Cabo Concepts Limited & Anor [2025] EWHC 1451 (Ch)

[2025] EWHC 1451 (Ch)

Case No HP-2020-000016

IN THE HIGH COURT OF JUSTICE

BUSINESS AND PROPERTY COURTS OF ENGLAND AND WALES

INTELLECTUAL PROPERTY LIST (ChD)

PATENTS COURT

Venue Rolls Building, Fetter Lane

Date 16 June 2025

London, EC4A 1NL

Before

MRS JUSTICE BACON

Between

CABO CONCEPTS LIMITEDClaimant

THE LICENCE WORLD LIMITEDClaimant

MGA ENTERTAINMENT (UK) LIMITEDDefendant

MGA ENTERTAINMENT, INCDefendant

Ronit Kreisberger KC, Mark Chacksfield KC, Stefan Kuppen, Alfred Artley& Greg Adey (instructed by Spector Constant & Williams Limited) for Claimants

Victoria Wakefield KC, Jennifer MacLeod & Richard Howell (instructed by Fieldfisher LLP) for Defendants

Hearing Hearing dates: 8 October–15 November 2024, 14–21 January 2025

Approved Judgment

This judgment was handed down remotely at 10 am on 16 June 2025 by circulation to the parties or their representatives by email and by release to the National Archives.

Introduction 1

The evidence of fact 9

Cabo’s witnesses of fact 9

MGA’s witnesses of fact 16

Mr Larian’s breaches of purdah 27

The expert evidence 38

The toy industry experts 38

The economic and valuation experts: preliminary comments 42

Assessment of the economic and valuation evidence 55

The Decision Tree Model (DTM) 58

Issues 62

Factual background 66

The UK toy industry 67

MGA and LOL Surprise 74

The founding of Cabo and development of Worldeez 79

The initial marketing of Worldeez 89

Discussions with the launch retailers 93

MGA’s intervention 104

Worldeez repackaging and relaunch 154

Demise of Cabo 177

Procedural background 182

Abuse of dominance claim 188

Overview of the legal framework 188

The relevant market definition 192

Whether MGA was dominant on the relevant market 255

Whether MGA’s conduct amounted to an abuse 306

Conclusion on abuse of dominance 357

Unlawful agreements claim 358

Overview of the legal framework 358

Agreements with the toy traders 367

Anticompetitive object or effect 375

Exemption under the VBER 387

Exemption under s. 9 / Article 101(3) 411

Conclusion on the unlawful agreements claim 415

Patent threats claim 416

Overview of the legal framework 416

Threats of patent infringement proceedings 421

“Person aggrieved”444

Conclusion on the patent threats claim 446

Causation and quantum 447

Preliminary comments 447

Legal principles 451

Causative effect of MGA’s conduct 478

Actionable damage and causation: Cabo’s heads of loss 493

Whether Cabo would have traded profitably in the counterfactual case 501

The parties’ quantum models 594

Declaratory relief 680

Conclusions 682

Introduction

[1]This claim concerns the sale of rival brands of small plastic toys. In 2017 the claimants (Cabo), a toy start-up company, launched a new toy brand called Worldeez, which was a line of surprise collectible figurines with a world travel theme, packaged in a plastic capsule designed to represent a globe. Cabo contends that the defendants (MGA) stifled the launch of Worldeez by claiming that the globe was a “knock off” of MGA’s very popular “LOL Surprise!” toy, and by threatening toy retailers that their supplies of LOL Surprise would be withheld if they stocked the Worldeez globe. The Worldeez toy went on to fail, and was discontinued in 2018.

[2]Cabo’s case is that MGA’s conduct amounted to(i) an abuse of a dominant position contrary to the prohibition in Chapter II of Part I of the Competition Act 1998 (the 1998 Act) and/or Article 102 TFEU;

(ii) unlawful agreements contrary to the prohibition in Chapter I of Part I of the Competition Act 1998 and/or Article 101 TFEU; and

(iii) unjustified threats of patent infringement proceedings within the meaning of s. 70 of the Patents Act 1977 (the 1977 Act). Cabo claims that but for MGA’s unlawful conduct Worldeez would have been a successful product and Cabo would have gone on to become a successful toy business. Cabo claims loss and damage in the form of lost profits. The initial claim was for in excess of £170m. By the end of the trial, the claim had reduced to £53–90m, depending on the assumptions used in Cabo’s quantum models.

[3]MGA denies any infringement of competition law. It denies dominance during the relevant period; contends that even if dominant, its conduct was not abusive; and argues that any agreements between MGA and retailers not to stock the Worldeez globe were not restrictive of competition and were in any event exempted from the Chapter I prohibition and Article 101 TFEU by virtue of the Vertical Agreements Block Exemption Regulation 2010/330/EU [2010] OJ L102/1 (the VBER). MGA further contends that its threats to retailers were not (properly characterised) threats of patent infringement proceedings, and that even if there were such threats they did not cause retailers not to stock Worldeez. More generally, MGA argues that irrespective of its conduct Cabo’s toy business would most likely have failed, on the grounds that the founders of the business were inexperienced and naïve, with inadequate operational control, insufficient capital support and an insufficiently appealing and innovative product to achieve commercial success in the highly competitive toy industry (both UK and worldwide). On that basis MGA says that Cabo’s damages claim should fail.

[4]The case was originally listed for trial in June 2022. A few weeks before the trial was due to commence, however, it was adjourned because of the discovery of significant flaws in MGA’s disclosure process. Further disclosure was then provided in 2023 and the trial was relisted for 2024. The main part of the trial took place over four weeks in October and November 2024. The trial was then adjourned by the court shortly before the written closing submissions were scheduled to be delivered, with the closing submissions ultimately filed in December 2024 and oral closing submissions in January 2025.

[5]Along with their written closing submissions, the parties filed various additional excel spreadsheets with further or updated figures and calculations supporting their respective submissions on the quantum assessment. The parties also filed an updated version of the agreed Decision Tree Model (DTM), an excel spreadsheet combining the parties’ central inputs for the quantum assessment, in a form which could be manipulated by the court. Numerous further materials relating to (in particular) the economic evidence, including the quantum models, were filed during the course of the oral closing submissions, including further excel spreadsheets on both sides, and a further iteration of the DTM.

[6]By the last day of the oral closing submissions, it became apparent that the parties’ positions on the quantum assessment were still evolving. The parties’ final positions on a number of specific points were subsequently filed on 17 February 2025, and a final version of the DTM was provided on the same date. In response to questions from the court, further figures and explanations were provided by the parties on 12 March and 9 and 14 April 2025. Those addressed specific points on the market shares, the DTM, and the parties’ profitability calculations.

[7]During the trial, submissions on behalf of Cabo were made by Ms Kreisberger KC, Mr Chacksfield KC and Mr Kuppen, with cross examination and re-examination divided between them and Mr Artley, and Mr Adey assisting as junior counsel. MGA’s submissions, cross-examination and re-examination were divided between all three of its counsel team, namely Ms Wakefield KC, Ms MacLeod and Mr Howell.

[8]I am very grateful for the assistance of all of the counsel instructed in these proceedings, noting not least the efforts on both sides to accommodate the adjourned closing submissions in the case. It will be apparent from the comments above and the remainder of this judgment that this was a case of considerable complexity, on both issues of substance and procedure. It is, in those circumstances, to the credit of the counsel teams that the trial was conducted with the utmost professionalism and courtesy on both sides. The evidence of fact Cabo’s witnesses of fact

The evidence of fact

[9]Cabo’s main witnesses were Mr and Mrs Michaelson and Mr and Mrs Cohen. These two couples were the founders of Cabo alongside two further individuals, Marc Sivner and Alexander Lazarus, who have since exited the company. The Michaelsons and Cohens are now the sole shareholders of Cabo; Mr Lazarus is, as discussed below, a witness for MGA; and Mr Sivner has not participated in these proceedings in any way. I will refer to the Michaelsons and Cohens in this judgment as the Cabo founders.

[10]Marc Michaelson had been a director of a sports equipment business prior to founding Cabo. Within Cabo he contributed to the product design of Worldeez and led the marketing of the product. Mr Michaelson provided five witness statements, giving extensive evidence about the background to the creation of Cabo, the design and launch of the Worldeez brand, MGA’s intervention in May 2017 and the retailer response, Cabo’s efforts to sell Worldeez despite the dispute with MGA, and the eventual demise of both the product and Cabo’s business. He was cross-examined on that evidence over one and a half days (with the evidence of Mr Hunter, one of MGA’s witnesses, interposed briefly on the second morning). His evidence was unfortunately very defensive, and evasive on numerous points. It was apparent that his answers were designed to advance Cabo’s case, even when those answers were inconsistent with the contemporaneous documentation or otherwise implausible. It was also apparent that he did not have a firm grasp of the details of Cabo’s business, including on points covered in his witness statements. I do not regard him as a reliable witness.

