and also across South and Central America. iii) When shown some polo shirts featuring the classic RL Polo logo on them, that the logo would be widely recognised by the brand-aware public and shirts so branded have been sold in their millions for decades in all the countries in dispute here. 88. Mr Silverleaf also took Mr Haddad to two other similar agreements, where the BHPC brand came into conflict with two other polo clubs which merchandised branded goods in the mid to late 1990s. The first agreement was between BHPC Marketing Inc. and Santa Barbara Polo & Racquet Club of 31 st May 1997 (E5/68). Mr Haddad preferred to call this a Co-operation Agreement rather than a co-existence agreement because it contained remedies if confusion did occur. This agreement is expressed to be binding on assigns. In connection with this agreement, Mr Silverleaf put to Mr Haddad a decision of the Supreme Court of Thailand (X2) (in which that Court overturned the decisions below and permitted registration of the BHPC mark in the face of the earlier Santa Barbara Polo Club mark). Mr Haddad agreed that the evidence in that case was that the two brands could work in the market, side by side without confusion, but he disagreed the same would apply in relation to BHPC and other similar polo club marks using the picture of a mounted polo player. Mr Haddad said they were currently in litigation with Santa Barbara in various venues around the world, including the UK. 89. The second agreement was with Greenwich Polo Club (‘GPC’) (E5/70) in March 1998. The conflict arose due to GPC’s application to register its logo which looked like this: 90. In essence, GPC agreed to withdraw that application and agreed not to file or use any marks which showed a horse and rider in side-view moving to the right. BHPC agreed that GPC was not prohibited from using or filing in the US any mark that includes a horse shown in side-view moving to the left or toward or away from the viewer at an angle. 91. More generally, Mr Haddad agreed that RL Polo, USPA and BHPC were all equestrian fashion brands and that it was necessary to ensure separation. He asserted that it was the ‘icon’ which distinguished BHPC, by which he meant the horse and rider motif. He was very sensitive about any comparisons with RL. For example, he denied that BHPC competed in any way with RL. A further distinguishing element, in his view, was that RL had an East Coast prep image, whereas BHPC was younger and West Coast-oriented. 92. Mr Haddad presented the BHPC brand in his witness statement as a global brand under which a wide range of goods are sold (which he called ‘BHPC Brand Goods’) including menswear, womenswear, childrenswear, footwear, perfume, watches, eyeglasses, bags and luggage, cosmetics and skin products. He cites this marketing statement which he wrote around the time he first became involved with the brand as a licensee, and which he says has featured in the Cs’ marketing materials since around 2012: ‘The Beverly Hills Polo Club brand was established in 1982. Inspired by both the elegance and glamour of Beverly Hills as well as the heritage the sport of polo, the Brand captures the excitement of this competitive sport along with membership in an exclusive social club.’ 93. Mr Haddad says the key element of the BHPC brand is the horse and rider logo. 94. Of course, what matters are the views of the average consumer, not Mr Haddad. 95. He set out a brief history of the BHPC brand in his first witness statement. He says all rights in the brand were owned by BHPC Marketing Inc in early 2001. In April 2001, the rights to the brand in various Asian countries were assigned to a Japanese company, Young Sangyo Co. Ltd. The countries involved were Japan, People’s Republic of China, Hong Kong, Republic of Korea, Indonesia, Malaysia, Singapore, Taiwan, Philippines, Thailand, Brunei, Vietnam, Laos, Union of Myanmar and Cambodia, although he says that the rights to the BHPC Brand in the Peoples Republic of China are now owned by Wah Sing. 96. By 2004, Mr Haddad says he had been a licensee of the BHPC brand for a few years and he decided to purchase all remaining rights in the brand. The purchase was made through a company owned by Mr Haddad and his two brothers which later changed its name to BHPC Associates Inc. In 2008, the brothers decided to divide the brand. His brothers retained ownership of BHPC Associates Inc and the rights to the BHPC brand in the United States. All remaining rights were assigned to the First Claimant. 97. The result is that globally, the ownership of the BHPC brand is split into four parts: i) BHPC Associates Inc. own the brand in the US; ii) Young Sangyo Co. Ltd (or its successor) own the brand across Asia; iii) Wah Sing owns the brand in China; iv) The Cs own the brand in the rest of the world. 99. Mr Haddad was asked about a pie chart in a presentation dating from about 2016 which purported to show the make-up of sales of BHPC branded products in various areas of the world. He indicated it was inaccurate, and stated that the split should be China & Asia: about 45%, US: 45% and the rest of the world (i.e. the Cs’ territory) about 10%. the Claimants’ “evidence” 100. As the inverted commas in the heading indicate, there were a number of problems with the ‘evidence’ put before the Court on behalf of the Cs in this case, and some of these problems were interrelated. I address them under the following headings. Marginal Annotations 101. Mr Haddad provided two witness statements, both relatively short. The second responded specifically to a paragraph in Mr Amoore’s first witness statement concerning a meeting with Mr Haddad in 2012. His first witness statement covered (a) the history of the BHPC brand; (b) complaints from the Cs’ licensees regarding the Ds brand and (c) his dealings with (some of) the Ds prior to the proceedings. In compliance with PD57AC, the appendix listed out seven specific documents to which he made reference in his witness statement, all of which were included in E/1/1-7. 102. In the version of his witness statement which was put into the trial bundles, a series of marginal annotations had been added. Those which referred to E/1/1-7 were unobjectionable, but were very much in the minority. By way of example, I take his paragraphs 7 and 8. I have inserted into [7] the marginal annotation: ‘7. The key element of the BHPC Brand is the horse and rider logo. The horse and rider logo has been used consistently together with the words “Beverly Hills Polo Club” since the BHPC Brand was established and is used in relation to all activities under the BHPC Brand. As well as featuring on the goods, the logo and the words are also shown on swing tags, labels, packaging in BHPC [F/1/12/79] stores and concessions, and in all advertising materials. 8. A wide range of goods are sold under the BHPC Brand, including menswear, womenswear, childrenswear, footwear, perfume, watches, eyeglasses, bags and luggage, cosmetics and skin products ("BHPC Brand Goods").’ 103. Against paragraph 8, the marginal annotations were: [F/1/15-18/85-156] [F/2/25-30/365-610] [F/3/31-32;34-40/611-656; 697-866] [F/4/41-42;44-46;48;54/867-950;975-1018;1067-1068; 1079-1090] [F/6/58;61-63;68/1425-1446; 1457-1600;1663-1738] [F/7/69;71;75-101;103-104; 106-123/1739-1814;1819- 1842;1895-1948;1951-2000; 2003-2038] [F/8/124-137/2039-2066] [F/10/171-172;174/2747-2808;2815-2896] 104. As was demonstrated in cross-examination, a number of these marginal references were either inaccurate or positively misleading. The marginal reference to paragraph 7 did not relate to the last sentence at all. The ‘document’ was to a picture of a box containing a digital movie presentation which was only ever made to a potential investor, not to the public at all. More seriously, these marginal annotations were inserted by some unidentified person, presumably at the Cs’ solicitors, but without any consultation or discussion with Mr Haddad. This only became clear in the course of Mr Haddad’s cross-examination. He did not select any of these documents. 105. A further example is Mr Haddad’s paragraph 21 (which I discuss in greater detail below). It is under the heading ‘PROMOTION OF THE BHPC BRAND IN THE UK AND EU’ and reads as follows, with the marginal references set out below: ‘21. The Claimants and their licensees advertise and promote the BHPC Brand through many means, including retail stores, websites, social media, print and digital advertising campaigns, sponsorship and attending trade shows. The BHPC Brand has also enjoyed substantial media coverage and won industry awards. The Claimants’ disclosure contains documents that show those activities and so I do not duplicate the contents of those documents by setting out what they say in this witness statement.’ [F/1/13;19;20/81;201;283] [F/3/33/673] [F/4/43;47;49-53;55/951-974;1041;1069-1078;1091] [F/6/59-60;64;67/1447-1456;1621;1659-1662] [F/7/70;72-74;102;105/1815-1818;1846;1860;1889-1893;1949;2001] [F/9/156-157;161-168/2365- 2378;2409-2642] [F/10/169-170;173;175-177/2643-2746;2809-2814;2897- 2927] 106. The final sentence of paragraph 21 displays a deep misunderstanding of CPR PD 57AC, I assume on the part of the Cs’ solicitors. As the paragraph 2.1 makes clear, the purpose of a trial witness statement is to set out in writing the evidence in chief that a witness would give if they were allowed to give oral evidence at trial without having provided the statement. Mr Haddad would have been entitled, in such oral evidence, to refer to having won media awards in certain years for example, and Counsel would have been entitled to put documents to him which evidenced the relevant media awards. The point is that Mr Haddad would have identified the documents which evidenced the media awards. 107. Translating that process into the written witness statement again would have required Mr Haddad to have identified the relevant documents himself in his witness statement. It is completely inadequate simply to insert ‘The Claimants’ disclosure contains documents that show those activities’, because the relevant documents need to be identified and, in many cases, explained. This does not involve duplicating the contents of the documents, a particular vice at which PD 57AC was aimed. The lack of explanation as to what each document comprises in Bundle F is a particular problem because I am left to infer what each document shows, mostly from its content and from other surrounding documents. 108. In any event it was entirely inappropriate for these marginal annotations to be inserted in this way. The Cs’ legal team would have known that, with the annotated witness statement being put in the trial bundle, Mr Haddad would be asked in his evidence in chief whether the contents of his (annotated) witness statement were true and that he would be expected to and would say yes. Their insertion and presence gave rise to a very substantial risk of the Court being misled as to the evidence which Mr Haddad was giving. The fact that the Court was not misled because Mr Silverleaf exposed this in cross-examination provides no excuse. 109. The problem with these marginal annotations is closely related to the next problem which concerns the status of Bundle F more generally. After Mr Silverleaf had elicited from Mr Haddad in cross-examination that he had had no role at all in the marginal annotations to his witness statements, I was given almost no assistance from the Cs as to what I was supposed to do with the documents in Bundle F or the marginal annotations in Mr Haddad’s witness statement. Bundle F 110. Bundle F is entitled ‘Claimants’ Other Disclosure Documents’. It comprises 10 volumes and 177 electronic tabs, each tab seemingly containing a single document. Overall, there are 2927 pages. I received Bundle F in electronic form. The documents are described in rather vague terms in the index and a number of tabs provided no date information at all, although the documents appear to be in chronological order. 111. The title suggests that the Cs’ solicitors simply put all of the Cs’ disclosure into the trial bundles, with seven documents at E1/1-7 and explicitly referred to in Mr Haddad’s witness statement (before and after annotation), and the remainder in Bundle F. Once again, that is not an appropriate way to proceed. 112. Bundle F was not an agreed trial bundle, and Mr Silverleaf gave some examples which, it seemed to me, fully justified why the Ds had refused to agree it as a trial bundle. As I pointed out to Mr Edenborough, that meant that the content of Bundle F did not benefit from the provision in CPR PD 32 paragraph 27.2. Furthermore, no Civil Evidence Act notice had been served in respect of any of the documents in Bundle F. Despite all of this, Mr Silverleaf did not suggest that Bundle F should be left out of account entirely. Indeed, he put a number of documents from Bundle F to Mr Haddad. 113. I propose to analyse first the sales figures put forward for the BHPC brand and, in the light of what they reveal, then to consider what, if any, weight to give to the documents in Bundle F. BHPC Sales Figures 114. I was presented with two schedules of BHPC sales figures, one from 2006-2016 and the other from 2010-2018 (E/1/6 & 7), each one headed ‘Sum of Retail Sales’. The second set of sales figures were also presented in a bar chart in F/1/14/p83. Where the two schedules overlapped (i.e. for 2010-2016) the figures appeared to be the same. It was unclear why no-one had taken the trouble to combine the two schedules into one. 115. In his first witness statement, Mr Haddad dealt with ‘Sales in the UK and EU’ in [17]-[20]. This section is preceded by [8], where he defines a term he uses later: ‘8. A wide range of goods are sold under the BHPC Brand, including menswear, womenswear, childrenswear, footwear, perfume, watches, eyeglasses, bags and luggage, cosmetics and skin products (" BHPC Brand Goods ").’ 116. In [18], he referred to the documents at E1/6-7 and described them as: ‘spreadsheets of estimated retail sales of the BHPC Brand Goods in the UK and EU between the years 2006 - 2019, broken down by country and category of goods . These documents were created by the Claimants and have been relied upon in my evidence in High Court proceedings previously, although it was helpful to refresh my memory as to the exact details. This information is commercially sensitive and highly confidential and so the Claimants will ask the court to keep this information secret. The headline retail sales of BHPC Brand Goods within the EU and UK are as follows:’ (my emphasis added) 117. He then sets out a table showing the totals for the UK and EU for each year 2006-2018 and for the whole period. He continues in [19]: ‘19. As shown from these documents, on average, between the years 2007 and 2018 (inclusive) retail sales of BHPC Brand Goods within the EU were $17.9mn per annum and within the UK $4.2mn per annum.’ 118. In cross-examination, Mr Haddad accepted that he had not calculated these averages himself but they do seem to be accurate (the figures for 2006 are not part of the averages because the data for 2006 is for December 2006 only). 119. Mr Haddad’s [17] reads: ‘Sales of goods under the BHPC Brand in the UK and the EU date back to the early 1990s. Data available to the Claimants show that there have been consistently high levels of sales of BHPC Brand Goods in the UK and EU since at least 2006.’ 120. The marginal note inserted next to this paragraph was to F/1/14/p83 which, as I have mentioned above, is simply the same information as in E1/7 but in the form of a bar chart. 121. Thus, the impression created by this evidence and the schedules is that a wide range of goods (including menswear, womenswear, childrenswear, footwear, perfume, watches, eyeglasses, bags and luggage, cosmetics and skin products) have been sold under the BHPC Brand consistently across the UK and EU for many years i.e. at least 2007-2018. 122. It would seem that the deployment of the sales schedules by Mr Haddad in previous High Court litigation has persuaded the Court to find that the Cs have a reputation in the BHPC Brand across the UK and EU. 123. The cross-examination in this case revealed a rather different picture. The following points should be noted: i) First, although reported as ‘Retail Sales’, the sales figures are inferred retail sales based on reports of wholesale sales by licensees. ii) Second, the sales are reported by the country of the licensee . For example: a) Belgium shows total sales of $3.75m of socks because the sock licensee (for the EU) is based there. There is no indication of where BHPC socks have actually been sold, although Mr Haddad mentioned in his evidence that socks had been sold in Finland. b) The UK shows substantial sales of luggage, because the licensee is based here. However, other countries are listed as having sales of luggage: Belgium, Denmark, Finland, France, Germany, Hungary, Romania, Slovakia, Slovenia, so, on Mr Haddad’s evidence it appears there were multiple luggage licensees. If so, one might expect most of each licensee’s sales of luggage to take place in their own country, albeit that their luggage products could be sold anywhere in what then constituted the single market. iii) Third, as Mr Silverleaf submitted, it is clear why Mr Haddad wished to rely on the averages cited in his [19]. These are overall averages across all product categories and across all years. The averages conceal significant variation across product categories and years. iv) Fourth, and perhaps most importantly, because the sales are reported by licensee and by country, it appears that Mr Haddad has little or no idea where the goods were actually retailed to consumers. Alternatively, if Mr Haddad does know more about where the sales actually took place he was not willing to divulge that information. However, it is possible to discern from some of the documents in Bundle F that it is likely that most of the sales reported in relation to Italy were sold in that country or possibly in close proximity to it. One can perhaps be slightly more certain of the corollary because of the list of outlets to which the Italian licensee distributed: the sales reported in relation to Italy were not distributed across the EU, and in particular, it is unlikely that any significant proportion or perhaps any of those sales were made in the UK. 124. The early years marked 2006-2009 bear three footnotes. First, all sales figures are in US$; second, that the 2006 sales are December only; third, that all these sales were Licensee 18 only. Over this period, the UK had the greatest sales, totalling just under $16m. In second place was ‘No Country Data’ with about $3.5m, indicating the sale could not allocated to a particular country. Germany was third and France a very distant fourth. Of the sales in the UK, luggage predominated, accounting for $9m, followed by underwear at $3m. 125. Moving to the more recent sales 2010-2018, the grand total of all sales across all countries totals $192.2m, averaging just over $20m per year over those 9 years. Indeed the annual totals vary around that mean with 2018 being the exception with a total of £30.3m. 126. In terms of the country to which sales were attributed, and the main categories, the pattern is roughly as follows: Country Total 2010-2018 Category 1 Category 2 Category 3 Italy $123.4m (~65%) Menswear $42m Footwear $29m Kidswear $19m UK $34.8m (~18%) Luggage $13.8m Underwear $11m Kidswear $5.6m Spain $16.5m (~8.5%) Home Textile $9.6m Handbags $6.8m - Belgium $5.5m Socks $3.7m Luggage $1.7m - Slovakia $4m Luggage $3.9m - - Germany $3.9m Luggage $1.6m Underwear $1.3m Watches $1m 127. In terms of categories of product, the order is Menswear Italy, Footwear Italy, followed by Luggage across all countries at $25m, of which half was allocated to the UK. 128. The sales were by no means uniform. For those country categories where the total over the 9 years was $6m or more, there were sales in each year (apart from Accessories in Italy, where the majority of the total of $13.7m took place in 2017 and 2018 with $4m in each year). 