[11]Hayley Michaelson is married to Mr Michaelson and is the sister of Lauren Cohen. She had a degree in Childhood Studies and experience working with children, and she and Mrs Cohen had the initial idea for the Worldeez toy, based on observing their own children’s interest in particular toys. Within Cabo, Mrs Michaelson was principally responsible for the Worldeez product design, alongside Cabo’s designer Helder Olivier. She provided two witness statements addressing mainly that issue, and was cross-examined for around half a day. She was a frank and straightforward witness, who sought to answer questions directly. However, while her passionate belief in the prospects of success of the Worldeez brand was evidently sincere, she came across as rather naïve regarding the commercial challenges faced by a startup toy company.

[12]Johnny Cohen had previously operated multiple successive businesses selling heat packs, several of which had been liquidated with significant outstanding unsecured liabilities. His role in Cabo was mainlyto oversee the manufacturing and import of Worldeez products. He was also (albeit to a more limited extent) involved with Cabo’s finances, product design and marketing. Mr Cohen provided five witness statements, mainly addressing the design of the Worldeez toy, the manufacturing process, marketing, costs and sales projections, and the reaction to MGA’s threats, as well as his initial discussions on international sales and licensing. His cross-examination took most of a day. He was another very defensive and sometimes combative witness, whose optimism regarding the success which he considered Worldeez would have enjoyed obscured a realistic assessment of the commercial position. He was very willing to blame others for problems, including internal logistical and administrative issues, and reluctant to take responsibility for matters that had been within his control. As with Mr Michaelson, I do not regard him as a reliable witness.

[13]Lauren Cohen is married to Mr Cohen. She practised as a solicitor before founding Cabo, and along with Mrs Michaelson her main responsibility was the design of the Worldeez products. She provided a very short witness statement addressing the alleged similarities between the Worldeez globe and LOL Surprise, and was cross-examined only very briefly. Her evidence was undoubtedly sincere, but (as with the evidence of the other Cabo founders) was coloured by her optimism in the likely success of the product.

[14]Nick Mowbray is the co-founder of Zuru Toys, which has marketed a number of commercially successful toys including “5 Surprise Mini Brands”. He provided one short witness statement and was cross-examined remotely from New Zealand for around two hours. His evidence addressed, in particular, his knowledge of and interactions with MGA and Mr Larian, his experience of bringing a successful toy to the market, and his views on the likely commercial success of Worldeez (although he was called as a witness of fact, rather than a toy industry expert). I have no reason to doubt that his views were honestly held. Equally, however, it was clear that he and Mr Larian were longstanding rivals, and that he was keen to support Cabo as having been a small new entrant to the toy business and potential challenger to MGA. I do not, therefore, consider his evidence to have been entirely objective.

[15]Richard Spector is the solicitor at Spector Constant & Williams with primary conduct of the case for Cabo. He provided a very short witness statement describing a brief exchange of messages on the social media platform LinkedIn, between himself and Kevin Macnab, the former president of TRU International, in which Mr Macnab said that he had no recollection of speaking to Mr Larian about Worldeez in 2017, and did not want to become involved in the dispute between Cabo and MGA. Mr Spector was not cross-examined at the trial.

MGA’s witnesses of fact

[16]Isaac Larian is the founder and CEO of MGA, and the main protagonist in the events of 2017 which form the subject-matter of this claim. He was therefore the main witness for MGA. He provided, prior to the trial, four witness statements, and was cross-examined over two days. His evidence was defensive and argumentative throughout, attempting to give speeches as to the evidence in the case rather than answering the questions put to him (to the extent that he had to be repeatedly reminded to address the question rather than trying to put his case). His mantra that a retailer stocking a “knock off” did not need the original product was repeated at least seven times during his cross-examination, frequently as a means to avoid giving a straight answer to the question put to him. He repeatedly refused to give proper (or any) answers to questions, instead making implausible denials or simply responding “no comment”. By the second day of his cross-examination he was increasingly irascible, giving irritated and facetious answers.

[17]On the basis of the content and manner of Mr Larian’s evidence during this part of the trial, I regret to say that I considered Mr Larian to be an unreliable witness who was less than candid in his evidence to the court. That impression was reinforced by Mr Larian’s multiple breaches of purdah between the first and second days of his evidence, and the subsequent evidence that he gave when recalled to be cross-examined on this point, which I address below. My conclusion is that I can give no weight to Mr Larian’s evidence where it is not corroborated by the contemporaneous documents or the evidence of other more reliable witnesses.

[18]Andrew Laughton was, during the events at issue in these proceedings, the managing director and senior vice president of MGA. He was a central figure in the implementation of MGA’s conduct in relation to Worldeez, acting as Mr Larian’s mouthpiece and enforcer. Numerous of the communications relied on by Cabo as constituting infringements of the competition rules and unjustified threats of patent infringement proceedings emanated from Mr Laughton, whether in the form of emails or telephone calls to retailers. Mr Laughton was therefore a central witness for MGA. He provided three witness statements and was cross-examined for a little less than a day. He sought to give the impression of being unwillingly caught in a conflict between Mr Larian and the retailers, with a rather grandiose suggestion that he played the part of a “Kofi Annan” peacemaker. In reality, it was apparent from the contemporaneous documents that he was a willing and proactive participant in Mr Larian’s conduct. His answers to questions about the events of 2017 wereoftenevasive, seeking to play down the impact of his communications with the retailers. I do not regard Mr Laughton as an entirely reliable witness, and have treated his evidence with caution.

[19]Alexander Lazarus is one of the two founders and directors of Singleton Trading Limited (Singleton), a successful toy company which sells “close-out” stock to toy retailers, and the related company Sinco Toys Limited (Sinco), a licensing and distribution business. His business partner in both companies is Marc Sivner, who is Mr Michaelson’s cousin. Mr Lazarus and Mr Sivner were initially shareholders in Cabo, alongside the Cabo founders, and Singleton provided both financial backing and logistical support for the launch of Worldeez. Mr Lazarus provided two witness statements, addressing Singleton’s involvement with Cabo, the development of the Worldeez product, the launch of Worldeez and MGA’s reaction, and the subsequent demise of Worldeez. He was cross-examinedfor almost a day, and his evidence was, in general, objective and candid. Cabo contended that Mr Lazarus’ credibility was undermined by the fact that Sinco entered into a licensing agreement with MGA in 2018, while Mr Sivner and Mr Lazarus were still directors of Cabo. It is fair to say that when Mr Lazarus was questioned about that agreement, his answers were somewhat defensive. Ultimately, however, the Sinco licensing agreement has no bearing on the facts relevant to the present proceedings; and in relation to the events material to these proceedings Mr Lazarus’ evidence was both measured and corroborated by the contemporaneous documents.

[20]Stuart Grant is a shareholder and director of The Entertainer, a family-owned toy business founded by his parents Catherine and Gary Grant. The Entertainer was one of the toy retailers threatened by MGA, and during that period Stuart Grant was the company’s buying director. Stuart Grant provided three witness statements describing The Entertainer’s initial order of Worldeez, MGA’s intervention, and The Entertainer’s subsequent decision not to stock the globe. He was cross-examined for around half a day, and was a straightforward witness, giving measured and candid responses. I consider him to be a reliable witness on the factual issues in dispute, albeit that (for the reasons discussed below) I do not entirely accept his characterisation of The Entertainer’s decision not to stock the Worldeez globe.

[21]Hayley Whiteis (and was, during the relevant period of time) the toy buyer at the retailer B&M Retail Limited (B&M). She provided a short witness statement describing B&M’s sales of Worldeez during 2017 and its relationship with MGA in 2018. She was cross-examined very briefly, and was a straightforward and candid witness.

[22]Darran Garnham is the founder and CEO of Toikido Limited, an entertainment company specialising in digital design and toys. He gave some advice to the Cabo founders and Singleton on the design of the Worldeez product in 2016, and provided a witness statement responding to the Cabo founders’ evidence about their discussions with him during that period. He was cross-examined very briefly, and his evidence was straightforward and entirely credible.

[23]John Hunter is a director and co-owner of AB Gee of Ripley Limited (AB Gee), a toy wholesaler which supplies smaller independent toy retailers. He provided a short witness statement addressing AB Gee’s position in the toy market, its sales of Worldeez and its interaction with MGA regarding the sale of Worldeez. MGA then issued a witness summons to compel Mr Hunter’s attendance at the trial. His (brief) cross-examination was interposed at the start of the second day of Mr Michaelson’s cross-examination, and his evidence was entirely straightforward.