129. The Cs (and Mr Haddad in his evidence) were keen to emphasise that the inferred retail sales figures were confidential. I am inclined to find that at least part of the reason for the claim to confidence is because these figures do not fit with Mr Haddad’s claims that BHPC is a successful global brand with sales across the UK and EU in all the categories of product he mentions in his witness statement. With Italy as the exception, the reality is a very patchy picture. 130. Whilst Mr Haddad gave three years of royalties paid by Onixco in his attempt to say that confusion caused by RCBPC was responsible for the decline in those payments, another curiosity is that I was presented with no sales figures at all in respect of the Overseas Territories. This despite the fact that it was clear that, at least through the BHPC outlet at the Albrook Mall, BHPC goods had been on sale for more than 10 years. This was explained on the basis that the Cs were not claiming to benefit from any reputation in the Overseas Territories. I do not consider this explanation to be satisfactory. However, I will proceed on the basis that the Mark has been used in the Overseas Territories, but not to such a degree that the Cs have a reputation in any of individual countries. 131. Even at this stage, one can already see that the impression (see paragraph 121 above) which Mr Haddad was evidently seeking to convey by his evidence was misleading. Bundle F in more detail 132. Although Mr Silverleaf only put a small number of the documents in Bundle F in cross-examination, I considered it right to review the whole of Bundle F. Review of just a small selection might present an inaccurate picture. In this judgment I do not propose to analyse every document in Bundle F, not least because there is a lot of repetition and a lot of marketing speak, which is not evidence. However the following are the results of my review. I will mention the most prominent categories first, and then mention some documents in the order in which they appear. Dream Project srl 133. A significant number of the documents emanate from Dream Project Srl, the Cs’ Italian licensee. With a few exceptions which I mention below, almost all of the Brand or Look Books are from Dream Project. Many of them are repeated and some appear three times in the Bundle - it is unclear why. These Look Books feature a wide range of menswear, womenswear, kidswear, luggage, bags and certain accessories. 134. The Photo Books seemed consistently to bear the following rubric on the back cover ‘THIS PHOTO BOOK HAS BEEN PROVIDED IN CONFIDENCE AND CONTAINS CONFIDENTIAL AND/OR PROPRIETARY INFORMATION OF BEVERLY HILLS POLO CLUB BY DREAM PROJECT SRL AND MAY NOT BE USED, COPIED, REPRODUCED, DISTRIBUTED OR DISCLOSED, IN WHOLE OR IN PART, TO OTHERS WITHOUT THE PRIOR WRITTEN CONSENT OF DREAM PROJECT SRL’. In other words, these Photo Books were not for the purchasing public, but represented the range of goods which a retailer could order from Dream Project. 135. There were also various lists of what were apparently retail outlets in Italy associated with Dream Project. 136. At F161 is a list of 17 stores in Italy, with photographs. None of the stores are named Beverly Hill Polo Club and I have the impression that these are stores (mostly men’s fashion shops) which have sections selling BHPC clothing, some of them being concessions in department stores. The same document is duplicated at F165. 137. At F166, there is a 2017 print-out from the Dream Project website ‘Store Locator’ webpage with a 16 page list of stores appended. Each page lists 21-23 stores, so it would appear Dream Project supplies a considerable number of stores. From the map, almost all the stores seem to be in Italy, with one in Serbia and one in Hungary. I infer that this is a complete list of all retailers supplied by Dream Project. Catalogues & Look Books from other licensees 138. At F1/15 the document is identified as ‘BHPC catalogue’. It bears the date 2010-2011 but there are some unexplained redactions. It features various items of luggage and handbags which are named in Spanish and English. On the last page, amongst the redactions appear the words ‘Valencia (Spain)’. It is fair to infer these are bags produced by the Spanish licensee with sales totalling $6.9m over the 9 years from 2010-2018. Although one can see bags bearing the Mark, there are also bags which bear a very different logo, the main features of which are BH PC, a shield device and a horse-and-rider motif. 139. At F1/16 the document is identified as Licensee 19’s catalogue. It features home textiles (duvet covers, pillow cases, towels, cushions) and one can see the Mark in use on some of these goods. The products are identified in Spanish, and this is consistent with the sales figures for home textiles being reported for Spain and totalling $9.6m over 2010-2018. A similar catalogue with many of the same products and images is at F1/17. F/106-137 contain a series of pages showing duvet covers and pillows from the ‘Home Textile 2019 Collection’, featuring a lot of US imagery. 140. At F/43 is a report dating from October 2015 of the launch of BHPC fragrances in Spain. At F/58 there is a Look Book of shoes. 141. F/75-101 feature images of ‘Watches - 2018 new’, but no indication as to the licensee responsible or where they were sold. The turnover figures for 2018 indicate licensees in Germany, France and the UK reporting sales of watches, with Germany predominating. At F/171 there is an undated but apparently older Watches Look Book, followed by a webpage from ‘Watchpro’ from September 2014 announcing BHPC watches enter the UK market, from the licensee Brand Avenue. Although the sales figures indicate relatively small sales of watches from a UK licensee prior to 2014, there was a jump in sales in 2014, but a noticeable peak in 2018 in the UK. 142. At F/104 there is a catalogue of eyewear, apparently from 2019, attributable to an Italian licensee. 143. Although I have not specifically mentioned every Look Book in Bundle F, overall, in comparison with the Dream Project Look Books, there were precious few catalogues and Look Books from other licensees. It is important to appreciate that these Brand Books did not necessarily represent the range of goods actually sold under the Mark. They were simply goods from which a particular retailer might select what it wanted to sell. Only the sales figures give an indication of the levels of sales, but only by licensee. ‘Industry Awards’ 144. Interspersed between the various catalogues and Photo Books were a series of extracts from ‘Global License’ magazine, with the sub-title of ‘The source for licensing & retail intelligence’, and published in New York. These were May 2013 (F/19), May 2014 (F/33), May 2015 (F/47), May 2016 (F/64) with an Award Photo at F/65, April 2017 (F/72), with an Award Photo at F/73. Each of the BHPC entries gives either S. Haddad (I think one of Mr Haddad’s brothers) or Dan Haddad (Mr Haddad’s son) as the contact, with a Manhattan telephone number. The Top 150 listings also include Ralph Lauren (where Polo Ralph Lauren is presented as the first of several sub-brands), BHPC and USPA. There was also a website extract which showed the rankings for BHPC for 2011, 2013 and 2014. The rankings and the turnover figures given are: Year Ralph Lauren BHPC USPA 48 25 : $2.5B (E) 39 : $1.3B (E)(private) 44 : $1.3B 24 : $2.5B (E) (NASDAQ: RL) 44 : $1.3B (E)(private) 48 : $1.3B (private) Currently operating 450 stores worldwide 24 : $2.5B (E) (NASDAQ: RL) 49 $1.2bn (E)(private) 59 : $1bn (private) Currently operating 450 stores worldwide 26 : $2.5B (E) (NYSE: RL) 44 : $1.35B (E)(private) 56 : $1bn (private) 27 : $2.5B (E) (NYSE: RL) 34 : $1.7B (E)(private) 48 : $1.3B (private) 829 mono-brand stores worldwide 145. The License magazine describes its Top 150 as ‘The most authoritative guide to the world’s largest licensors and properties’. 146. By way of example, in the 2016 listing, Beverly Hills Polo Club is listed at number 44 with a turnover of $1.35bn. and the following text: ‘The name Beverly Hills suggests luxury, elegance and comfort. The game of polo, known as the sport of kings, reinforces these aspirational characteristics with the added dimensions of both the challenge of competitive sport as well as individual physical strength. The company’s focus is on international expansion with a new lifestyle shop concept. The company’s performance as a consumer brand has been recognized by the international retail community in several major venues with its nomination as International Emerging Market Retailer of the Year 2015 as well as International Lifestyle Retailer.’ 147. In the Bundle, the magazine is followed by an award photograph featuring Mr Haddad holding the ‘Highly Commended’ award in the ‘Global Emerging Retail Brand of the Year 2016’. BHPC was the winner of the 2017 Award for Emerging Market Retailer. 148. In cross-examination, Mr Silverleaf put the 2022 entry criteria to Mr Haddad. Entries cost £275 + VAT but all ‘shortlisted’ companies are required to book a table of ten at the live awards dinner event for £2,995 and all winners and highly commendeds must take an A4 advertisement in the Winners Brochure for £1,500. Although there is a judging panel, the notion that any of their awards are independent recognition of the actual merit of a licensing business must be treated with caution. The rankings appear to depend entirely on the turnover figure provided by the business in question. Of course, the turnover of a publicly listed organisation can be checked, but those from private companies cannot. Estimated (E) turnover figures may be an additional reason for caution. 149. Mr Haddad was asked where the 2017 $1.7bn figure came from. His answer was ‘ From estimates based on retail locations, licensee reports, USA royalty revenues, similar to that pie chart that you referred to before. ’ He was also asked what proportion of the $1.7bn total that his part of the business was responsible for. He said it was probably around 100m, maybe 80m. ‘ In that area. It is hard to come up with actual figures based on we do not know the square footage of the stores in China. There were 600 stores at the time. The US sales, at the time, I can only project based on royalty revenues or advised to me informally, but are credible, and based on our own royalty collection, royalty collections .’ Mr Haddad’s estimates for the Cs’ contribution to the global figure were in marked contrast to the actual figures presented in E/1/6 & 7. 150. The pie chart previously referred to was from the presentation to Blue Inc in about June 2016 and it showed USA: 40%; China & Asia: 25%; Middle East: 15%; Europe: 10%; South Americas: 6% and ROW: 3%. The text underneath the pie chart said that the BHPC EU target is the increase sales up to 20% of Global target and the UK is key to this plan. In the accompanying letter, Mr Ball said the BHPC brand had a turnover of $1.5bn (i.e. that was the size of the total pie). Earlier Mr Haddad had indicated that the pie chart was wrong. Where he ended up in his answers was that China and the US should each account for 45%, with the remainder (being his part of global BHPC business) 10% of the global turnover pie. Although the questions were put as at 2016, Mr Haddad also indicated that 10% was the appropriate share in 2010 when he took control of his share of the rights. 151. Mr Haddad disowned a number of statements made in this presentation and accompanying letter, written by Mr Ball of Harvest Group Ltd. When asked why Mr Ball got the pie chart so badly wrong, Mr Haddad indicated it was because Mr Ball was a salesman. 152. Reverting to the Global License Top 150 rankings, in relation to the Cs, the cross-examination established to my satisfaction that: i) BHPC (i.e. including the Cs) entered itself for inclusion in the list. ii) For the BHPC entries, it would seem the turnover figures were provided by either Mr Haddad or one of his brothers. I remain unconvinced that I can place any reliance on those turnover figures. 153. Accordingly, I intend to place no weight on these rankings or the awards. The following reasons will suffice: i) First, they are not an independent recognition of the success of the Cs’ licensing business. The rankings seem to be based primarily on the estimated turnover figures supplied by the Cs themselves. ii) Second, I do not find those estimated turnover figures reliable. Although I do not have any figures for the turnover of the US or Chinese parts of the BHPC brand, I found all of Mr Haddad’s evidence as to the global turnover in the BHPC brand to be vague and unsupported. None of the figures match up or support each other. Take Mr Ball’s $1.5bn global turnover figure in 2016. 10% would be $150m. Yet the highest turnover figure for the Cs was £30m in 2018, nowhere near $150m. Even though Mr Haddad disclaimed anything said by Mr Ball on the basis he was a salesman, Mr Haddad is himself an enthusiastic salesman for the BHPC brand. If the figures given to the Global Licence magazine were accurate, then the Cs would account for less than 2% of those global sales. By contrast, if the inferred retail sales relating to the Cs’ and Mr Haddad’s 10% estimate are accurate, that would imply global sales of a maximum of around $300m. 154. Having considered Mr Haddad’s answers in cross-examination carefully, I have come to the conclusion that there is a disparity between what he thinks the brand should be achieving in terms of turnover, reach and popularity amongst the consuming public and what it actually achieves and has achieved in recent years. He is programmed to talk up the brand, even to exaggerate its performance. When challenged on what he has said about it, he blames the lack of performance on others. Promotion of the BHPC brand in the UK and EU 155. Under this heading, Mr Haddad said this: ‘21. The Claimants and their licensees advertise and promote the BHPC Brand through many means, including retail stores, websites, social media, print and digital advertising campaigns, sponsorship and attending trade shows. The BHPC Brand has also enjoyed substantial media coverage and won industry awards. The Claimants’ disclosure contains documents that show those activities and so I do not duplicate the contents of those documents by setting out what they say in this witness statement.’ 156. The Cs’ solicitors placed a large number of marginal references to documents in Bundle F against this paragraph. The limited cross-examination on some of those documents but more particularly my own review of Bundle F demonstrated that this evidence is materially misleading and untrue, particularly since it attempts to portray a general picture applicable across the UK and EU. 157. In amongst the documents in Bundle F, it is possible to find evidence of BHPC presence at various trade shows including Cosmoprof, Mipel, Micam, Piti Bambino (Italy), Premium & Heimtextil (in Germany), MAPIC, Who’s Next (France), Intergift (Spain), Brand Licensing Show (UK), although I am not inclined to accept that attendance at all these shows was as consistent as some of the pages indicate. However, the point is that these were all trade shows. These were attempts to attract licensees. They did not contribute to the generation of reputation amongst the purchasing public. 158. Bundle F does contain a few posters and press articles which would have been seen by the public. Thus, F/46 features a large poster in Florence railway station, with the document indicating that 160,000 passengers pass through the station daily, but only for 4 days in June 2015. Also in 2015, there were full page BHPC advertisements (featuring the same image) in Corriere della Sera Style magazine, GQ Italia, L’Uomo Vogue, Alitalia inflight magazine, Wired, and La Gazzetta dello Sport SW magazine (these selections appear three times at F/173, F/157 and F/49-53). There are some Spanish magazine pages featuring BHPC fragrances (exclusively in El Corte Ingles), apparently also from 2015. 159. In addition, there is evidence of sponsorship of a volleyball team in Italy in 2015. 