[24]Kathy Brandon is vice president of research and consumer insights at MGA. She provided a short witness statement to explain two viability studies conducted by MGA in 2017 and 2018, addressing the girls’ collectibles market (with the focus apparently being on the US market) and MGA’s products in that market. She was not cross-examined, but it was agreed that her evidence was admissible as hearsay evidence, and should be given the same weight as if she had been called and given evidence in court.

[25]Nicholas Pimlott is one of the partners at Fieldfisher with primary conduct of the case for MGA. He provided a very short witness statement recording a telephone call with Sally Hunter, who was the sales director at Sinco between July 2017 and March 2018, and subsequently worked at MGA. She had provided information to Mr Pimlott regarding the sales of Worldeez during the time of her employment at Sinco, but was not thereafter willing to give evidence in these proceedings. Mr Pimlott was not cross-examined on his witness statement and nothing ultimately turned on his evidence.

[26]Patrick Smyth is the co-founder and a director of Smyths Toys (Smyths). He provided two witness statements referring to his interactions with Cabo regarding the purchase of Worldeez, and denying that MGA had pressured him to drop the product. Shortly before the trial was due to start, Mr Smyth said that he was no longer able to attend court. It was therefore agreed that his evidence was admissible as hearsay, but I have treated it with appropriate caution where his account was inconsistent with the contemporaneous documentation and other witness evidence.

Mr Larian’s breaches of purdah

[27]During the first day of Mr Larian’s cross-examination on Thursday 17 October 2024, he was twice warned by the court about the purdah rules for witnesses giving evidence, in terms which reminded him that he could not talk to his legal team or anyone else about the case until the end of his cross-examination the following day. He confirmed that he was aware of the rules.

[28]Before the start of the hearing the next day, however (i.e. Friday 18 October), the court was informed that after court on the previous evening (at 17:59) Mr Larian had sent an email concerning the case to his counsel Ms Wakefield, two of his solicitors, Stephen May and Nicholas Pimlott, and the head of litigation at MGA, Richard Grad. That email carried the subject line “Nick Mowbray” and set out a series of questions which Mr Larian proposed should be put to Mr Mowbray in his cross-examination which was scheduled for the following Monday. That email was, quite properly, immediately deleted by each of the recipients, and a reply was sent to Mr Larian from Mr Pimlott (at 18:02) saying “We cannot read this email. As explained, please do not email, text or otherwise contact us while you are in purdah before you have finished giving your evidence tomorrow.”

[29]In discussion with Ms Wakefield and Ms Kreisberger at the start of the hearing on the Friday morning, it was agreed that Mr Larian would hand over his mobile phone to his solicitors until the end of his cross-examination that day, and that at the end of the day it would be examined by Ms Wakefield and his solicitors to ascertain whether Mr Larian had communicated (or had sought to communicate) with anyone else during the period while he was being cross-examined. The court could then consider what further steps to take in relation to this matter. During the course of that discussion, Mr Larian said that his email to his solicitors had been a mistake. He did not disclose that he had indeed sent further messages to other people.

[30]Following the review of Mr Larian’s mobile phone by his counsel and solicitors, MGA filed over the weekend a further (fifth) witness statement from Mr Larian, and a witness statement from Mr May, with exhibits to both. That evidence revealed that Mr Larian had sent a number of messages concerning the case during the evening after the first day of his cross-examination on 17 October, in addition to the email to his legal team. These included the following WhatsApp message sent at 17:44 to Manny Stull, the chairman of Moose Toys, asking for a letter relating to a dispute between Moose Toys and Zuru Toys:

“Hi. I am in court and next week Nick is testifying. Please send me a copy of your cease and [desist] letter. I want a loyal to show that to Nick and get his reaction on the record. Then I will send you that testimony for you can use yours.”

[31]In addition, between 08:46 and 09:54 the next morning, Mr Larian had sent a series of WhatsApp messages to Sara Taylor, the managing director of MGA UK, asking for information as to the sales of the Little Tikes brand by The Entertainer and Smyths, for the most recent year and also for 2017 and 2018. He also specifically asked whether The Entertainer was buying a particular Little Tikes toy called Story Dream Machine. Ms Taylor answered Mr Larian’s questions, although by the time he recommenced his cross-examination he had not read her response to the last of those questions. Mr Larian’s counsel and solicitors had not been aware of these further messages (or any of the other messages sent by Mr Larian) during the discussion at the start of the hearing on the Friday morning.

[32]It was also apparent that Mr Larian had used the information he had received in his evidence on the second day of his cross-examination. After the initial discussion about his breach of purdah on the Friday morning, that cross-examination had recommenced, and as soon as it did so Mr Larian said that he wanted to add one comment to his testimony from the previous day. He then went on to make comments about The Entertainer’s sales of the Little Tikes brand in 2017/18 and 2024, with sales figures given purportedly “to the best of my recollection”, asserting that these proved that The Entertainer and other retailers were more powerful than the toy vendors. Later on the same morning, Mr Larian twice repeated his comments about The Entertainer’s failure to purchase the Little Tikes brand. What Mr Larian did not reveal, at the time, was that his comments about Little Tikes were based on the information he had received from Ms Taylor before court that morning.

[33]Mr Larian’s subsequent witness statement explaining his breaches of purdah sought to explain his email to his legal team as having been sent by mistake: he said that he had intended to save it in draft, to send later, but then sent it by accident. He sought to explain his messages to others, including Mr Stull and Ms Taylor, on the basis that he had not properly understood the purdah rules, and did not consider those messages to be in breach of the instructions he had been given.

[34]Cabo considered Mr Larian’s explanations to be unsatisfactory, and applied to recall him for further cross-examination on this issue. As he had by then left the country, he was cross-examined remotely from the offices of a law firm in Los Angeles on the afternoon of 4 November 2024, the last day of evidence before the break for closing submissions to be written. His answers were defensive and maintained the position set out in his witness statement.

[35]Mr Larian’s evidence on this issue was not, in my judgment, entirely candid. Mr Larian is an intelligent businessman, who accepted that he had been warned about the purdah rules by his solicitors prior to his cross-examination, as well as by the court during his cross-examination. Mr Larian’s communications with Mr Stull and Ms Taylor were quite obviously discussions about the case, and I do not accept his claim that he did not understand that these were prohibited by the purdah warnings he had been given. In Mr Larian’s message to Mr Stull, he explicitly said that he was in court and wanted a document to use in Mr Mowbray’s cross-examination. In the case of Ms Taylor, he asked for information which he went on to relay (repeatedly) in his evidence. His claim to have emailed his legal team accidentally was likewise unconvincing. The more likely explanation is that Mr Larian understood the purdah rules but did not take them seriously, and did not consider the implications of those rules for his communications with his lawyers and others during his cross-examination. It probably did not even occur to him that these communications might be revealed to the court.

[36]As Patten LJ noted in Jarvis v Searle

[2019] EWCA Civ 1

, §§23–24 and 28, witnesses are commonly given warnings by the trial judge not to discuss their evidence until after it has been completed, the purpose being to ensure that the evidence given by the witness is their own best recollection, untainted by any influence or assistance from a third party. Where a witness does, notwithstanding such a warning, attempt to communicate with a third party, but does not get a response, there is no damage to the integrity of the trial process. If the witness does, however, discuss some matter relevant to their evidence with a third party, that may (if appropriate) lead the court to discount or give no weight to that evidence.

[37]In the present case, MGA invited the court to deal with the matter by striking from the record Mr Larian’s repeated comments about Little Tikes, on the second day of his cross-examination; but said that Mr Larian’s breaches of the purdah rules did not detract from his general credibility. I agree that the comments about Little Tikes must be disregarded. They were made on the basis of the information obtained by Mr Larian that morning before the court hearing recommenced. I do not, however, agree that this incident is irrelevant to the assessment of Mr Larian’s credibility. On the contrary, Mr Larian’s conduct and his unsatisfactory explanations for that conduct reinforce the conclusion that he was an unreliable witness. The expert evidence The toy industry experts

The expert evidence

[38]The parties’ respective toy industry experts were Wendy Munt (Cabo) and John Harper (MGA). They produced initial reports and reply reports in 2022, followed by a joint report. After additional disclosure was provided by MGA (as explained below), both experts provided further individual expert reports and a further joint report during the course of 2024, prior to the trial. There were eventually a total of 15 toy industry reports: six reports from Ms Munt, seven from Mr Harper, and two joint reports.

[39]The joint expert reports identified a large measure of agreement regarding the characteristics and structure of the toy industry, the costs and operational requirements of bringing a toy product to the market, and the likely shape of the profit and loss (P&L) account for a toy company. The experts’ main areas of disagreement were their assessment of the substitutability of LOL Surprise with other toys (such as Barbie), and their assessment of the likely success of Worldeez. The experts were cross-examined for a day each.