160. Although, as I have said, the Cs gave me no assistance as regards the content of Bundle F, I will assume they would invite me to infer that the press articles included in Bundle F were just a selection of a much greater set. In normal circumstances such an inference would be readily drawn. However, in this case, due to the degree of exaggeration in Mr Haddad’s evidence as to the scale and reach of the BHPC brand, I am not prepared to make any inference that there has been other promotion which would materially alter the situation presented. Interim conclusions as to the status of the BHPC brand 161. Overall, my findings based on my own review of Bundle F, the sales figures in E1/6&7 and Mr Haddad’s witness statement are as follows: i) In his witness statement, Mr Haddad attempted to present a picture of BHPC as a successful global brand, with the Cs selling the BHPC Brand Goods (as defined in his [8]) across the UK and EU. The reality is significantly different. ii) It is impossible to say what the reality is, due to the highly selective and incomplete information provided in the Cs’ evidence and disclosure. It is clear that the Cs’ licensees could have provided a lot more information about the position in particular markets, but Mr Haddad is secretive about even the identities of his licensees, let alone their performance. For example, the presentation to Blue Inc mentioned that an entity called The Harvest Group Ltd owned the rights (to BHPC) in the UK, but, apart from the turnover figures which Mr Haddad presented, no additional information was forthcoming from that licensee. In the circumstances, however, I must do the best I can based on the information I have. iii) Primarily due to the efforts of Dream Project and the materials emanating from them, I am satisfied that a good range of BHPC goods have been sold over the years in Italy, comprising menswear predominantly, with some kidswear, footwear, bags and accessories, supported by some promotional activity, giving rise to the BHPC mark having a reputation in Italy for such goods. iv) I am prepared to find the BHPC brand has a reputation in Spain for bedding and bags. v) So far as the UK is concerned, I find the BHPC brand has a reputation for Luggage, Men’s Underwear and, due to sales in 2017 and 2018, for children’s clothing. I saw no evidence of sales of women’s underwear in the UK. vi) The position as regards other goods and in other countries of the EU is too obscure and uncertain for me to find any reputation. The market for clothing and the other goods of the types sold by BHPC is huge across the UK and EU and the total inferred retail sales of BHPC branded goods over 9 years (2010-2018) for various categories of goods are small. vii) When one looks at the individual years, other than where I have found reputation, the sales figures indicate sales took place in a rather sporadic fashion, consistent with a pattern of new licensees being signed up, putting in early effort but then losing heart and the licence terminating. This pattern is not consistent with a successful brand. viii) Overall, I find Mr Haddad’s evidence significantly exaggerated the position. His evidence was misleading. The history of RCBPC 162. As I related above, the Royal County of Berkshire Polo Club was founded in 1985 by the late Bryan Morrison and his business partner, Norman Lebel. The then Prince of Wales was a founding member of the Club and Princes William and Harry have played polo there at various times in the past. Mr Amoore described it as follows. The Club is set in a 200 acre estate near Windsor in Berkshire and has world class sporting facilities including six polo fields, the world’s first all-weather polo arena, stabling for 175 horses and a highly regarded polo academy along with other sports activities. It is one of the premier polo clubs in the UK, ranked in the top 5. It hosts approximately 70 polo events each year including a number of prestigious international tournaments. It is extremely well-known in the world of polo both in the UK and elsewhere. None of this evidence was challenged and I accept it. 163. Mr Amoore also said that the Club has sold merchandised goods bearing some of the Ds’ logos since its inception, at first on a small scale in the UK only. I infer that these early sales were made from the Club itself. In around 2005, sales expanded to other countries and thereafter on an increasing scale. Mr Amoore says that since 2010 its global sales have been substantial. By 2015, his evidence establishes that annual sales exceeded $50m, of which $6m were in the EU. He says that sales have declined since then, which he attributes to the litigation campaign being waged by the Cs against the Ds. 164. I have set out Sign 1 above. Mr Amoore said that Sam Houston designed this logo for D1 in 1985. He says it was used as the Club’s logo from then on letter headings, other business papers, flags, banners and other signage as well as on merchandise sold through the Club shop. Sign 3 was developed in 2015, as Mr Amoore put it, to establish a modern and more distinctive horse and rider logo. At that time, D1 had been in a long-running dispute with Ralph Lauren over its use of Sign 1. That dispute was resolved by a co-existence agreement between Ralph Lauren and the D1 in 2018. Following that agreement, Sign 1 was replaced by Sign 3 in all licensed territories covered by the agreement. Mr Amoore explained that the form of Sign 3 was developed with the Cs’ previous complaints in mind and with a view to distinguishing from the Cs’ mark as well. 165. It is perhaps not surprising that D1 also has a co-existence agreement with RL, which is dated 2 nd May 2018 and was made ‘in order to settle their current disputes … and to avoid any disputes in future’. The agreement is detailed and comprehensive but in essence, D1 agreed to cease use of its existing marks, all based around Sign 1, and to commence use of the new mark which was Sign 3. There were corresponding provisions for D1 to withdraw registrations relating to Sign 1 and to apply for registrations for Sign 3. I also mention clause 5.1 entitled ‘Respect for each parties rights’ which provides: ‘Without prejudice to POLO’s acceptance of the RCBPC New Mark under the terms of this Agreement, each party agrees to respect the trade marks and other intellectual property rights of the other including but not limited to trade dress, copyright, designs and get up rights and to use all reasonable endeavours to avoid consumer confusion in relation to their respective brands.’ 166. If one compares the RL Polo horse and rider logo with the horse and rider in Sign 1, although there are differences, one can see the logos feature the horse and rider at a similar angle to the viewer, both with mallet raised. The change to Sign 3 was evidently considered acceptable by RL as sufficient to eliminate any material degree of confusion between the brands. RL also had the comfort of clause 5.1, so that if any confusion occurred for whatever reason, it could be addressed in a sensible way. 167. The two co-existence agreements made by RL with each of the parties provide, in my view, a very useful and practical insight into the market for ‘polo’ brands and especially those which feature a polo horse and rider motif. Indeed, it should be noted that RL’s logo comprises just its horse and rider motif. RL trusts that logo to act as the indicator of origin, particularly on garments, without any additional words. RL was and remains a well-established and well-resourced brand owner and can be taken to have an excellent understanding of its brands and the importance of protecting them. Indeed those attributes are reflected in the professional drafting of the co-existence agreements. 168. These co-existence agreements indicate that RL considers the combination of differences in the appearance of the horse and rider motif and the accompanying words are sufficient to avoid consumer confusion. Of course, RL has not had to assess or consider the comparison I have to make in this action, but the point is that the comparisons RL had to consider involved less distinguishing matter than I have to consider. The views of RL which I infer from these co-existence agreements are in no sense determinative and I must still carry out the process of comparison and global assessment required by the caselaw. 169. I should add that Mr Amoore gave unchallenged evidence about the existence of further co-existence agreements in this field: i) Between RL and USPA. ii) Between RL and Santa Barbara Polo & Racquet Club. iii) Between the Cs and Santa Barbara & Greenwich Polo Club. iv) Between D1 and California Polo Club. Dates of first use of the RCBPC brand in various countries 170. As I mentioned above, in cross-examination, Mr Amoore was asked about the dates of first use by reference to the dates when licence agreements were signed. His evidence can be summarised as follows: i) The sale of branded clothing from the shop on D1’s premises started very early in the life of the Club. ii) The international licensing business started before 2005. Mr Bryan Morrison engaged an agent and D1 had a licensee which operated at least in Argentina. iii) In about late 2010, D1 first engaged D4 to promote the exploitation of D1’s brand, particularly abroad. This resulted in various licence agreements being signed: a) Panama: late 2011/early 2012 with Mays Zona, which Mr Amoore identified as ‘our biggest licensee’. He said their agreement covered a number of countries and Mays would get local sub-licensees in particular territories such as Chile. He also said that the local licensees would produce a clothing line to suit the market in their territory so, for example, there was not a standard polo shirt which was sold by all licensees. b) Chile, Peru: agreement extended in about early 2012; c) Mexico: May 2014; d) UAE: about May 2015. iv) It seems that in about January 2018, D1 began working with a new licensing agent, CAA-GBG, which had been selected by Mr James Morrison, although there seems to have been a period of overlap with D4 until about November 2018. Mr Amoore was clear that the Cs’ letter before action on 24 th July 2018 was not the cause of D4 ceasing to act as licensing agent for D1. 171. Mr James Morrison indicated that during D4’s tenure, the overseas licensing activity was largely managed by Mr Amoore, under the supervision of Mrs Greta Morrison. Mr James Morrison took more of an active interest in that side of the business after CAA-GBG took over, and had more regular contact with CAA than he had ever had with D4. 172. In the absence of any more precise information, I will work on the basis that sales of clothing bearing D1’s logo(s) commenced within a few months of the date of each licence agreement. I address the situation in each country in further detail below. Overall use 173. Mr Amoore presented global retail sales figures of D1’s branded goods since 2015, as follows: 2015: $61.8m; 2016: $57.7m; 2017: $53.4m; 2018: £14.2m; 2019: £15.9m; 2020: £10m; 2021: £13.8m, of which £1.27m were sales in the EU. 174. Mr Amoore gave unchallenged evidence that D1 had traded RCBPC branded goods since its opening in 1986 through the Club ‘shop’, which he describes as a simple area in a barn at the Club, an area within existing third party shops run by Argosy and then Asprey. He said the trade has been in various items, like polo shirts, rugby shirts, sweatshirts and hoodies, baseball caps and shirts, all branded with Logo No.1 until about 2018 and then with Logo No.3. He said the purchasers were all members of the Club or their guests. He also listed men’s jackets, ladies’ jackets, men’s and ladies’ t-shirts, cardigans, long-sleeved polo shirts and numbered shirts. 175. Overall, Mr Amoore said sales in the UK are small and have only been from the club shop and certain pop-up shops at particular events like on International Days. He says that D1 intends to appoint a UK licensee ‘when practicable’ and has plans to increase sales through the Club via its website and a physical shop. He attributes the gradual fall in global sales since 2015 to the Cs’ campaign of oppositions and litigation. He says the threat of litigation in 2018 caused a major candidate licensee to decline to go forward. None of this evidence was challenged and I accept it. 176. So far as sales in the EU are concerned, Mr Amoore only made mention of D1’s licensee in Germany, Silag AG. In an email, Sam Fisher of D4 described Silag as having excellent warehousing and logistical capabilities but as difficult to manage. I was given no further information about sales in the EU, albeit that the figure given for 2021 sales is low and rather indicates that sales were likely to be restricted to the territory of the licensee, i.e. Germany. 177. During the currency of this litigation, I infer that D1 has not been pushing to expand the existing business in RCBPC branded goods, largely because potential licensees for the brand have been put off by the litigious attitude of the Cs and are unlikely to commit until this litigation has been resolved. However, I was struck by the fact that D3 and Mr Amoore are sensible people who have no desire to cause any confusion amongst consumers. They too are awaiting the outcome of this litigation to find out what they can and cannot do with the RCBPC brand. 178. Mr Amoore also gave details of numerous trade mark disputes between the Cs and D1 in the countries in issue in this action. In addition to those, Mr Amoore also gave details of the registrations which D1 has secured. So far as the UK is concerned, these were as follows: i) D1 applied for and was granted four UK registrations for Logo No.1, in applications granted on 11.8.1995 (from application filed 25.1.1989), 2.2.2001, 22.3.2002 and 14.6.2002. ii) D1 applied to register, in essence, Logo No.2 on 15.8.2014 and this was granted on 30.12.2016. iii) On 18.3.2016 D1 applied to register Logo No.3, and this was granted on 3.8.2018. Two further applications for Logo No.3 were filed on 16.6.2016 and granted on 3.8.2018. iv) None of these applications were opposed by the Cs. The market in which the two brands operate 179. There are a number of aspects to this. The first is the relevance of the sport of polo and the second is the prevalence of polo-themed brands in various markets. The relevance of polo as a sport 180. There was much debate about the relevance of polo as a sport. The Cs were keen to portray it as a highly niche sport which average consumers have neither played nor watched. 181. The Ds were keen to emphasise that polo is often said to be the sport of kings. They provided some details of its origins: it is said to be one of the oldest known team sports, reputedly having its origins in Persia, originally as a training game for cavalry units, from the sixth century BC. The rules of modern polo were drawn up by the British in India in the 19 th century and the first polo club, Hurlingham, was founded in the UK in 1876. It was an Olympic sport from 1900 to 1936. The Ds characterised polo as a global sport played in over 50 countries, the dominant ones amongst them being Argentina, the USA, and Britain, those three countries accounting for more than 50% of the players and clubs worldwide. 182. The evidence indicated that there are more than 1100 polo clubs in countries across the world, which include countries in which both the Cs’ and the Ds’ goods are sold and perhaps 20,000 people worldwide who actually play polo. The Ds characterised polo as a prestigious and popular sport, particularly in South America, and that it is known throughout the world to be a sport practised by the British Royal Family and consequently attracts attention everywhere. 183. I did not get the impression that the Cs really disputed any of the detail as to the history and existence of polo as a sport. I agree with the Cs that the detail does not matter and that average consumers have not played polo and only a very small fraction will have ever seen it actually played. However, it is very likely that average consumers have seen photographs of people playing polo and are well aware that the small teams of players on each side are on horseback and use a mallet to strike the ball. For this reason, the various horse and rider motifs used by the various polo brands will be familiar and readily understandable images to average consumers. 184. Notwithstanding the relatively tiny number of people who actually play polo, it seems to be a sport which is known by a disproportionately large number of people. It also seems to be a sport which is inordinately attractive to the creators of clothing brands, due to its ‘upmarket’ image. Mr Haddad gave a flavour of why it is attractive in this answer in which he waxed lyrically over the image of the horse and rider in the BHPC mark (T2/p223): ‘It is an icon which promotes an image of glamour, confidence, strength, competitiveness. When you see that rider on the horse waving that mallet, playing a game that only the upper class plays, you get that feeling of strength and competition, even in the horses’ muscles that are articulated in the design, you see the flexing of the strength of that horse’s muscle, you feel more than any other, of the other two polo brands that exist, you feel the drama and the spirit of the strength of that animal and competitive spirit of the guy on that horse waving that mallet. That is an incredible, incredible image. Better than any equestrian image out there.’ 185. For their part the Ds submitted that the sport of polo is well-known by an audience far beyond those who are involved with it, that its upmarket image is ‘universally recognised’ and, they submitted, that is no doubt why polo and polo club themed clothing has long been popular with the public. Whilst ‘universally recognised’ is somewhat of an exaggeration, the evidence persuaded me that these submissions are basically correct. I turn to consider the evidence which, in my view, supported that finding. The growth and popularity of polo-themed brands 186. Whilst I do not doubt that various polo clubs had produced various pieces of ‘polo’ clothing for their teams over the years, it seems to be common ground that it was Ralph Lauren which was responsible for the first major commercialisation of polo-themed clothing, using this style of branding: 187. The company behind the brand was/is Polo Fashions Inc. founded by Ralph Lauren in 1968. The logo set out above was adopted and started to be used in 1971. Over the years it has been used alone, as an external brand indication on clothing or accompanied by the words ‘Polo by Ralph Lauren’ or simply ‘Polo Ralph Lauren’. The size of POLO relative to Ralph Lauren varies. Subject to the limitations of stitching the logo onto garments, as far I could see the logo never varies, although the size varies when used externally on clothing. For example: 188. Indeed, in closing, the Cs did not dispute that Polo Ralph Lauren is a successful brand, that it has made extensive sales of clothing, and that it has a reputation in respect of clothing (i.e. that it is known by a significant part of the public concerned with clothing in a significant part of each of the relevant territories). 189. As an aside, I note that in FGC-3 (on which more below), the global turnover of RL Polo was reported to be in excess of $6bn per annum. The source of this figure was not disclosed and there was no cross-examination about it, but it is consistent with other evidence indicating RL Polo is plainly the largest ‘polo’ brand by some distance. It is not inconsistent with the estimates in ‘Global Licensing’ of the scale of the licensing business in RL Polo if (as is not uncommon) RL conducts its own business operations in RL Polo in conjunction with its licensees. 190. Another ‘polo’ brand relied on by the Ds, particularly as regards Latin America is the US Polo Association (‘USPA’). The Cs acknowledged USPA as No.2 (as I understood it, globally), but care is required because, for example, the USPA has only recently begun to establish a presence in the UK. 191. In an ideal world, the Court would have received evidence of the sales volumes of each ‘polo-themed’ brand in each relevant territory. It is clear this type of evidence is not publicly available and that the cost of attempting a detailed analysis would be prohibitive, if possible at all. So although the Cs were correct to point out that there is no evidence of the extent of trade by Polo Ralph Lauren or US Polo Association (by way of example) in any of the territories in issue, that does not mean that I can ignore other ‘polo-themed’ brands which are obviously in the market. Nor can I ignore the fact that RL and USPA are likely to dwarf both the Cs and the Ds in terms of likely sales volumes and presence. 192. Perhaps the most detailed picture was provided in relation to the Overseas Territories in the evidence presented by Señor Garcia, which I must now discuss in some detail. It too suffered from some problems. The evidence of Señor Garcia 193. In my judgment from the PTR, the neutral citation of which is [2022] EWHC 1244 (Ch) , I had to consider the content of Señor Garcia’s witness statement in some detail - see [46]-[75] - and I have reminded myself of what I said there. 194. In certain key respects, it is fair to say that in cross-examination, Señor Garcia did not come up to proof and I need to explain why. 195. He is the Vice-President of the Latin American arm of CAA-GBG and has occupied that role since June 2015. He has managed the RCBPC brand in Latin America since 2015, when his arm of CAA-GBG was appointed as sub-agent by D4. He is based in Mexico City and his company has various offices in Mexico, Columbia, Brazil and Chile. His evidence was that Mexico is the territory in which RCBPC currently earns the most revenue, with 9 licensees accounting for 84% of its licensing revenue globally. There was some cross-examination about these licensees and the outcome was that there were now 8 including Sears as a direct retailer. He identified D6 as the sole remaining licensee operating in Chile, Panama and Peru and D9 (which operates the Oeschle chain of department stores in Peru) as previously a licensee but which no long sells RCBPC goods. 196. Señor Garcia’s evidence can be divided into the following sections: i) First, he exhibited at FGC-3 a document compiled by D5 and his team in or around May 2019 in connection with a possible application for interim relief. Even though he did not compile FGC-3, Señor Garcia made some observations on the contents in his witness statement. He indicated that the picture presented in FGC-3 was consistent with his own knowledge of the market. ii) Second, Señor Garcia included a substantial section in his witness statement designed to show that there are ‘many’ polo-themed brands on the markets in Latin America. He identified RL Polo as the largest. This section covered: a) physical stores in Mexico, Chile, Panama and Peru which carried various polo brands which he listed in a table: RL Polo, RCBPC, USPA, York Team Polo Club, Rock Hampton Polo Club, New Forest Polo Club, Long Beach Polo Club, Oxford Polo Club, HPC, BHPC, La Martina and four others. The first three he identified as selling in all four territories (but with the qualification that RCBPC in Peru was now historical. Beyond those three, he identified BHPC as present in Chile, Panama and Peru but not Mexico. Most of the others were present in Mexico but not the other three countries. All the ones I have named featured some form of horse and rider logo, all different. b) Online marketplaces in Latin America. He said Mexico has one of the largest e-commerce markets in Latin America and a number of the physical retailers he had already mentioned have e-commerce selling sites (Walmart, Coppel, Liverpool, Sam’s Club and Sears). In addition he identified several online retailing sites which sold more than one polo brand, including: (i) Mercado Libre as the largest online marketplace in Latin America with approximately 139m active users in 2021, selling 14 polo brands on mercadolibre.com.mx: RL Polo, RCBPC, HPC Polo, York Team Polo, Long Beach Polo Club, Rockhampton Polo Club, BHPC, New Forest Team Polo, Oxford Polo Club, La Martina, USPA and 3 others; RL Polo, USPA, BHPC, RCBPC and Wellington Polo Club on mercadolibre.cl; no polo brands on its Panamanian website and just two, RL Polo and BHPC on its Peruvian website: mercadolibre.com.pe. (ii) Claro-shop.com, selling 8 polo brands. (iii) Amazon, selling RL Polo, BHPC, USPA, and Royal London Polo Club in Mexico, by way of example. (iv) Dafiti, which he described as ‘a major fashion e-commerce group in Latin America selling in Brazil, Argentina, Chile and Colombia’. (v) Linio (owned by the Chilean retail store chain Falabella), selling 4 polo brands on its Mexican site (including RCBPC) and 3 on its Chilean website (RL Polo, BHPC and USPA). c) He also gave evidence of various department stores carrying more than one polo brand, including, in Mexico: Coppel, Chedraui, El Nuevo Mundo, El Palacio de Hierro, Liverpool, Sears/Sanborns, Soriana, Walmart/Sam’s Club; in Chile and Peru, Falabella, Paris and Ripley; and of a number of polo-brands co-existing in shopping malls in Mexico, Chile, Peru and Panama. The mall in question in Panama was the Albrook Mall, in respect of which he said USPA has 3 own brand stores with multi-brand retailer Stevens stocking RL Polo (this evidence was consistent with neither BHPC nor RCBPC having a store there in early 2022). iii) Third, he addressed the issue of whether there had been confusion between BHPC and RCBPC in Latin America generally and in Mexico, Chile, Panama and Peru in particular. First he said that all licensees of RCBPC in Latin America are happy with the brand and with where it fits within the ‘relatively crowded and popular market for polo related brands’. Second, he also said that if any instances of confusion had been reported to RCBPC’s licensees of buyer or consumer confusion, he would have heard about them. He said he has not heard of any issues with members of the public being unable to distinguish between polo brands. He was not challenged on these first two points. Third, he was invited to comment on documents in the Cs’ disclosure which might be said to evidence confusion. These included the email from Ms Borycz I have already discussed, and some screenshots which purport to be from the Dafiti website but which he identified as from a fraudulent or scam website. 197. In closing, Mr Edenborough was very critical of Señor Garcia’s evidence, submitting he was not independent and had a vested interest in the outcome of this claim. Those points are correct, but they do not mean I can or should discard his evidence. Particular points of criticism were as follows: i) Dafiti: I have quoted what he said about Dafiti in his witness statement. Counsel submitted that he contradicted himself in cross-examination, alleging that he said ‘Dafiti is a very small retailer’. Although he said those words, one needs to read on to understand what he was saying: ‘Very small for our market, for our brand.’ There was no contradiction in my view. ii) That he appeared not to know his own evidence or had been sloppy. Mr Edenborough established that there were a number of mistakes in Señor Garcia’s evidence, but these were on points of detail and Señor Garcia readily acknowledged any mistakes. iii) That, when asked questions that he considered problematic about documents in his exhibits, his approach was to distance himself (e.g. saying he did not control the website). What this point goes to is that the documents speak for themselves, to the extent that they are relevant. iv) That FGC-2 was simply ‘state of the register’ evidence, with no sufficient detail as to commencement, duration of use, turnover, penetration, distribution or goods to be able to ascertain the market presence of any of the brands listed. It is true there is little detail, but this is far more than ‘state of the register’ evidence, since Señor Garcia gave specific evidence about co-existence of these ‘polo-themed’ brands in physical stores, malls and on websites. The fact, for example, that a particular brand is carried in department stores like Falabella (100 across Columbia, Chile and Peru) or El Nuevo Mundo in Mexico or Liverpool (131 stores across Mexico), gives an impression of scale and market presence. 198. Whilst acknowledging that Señor Garcia worked in the fashion sector, Mr Edenborough submitted that he did not have sufficient knowledge to assist the Court with knowledge-based opinions. The examples relied upon in support of this submission were: i) That he did not know whether Polo Ralph Lauren was a real polo club or not. Counsel contended this was an example of ‘staggering ignorance’. ii) That he was not familiar enough with the market to know the difference between polo clubs and polo-themed clothing brands. iii) That he did not know where different physical stores he had mentioned were located in Mexico City. iv) That his near constant refrain when shown a document was that was what ‘it shows’. v) That he did not know the turnovers of different brands, save that he thought RL Polo was the largest. vi) That he gave no written evidence about when any particular brand had commenced trade. vii) That he presented no evidence of the marketplace in 2018 and all his evidence concerned the position in early 2022. 199. It is true that his evidence related to the position in early 2022 (I deal with FGC-3 separately below), but I consider the criticisms at i), ii), iii), v) and vi) to be largely unfair. Someone giving evidence about the position in the market does not need to be familiar with when a brand commenced use, or whether a polo brand is or is not linked to a real-life polo club. 200. Of course, these criticisms were made in an attempt to damage the evidence and indications in his evidence of co-existence of many polo-themed brands in various countries. The ultimate conclusion invited by the Cs was that Señor Garcia’s opinion that the consumer in any of the territories is conditioned to distinguish between polo-themed brands should be rejected because he does not have the requisite expertise to opine on such matters. 201. I am inclined to place minimal weight on the opinions and conclusions expressed by Señor Garcia in his evidence, largely because the factual premises presented in his written evidence and supported by the documents he exhibited allow me to reach the same or similar conclusions myself. The power of his evidence lies in the basic factual picture which was presented (which, subject to the mistakes identified in Señor Garcia’s cross-examination) was left very largely unchallenged and unaffected. The precise details do not matter. His evidence and exhibit FGC-2 demonstrated that there is extensive co-existence of many polo-themed brands in Latin America. The greatest degree of co-existence is in Mexico, where, even the Cs were prepared to accept that, taking Señor Garcia’s evidence at it highest, in 2022 there were 20 polo-themed clothing brands in existence. The Cs submitted that there were far fewer such brands co-existing in Peru, Chile and Panama, but that all this evidence was lacking in any useful or probative detail. 202. Having considered all of the Cs’ criticisms carefully, I am satisfied that Señor Garcia’s evidence, but largely the documents he exhibited, established the following: i) That there is and has been for several years, extensive co-existence of ‘polo-themed’ brands in Latin America. Certainly, the greatest number of such brands co-existing is in Mexico with somewhat lesser numbers of such brands coexisting in Chile, Panama and Peru. ii) That physical and online retailers operating in Mexico, Chile, Panama and Peru have no qualms about stocking and selling more than one ‘polo-themed’ brand and often several of them. To the same end, the operators of shopping malls do not see any problem in having multiple ‘polo-themed’ brands being sold in the same mall, whether in their own units or in department stores. iii) All of the polo-themed brands of any significance mentioned in his evidence (and I have discarded some of the apparently minor brands) feature some form of horse and rider logo in their branding. As I mentioned, all of these horse and rider motifs are different. Only one (that of RL) is trusted to, and does in fact operate on its own without any additional wording. iv) That RL Polo is the number one in all these markets, probably followed by USPA. It is debatable whether RCBPC or BHPC ranks at number 3. In Peru, BHPC appears to be number 3 (and RCBPC is no longer sold in Oeschle), whereas in Mexico it is RCBPC which is definitely ahead (and BHPC was said not to be present in physical stores). v) Although his evidence related to early 2022 and the documents in FGC-3 to early 2019, I am satisfied that the general picture presented is accurate going further back in time, although it is likely that the number of ‘polo-themed’ brands has increased over time. 203. All these points favour a conclusion that there is no confusion between these ‘polo-themed’ brands in South America and that consumers have learnt that the presence of a horse and rider motif is generic for such brands such that they must rely on other aspects of the branding and in particular the name as indicating origin. 204. That leaves FGC-3, which was directed at the position of polo globally, the position in the UK and EU, US, Canada, UAE and South Africa, with a little bit of evidence directed to Mexico and Peru. Compiled in early 2019, FGC-3 shows a number of polo clubs, especially in the UK and Europe, selling their own merchandise. All of this was far too small scale to have any effect on the average consumer. By contrast the details presented for RL Polo and USPA were consistent with other evidence - that they were number one and two globally. So far as USPA is concerned, there was some evidence that the brand was first launched in the UK in Spring 2018 with a range of kidswear. The relevance of other polo-themed brands 205. Under this heading I deal with a number of ways in which the Cs tried to contend that the existence of other ‘polo-themed’ brands was irrelevant. 206. In closing, the Cs posed the question: ‘Are consumers accustomed to horse and player marks?’ and submitted the answer was ‘No’. I do not regard that as the correct question and furthermore the answer is and needs to be far more nuanced - it depends on the country concerned. 207. More generally, the Cs were keen to stress that virtually all of the goods sold by any of these ‘polo-themed’ brands had nothing to do with polo - by which they meant the playing of polo. Many of these brands feature polo shirts but I can proceed on the basis that none of them are used in actual polo playing. This does not mean that all the other ‘polo-themed’ brands can be ignored. 208. Then the Cs contended (based on [87] from Jack Wills, quoted above at [71], and apparently picking up on the reference in that paragraph to inherent distinctive character) that any attack on the distinctiveness of a trade mark is one that must be considered as of its application date. That is true in the context of an attempt to invalidate, but it is not true when it comes to assessing the distinctive character (inherent and acquired) of a registered trade mark for the purposes of infringement. 209. The Cs contended that it is of no consequence that there might be another entity with relevant rights which pre-date those of the Cs - a submission made with RL Polo in mind, no doubt. But if by that submission the Cs are contending that the existence of the RL Polo brand should be ignored, I believe it to be incorrect. 210. In a similar vein, the Cs submitted that if an element of a composite mark is inherently distinctive for the goods in issue, it cannot be disregarded, nor given little prominence in the assessment of the likelihood of confusion. They also submitted that ‘ It is not relevant that other traders who use a similar element can be identified’ , but the existence of other similar elements in the market is likely to affect the distinctive character of the element in question. 211. Finally, under this heading, I mention the way in which the Cs positioned themselves in closing in relation to RL Polo and USPA, and the relevance of the Ds co-existence agreement with RL: i) The Cs said they are not and do not consider themselves to be in conflict with either RL Polo or USPA (in the case of RL because of their co-existence agreement). ii) By contrast, because the Ds coexistence agreement with RL required the Ds to move away from Sign 1 (except in Mexico), the Cs say that recognises a conflict between Sign 1 and another set of polo-themed marks. iii) The Cs also suggested that by moving to Sign 3, that put the Ds into clearer conflict with the Cs. 212. The Ds agree with point i). On ii) they say their co-existence agreement with RL represents a sensible commercial compromise and does not evidence actual conflict - I agree. As for point iii) that begs the question. 213. Overall, the existence and prominence of RL Polo and of USPA in certain markets (in particular not the UK), and of other ‘polo-themed’ brands in varying degrees in other markets, all using variants of horse and rider motifs, indicates that, with RL Polo as the exception, the average consumer cannot rely on the motif as reliably indicating trade origin. In a sense, for a polo-themed brand, a horse and rider motif of some sort is almost de rigeur . Therefore, the average consumer has to rely on other material in the branding as well as indicating origin. This, of course, is consistent with the notion that the average consumer normally views the Mark and the Sign each as a whole and does not dissect either into its constituent elements. The Cs’ case on the words ‘POLO CLUB’ 214. In direct contrast to their attempts to say that all the other ‘polo-themed’ brands were irrelevant, the Cs relied on at least some of them in support of their contention that the words ‘POLO CLUB’ in the Mark and in the Sign were distinctive and, indeed, ‘the only relevant verbal element’. On the other side of that coin was the Cs’ proposition that the words ‘Royal County of Berkshire’ are ‘irrelevant or certainly of significantly lesser importance’. Although not made expressly, this proposition presumably also applies to the words ‘Beverly Hills’ in the Mark. This point emerged in three respects. The point was raised generally in their Opening Skeleton, but it is only necessary to consider it in relation to the South American Overseas Territories. The point is plainly unarguable so far as the UK is concerned and, to be fair, I think it was always aimed at the South American countries. 