[40]Wendy Munt, Cabo’s industry expert, has over 30 years’ experience in the toy industry, as a toy buyer for Argos and subsequently working as a consultant to toy suppliers, retailers and licensors. She fairly accepted that she was not able to give evidence as to the financial metrics relevant to the profitability of a toy supplier; rather, her previous work with toy suppliers had primarily focused on product development. Within the scope of her experience, her evidence as to the toy market dynamics and operational requirements was objective and largely uncontroversial. However, on some key issues of disagreement between the experts, while I have no doubt that Ms Munt was genuinely seeking to assist the court, her evidence came across as rather partisan, leading her to adhere to positions that were not very convincing or well-explained.

[41]John Harper, MGA’s industry expert, has nearly 40 years’ experience in the toy industry, with executiveand management positions at Mattel and Hasbro, and subsequent advisory roles at several toy companies. By contrast with Ms Munt, therefore, his experience was on the supplier rather than the buyer side of the industry. He was an impressive and knowledgeable witness, fairly accepting the limitations of his experience (for example the fact that he had no direct experience of surprise collectible toys), but demonstrating a very detailed knowledge of the areas within his expertise. His evidence was mostly balanced and cogently explained, although he strayed into arguing MGA’s case on the market definition issue.

The economic and valuation experts: preliminary comments

[42]The economic evidence covered two areas: first, the question of market definition and dominance, and secondly the issue of quantification of damages. The valuation evidence solely addressed the quantification of damages. The evidence on these issues evolved considerably during the course of the proceedings.

[43]Initial expert reports and reply reports were produced in 2022 by Liam Colley for Cabo and David Parker for MGA, covering both market definition/dominance and quantification of damages. There was then a dispute as to whether Cabo could advance, in addition to a claim for lost profits in relation to Worldeez, a further (then unpleaded) claim for the lost opportunity to develop a valuable business. At a hearing before Joanna Smith J on 24 March 2022, with an ex tempore judgment given at the hearing

[2022] EWHC 702 (Pat)

, Cabo was given permission to amend its case to include a claim for loss of the value of the business that it would have been able to build. MGA was given permission to call an additional expert witness (Mr Davies) to address the valuation claim. Mr Colley’s second report had by then already addressed the valuation issue, and Cabo was therefore not permitted to call an additional valuation expert. Mr Colley and Mr Parker then filed a joint expert report on the issues of market definition, dominance and quantum of damages other than the valuation claim. Mr Colley and Mr Davies filed a joint report on the valuation issues.

[44]After MGA’s additional disclosure in 2023, all three experts provided (in the course of 2024) updated individual reports and thereafter further joint reports, again with separate joint reports from Mr Colley/Mr Parker and Mr Colley/Mr Davies, respectively. During the course of preparing their revised and updated reports, all three experts collaborated on the production and population of the DTM (which is described further below). That led Mr Parker to revise his assumptions, as set out in a further report. By the end of August 2024 there were a total of 19 economic and valuation reports that had been filed: six from Mr Colley, seven from Mr Parker, two from Mr Davies and four joint expert statements.

[45]As noted above, matters did not stop there. During September 2024 (i.e. prior to the start of the trial), and then during the course of the trial in October, up to the point at which the economic and valuation experts were cross-examined, there was a flurry of further reports and letters filled by all three experts, including documents filed by all three experts on 27 October 2024, only two days before Mr Colley was cross-examined. That gave a total of 33 economic and valuation expert reports and letters by the close of evidence in the trial.

[46]Thereafter, since the experts had given their evidence, no further reports or letters were filed. Instead, the parties continued to update the evidence by way of legal submissions and additional calculations and models (including updates to the DTM and other excel spreadsheets) filed alongside their written closing submissions and continuing during the course of the oral closing submissions. As set out above, the parties’ final submissions on various of the quantum issues were not filed until 17 February 2025, almost a month after the end of the trial. Those submissions made further detailed comments on the economic evidence.

[47]It is an understatement to describe this process as unsatisfactory. In the first place, the volume of expert evidence filed by the economic and valuation experts, and the multiple amendments, revisions and updates to that evidence during the course of the trial, was such that it was impossible properly to digest the new material as it came in. That made following the evolution of the evidence very challenging for both the court and the parties. It was also difficult, in these circumstances, to distinguish material issues from ancillary points of detail.

[48]Secondly, significant shifts in position by an expert during the course of a trial may well lead the court to scrutinise the objectivity and independence of that expert’s initial opinion. That does not of course mean that an expert should stubbornly maintain their initial position simply for the sake of consistency. As the Court of Appeal has emphasised, the court will expect and encourage experts to adjust their opinions during the course of the proceedings, including during the course of the trial, in the light of emerging evidence (see for example the comments of the Court of Appeal in Royal Mail Group v DAF Trucks

[2024] EWCA Civ 181

, §177). Many an expert will, quite properly, concede ground when their evidence is exposed to the spotlight of cross-examination and the hard stare of the judge. Where, however, an expert completely abandons a particular part of their evidence, not as a result of any new evidence available at the trial, but rather because it is plain that the original position was untenable from the outset, that may suggest that the expert’s initial position was unduly influenced by a desire to advance their client’s case, rather than representing evidence which was the independent product of the expert, uninfluenced by the pressures of litigation, as required by CPR PD 35 §2.1.

[49]Thirdly, and related to the second point, expert evidence at a trial is not and cannot be seen as a negotiation process, where the experts start from extremely polarised and partisan positions, only to edge incrementally towards the centre ground as the trial progresses. That would make the trial unworkable, for the parties as well as the court. The proper course is for each opposing expert to start from a position that is objective and defensible. Any differences in opinion between the experts should be discussed fully at the stage of a joint meeting of experts (if there is one). The experts should revise their opinions as appropriate following that meeting, with the joint statement reflecting their revised opinions. Any residual differences can then properly be tested through the experts’ oral evidence at the trial.

[50]That process requires a willingness by the experts to engage with the evidence of the other side in a manner that reflects objective consideration as to the strengths and (importantly) weaknesses of their position. Where an expert fails to do so, and maintains instead an entrenched and polarised position right up to trial, that may again indicate a lack of objectivity in their approach, thereby undermining the credibility and reliability of their evidence.

[51]In the present case, prior to the trial, the positions of Mr Colley and Mr Parker were about as far apart as could be imagined on both the questions of market definition/dominance and the quantum assessment. The result of the evolution of their positions, during the course of the trial, was that significant parts of the evidence on both sides were essentially abandoned by the end of the trial. More than that, however, on the quantum analysis both parties not only updated their figures during the trial (for example on the costs stack), but materially changed or developed their positions on significant points, with the result that on several points the position finally advanced in closing submissions was one that had not been set out in Mr Colley and Mr Parker’s expert reports, and therefore had not been tested in cross-examination.

[52]The court is obviously required to do the best it can with the material before it, and counsel on both sides went to considerable lengths to provide submissions that were as comprehensive and helpful as possible, including (as noted above) providing further written submissions after the hearing on points which had not been fully addressed in the closing submissions. But in a case where the parties and their experts have had over four years to prepare for the trial, with multiple rounds of expert evidence over the course of a three-year period, it is very unsatisfactory for the court to be asked to consider and make findings on material which could and should have been addressed by the experts, but which did not emerge until after the hearing of the expert evidence in the trial. I consider that considerably more progress could and should have been made in the succession of expert reports and joint reports prior to the start of the trial, so that the parties’ final positions could be fully considered by the experts and properly tested in their oral evidence, rather than cobbled together in haste in the very last days of (and indeed after) a lengthy trial.

[53]Finally, even by the end of the trialthe distance between the experts’ positions, and the evolution of the arguments, was such that there were inevitably some lacunae in the evidence. There are several different approaches which the court can adopt in that situation, depending on the nature of and reasons for the evidential gap. One approach might be to rely on the burden of proof, and to say that if there is no evidence on a particular point then the party which has the burden of establishing that point should be taken not to have proven its case. In certain types of cases, as discussed further below, evidential presumptions arising from the conduct of one or other party might also come into play. In many cases, the court will simply do the best it can on the evidence, albeit incomplete. Where appropriate the court can, however, ask for further information to be provided, particularly where that simply requires the experts to carry out further calculations on the basis of the existing underlying data.

[54]All of those options were canvassed by the parties in the course of the trial, in relation to different issues. In most cases I have felt able to reach a conclusion on the evidence before the court, without needing to rely on the burden of proof or any evidential presumptions. On two issues, however, I asked for further calculations to be provided by the parties following the trial:(i) the market share figures implied by an extended version of Mr Colley’s market definition (§271 below); and

(ii) a further version of one of Mr Parker’s quantum calculations, using a combination of Mr Parker’s price assumptions and Cabo’s cost stack (§661 below).