215. In support of this argument, in their Opening Skeleton the Cs relied on a number of mainly online listings contained in FGC-2 at E2/10/pp85, 102, 181 and one Sears catalogue from the Ds’ disclosure at E2/12/p449 (all of which feature goods bearing Sign 1) and at pp199 and 248 which feature goods from Buckingham Polo Club and Long Beach Polo Club. The Cs contend that in all these listings the name used to identify the brand is ‘Polo Club’. It is true that the headline listing on each webpage includes the words Polo Club viz. p85: Zapatillas Polo Club (shoes); p102 Polo Club Majesty EDT 100ml (fragrance); p181 Bota Polo Club (boots); p199 Marca Polo Club (shirt); p248 Polo Club (shirt). However, even in the reduced size copies of these listings, the branding is clearly visible on the goods in question - whether it is Sign 1 or the brand of Buckingham Polo Club or Long Beach Polo Club. Certainly, anyone viewing the listing on screen and considering a purchase would be able to see the branding. 216. The document at E3/12 is a Sears Mexico 2015 catalogue of 20 plus pages which is evidently designed to promote a range of RCBPC goods, and features photographs of a RCBPC store presumably in a Sears department store. The Cs drew attention, on the ‘Back to School’ page, to a tennis shoe bearing Sign 5 listed as ‘Tennis tipo chocio POLO CLUB’ and a pile of polo shirts all bearing Sign 2 listed as ‘Camisas tipo polo POLO CLUB’. However, the Cs’ point is destroyed by the fact that the page in the Sears Catalogue has a banner across the top with ‘Sears Catalogue Back to School 2015’ with ‘Royal County of Berkshire Polo Club’ on the left and Sign 1 on the right, and the whole catalogue promotes the RCBPC brand even though it features other products as well. 217. In Closing, the Cs added some other pages which, it was submitted, evidence an ‘obvious desire on the part of brands to emphasise the ‘POLO CLUB’ element of their signs and to lessen the impact of any other element of their names. The examples given were E1/10/p66 York Team Polo Club; p67 Long Beach Polo Club; p111 & 112 New Forest Polo Club and p128 Royal London Polo Club. In each instance it is true that the words POLO CLUB appear in a larger font below the first two words which appear in a smaller font, apparently to make them fit into the same horizontal space (although York Team is the exception, being in a smaller font still). 218. The third aspect of this point concerns some of the evidence given in relation to the Albrook Mall. On one of the Albrook Mall listings (from 2015) a store was listed as ‘Polo Club’ in the Leon section at 41. Señor Garcia accepted this was a reference to the RCBPC store. Another Albrook plan identified the RCBPC store as ‘Polo Royal C Berkshire’. What was clear from all the various Albrook Mall plans I was shown was that, due to the number of stores/outlets which had to be included, there was insufficient space to set out ‘Royal County of Berkshire Polo Club’ in full. Hence the need for abbreviation. There was sufficient space (just) to set out U. S. Polo Association and Beverly Hills Polo Club. Anyone who arrived at the ‘wrong’ store would see the clear branding on the fascia and inside - only a consumer not paying any attention or who just wanted the generic ‘polo’ look could miss it. 219. The Cs seemed immune to the irony of attempting to dismiss Señor Garcia’s evidence (and exhibits) in their entirety (and all the ‘polo-themed’ brands) yet relying on their chosen selection. To be clear, I do not consider the few isolated examples identified by the Cs in Bundle E establish their contention that the words POLO CLUB were or are distinctive, whether in South America or elsewhere. In the online listings, it appears the words POLO CLUB were used to attract purchasers i.e. they were used as indications of the type of goods - goods associated with the ‘polo club’ vibe. The pages in [217] above are slightly different but it would appear the emphasis on the words ‘POLO CLUB’ was not because those words are distinctive: quite the opposite, in my view: they were emphasised to emphasise the ‘polo club’ vibe. Generally, I reject the Cs’ contention that the words POLO CLUB are distinctive, whether in South America or elsewhere, along with the related proposition that the other words in the Sign and Mark are either irrelevant or of lesser importance. Apparent Conflicts between the brands Complaints from Licensees 220. Mr Haddad’s first witness statement contains a short section with this heading. He says that in around 2011 he started to receive complaints from his licensees in relation to the RCBPC brand. He says the complaints started in South America and then licensees in Europe started to complain. He says our licensees were ‘up in arms’ because of confusion in the market, as a result of the RCBPC’s Brand’s use of a logo similar to the Cs’. He refers to complaints from a long-established licensee, Onixco, which operated in Central America and Panama in particular and he says that following the arrival of RCBPC in those territories the royalties received from Onixco reduced substantially, and he gave the figures. 221. Mr Haddad refers to the email from Mr Dayan of Onixco dated 15 th December 2011, which I have already discussed. He also refers to complaints from two other South American licensees who terminated their licences for the BHPC brand following their complaints about the RCBPC Brand. One was a licensee from 2009-2013 and the other from 2011-2014. Mr Haddad says he received these complaints in separate phone calls. The complaints were that they felt the RCBPC Brand was infringing the BHPC brand and it was also undercutting their prices. One particular complaint in 2013 concerned RCBPC copying BHPC’s luggage designs. 222. I have already reviewed the documentary records of these complaints. Generic complaints from licensees need to be taken with a pinch of salt. They take a license to make money and they are liable to complain about anything which reduces their profitability in order to improve their bargaining power with their licensor. The negotiation with the licensor is either to persuade the licensor to reduce the royalty, or to provide more marketing support, or support for legal action or something else of benefit to the licensee. 223. I am inclined to contrast generic complaints from licensees with specific, documented complaints which provide evidence that confusion attributable to the branding has occurred in real life. F/9/160/pp2401-2408 224. This document (which I will refer to as F/160) appears as a margin reference in both the witness statements of both Mr Haddad and Ms Borycz. Ms Borycz was clearly well aware of it because she spontaneously brought it up in the course of her cross-examination. It appears she supplied some of the images used in it. It was never made clear who had actually created this document, but the implication was that it was prepared by Cs’ solicitors. It comprises 14 comparisons of products. The products have been carefully chosen to look as similar as possible. 225. The first three entries look like this: 226. As no doubt intended, this is an arresting image, even though it is very unlikely these goods would be seen side by side. However, the impact of this image is severely dented when one sees the range of fragrance sprays RL produced from 2010 onwards (images of which were in X/3). These were produced in groups of 4 products, each product in a bright colour (e.g. blue, red, green & orange) and bearing a large number (1, 2, 3, 4, a nod to the number of polo players in a team). The Cs’ deodorant spray products were produced from 2012 onwards and the Ds’ from 2018 onwards. It is plain that the Cs’ products were at the very least ‘inspired’ by the RL products and, at worst, they were copies. However, Mr Silverleaf made the general point that if the BHPC products were distinguished from the RL products, it was by reasonably close attention to the branding and if that was the case, reasonably close attention to the Ds’ branding would also serve to distinguish their products. 227. Other products set out in the F/160 comparison were leather wallets (both with relatively small branding in the corner), leather rucksacks, shoes, polo shirts and shirts. The products for each comparison had been carefully chosen to be as similar as possible. Thus, there were polo shirts from each brand with a bold diagonal stripe from right shoulder to left waist, polo shirts in white with dark branding, polo shirts in dark blue with red branding, polo shirts with contrasting colours left and right and female polo shirts with colourful hoops in contrasting colours. All these combinations mimic the shirts worn by real-life polo teams or are staple products in the market. All the products (save for the particular appearance of the fragrance/deodorant sprays) were essentially staple products in this market. 228. Since the person who created the document did not give evidence about it, I am left to draw inferences. The emphasis in F/160 was on similarity of the products, not the branding. In some of the images it was difficult to make out the branding but this was because of the low quality of the images and/or the fact that some had been reduced in size to fit into the document: in other words, these images were not a reliable guide to what the consumer would actually see or experience. A bigger problem with many of these comparisons was that the products attributed to the Ds did not derive from any of the countries in issue. A number of the images were plainly from Malaysia, bearing prices in Malaysian Ringgits. Mr Haddad defended the comparison on the basis that it showed the sorts of products which could carry the Ds’ branding. 229. Overall, I find the document at F/160 to be a rather clumsy piece of advocacy. It has very little relevance to what I have to decide. It comprises a small selection of products deliberately chosen because they look similar (and presented side by side) from much larger ranges of product. As I have already indicated, most of the products shown are generic in this market. That fact indicates that if a consumer cares about the brand they are purchasing, they must pay attention to the branding. Alternatively, if the consumer just wants the ‘polo’ look, they may not pay much attention to the branding. Ms Borycz’s evidence of confusion 230. The Cs submitted that Ms Borycz’s evidence supported three distinct and repeated types of actual confusion in the Overseas Territories: i) First, that consumers attempted to return RCBPC goods to BHPC stores. ii) Second, that in department stores workers placed RCBPC goods on shelves intended for BHPC goods, alleged to have been reported to Ms Borycz at least once a year. iii) Third, that when people asked for directions to the BHPC store in the Albrook Mall, they were directed to a RCBPC store. 231. In her witness statement Ms Borycz gave evidence about the position in the Albrook Mall in Panama in 2014 and an email she sent to Mr Haddad on 2 nd June 2014. It reads as follows: Hello Eli, See attached pictures. I didn’t make it to the Allbrook mall as traffic on Friday was ridiculous but I had my people go there and they took pictures. They gave them a hard time to take pictures at stores but they did. Pictures taken Monday May 19. This is our one store in Allbrook mall you can see collection is correct but display and pictures horrible. Also I heard Royal County have 2 stores in this mall and when you ask for bhpc store people getting confused and sending people to RYOB store [which she clarified as a typo for RCBPC]. 232. In her witness statement, Ms Borycz referred to the last sentence of her email and said the store manager of the BHPC store ‘told me this verbally in the store in May 2015 that he had received complaints from customers looking for the BHPC store who were directed by staff working at the mall (for example staff at the information desk and security guards) to the RCBPC store. The store manager also told me that consumers tried to return RCBPC goods to them.’ 233. Ms Borycz says she visited the Albrook Mall herself in September 2015 and ‘asked the security guard where the BHPC store is’. The conversation took place in Spanish. The security guard ‘specifically pointed me to the RCBPC store. I knew this was not correct as the BHPC store is on a different floor to the RCBPC store.’ In cross-examination, Ms Borycz acknowledged she was mistaken in thinking the RCBPC store was on a different floor. Mr Silverleaf also suggested she might have had one of the USPA stores in mind, because they were on a different floor, but I do not need to resolve that dispute. The general point was that Ms Borycz’s recollection was not entirely reliable, but that is not surprising in relation to events some 7-8 years ago. 234. She also went on to say ‘The BHPC licensee still has the same store in the Albrook Mall. That licensee has often complained to me about RCBPC that it is affecting its business because of the confusion.’ She acknowledged that first sentence was wrong even when she made the witness statement in April 2022. 235. As a result of her written evidence, in her cross-examination there was a detailed examination, based on floor plans of the Albrook Mall with various dates, of where various ‘polo’ brands were positioned. I can summarise all this evidence as follows. Unless otherwise stated, these findings relate to the position in 2015: 236. The Albrook Mall is a very large shopping mall on the outskirts of Panama City. It is reputed to be the largest mall in South America. It has its own station on the Panama Metro, its own bus station and even its own chapel. It has at least 11 entrances/exits. It comprises two floors. There are so many shops that the mall is divided into 14 sections, each named after an animal. There are hundreds of shop units, from large department stores (Stevens, Conway, Titan, El Costo, Madison, Collins, Dorians, Super 99, Felix B. Maduro) but with a preponderance of smaller retail units, varying in area. It contains three food courts, with about 12, 50 and 36 outlets respectively, and a hotel on one side. In between the fixed retail units there are a large number of kiosks, perhaps as many as 100. It is said there were 700 shops in total. 237. A number of the store names are not familiar, but a good number are. These are what I would call middle or mass market brands - Columbia, Tommy Hilfiger, Nike, New Balance, Vans, Skechers, Tommy Jeans etc. 238. At all material times, the BHPC store was in the Rhino section on the ground floor, store 93, in one of the smallest fixed retail units. Above it (on the upper floor), but slightly to the right as one looks at the store, there was a large US Polo Association store also in the Rhino section, at least 10 times larger in terms of floor area. There were two other USPA stores. One in the Dino section was also on the ground floor perhaps 100m away towards the left hand exit (about twice the area of the BHPC store). The other was in the Gorilla section on the upper floor, probably more than 500m away, about 4 times the surface area of the BHPC store. 239. One RCBPC store was between 300-400m away from the BHPC store, on the ground floor in the Panda section. Ms Borycz remembered that the BHPC store was always in Rhino and the RCBPC store was always in the Panda section. At certain times there was another RCBPC store in the ‘Leon’ or Lion section. 240. From the detailed floorplans, it did not appear that RL Polo operated an outlet. I consider it is highly likely and I so find that RL Polo clothing was on sale in at least some of the department stores operating in the Albrook Mall, indeed Señor Garcia’s evidence suggested RL Polo was on sale in the Stevens department store in the Albrook Mall, co-existing with 3 USPA stores and, at various dates, one or two RCBPC stores and one or two BHPC stores. 241. There are the following general categories of ‘confusion’ in this evidence. 242. The first, and the most specific category, is concerned with people being misdirected at Albrook Mall. These incidents have nothing to do with the horse and rider part of the respective brands and can only have been based on some version of the name used. The most likely explanation for these mis-directions is that the person at the information desk or the security guard picked up on the mention of ‘polo’ or ‘polo club’ and gave directions to the store with which they associated the word or words they picked up on. Although there was no specific evidence of this, if these mis-directions occurred, people might have been directed to the BHPC, RCBPC or USPA stores but anyone interested in the BHPC brand in particular would be able to see, even before they entered the store to which they had been (mis-)directed, whether it was the store they were looking for. Alternatively, if all they wanted was a ‘polo’ type brand, they might have been content to proceed into the store in front of them and look at what was on display. 243. I find that these mis-direction incidents are not probative of a likelihood of confusion for at least the following three reasons. First, a security guard or person on an information desk is not representative of any average consumer I have to consider. Second, they are not presented with either the Mark or the Sign. Third, to the extent that the mis-directions occurred because of the words ‘polo club’ or ‘polo’, they amount, at best, to administrative inconvenience which ‘polo’ brands must put up with in this market. 244. The second category is concerned with staff putting delivered goods in the wrong place (see Mr Dayan’s email of 15 th December 2011). There are perhaps two aspects to this. The reference in that email to ‘corner’ is suggestive of goods being placed in the wrong concession in a department store. Counsel speculated as to the reasons how this occurred, inviting the conclusion that any realistic explanation supports a finding of a likelihood of confusion. I am inclined to disagree. Cartons containing clothing are unlikely to have carried either the Mark or the Sign. At best, they would have been addressed by name: Beverly Hills Polo Club or Royal County of Berkshire Polo Club or the name of the operator of the store or concession. Even if the goods were delivered to the wrong concession or store, such incidents have little to do with the horse and rider part of the branding and, again, must have been based on the names and again, are likely to have occurred because of the common use of ‘polo club’. 245. The second aspect arises from something said by Ms Borycz in her cross-examination where she suggested the wrong goods were put on the shelves, meaning, as she said, the staff were confused. Again, this was very second-hand evidence lacking specificity. If such incidents occurred, the staff must have been in automatic pilot, and not paying any attention to the branding, neck labelling or swing tags. 246. On either aspect, such incidents amount, in my view, to nothing more than administrative inconvenience. There is nothing to indicate that the persons making the deliveries or putting the wrong goods on the shelves had any interest in either brand which approximated to the interest of any category of average consumer as purchaser. 247. The third category is concerned with the behaviour of consumers and concerns an alleged incident where the BHPC store manager told her that consumers tried to return BHPC goods to them. Although it was not an answer to any question put to her, in cross-examination Ms Borycz said ‘there is a lot of people coming and keep confusing the brands and returning the stuff, which is not with BHPC brand.’ This category is not clear cut. If there was other evidence indicating actual confusion, the incident or incidents might lend some support a finding of a likelihood of confusion. However, the evidence that such incidents occurred was at best second-hand and unspecific in the sense that no records were kept as to when any such incidents occurred or how frequently. On their own, they might well amount to administrative inconvenience. In addition, I must take into account, for the reasons explained below, that there was a degree of exaggeration in Ms Borycz’s oral evidence. 248. The fourth category comprises the general assertions of ‘confusion’. In Ms Borycz’s cross-examination it became increasingly apparent to me that she was essentially programmed to say that anything associated with RCBPC was confusing. For example, by reference to photographs of the front of the BHPC and RCBPC stores in the Albrook Mall, taken in 2014, she thought they looked similar, especially because of the horse and rider logo on each fascia. The BHPC store was refitted in around May 2015 so it no longer had two glass windows either side of the double doors, but had an entirely open front. She accepted the furniture and set-up inside the two stores were different but she maintained that the look, the feel, ‘it is very confusing’. She also pointed to the luggage shown in the windows in the two photographs and said the colours (which were e.g. standard colours blue, red, turquoise, pink and khaki) and the placement of the logos (on the front) were, ‘for me, similar and can be confusing’. 249. This exchange prompted Ms Borycz to recall the document at F/160. In relation to those comparisons she said ‘I know the styles which are being confused…’. She said she supplied the pictures ‘which I believe were confusing’. She said F/160 was ‘just to bring attention to how much confusing can be products.’ 250. In re-examination, she was asked again about a photograph of the front of the RCBPC store. She immediately focused on the ‘luggages’ shown in the window and said ‘which I know they were at the same time similar to ours. That is why our partners had a problem because customers are very confused’. Taken to an earlier photograph of the RCBPC store, she said ‘I see the horse and rider everywhere, I mean on the top’ [by which she meant the fascia], but that answer ignored the name. 251. It was her evidence in this fourth category which led me to the realisation that she approached her evidence programmed to find anything relating to RCBPC confusing. In her attempts to justify her answers, she ended up focussing on irrelevant matters (similarities in types of product, especially the luggage) or focussing just on the horse and rider logo (itself an implicit recognition that the names were different and distinct). 252. Having considered these categories separately, I must step back and consider them collectively. Having done so, I was struck by how insubstantial this evidence was, bearing in mind the fact that these two brands had traded in the same market (in Panama in particular) for many years, possibly as much as 10 years (2011-2021). Even if the period was only 7 years (say 2011-2018), I would still have expected much more evidence of confusion to have come to light, if it had really been occurring. The circumstances in this case are not ones where instances of confusion would not come to the attention of Mr Haddad or Ms Borycz. Evidently, licensees were not shy of making complaints. Furthermore, Mr Haddad and the Cs have been involved in various trade mark disputes for a number of years. Even if the early years were dominated by oppositions and not infringement litigation, the Cs must have understood the value of recording and documenting instances of confusion which came to their attention. 253. In these circumstances, if there really was confusion occurring between these brands by purchasing consumers, I would have expected a much more substantial body of incidents to have been reported and recorded. The section 10(2) case Overview - the Cs’ principal contentions 254. Before I undertake the required assessment of the (degree of) visual, aural and conceptual similarity between the Mark and the Sign, it is worth discussing the principal contentions made by the Cs in various guises. I start with the Cs’ case of direct infringement (under s.10(2) and equivalents). The principal contentions were that (a) the horse and rider logo in the Cs’ Mark is distinctive, indeed that it has an independent distinctive role in the Mark (b) the main distinctive verbal element comprised the words ‘POLO CLUB”, (c) because none of the goods in issue are for playing polo or likely to be used by those watching polo matches, the Mark has a highly distinctive character and (d) bearing in mind imperfect recollection, the high level of visual similarity in the horse and rider motifs, along with the words Polo Club, are sufficient to give rise to a likelihood of confusion. Although in opening the Cs contended that the average consumer’s level of attention was ‘moderate’, by the time of closing, the Cs were contending the level of attention was ‘low’ (without any explanation for the change). 255. It is not difficult to discern the purpose or the intended consequence of these submissions. Through them, the Cs seek to downplay the significance of ‘Beverly Hills’ in their Mark, and the significance of ‘Royal County of Berkshire’ in the Ds’ Sign. Those submissions can also be taken to be addressed at a level of generality removed from the actual Mark and Sign i.e. there is a risk that other elements of the Mark and Sign which make an impression on the average consumer are left out of account. 256. However, I remind myself that the Court is required to assess the visual, aural and conceptual similarity between Mark and Sign. In SABEL v Puma (from which point (d) (quoted in paragraph 48 above) amongst others is derived, and where the goods were identical), it is understandable that the CJEU concentrated on these similarities because, in such a case, it is the similarity between Mark and Sign which is going to give rise to the existence of a likelihood of confusion, if one is held to exist. However, it is important not to assess just the similarities. One must also assess the differences and their significance. This is inherent in the requirement to undertake the assessment of similarity as a matter of overall impression i.e. one is assessing degrees of similarity from a visual, aural and conceptual viewpoint. It is also inherent in the CJEU’s later identification of the interdependency principle which is summarised in point (g) above. 257. In general terms, the goods are identical, but precisely which goods I should take into account may vary by the countries I have to consider. 258. I will start by setting out my findings as to the relevant characteristics of average consumers in each relevant market and proceed to set out my findings as regards the assessments of similarity and distinctive character. Average consumers 259. I did not receive direct evidence on these issues but these findings are based on my assessment of the evidence as a whole. 260. In all relevant countries, the goods of both brands have a price premium over unbranded goods. They are not in the prestige sector of the fashion market, but sit in what is sometimes called the mass-tige sector. The relevance is that consumers of these goods are prepared to pay more for the brand, and therefore have a degree of brand awareness. This means that, contrary to the Cs’ position in closing, the level of attention is not low but at least medium if not slightly higher. 261. For the reasons already explained, I find that this brand awareness extends to a general recognition of other ‘polo’ brand(s) in the same territory. For all territories in issue, this means at least RL Polo. In some territories it includes other ‘polo’ brands. 262. In saying this, I recognise that consumers do not have the overview that the evidence in this case presents to me. In any event, I also recognise that the average consumer of the goods in question will not have encountered most of the other ‘polo-themed’ marks present in the relevant territory. However, even if they have never purchased a RL Polo product, the average consumer for these goods will be aware of that brand and its horse and rider logo at least and is likely, depending on the territory, to have encountered other ‘polo-themed’ brands as well. 263. Beyond those points, the familiarity with English varies across the countries in issue. It is convenient to assess that familiarity and its consequences in the context of the assessment of the similarity between Mark and Sign. UK - analysis of similarity 264. The Mark and Sign in question here are complex or composite marks. The overall impression of each one which is carried in the mind of the average consumer is of the whole, but it is not photographic recall since it is necessary to take into account imperfect recollection and the fact that Mark and Sign are not seen side by side by the average consumer. Furthermore, the branding is relatively small in each case. However, the key point I decide here is that no particular component dominates in either the Mark or the Sign. In particular the horse and rider motif does not dominate in either the Mark or the Sign. The reason is simple. It is because the Cs’ horse and rider motif was not the first in this market, is not unique, and has had to co-exist with others. In fact, as I have already mentioned, one might say that some form of horse and rider motif is almost de rigeur for one of these ‘polo’ brands. This chimes with the finding made by Marcus Smith J. in Greenwich Polo that the horse and rider motif signifies a generic link to polo and to sport (see Lifestyle Equities CV v The Copyrights Group [2021] EWHC 1212 (Ch) at [51(1)]). 265. I also find that the horse and rider motif does not retain an independent distinctive role in the Mark. 266. My assessment of the degrees of visual, aural and conceptual similarity between Mark and Sign is as follows. I emphasise that the detail contained in the following analysis is my attempt to explain in words what the mind sees and recalls in each of the Mark and Sign. Necessarily the explanation is more detailed than the impression which is left. 267. So far as the visual elements of the Mark is concerned, the most important and distinctive element is at the top - BEVERLY HILLS - in clear block capitals in the curved shape. The eye is drawn to that element. It wraps around the horse and rider motif which is also encased by the words POLO CLUB at the foot. It is the careful juxtaposition of the three elements which means they are taken together. 268. So far as the Sign is concerned, again it is composed of three elements but they do not hang together in the same way and are more separate. The incomplete horse and rider motif floats free above the words. The second element comprises the words Royal County of Berkshire in the particular italicised script which is more difficult to read, with the third element POLO CLUB in block capitals at the foot. 269. The contributors to visual similarity are that both feature POLO CLUB in capitals at the foot and both feature a horse and rider logo. However, the names are different - BEVERLY HILLS vs Royal Country of Berkshire, the overall shape is different and there are some obvious differences in the horse and rider motifs: direction of movement, complete vs partial and in the degree of shading. 270. Overall, I assess the degree of visual similarity as low to medium. 271. In terms of aural similarity, the figurative elements cannot contribute. The average consumer would refer to the Mark as ‘Beverly Hills Polo Club’ possibly abbreviated to Beverly Hills. The Sign would either be referred to as ‘Royal County of Berkshire Polo Club’ but since that is a bit of a mouthful, more likely to be abbreviated to ‘Royal County’ or possibly Berkshire Polo Club. In other words, the only aural similarity exists in ‘Polo Club’. Overall I assess the degree of aural similarity as between low and medium. 272. The principal conceptual similarity lies in the fact that both Mark and Sign signify a polo club. This is signified via the words POLO CLUB and the horse and rider motifs. The Mark and Sign are by no means conceptually identical because the Mark signifies a polo club in Beverly Hills, which is reasonably well known as a very affluent area in the western US, whereas the Sign does not have a US feel at all, the words Royal and Berkshire having a much more British feel. I assess the degree of conceptual similarity as slightly above medium. 273. In terms of the relative importance of each, I agree with the Cs that, in this market, the visual impression is the most important since these goods will be selected primarily on the basis of visual inspection. Aural impression is the least important. 274. It can be seen from my analysis that I reject the Cs’ attempt to ignore the words ‘BEVERLY HILLS’ in the Mark and ‘Royal County of Berkshire’ in the Sign. Both are effectively the start of the Mark and Sign respectively and are most unlikely to be ignored or missed by the average consumer. Even with imperfect recollection in mind, these elements have a material effect on the degree of similarity between Mark and Sign. 275. So far as the UK is concerned, I can proceed on the basis that the goods are identical, comprising items of menswear (e.g. polo shirts), womenswear, children’s clothing, wallets, bags and luggage. I will assess later the influence of the Cs’ reputation for luggage, men’s underwear and children’s clothing. EU - similarity analysis 276. So far as the EU analysis is concerned, I must take account of the fact that English is not so well understood in certain countries in the EU. However, Italy represents the Cs’ best case in the EU. There, I consider there to be sufficient familiarity with English in branding. There will be slightly less recognition of Berkshire than for the UK average consumer. Overall, however, I do not consider those differences make any material difference in the analysis of similarity. Panama - similarity analysis 277. The issue is whether the analysis is affected by (1) the fact that most consumers in Panama will have Spanish as their first and possibly only language and (2) whether such consumers have a different perception of the place names. In relation to Panama, the Cs emphasised their argument that POLO CLUB would be seen as distinctive. This argument seems have arisen from some evidence given by Ms Borcyz which indicated that it was the use of POLO CLUB which gave rise to some of her instances of confusion. 278. I take judicial notice of the fact that even Spanish speakers in Central America will have been exposed to English terms in advertising. Thus, I find that average consumers in Panama understand the words CLUB and POLO, and understand the Mark and Sign are badges of ‘polo clubs’, but they do not pause to consider whether these signify actual polo clubs. I also find that they will be aware of the exclusive and affluent aura of BEVERLY HILLS through the media i.e. magazines, TV and film. They will also have some awareness of the British Royal family. Whilst they are unlikely to be able to place Berkshire, they associate ‘Royal County of Berkshire’ with Britain. I also take into account the fact that the market for the relevant goods in Panama is likely to attract a significant number of tourists from other countries who are likely to have greater familiarity with English words. These are all influences which operate in the mind of the average consumer but I emphasise they are not conscious thoughts when either Mark or Sign is encountered. If it is not already clear from the above, I reject any notion that POLO CLUB is a distinctive or important element of the Mark for average consumers in Panama. If mistakes arose due to the use of the term POLO CLUB for either of the stores in the Albrook Mall, that is nothing more than administrative inconvenience. 279. Overall, although average consumers in Panama have a slightly different impression of each mark, since the Mark and Sign are primarily visual and the concepts involved in both Mark and Sign are essentially the same as in the UK, I do not find there is any material difference in the overall assessment of similarity. Mexico, Chile, Peru 280. So far as the other countries in South America are concerned, I proceed on the basis that familiarity with English words amongst average consumers is somewhat less in Chile and Peru than in Panama and Mexico, but also that awareness of polo (and the concept of polo clubs) may be slightly greater in South America than in the UK. 281. Overall, having considered the rather limited information available to me about the situation in these other countries, once again I do not find there is any material difference in the overall assessment of similarity. UAE 282. I was provided with almost no information at all about the situation in the UAE. Familiarity with English words will be high in the UAE and the market for these goods in the UAE is likely to include a significant number of tourists with a good understanding of English words. The average consumer in the UAE will not have as precise an understanding of Berkshire as someone in the UK, but s/he still understands the aura of ‘Beverly Hills’ and the association of Royal and Berkshire with the UK. 283. Once again, I do not find there is any material change to be made to the assessment of similarity. 