Assessment of the economic and valuation evidence

[55]Liam Colley is a partner at the economic consultancy firm Cornerstone Research, and has over 25 years’ experience as an expert in competition disputes. He was cross-examined over two days, showing impressive understanding and depth of knowledge of the economic issues in the case. For the reasons set out below, I have largely accepted his market definition and dominance analysis. On the quantification of Cabo’s counterfactual profits, however, Mr Colley provided models that were unrealistically slanted towards a picture of huge sales in the counterfactual scenario, without addressing the question of Cabo’s profitability if Worldeez had enjoyed more modest levels of success. That was neither objective nor balanced, and left the court without any useful basis for the assessment of quantum if the very high levels of success assumed in his models were rejected.

[56]David Parker is a director of the economic consultancy firm BRG, with over 20 years’ experience as an expert in competition disputes. He was cross-examined for almost two days. Again, unfortunately, while Mr Parker was clearly very knowledgeable in his area of expertise, I consider that some parts of his evidence crossed into advocacy for MGA and lacked the objectivity which I would expect from an expert witness. That was the case, in particular, for his evidence on market definition and the dominance assessment, in relation to which he had focused on a statistical analysis, with little or no regard to the qualitative evidence and the evidence of MGA’s market conduct. Mr Parker’s quantum analysis was ostensibly more balanced, by considering a range of comparators, but Mr Parker noticeably placed greatest reliance on a model of “moderate success” which was unreasonably unfavourable to Cabo. In addition, significant costs assumptions in Mr Parker’s model were wholly unrealistic and indicated a lack of objectivity. These problems led to substantial amendments being made to Mr Parker’s model by MGA in its closing submissions.

[57]Gary Davies is a senior managing director of the management consultancy firm Ankura, with over 25 years’ experience of advising and providing expert evidence on valuation issues. He was cross-examined for around an hour, and it was common ground that he was a fundamentally honest and helpful witness. There was ultimately very little dispute between him and Mr Colley on the valuation issues, and in the light of the conclusions I have reached the relatively minor residual areas of difference between them do not arise for determination.

The Decision Tree Model (DTM)

[58]The DTM was, over the course of the trial, a central tool for the parties and their experts to use, to set out their respective primary and alternative positions on the quantum modelling, and for the court to use, to understand the implications of the different approaches addressed in the expert evidence on each side.

[59]The purpose of the DTM was to allow the court to calculate Cabo’s losses based on the variety of different assumptions that were disputed as between the economics and valuation experts. It operated through two basic functions. First, it enabled the court to select different primary input options from the alternatives presented by the experts, to model (in terms of cashflows) the likely success of Worldeez in the counterfactual scenario in which the impugned conduct of MGA had not occurred, as well as the parties’ alternative estimates of Cabo’s costs. Secondly, the DTM provided probabilistic functionality, enabling the court to adjust the key inputs for domestic and international sales, licensing, and the development of further products, by reference to an assessment of the probability of Worldeez achieving the success levels modelled by the experts.

[60]In broad outline, the DTM worked as follows: i) The DTM started with two input points requiring an assessment by the court. The first input point was an estimate of the cashflows which Cabo would have generated in the UK market but for MGA’s conduct, derived by reference to the various quantum models provided by Mr Colley and Mr Parker. The models available for selection under this input point changed during the course of the trial, with the parties’ cases evolving as described further below. A probabilistic adjustment enabled the court to calibrate the cashflow model selected, by selecting a probability of that outcome of less than 100%. (For this and the other probability adjustors described below, there was no mechanism for adding in a model regarded as reflecting the remaining probability, if a probability of less than 100% was selected. The adjustments were therefore not true probabilistic adjustments in a mathematical sense, but simply a means of reducing the cashflow figures which had been selected on the basis of the input option chosen by the court.) ii) The other starting input in the DTM was the cost stack for UK costs, with options to select between Cabo’s cost stack and two versions of MGA’s cost stack (one with the average wholesale prices assumed by Mr Parker, and one with the average wholesale price set to £1.20). iii) The DTM then provided for the input of various assumptions as to the likelihood of international expansion, licensing revenues and the development of further products (i.e. beyond the Worldeez range). The cashflows predicted for each of those followed mechanically from the initial selection of the appropriate quantum model, but the DTM (again) allowed for the court to discount the cashflows arising under each of those elements on the basis of a probabilistic assessment of the likelihood of each of those being achieved by Cabo. iv) Further options allowed for the input of actual losses, an adjustment of the agreed discount rate for the small company risk premium, and the selection of simple or compound interest. v) Once all relevant input options were selected, the DTM calculated damages on the basis of undiscounted cashflows up to the date of demise of Cabo (agreed for the purposes of the DTM to be January 2018), plus cashflows thereafter discounted back to the date of demise with an agreed discount rate of 11.19%, reflecting the inherent uncertainty of cash flows after the demise of Cabo.

[61]The DTM was a very useful analytical tool during the trial, because it enabled the parties to interrogate, with the experts, the implications of different permutations of the model, and to explain those implications in their submissions to the court. It also had the benefit of enabling the outputs of the experts’ modelling to be presented to the court on a consistent and (reasonably) accessible basis. While I have not ultimately needed to use the DTM to calculate the quantum of Cabo’s losses, in light of my conclusions as to the likely profitability of Worldeez, that does not, however, diminish the helpfulness of the model as a tool for understanding the parties’ positions. I am grateful for the care taken by the experts and counsel to construct it and then to update it during and after the trial.

Issues

[62]Cabo claims damages against MGA on three bases:(i) abuse of a dominant position contrary to the prohibition in Chapter II of Part I of the 1998 Act and/or Article 102 TFEU;

(ii) unlawful agreements and/or concerted practices contrary to the prohibition in Chapter I of Part I of the Competition Act 1998 and/or Article 101 TFEU; and

(iii) unjustified threats of patent infringement proceedings within the meaning of s. 70 of the 1977 Act. The claims give rise to disputed issues of both liability and quantum. In the case of the patent threats claim there is also a question of whether other relief should be given.

[63]The liability issues are as follows: i) In relation to the abuse of dominance claim: a) The definition of the relevant market on which LOL Surprise was sold by MGA to retailers. b) Whether MGA was dominant on that market, having regard to its market share and other relevant factors. c) If so, whether MGA’s conduct amounted to an abuse of that dominant position. ii) In relation to the unlawful agreements claim: a) Whether there were any agreements or concerted practices between MGA and the main toy retailers or AB Gee in relation to the stocking of Worldeez. b) If so, whether those agreements (or concerted practices) restricted competition by object or effect. c) If so, whether the agreements (or concerted practices) are exempted from the Chapter I prohibition and Article 101 TFEU by virtue of the VBER. d) If not exempted by the VBER, whether the agreements (or concerted practices) are exempt under s. 9 of the 1998 Act and Article 101(3) TFEU. iii) In relation to the patent threats claim: a) Whether MGA’s emails to The Entertainer and TRU constituted threats of patent infringement proceedings within the meaning of s. 70 of the 1977 Act. b) If so, whether Cabo is a person aggrieved by the threats.

[64]The causation and quantum issues arise across all of the claims for which liability is found, and are as follows: i) Whether the patent threats (if established) had any causative effect on the decisions of the retailers in relation to the stocking of Worldeez. ii) The identification of the heads of loss claimed by Cabo. iii) Whether Cabo would have traded profitably but for MGA’s conduct, in the counterfactual case. iv) The quantification of Cabo’s claim, if causation of loss is established.

[65]The final issue is whether declaratory relief should be given on the patent threats claim, irrespective of whether damages are awarded on that claim or the competition claims.

Factual background

[66]The factual events which have given rise to these proceedings are largely uncontested, with the disputes focusing on the characterisation of some of the individual incidents. The description which follows is therefore mostly common ground, with findings of fact where necessary on specific disputed points.

The UK toy industry

[67]The UK’s toy market is one of the largest in the world. NPD, which the parties agree is the authoritative source of market research in the toy industry, valued the UK toy market at £3.4bn in 2017. The toy experts broadly agree on the characteristics of the market. Two common themes emerge from their evidence.

[68]The first is that the toy industry is a fashion industry. While there are established brands that enjoy consistent consumer loyalty, the success of new products is determined by consumer trends, and the fickle tastes of children cause the market to be volatile. It is thus difficult to predict whether a new toy will be successful. The experts agreed that having an innovative product was critical, although that alone was not a guarantee of success. It was common ground that most new toys fail, and the majority of new toys launched in Europe do not last in the market for more than a year. That gives rise to a “close-out” industry of wholesalers (such as Singleton) and retailers, who specialise in purchasing and re-selling underperforming stock at discount prices.

[69]The second theme is the competitiveness of the market. A large number of manufacturers compete by offering differentiated and innovative products. The three largest manufacturers, Lego, Hasbro, and Mattel, had a combined market share on the overall toys market of only 28% in 2017.

[70]For the purposes of its data collection in the toy market, NPD segments toys into supercategories, segments and subsegments, classes, and subclasses, allowing the performance of individual products to be measured. The present case concerns the “dolls” supercategory. That supercategory is then divided into four segments as set out in the table below: nurturing dolls and accessories (nurturing dolls), fashion dolls/accessories and role play (fashion dolls), playset dolls and accessories (playset dolls), and large dolls and accessories/furniture (large dolls). The playset dolls segment is divided into two classes: playset dolls and collectibles, and playset doll accessories. Both LOL Surprise and Worldeez fell within the playset dolls and collectibles class, shown in bold in the table. 71. Table 1: NPD dolls classifications DOLLS SEGMENT SUBSEGMENT CLASS SUBCLASS Nurturing dolls & accessories Nurturing dolls & accessories Nurturing dolls Traditional nurturing dolls Special feature nurturing dolls Nurturing doll clothes Nurturing doll clothes Nurturing doll carriages Nurturing doll carriages Nurturing doll other accessories Nurturing doll other accessories Fashion dolls/ accessories & role play Fashion dolls & accessories Fashion dolls Fashion dolls Fashion doll clothes/ accessories Fashion clothes Fashion accessories Fashion role play & dressup Fashion role play & dressup Fashion role play & dressup Playset dolls & accessories Playset dolls & accessories Playset dolls & collectibles Playset dolls & collectibles Playset doll accessories Playset doll accessories Large dolls & accessories/ furniture Large dolls & accessories/ furniture Large dolls & accessories Large dolls & accessories Doll houses/furniture Doll houses/furniture

[71]During the relevant period, the UK market for playset dolls and collectibles was dominated by three specialist retailers: The Entertainer, Smyths and Toys R Us (TRU). General grocers such as Tesco and Sainsbury’s also sell toys including playset dolls, but toy sales only constitute under 1% of their revenue. The toy experts agreed that a successful launch of a new toy product requires support from at least several of the key toy retailers, and that take-up by grocers such as Tesco is generally dependent on initial successful sales in the specialist toy retailers.

[72]During the period relevant to this dispute, toy sales were influenced by two particular trends. The first was the growing popularity of unboxing videos on platforms such as YouTube(and subsequently TikTok, which launched internationally in 2017). Unboxing videos consisted of child “influencers” unboxing various toys on camera. Many such influencers grew to become incredibly popular. The influencer of common interest to Cabo and MGA was a girl called Tiana Wilson (Tiana), who amassed some 16.5m subscribers on her YouTube channel. Both Cabo and MGA engaged Tiana to promote their products.

[73]The second trend was a spike in the popularity of collectible toys. Although the collectibles market is an established toy market, there were several brands which enjoyed particular success in and around 2016–17, including in particular Shopkins (produced by Moose Toys) and Hatchimals Colleggtibles (produced by Spin Master).

MGA and LOL Surprise

[74]MGA is North America’s largest privately owned toy manufacturer. It owns a number of successful toy brands including Bratz, Num Noms, Little Tikes and LOL. The company was founded by Mr Larian in 1979. It began life as a licensing business, before starting to design and release its own toys in the mid-1990s. MGA’s first major commercial success was the Bratz range of dolls, released in 2001, which became the main competitor to Mattel’s Barbie range of dolls. This led to litigation between MGA and Mattel in the US between 2006 to 2010. MGA has subsequently engaged in litigation with other manufacturers, including a dispute with Mr Mowbray’s company Zuru.

[75]MGA’s LOL Surprise toy was launched in the US in December 2016, and continues to be a popular toy. The original version is pictured below. It comprises a small plastic collectible doll, approximately nine cm high, packaged in a spherical plastic container which is then wrapped in multiple layers of plastic wrapping, as depicted below.

[76]The removal of the first two layers of wrapping reveals paper “teasers” to the contents of the sphere. As further layers are removed, recesses in the sphere reveal miniature toy accessories (themselves in wrapped packages), such as a baby bottle, shoes and clothes. The final layer is the plastic sphere itself which opens revealing the doll. The container can be arranged to stand open as a presentation platform for the doll, and an arched handle is attached to the container, allowing the container and doll within to be carried like a handbag. An example of the container and doll, fully assembled, is shown below.

[77]LOL Surprise was an immediate and phenomenal success, selling out in weeks following its US launch. By January 2017 it was the top selling doll in the US. The industry experts agreed that toys that perform successfully in the US generally go on to repeat that success in the UK market. That was the case for LOL Surprise, which launched in the UK in February 2017. It is common ground that in the UK market, as in the US, the product was an outstanding commercial success. By May 2017 it was the second best-selling toy in the UK across all toy categories (the best-selling toy at the time being one of the Lego minifigures series). Mr Larian described it as a “once in a lifetime product” and a “billion dollar, lightning-in-a-bottle success stor[y]”. Mr Laughton said that “as far as I am aware, no MGA toy (or indeed any toy) has sold as quickly in such a short period of time”. Mr Hunter commented, similarly, that “I have dealt with collectibles for my entire career, and nothing has ever struck a chord with girls like that did”. Stuart Grant said that at that time “LOL was the number one brand in the industry”.

[78]Following the success of LOL Surprise, MGA expanded the range to include other LOL toys, releasing LOL Lil Sisters in August 2017 and a Big Surprise Ball in October 2017. These were also hugely successful toys, and by the end of the year all three toys were ranked within the top ten toys in the UK across all toy categories, with the original LOL Surprise as the best-selling toy. In the course of 2018, further products were added including Pets and Confetti Pop. Numerous further products have been added to the range since then, to maintain consumer interest in the brand.

The founding of Cabo and development of Worldeez

[79]The Cabo founders came up with the idea for a new collectible toy, Worldeez, on a family holiday together in 2014. They were inspired by watching their own children’s interest in unboxing videos and surprise collectible toys such as Kinder Eggs and capsules sold in vending machines. Their idea was for a range of collectible dolls from countries around the world, complemented by other objects associated with each country.

[80]The design process started in early 2016, with the help of an external designer, Mr Olivier. The initial range featured eight countries (UK, France, Italy, Brazil, India, Egypt, Japan and the US) with fifteen figurines for each country: a boy and girl figurine, plus thirteen figurines of associated cultural items. The UK range included, for example a teacup and a Loch Ness monster. The France range included a croissant and an Eiffel Tower figurine.

[81]The Cabo founders did not themselves have any experience of the toy market: for that they relied on Singleton. Mr Sivner had already expressed an interest in Singleton expanding into having its own “intellectual property”, and was aware of the growing market for collectibles. Mr Sivner and Mr Lazarus agreed to enter into a business partnership with the Cabo founders to bring the product to market.

[82]Cabo was incorporated in August 2016 with an initial capital investment of £25,000 from each of Mr Sivner and Mr Lazarus, and investments of £12,500 from each of Mr and Mrs Michaelson and Mr and Mrs Cohen. Mr Sivner and Mr Lazarus each held 25% of the shares in Cabo, with the four Cabo founders each holding 12.5%. The directors of Cabo were Mr Michaelson and Mr Lazarus. Singleton agreed to provide the further funding required to market the products and to fund (at least) the initial production costs. The extent of its willingness to provide ongoing funding was, however, later a source of disagreement within Cabo, and remains a disputed issue in these proceedings.

[83]It was agreed that the Cabo founders would manage the design, marketing and sales of Worldeez. In addition to providing financial support, Singleton was to provide the use of its operational infrastructure and contacts within the toy business, including contacts at a number of leading retailers. In August 2016 Mr Sivner introduced the Michaelsons to Mr Garnham, who had been involved in the development of the Moshi Monsters brand. He offered advice on the design of the Worldeez characters, suggesting the addition of eyes to all of the figurines. Mr Garnham said that he did this as a favour to Mr Sivner, and because he liked to encourage entrepreneurs in the toy industry. He thought that the initial design of the product was too basic, but that it had a better chance of success once the design had been changed.

[84]A factory in China was identified to manufacture the products, and Mr Michaelson and Mr Cohen attended the Hong Kong toy fair in January 2017 where they showed samples of the product to Stuart Grant. He was enthusiastic about the product, and offered advice on the size and packaging. Initially, the Cabo founders had considered packaging the figurines in a toy suitcase, but Stuart Grant felt that this lacked the necessary “wow factor”. Mr Cohen then came up with the idea of a globe design, which fitted the travel theme and was already present in the “O” in the Worldeez logo (which was drawn as a globe). Mr Cohen further suggested the use of a key to unlock a compartment in the globe. The Cabo founders then finalised the artwork for the packaging and wrapping. They were told by the packaging company that they would need two layers of wrapping (one vertical and one horizontal) to cover the globe.