284. Standing back, I do not find the lack of variation between the similarity analysis in the various countries surprising. This is because the Mark is international in nature, capable of being understood in many countries, not just those where English is the primary language. As I have indicated, the Sign is very slightly more difficult to understand in countries where English is not the principal language, but not so as to alter in any material way the overall analysis of similarity. Assessment of distinctive character 285. In the past I have noticed the slight lack of clarity in the CJEU jurisprudence as to at which stage of the analysis one takes account of the distinctive character of the Mark. On the one hand, it can be conceptually difficult to put out of one’s mind the distinctive character of a Mark when making the similarity comparison, even if one ignores distinctive character acquired through use. On the other hand, if, as directed, one takes the distinctive character of the Mark (both inherent and acquired) into account specifically when conducting the global assessment of the likelihood of confusion, it is important that distinctive character is not taken into account twice in the overall analysis. 286. I refer to point 8 from the Judgment of Kitchin LJ in Comic Enterprises cited in paragraph 58 above. It is useful to remind myself specifically of the passage: ‘(8) The issue of a trade mark’s distinctiveness is intimately tied to the scope of the protection to which it is entitled. So, in assessing an allegation of infringement under art.5(1)(b) of the Directive arising from the use of a similar sign, the court must take into account the distinctiveness of the trade mark, and there will be a greater likelihood of confusion where the trade mark has a highly distinctive character either per se or as a result of the use which has been made of it. It follows that the court must necessarily have regard to the impact of the accused sign on the proportion of consumers to whom the trade mark is particularly distinctive. 287. That passage provides a strong indication that distinctive character (inherent and acquired) feeds into the global assessment of the likelihood of confusion. 288. The further point I take into account is the influence of a mark having a reputation for certain goods. Normally, a reputation does translate into some enhancement of the distinctive character of the mark. In this case, the Cs rely on their reputation as conferring enhanced distinctive character on the horse and rider motif and, in the Overseas Territories or some of them, in the words POLO CLUB. 289. In order to do that, I must assess the distinctive character of the Mark in the various countries in issue. 290. So far as the UK is concerned, although I proceed on the basis that the Cs have a reputation for luggage, men’s underwear and childrenswear, and I will proceed on the basis that the Mark as a whole has some enhanced distinctive character for those goods, I remain unpersuaded that the trading under the Mark in the UK has been sufficient to give rise to enhanced distinctive character in the horse and rider motif, as the Cs alleged. 291. In the EU, as I have indicated, the considerable level of trade conducted by Dream Project in Italy has generated reputation in the Mark in Italy for a range of goods. That trading has also given rise to some enhancement of the distinctive character of the Mark as a whole, but since that trading has taken place in conjunction with RL Polo, the enhancement is limited. Specifically, I am not persuaded that that trading has enhanced the distinctiveness of the horse and rider motif in the Cs’ Mark so as to render it either a dominant or important element in the Mark. 292. The Cs did not rely on reputation in any of the other countries in issue. Nonetheless they contended that the horse and rider motif in particular possessed enhanced distinctive character in all relevant territories. So far as the Overseas Territories are concerned, I reject that contention. The evidence indicated there are more ‘polo’ brands on the market in Panama and Mexico than in the UK or EU. In any event, RL Polo is, in a very real sense, the dominant ‘polo’ brand in every relevant territory, even if average consumers have never purchased a RL Polo product. 293. It is difficult to put into words one’s assessment of the distinctive character of a Mark. I know that UK IPO hearing officers tend to use terms like low, medium, average or high, but they are imprecise indicators of what the decision maker has in mind. I find the Mark has a reasonable degree of distinctive character, but that resides in the whole. Specifically, I do not accept the Cs’ contention that their horse and rider motif has particular distinctive character because it has been used in relevant markets with RL Polo and, in certain territories, other ‘polo’ brands as well. Global assessment of the likelihood of confusion 294. At this stage of the analysis it is necessary to take a step back from the detail and to ensure one is looking through the eyes of the average consumer, who has an imperfect recollection of the Cs’ Mark and who encounters Sign 3 on the goods in question. The average consumer does not have the opportunity to make a side by side comparison of the type set out at the start of this Judgment. Indirect Infringement 295. I start by mentioning the Cs’ fallback case on indirect infringement (in the event that their case on direct infringement did not succeed). The Cs contended as follows: ‘To the extent that the average consumer (with their imperfect recollection) recognises there is a difference between those words [sc. between ‘Beverly Hills’ and ‘Royal County of Berkshire’], and recognises that it might be important as an indicator of origin of the goods, they (or at least a significant proportion of them) will likely consider that they are seeing a sub-brand. Similarly, to the extent that difference in the presentation of the horse-and-player is noticed and recalled, that will be considered as an indicator of a sub-brand, or maybe more likely of a brand update / refreshing.’ 296. Although in the analysis above, and below (which focusses on direct infringement) I had in mind this case on indirect infringement, it is convenient to dispose of the case on indirect infringement at this point. With the principles applicable to indirect infringement in mind (see [54]-[55] above), I am not persuaded by either argument. The circumstances in this case, with the crowded market along with the elements in the Mark and the Sign, militate against anyone viewing the Sign as a sub-brand or as a brand refresh. Accordingly, I reject the case on indirect infringement. Direct Infringement 297. At this global assessment stage of the analysis, I propose to start with the claim of direct infringement relating to Panama, then Mexico, Peru and Chile. My reason for starting with Panama is because of the long period of concurrent trading in both brands in the Albrook Mall. The form of that trading was such, in my view, as to be very likely to bring instances of customer confusion between the brands to light. As I have already indicated, what struck me was how insubstantial the supposed evidence of confusion was. Panama 298. I have already indicated the characteristics of average consumers in Panama. The respective stores in the Albrook Mall both sold a range of goods which I will take to include menswear, womenswear and children’s clothing and luggage. It is sufficient to proceed on that basis - i.e. these goods were identical. 299. I have already assessed the similarity between the Mark and the Sign and the distinctive character of the Mark, noting that in the Albrook Mall by way of example, RL Polo goods were on sale in the department stores and there were, at some times, three USPA stores in the Mall. 300. My assessment, taking all relevant considerations into account, is that no significant proportion of the relevant public was or is likely to be confused so as to warrant the intervention of the Court. I find no infringement in Panama. Mexico, Chile, Peru 301. Although I have noted some slight differences between the considerations applying to Panama and these countries, I do not find they are such as to require a different conclusion. I find no infringement of the Mark in these countries. UK 302. The situation I have to consider in the UK is rather different. The Ds have sold very small quantities of goods in the UK, essentially from the Club shop or pop-up shops at events on D1’s grounds. Due to the inadequacies of the Cs’ evidence, in relation to most of the goods in issue, I am assessing an essentially unused Mark. 303. The Ds have made it clear that they intend to appoint a UK licensee but it is clear that will only occur if the Ds prevail in this action. I proceed on the basis that a UK licensee would be likely to introduce a range of clothing, bags and luggage. For that reason, the UK action is essentially quia timet. Again, the goods are identical. 304. Although there has not been the type of side by side trading in the UK as has occurred in Panama, that side by side trading sheds some light on what would occur in the UK. In addition, I consider that UK average consumers have a greater ability and propensity to distinguish between these brands because they will more clearly recognise the place name contained in each of the Mark and the Sign. 305. In all the relevant circumstances in the UK, I find there is no likelihood of confusion. Infringement of the EUTMs 306. I should first mention Brexit. These proceedings were commenced before the end of the transition period so by virtue of Article 67 of the Withdrawal Agreement, the provisions of the EUTMR continue to apply and the geographical scope of the Cs’ EUTMs include the UK for the purposes of the Cs’ allegations of infringement of their EUTMs. The Court therefore retains the power to grant pan-EU relief pursuant to Article 130 of the EUTMR. 307. Turning to the facts, I received very little evidence as to where in the EU (the UK aside) trading had actually occurred under the Ds’ brand. There was some mention of trading in Germany, but no detail was provided. As I related above, the global (inferred retail) sales figures for goods bearing D1’s brand have shrunk considerably since 2015 such that he indicated that sales in the EU in 2021 were only £1.27m. This is consistent with the downward trajectory of such sales in view of the campaign of oppositions and litigation which the Cs have waged against the Ds. 308. I have found that reputation attaches to the Cs’ Mark in the following countries and for the following goods in the EU: i) In Italy, for menswear predominantly, with some kids wear, footwear, bags and accessories; ii) in Spain, for bedding and bags; iii) in the UK, for Luggage, Men’s Underwear and, due to sales in 2017 and 2018, for kids wear. 309. The Cs’ best case is in Italy. However, the only evidence as regards the sale of the Ds’ goods in Italy is a single webpage from amazon.it which was printed in September 2018. The listing concerns a grey polo shirt with a broad red stripe running diagonally from left shoulder to right waist. In addition, there are two shield devices on the front, neither of which is remotely like any of the Signs. It is difficult to tell, but there might be a representation of Sign 3 in the neck label. I was not provided with a translation of the text, which is in Italian but I can make out that the page says the item is no longer available, but as having been added to Amazon in May 2016. There are no reviews. This listing has all the hallmarks of relating to a single one-off sale. 310. In the circumstances, I question whether there is any realistic threat of trading under the Ds’ mark in Italy. I am inclined to believe that any licensee of D1’s brand would be cautious before launching a range of clothing bearing D1’s brand in Italy, in view of the reasonably extensive trading under the Cs’ Mark in Italy. 311. Notwithstanding the trading in Italy under the Mark, it has taken place against the backdrop of RL Polo, of which the average consumer in Italy would be aware. In the circumstances, I find that there has been no infringement of the Cs’ EU Marks under Art.9(1)(b), whether in Italy or anywhere else in the EU. It also follows there is no threat to infringe any of the EUTMs. 312. For the avoidance of any doubt, my findings in relation to Sign 3 apply equally to Sign 1. 313. Finally, I mention that, in Opening, the Cs relied on post-sale confusion, citing Anheuser Busch at [60] . Although I have considered this citation, it does not cause me to alter my conclusions. Infringement under section 10(3) and/or art 9(1)(c). 314. I will deal with this relatively briefly. I have made findings as to reputation in Italy, Spain and the UK. Once again, the Cs’ best case is in Italy, assuming there is a threat to use Sign 3 in Italy. Although all three forms of injury were pleaded, in Opening, the Cs identified dilution of distinctive character as the relevant form of injury. Reliance on this form of injury is consistent with the Cs’ case that their horse and rider motif has an independent distinctive role and character. 315. Bearing in mind all the relevant circumstances, I find that the average consumer would make no link between Mark and Sign. Even if I assume that a link would be made, it would not, in my judgment, give rise to any of the three forms of relevant injury or any damage to the Cs’ Mark at all. At best, a consumer might wonder whether there was a connection and conclude not due to the differences between Mark and Sign. Passing off 316. It was common ground that if the Cs did not succeed on infringement of their registered trade marks, they would not succeed in passing off. In any event, I find the Cs’ case in passing off fails. What I have not relied on 317. For the avoidance of doubt, I list here the material which I have not relied on and not taken into account at all. 318. The first category concerns all the various oppositions and cancellation actions between the Cs and D1, including appeals from the EUIPO to the General Court and the CJEU. Although D1 has won some of these encounters, it is fair to say that the Cs have prevailed in most of them. I can dismiss all of them as irrelevant to my analysis because they were not presented with the evidence which I have. 319. The second category concerns other decisions involving polo brands. The Cs placed particular reliance on the decision of the General Court in Polo Club Dusseldorf v EUIPO T-355/21, 8 th June 2022. The assessments of distinctive character in that case and the distinction between references to polo clubs and football clubs are of no relevance or assistance. 320. Third, it became apparent that the Cs had conducted two surveys which were not included in their disclosure. The Cs said they were not obliged to disclose them since disclosure was limited to instances of actual confusion. The two surveys were not in evidence but the Cs attached copies to their Closing Submissions. I have reviewed them, de bene esse : i) The first survey was conducted in Panama in October 2011, comprising a 10 minute interview with 419 respondents. The order of questions is not entirely clear, but at some stage respondents were shown the Mark and perhaps one of Sign 1 and Sign 5 (in reverse colours) and they were asked (in effect, because the translation is rough) ‘Which of the following affirmatives [sc. Statements] are you agreed with?’ The statements and reported percentages are: Logos are nothing similar 15%. Logos are a little similar 44%. Logos are pretty much similar 30%. Logos are a lot similar 11%. ii) The second survey was conducted in Warsaw in July 2012 using a ‘Computer Assisted Web Interview’ of 200 respondents in which, after assessing the degree to which the Mark or Sign 1 had been seen before (the awareness was low), those two logos were presented on screen side by side for 5 seconds and respondents were asked: Q5: Please rate on a scale whether the two logos are similar and Q6: what elements of the logos are similar in your opinion? For Q5, the choices were: not similar at all 3%; slightly similar 31%; quite similar 47% and very similar 19%. The top rated answers to Q6 were: The horse 68%; the rider 47%, polo player 31%; name/lettering Polo Club 28%; similar font 21% and so on. Q7 was ‘Please rate the likelihood of confusion products with such logos in the shop’. The ‘confusion scale’ was: ‘I would certainly not confuse these products 9%; I would probably not confuse these products 19%; I would confuse these products 60%; I would certainly confuse these products 11%; DK 1%. 321. Both surveys are perfect examples of a survey which asks leading questions of a scenario which would never be encountered in real life by a consumer and which yields results of absolutely no use or probative value. It is not surprising that the Cs placed no reliance on them. The only point of interest which emerges is the date of these two surveys. They indicate that the Cs were interested in taking action against D1 as early as 2011/2012. 322. Finally, it is common ground that a meeting took place between Mr Haddad and Mr Amoore and D3 at D1’s premises on 12 th March 2012. Mr Haddad gave his account of this meeting in his first witness statement at [26]-[29]. D3 responded in some detail, giving a rather different account, with which Mr Amoore agreed. D3’s account was not put to Mr Haddad in cross-examination, so I assume that the Ds concluded that whatever happened at the meeting would not assist me in resolving the issues I have had to decide, a conclusion that I agree with. There is no reason to resolve the dispute about what happened at that meeting. 