[85]The final “hero” or anchor product of the Worldeez range was the Worldeez globe, consisting of a pink spherical capsule covered in two layers of plastic wrapping which (originally, before it was changed in the circumstances described below) displayed an image of the Italian girl figurine “Bella” riding on a scooter down a rainbow through the sky, the sky being represented with a blue background.

[86]The sphere was divided into two compartments, with one covered by a plastic lid with a locking mechanism, the key for which was in a “blind bag” in the other compartment.

[87]Once unlocked, the covered compartment revealed two blind bags containing Worldeez figurines, each around 2cm high, plus two small fact cards featuring a fun fact about a traditional cuisine or another cultural highlight from one of the countries included in the Worldeez range, and a pamphlet listing the range of figurines to collect. The pictures below show three of the figurines – “Lottie” (part of the England range), “Gina Gelati” (part of the Italy range) and “Chevy Cheese” (part of the France range), as well as an example of one of the fact cards.

[88]Alongside the globe, Worldeez figurines were also available in 5- and 10-piece blister packs. The RRPs for the Worldeez products were set at £2.99 (globe), £5.99 (5-pack) and £9.99 (10-pack).

The initial marketing of Worldeez

[89]Cabo planned to launch Worldeez in late May and early June 2017. Mr Michaelson took the lead on marketing. He engaged a marketing agency, Evolution PR, which produced an initial marketing plan. Cabo eventually moved to using Weird Lime for their marketing, but the original marketing plan remained largely the same.

[90]Nickelodeon, which was then the most popular children’s television channel in the UK, was approached to provide television marketing. Nickelodeon proposed a revenue-sharing partnership in return for advertising airtime. Cabo engaged Diaframma, an Italian production company specialising in children’s television, to produce its advert.

[91]As part of the contract negotiations with Nickelodeon, Nickelodeon requested financial projections from Cabo. An initial set of projections over a three-year period (i.e. 2017 to 2019) was prepared by Mr Avrom Bishop, Singleton’s Finance Director. He emailed those projections to Mr Cohen on 1 March 2017. Mr Cohen replied later on the same date with amended projections, which included increasing the predicted sales for 2019. In cross-examination, Mr Cohen said that he had done so because he expected Worldeez sales to grow year-on-year. The amended projections were sent to Nickelodeon, and a draft contract was subsequently drawn up.

[92]Cabo’s digital marketing plan centred on the unboxing influencer Tiana. Her Worldeez unboxing videos proved to be popular, with the first, which launched on 19 May 2017, reaching around 500,000 views within a few days. Cabo also marketed on the children’s social media platform Popjam. The Cabo founders and Singleton did not, however, agree a specific launch marketing budget. Again, this became a point of disagreement between them.

Discussions with the launch retailers

[93]Cabo intended to launch Worldeez in The Entertainer, Smyths and TRU, followed by the UK grocers (especially Tesco, Asda, Sainsbury’s and Morrisons) and independent toy stores, followed by international markets. Initial discussions with the intended launch retailers were organised by Mr Sivner through his contacts. The Entertainer

[94]Stuart Grant, the buying director at The Entertainer at that time, was a friend of Mr Sivner. As noted above, he had already seen and discussed the Worldeez product at the Hong Kong toy fair in January 2017. He was keen to have the product before other retailers, and emailed Mr Sivner on 26 April 2017 saying “Marc we are 100% behind you to make this a success!! Love the final execution … let’s make this a winner!” The Entertainer placed an order for 40,000 items, including 30,000 globes, on 19 May 2017, and arranged a meet and greet event with Tiana to launch the products at its Birmingham Bull Ring store.

[95]The agreed wholesale prices were £1.20 for the globe, £2.40 for the 5-pack and £4.00 for the 10-pack. That gave The Entertainer a margin of around 52% for the globe, which Stuart Grant said was The Entertainer’s “absolute minimum requirement”. Toys R Us

[96]Mr Sivner and Mr Michaelson had a meeting with TRU on 19 April 2017, following which TRU agreed to hold a meet and greet with Tiana at its Nottingham store. Mr Michaelson and Mr Cohen both initially claimed that TRU placed an order, and Mr Michaelson gave evidence of the price that he said Cabo had “confirmed” with TRU.

[97]There is, however, no contemporaneous evidence of an order from TRU. Moreover, when cross-examined on this point, Mr Michaelson accepted that by the time of MGA’s intervention, more than a month later, TRU had not placed even a verbal order for Worldeez (although he said that it would have done so but for MGA’s conduct). Mr Cohen also accepted that neither a verbal order nor a formal purchase order had been placed by then. The evidence therefore clearly indicates that TRU did not, prior to MGA’s intervention, place an order for Worldeez, whether verbal or in writing. Smyths

[98]There was a meeting with Smyths at its head office in Ireland on 10 May 2017, attended by Mr Sivner, Mr Michaelson, Mr Smyth and some of their buyers. Cabo’s case is that Smyths placed a verbal order for 16,000 items, including 8,000 globes. There is no doubt that there was a discussion of initial order quantities at the meeting, as evidenced by Mr Michaelson’s email shortly after the meeting saying: “We are on the way to the airport but will firm everything up when we get back to Manchester for the following

All above to be delivered in mixed FSDU

[99]Two days later, Mr Michaelson sent an email to Mr Cohen and Singleton saying that Smyths had placed a “verbal order” for the quantities set out above, noting that it would be necessary to “officially agree the order”. Mr Lazarus replied saying that Smyths would need to raise a purchase order. Mr Michaelson then acknowledged that he had not heard back from Smyths following his email to them.

[100]Mr Michaelson claimed in his oral evidence that an order from Smyths “must have existed. It did exist.” There is, however, no evidence of a purchase order being raised by Smyths. As at 16 May 2017, an email between Singleton and Mr Michaelson said that they were expecting a purchase order “soon”. There were then various emails setting up a meet and greet event with Tiana at the Tamworth store (scheduled for the end of July). Smyths emailed Mr Michaelson on 23 May 2017 enquiring when stock would be available, and asking “Can we get stock now?”. On the same day Singleton asked Mr Michaelson and Mr Cohen for further information to send to Smyths, “so they can send PO”. It is apparent, therefore, that a purchase order had not been raised by Smyths at that point.

[101]By 13 June 2017 (after the events described below) in an email exchange between Mr Smyth and Mr Sivner, Mr Smyth said that “we have none on order and not sure if are going to buy”. The evidence of both Mr Smyth and Mr Lazarus was that no formal purchase order was ever placed by Smyths.

[102]Mr Michaelson’s claim that a final order was placed is therefore not supported by any of the contemporaneous documents or the other witness evidence. The evidence indicates that while provisional initial order quantities were apparently discussed with Smyths at the 10 May 2017 meeting, no purchase order was subsequently raised by Smyths either prior to MGA’s intervention or at any time after that. Other retailers

[103]Tesco was approached by Cabo on 11 May 2017, but rejected the product saying that “it looks quite similar to a toy we ran before and unfortunately the world theme didn’t work for us, so we wouldn’t select Worldeez”. Mr Michaelson later reported on one of the WhatsApp chats that this was a reference to Gift’ems. There was also an initial approach by Cabo to Argos, which did not result in any orders. Mr Lazarus’ explanation was that Argos was not interested in selling products with a price point of £2.99. That was consistent with Ms Munt’s evidence that collectibles were not a priority category for Argos, as it focused on the higher priced products.

MGA’s intervention

[104]MGA’s conduct, which is the basis of Cabo’s claim in these proceedings, commenced on 23 May 2017. As set out below, the majority of MGA’s disputed communications occurred in the last week of May 2017. Cabo also relies, however, on various further discussions between MGA and the toy retailers later in the summer of 2017 and (in one case) in the spring of 2018.

Contacts with Cabo and Singleton

[105]On the morning of 23 May 2017, whilst Mr Larian was attending the Las Vegas toy fair, he received an email from Mr Andrew Laughton (at 07:29 local time, i.e. Pacific Daylight Time, or PDT) with the subject line “Worldeez – lol copy – even using Tiana (Toys & Me) who we used to launch”. The email contained a link to the first Tiana Worldeez unboxing video. Mr Larian responded two minutes later (07:31) asking “Who are they?”. Mr Laughton immediately sent a text message to Mr Sivner to ask whether he was behind Worldeez. Mr Sivner responded confirming that he had invested in the product, to which Mr Laughton replied “You do realise this infringes on our patent with LOL – we will protect so expect contact from MGA legal today Marc”.

[106]Mr Laughton confirmed to Mr Larian that Singleton had invested in Worldeez. Mr Larian replied instructing Mr Laughton to “send the correspondence”. What that meant became clear when later that day (by then the early hours of the morning of 24 May 2017 in British Summer Time, BST) Mr Benjamin Johnson, MGA’s Senior Litigation Counsel, emailed Mr Sivner a cease and desist letter.

[107]The first paragraph of that letter stated that MGA had a patent pending for LOL Surprise. The letter went on to contend that Worldeez was marketing and/or selling toys in “packaging confusingly similar to that of LOL Surprise!, including color schemes, shape, art work, lettering, and play pattern replicating the unique unboxing aspect of L.O.L. Surprise!” After referring to MGA’s “substantial goodwill” in its LOL Surprise products, the letter contended that:

“The law prohibits conduct that is likely to mislead or deceive consumers, and it is likely that Worldeez’ packaging, shape, color, and art work, as well as the product concept, marketing methods, features, and play pattern will lead the public to mistakenly believe that Worldeez originate from, are associated with, licensed by, or connected with L.O.L. Surprise!”

[108]The letter then demanded that Cabo immediately cease offering the product for sale in packaging that resembles LOL Surprise packaging, and in general refrain from “any infringement on L.O.L. Surprise! products or trade dress, or any other intellectual property of MGA”, failing which MGA reserved the right to take all available legal action to enforce compliance. It concluded by saying that this was “not a complete statement of MGA’s rights in connection with this matter and/or L.O.L. Surprise!” and that MGA reserved all further rights.

[109]As it happened, Mr Sivner and Mr Lazarus were also attending the Las Vegas toy fair. On 24 May 2017, the day after receiving the cease and desist letter, they had a meeting with Mr Larian, at which they showed him samples of the Worldeez globe, and compared them with Mr Larian’s samples of LOL Surprise.

[110]The reaction of Mr Sivner and Mr Lazarus at that meeting with Mr Larian is disputed. Mr Larian said that Mr Lazarus admitted the similarity between LOL and Worldeez. In email exchanges with Stuart and Gary Grant between 24 May and 1 June 2017, he claimed that Mr Sivner and Mr Lazarus had “agreed it is a knock off”, and that they had agreed to change the design of the shrink wrap packaging to make it different to that of LOL Surprise (see §§125 and 129 below).

[111]Mr Lazarus in his evidence essentially confirmed that he had acknowledged that the Worldeez globe and LOL Surprise looked similar, and had suggested that Cabo could change the packaging design. At the time, however, when shown Mr Larian’s claims about what had been agreed at that meeting, Mr Lazarus was rather coy. He did not deny acknowledging the similarity of the products, but suggested to Mr Cohen and Mr Michaelson that they should discuss the matter with a lawyer, and that an email should be sent from the lawyer “clarifying what we said how we said we would consider looking at things”.

[112]It seems likely, on the evidence, that Mr Lazarus and Mr Sivner did indeed acknowledge that the products looked similar, and agreed to reconsider the design of the shrink wrap packaging around the globe, but were then reluctant to admit to the Cabo founders that they had done so. It is, however, very improbable that they agreed that the Worldeez globe was a “knock off”: that was Mr Larian’s mantra, repeated pejoratively throughout his email exchanges on this issue as well as in court, but there is no evidence other than his account that Mr Lazarus and Mr Sivner agreed with that description of their product (and I have already found Mr Larian to be a thoroughly unreliable narrator).

[113]On 25 May 2017, Mr Larian emailed Mr Sivner and Mr Lazarus, copying Ms Elizabeth Risha (MGA’s general counsel) and MGA’s solicitors Mishcon de Reya: As we discussed, your product look and feel and design (as well as your website) is a blatant knock off of LOL dolls (the NO 1 toy in the industry) and your intend has been to palm off of MGA’s good will and trademarks and trade dress. I also advised that we have a patent pending on this concept. YOU ARE ON NOTICE. … We will take full action if these goods in the current look (and the sample you gave me which even you agreed is confusingly similar to LOL) sold at ANY retailer anywhere in the world. You asked that we come back to you with changes that are acceptable. I have copied our lawyers here (including Mr Ray Black in the UK) on this email. They will send you an undertaking. Please review with your lawyer and sign. Ray: Hold the law suit for 5 days please.”

[114]Cabo’s solicitors wrote to MGA the next day, 26 May 2017, responding to the cease and desist letter, and asking MGA to identify the patent or patent application on which it relied. MGA did not ever respond to that letter. Instead, on the same day Mishcon de Reya emailed a letter and draft undertaking to Singleton. The draft undertaking required all Worldeez stock to be withdrawn from sale and destroyed, along with any “components, moulds, prototypes and any other product-specific material”, as well as all advertising materials. It also required any altered designs to be submitted to Mr Larian and MGA for approval, and that Singleton, Sinco and Cabo each admit liability for “in respect of MGA’s claim for infringement of its intellectual property rights in the LOL Surprise! products”. The nature of these rights was, however, not explained, either in the draft undertaking or in the covering letter from Mishcon de Reya.

[115]Cabo (unsurprisingly) did not sign the draft undertaking. It did, however, take MGA’s threats seriously. Mr Lazarus wrote in an internal WhatsApp message “This guy will tie us up in a million dollar lawsuit … He isn’t a joke. He is not afraid of courts … he will spend million dollars on court costs even if he [loses] just to fight for his brand. This is a serious and real issue now … The guy will do everything to try stop this. And throw all his resources at it. It’s what he does”. Cabo therefore changed the packaging of the Worldeez globe, as described further below, and took steps to protect its intellectual property rights in the event of litigation by MGA.

[116]In the event, notwithstanding the threats made by MGA in correspondence, MGA did not ever bring proceedings against Cabo for infringement of any intellectual property rights related to LOL Surprise.

The Entertainer

[117]At 07:33 PDT on 23 May 2017, two minutes after receiving Mr Laughton’s first email regarding Worldeez, Mr Larian emailed Stuart Grant, forwarding him the Tiana unboxing videoand saying “Please don’t buy knock off. We have a patent and will take action against the copycats”. He then emailed Mr Laughton forwarding his message to Stuart Grant and instructing Mr Laughton to “Cut their shipment now until we get an answer”. It is not disputed that this was an instruction to suspend MGA’s deliveries of LOL Surprise to The Entertainer, which at the time amounted to around $2m worth of the product. Mr Laughton responded confirming that he had done so. Mr Larian replied by repeating his instruction to block supplies to The Entertainer. He followed this with a general instruction to Mr Laughton, stating that “Any retailer that buys knock off will be black listed and not get goods. This will be our policy”.

[118]Mr Larian then sent a long email to both Stuart and Gary Grant (at 08:22 PDT), copying in Mr Laughton and Ms Elizabeth Risha (MGA’s general counsel). He informed the Grants that “We do have patent and copyright and we will take legal action and protect our IP. Andrew [Laughton]: stop LOL shipments to Entertainer please”. That latter instruction was, of course, entirely performative and included purely for the purpose of intimidating the Grants, given that Mr Laughton had already confirmed that he had cut deliveries to The Entertainer.

[119]Stuart Grant was (like Mr Larian) in Las Vegas to attend the toy fair, and replied at 09:01 PDT:

“I’m shocked to read your reaction below, the worldeez product is completely different to lol. The ball opens with a key to unlock 2 surprise 2cm … of course you are completely entitled to pull deliveries of your product but with all the plans we have to support the series 2 launch to pull at this stage would be a great shame. I am more than happy to provide a sample to you of worldeez so you can see the difference for yourself.”

[120]Mr Larian’s response at 09:24 PDT was as follows:

“We have a patent for this concept of layers of surprise in a ball. And this item has even copied the color of our balls as well as the shrink wrap in clear intent to palm off our good will and trade mark and trade dress. It’s really disappointing. … We invest millions in creating original ideas and innovations and will vigorously protect them.” … We invest millions in creating original ideas and innovations and will vigorously protect them.”

[121]Later that day, Mr Laughton spoke to Gary Grant, who Mr Laughton described (in his evidence to the court) as “very emotional”. Following the call Mr Laughton confirmed in an email to both Stuart Grant and Mr Larian (at 20:03 BST, 12:03 PDT) that he was cancelling all of The Entertainer’s orders for LOL Surprise, in order to “protect our property and our development”. It was, however, apparent that by that stage Mr Larian had not seen the Worldeez product and was not at all sure that it was, indeed, an infringement of any of MGA’s rights, because he replied to Mr Laughton’s email (at 12:06 PDT, 20:06 BST) saying “Ask for a sample to see if it’s really a knock off”. Mr Laughton simply replied within minutes saying, “It so is Isaac. Totally unacceptable”.

[122]Stuart Grant replied to Mr Laughton and Mr Larian (17:17 PDT, 01:17 BST on 24 May 2017) complaining that The Entertainer was being “picked on”, and noting that it was one of many retailers stocking the product. He asked that Mr Larian make contact directly with Mr Lazarus. Mr Larian responded (18:15 PDT) that MGA was not picking on The Entertainer, and would “take action against Sinco Toys and any other parties involved to protect our IP. If other