323. In view of my conclusions on infringement and passing off, I will deal with the remaining issues briefly, in case this litigation goes further. Joint and several liability Applicable Principles 324. In their Opening Skeleton, the Cs relied on reviews of the applicable principles in Lifestyle Equities CV v Santa Monica Polo Club Ltd [2020] EWHC 688 (Ch) at [25]- [27], the Deputy Judge, Recorder Douglas Campbell QC, citing for authorisation CBS v Amstrad [1988] AC 2013 and also referred to Twentieth Century Fox v Newzbin [2010] FSR 21 , and Fish & Fish v Sea Shepherd [2015] AC 1229 on common design. The appeal from his decision by the Ahmeds is reported as Ahmed v Lifestyle Equities [2021] EWCA Civ 675 where Birss LJ reviewed the earlier CA decision in MCA v Charly , setting out the four principles stated by Chadwick LJ at [49]-[52] '[g]iven their centrality to the issues on this appeal'. Those four principles are equally central to the issues in this case, so it is appropriate to set them out: '49. First, a director will not be treated as liable with the company as a joint tortfeasor if he does no more than carry out his constitutional role in the governance of the company—that is to say, by voting at board meetings. That, I think, is what policy requires if a proper recognition is to be given to the identity of the company as a separate legal person. Nor, as it seems to me, will it be right to hold a controlling shareholder liable as a joint tortfeasor if he does no more than exercise his power of control through the constitutional organs of the company—for example by voting at general meetings and by exercising the powers to appoint directors. Aldous L. J. suggested, in Standard Chartered Bank v. Pakistan National Shipping Corporation (No. 2) [2000] 1 Lloyd's Rep 218 , 235—in a passage to which I have referred—that there are good reasons to conclude that the carrying out of the duties of a director would never be sufficient to make a director liable. For my part, I would hesitate to use the word "never" in this field; but I would accept that, if all that a director is doing is carrying out the duties entrusted to him as such by the company under its constitution, the circumstances in which it would be right to hold him liable as a joint tortfeasor with the company would be rare indeed. That is not to say, of course, that he might not be liable for his own separate tort, as Aldous L. J. recognised at paragraphs 16 and 17 of his judgment in the Pakistan National Shipping case. 50. Second, there is no reason why a person who happens to be a director or controlling shareholder of a company should not be liable with the company as a joint tortfeasor if he is not exercising control though the constitutional organs of the company and the circumstances are such that he would be so liable if he were not a director or controlling shareholder. In other words, if, in relation to the wrongful acts which are the subject of complaint, the liability of the individual as a joint tortfeasor with the company arises from his participation or involvement in ways which go beyond the exercise of constitutional control, then there is no reason why the individual should escape liability because he could have procured those same acts through the exercise of constitutional control. As I have said, it seems to me that this is the point made by Aldous J (as he then was) in PGL Research Ltd v. Ardon International Ltd [1993] F. S. R. 197. 51. Third, the question whether the individual is liable with the company as a joint tortfeasor—at least in the field of intellectual property—is to be determined under principles identified in C. B. S. Songs Ltd v. Amstrad Consumer Electronics Plc [1988] AC 1013 and Unilever Plc v. Gillette (U. K.) Limited [1989] R. P. C. 583. In particular, liability as a joint tortfeasor may arise where, in the words of Lord Templeman in C. B. S. Songs v. Amstrad at page 1058E to which I have already referred, the individual "intends and procures and shares a common design that the infringement takes place". 52 Fourth, whether or not there is a separate tort of procuring an infringement of a statutory right, actionable at common law, an individual who does "intend, procure and share a common design" that the infringement should take place may be liable as a joint tortfeasor. As Mustill L. J. pointed out in Unilever v. Gillette , procurement may lead to a common design and so give rise to liability under both heads.’ 325. I also take account of the discussion by Birss LJ in the paragraphs following the citation from MCA v Charly, and in particular his conclusions at [49]-[51], especially at [49], on which the Cs relied: ‘49. Pulling things together, in my judgment MCA v Charly represents the law on this topic in England and Wales. The protection for a director identified in Chadwick LJ's first principle is not strictly confined merely to voting at a board meeting but it is a narrow protection limited to exercising control through the constitutional organs of the company, albeit that may be something which can be delegated. A director seeking to avail themselves of that principle of delegation in order to escape liability, will need to prove that that has taken place. In any event showing simply that the director has acted properly, in the sense of not acting in breach of their fiduciary duty, is no defence. ’ 326. None of the principles applicable in the UK context were in dispute, so I can turn to consider the various allegations of joint liability. D1 as liable for the acts of its licensees 327. Mr Silverleaf’s starting point was to note that the only primary tortfeasors were D1’s licensees. This is not correct since D1’s own sales from its Club shop were also alleged to be infringements. However, it is true that the major trading complained of was by D1’s licensees. Mr Silverleaf accepted that if D1’s liability for authorising their acts was governed by English law, then D1 would ‘probably’ be liable. He argued however that the issue of joint liability with the licensees must be decided under the local law and since no case had been asserted under the relevant local laws, there was no liability. 328. At [16i)] above, I set out the part of the agreed simplification of this case which was not disputed. The agreement that all issues of trade mark infringement (including in the Overseas Territories) are to be determined by the application of UK/EU law, in my view, carries with it the application of the UK principles of joint liability for such infringement. I reject Mr Silverleaf’s first point. If I had found any acts of infringement, D1 would have been jointly liable with its licensees for those acts, having authorised them. The other allegations of joint liability - summarised at [11 iv)] above. 329. I will deal briefly with the remaining allegations of joint liability. 330. In his closing, Mr Silverleaf argued against joint liability on the facts, submitting as follows: i) That D2 and D3 were carrying out their duties as directors to enable D1 to function, and their acts were the acts of D1. He contrasted the situation in this case with MCA v Charly , where he said the individual was a shadow director operating through a corporate stooge, and with Ahmed, where he said the directors were personally involved in carrying on the infringing activities and thereby made them their own. ii) D4 and D5 are not liable for the acts of D1 or its licensees because they can only introduce prospective licensees and recommend to D1 that it grants a licence. Even after a licence has been granted, it is suggested on behalf of D4 and D5 that they can only act as the agent of D1 and act merely as the means by which D1 exercises its rights and carries out its obligations under the licence. 331. However, largely for the reasons developed by the Cs in their Closing Skeleton Argument at [89]-[97], if I had found infringement, I would have been inclined to find: i) D2 and D3 to be jointly liable with D1, since both carried out acts outside the limited exception described in MCA at [49]. ii) D5 to be jointly liable with his company D4, for essentially the same reason. iii) D1 to be jointly liable for the acts of D4&5 and for the acts of D6-D11 (as above), on the basis of authorisation. iv) D4&5 to be jointly liable with D1 for the acts of D6-D11, on the basis of not only agreeing but actively implementing their common design to licence D1’s brand. 332. Notwithstanding my (contingent) findings of joint liability, I question whether it was necessary to include the allegations of joint liability against the personal defendants i.e. D2, D3 and D5, let alone pursue them to judgment. The bringing and the maintaining of those allegations, plus the allegation of conspiracy to injure by unlawful means, implies that the Cs adopted an approach to this litigation which was as oppressive as possible and not proportionate. It was hardly likely that D1 would disappear by way of a response to this action, and the Cs’ experience with the Ahmeds did not, in my view, justify such an aggressive approach. validity of marks 333. Since the Cs’ allegations of infringement have failed, the Ds’ conditional attack on the validity of the Cs’ registrations does not arise. 334. During the trial the Cs dropped their bad faith attacks against D1’s registrations. The remaining validity attacks were on relative grounds and the parties agreed they stand and fall with the infringement allegations. I dismiss them for the reasons I gave on the infringement allegations. 335. That leaves the Cs’ application for revocation for non-use. The Ds admit that the UK 617 and UK 406 registrations have not been put to genuine use in the UK and agree revocation as from 03.02.2006 and 15.06.2007 respectively. That leaves the allegations of non-use in relation to two registrations for Sign 1, UK 908 for clothing in class 25 and UK 096 for printed matter in class 16 respectively. 336. The Ds say that the unchallenged evidence of Mr Amoore and Mr Morrison show that Sign 1 was used in commerce in the UK up to 2018, so they say there has been genuine use within the relevant 5-year period. They refer to the variety of items of clothing sold by D1 (albeit on a small scale) and various publications of D1 concerning its business running its polo club, as well as its brand book. Although there was no cross-examination on this evidence, the Cs respond by saying that there is no evidence of the volume of any sales of goods in the UK, and no evidence of any use for any goods other than clothing. They suggest the two registrations should be revoked from their earliest available date. 337. The sales of clothing were on a small scale, but were quite clearly genuine use, aimed at maintaining or creating an outlet for the goods or a share in the market for the relevant goods. Similarly, the use on printed matter of various kinds was also genuine use. Conspiracy to injure by unlawful means 338. I remain bemused as to any genuine reason for the Cs to plead, let alone pursue this head of claim down to judgment, other than in an attempt to oppress. Counsel had no answer to the question of what this added to the primary claim of infringement of registered trademarks. I will address it very briefly. 339. In terms of the elements of the tort, I was taken to JSC BTA Bank v Ablyazov [2020] AC 727 at [10]-[11]. I have also had regard to the ensuing discussion of Total Network at [12]-[15]. 340. This cause of action fails because the unlawful means relied on by the Cs are the infringement of Cs’ trade marks, and I have found that no acts of infringement have occurred or are threatened. In the circumstances, it is not necessary to decide whether the Ds had the necessary intent to cause damage to the Cs. I will say, however, that a number of the points relied upon by the Cs in this regard seemed to me to be dubious (e.g. engaging in trade mark oppositions) and probably unnecessary since if infringement had been established, damage would have been caused to the Cs. Conclusion 341. I dismiss the Cs’ action for infringement of registered trade mark, for passing off, and for conspiracy to injure by unlawful means. I will make an order for revocation of certain registrations owned by D1 to the extent indicated above. I will hear Counsel as to the form of Order if it cannot be agreed. I direct that time for seeking permission to appeal shall not run until after the hearing on the form of Order (or the making of such Order if it is agreed) but the parties must co-operate to fix a hearing, if it is required, as soon as possible. 342. Finally, I must apologise to both parties for the length of time it has taken to prepare this judgment. This was caused by the extended time it took me to write the FRAND judgment in InterDigital v Lenovo and the pressure of having to hear other trials and applications in the intervening period since the trial concluded. Postscript regarding CPR 31.22(2) 343. After the provision of corrections to a draft of this Judgment, the Cs’ solicitors wrote to request that the confidentiality attaching to the Cs’ sales figures should not be waived in the public Judgment, and asked that [124]-[128] above should be suitably redacted. They reminded me that at the start of the trial I made a CPR31.22(2) Order pro tem which covered, inter alia, the sales figures at E1/6 & 7, even though I questioned at that stage whether they were truly confidential. Applicable Principles 344. The Cs helpfully drew my attention to the very recent decision of the Court of Appeal in J. C. Bamford Excavators Limited v Manitou UK Ltd & Anor [2023] EWCA Civ 840 (17 th July 2023) where Arnold LJ (with whom the President of the Family Division and Elisabeth Laing LJ agreed) gave a thorough analysis of the balance to be struck between the protection of trade secrets and the requirement for open justice in the operation of CPR 31.22(2). The appeal concerned a trade secret (Criterion X) which was part of a wider package of information which Manitou had disclosed in their Product and Process Description, in response to JCB’s allegations of infringement of four patents by Manitou’s telehandlers which used five different configurations of control systems. Criterion X concerned Configuration C, which HHJ Hacon found did not infringe any of the JCB patents. After evidence and argument on the point, HHJ Hacon found the balance favoured including the identity of Criterion X in the public Annex to his Judgment. 345. Arnold LJ found HHJ Hacon’s decision to be entirely understandable, due to the way in which the point had been argued before him but overturned his decision because the information which Manitou sought to protect was properly characterised as technical trade secrets. He also viewed Criterion X as part of a package of information which Manitou was entitled to keep confidential. 346. At [93]-[97] Arnold LJ reviewed the caselaw on ‘Final Orders under rule 31.22(2)’ and said that the leading authority on the principles to be applied when considering an application for a final order under CPR31.22(2) remains Lilly Icos ( Lilly Icos Ltd v Pfizer Ltd (No.2) [2002] EWCA Civ 2 ), but added that ‘As with any authority, the statements made in that case need to be seen in their factual context’. 347. At [95] Arnold LJ cited the well-known passage from Lilly Icos at [25] where Buxton LJ set out a number of considerations which guided the Court in that case. Pfizer had pleaded commercial success in response to Lilly’s allegation that the claimed invention in Pfizer’s patent was obvious. Pfizer served a schedule setting out its monthly sales figures (on page 1) and advertising and promotional expenditure (on page 2) for the patented product since launch. The only issue in the Court of Appeal concerned page 2 of the schedule. The Court of Appeal allowed Pfizer’s appeal and made a CPR 31.22(2) order in respect of page 2. As Arnold LJ said at [96]-[97]: ‘96. Buxton LJ went on to say that the most important feature of the case was the very limited role that page 2 had played at trial. Although it was covered by rule 31.22(1)(a) because it had been referred to in passing in a witness statement, it was not necessary, or even relevant, for the interested spectator to have access to page 2. It was common ground that the information in page 2 was confidential to Pfizer. In those circumstances the Court of Appeal allowed the appeal and made a rule 31.22(2) order in respect of page 2. 348. The last judgment cited by Arnold LJ was one of my own, in which I had to strike a balance between the protection of private and confidential financial information (much belonging to third parties) and the need for the public to understand important parts of my reasoning in my main FRAND Judgment in that litigation. 349. The final part of Arnold LJ’s reasons for allowing Manitou’s appeal was in [111]: ‘111. Thus the present case is quite different from Lilly Icos . In that case there was no departure from the open justice principle because there was no need for such a departure. As Buxton LJ explained, it was appropriate for the court to make an order under rule 31.22(2) in respect of page 2 of the schedule precisely because it was not relevant to the issues argued at trial and therefore the making of the order did not detract from open justice. It is doubtful whether the information in page 2 was properly characterised as a trade secret, but that was not the decisive consideration. Even lower grade confidential information is entitled to protection where its publication is not necessary for open justice. ’ 350. The Cs’ solicitors placed particular reliance on the first part of that final sentence. Analysis 351. With the applicable principles in mind, I have come to the conclusion that I should not redact any figures from [124]-[128] above. My reasons in summary are as follows: i) First, I acknowledge that the Cs’ sales figures as set out in E/1/6 & 7 are confidential to the Cs. ii) Second, however, I have not revealed those actual sales figures in [124]-[128]. I have set out aggregated figures over 2010-2018, albeit those aggregated figures also constitute the Cs’ confidential information. iii) Third, these sales figures, whether actual or aggregated are not technical trade secrets. They are, in the words of Arnold LJ, lower grade confidential information, the more so because they are now somewhat out of date. iv) Fourth, in view of my reasoning in and around [124]-[128], I have concluded that publication of the figures in those paragraphs is necessary for open justice and specifically to enable the public to understand my reasoning. 352. The paragraphs set out above in this Postscript were sent to the Cs so they could consider their position in advance of hand down (which was postponed until 2pm). In the event, the Cs decided not to challenge my conclusion in this Postscript. BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII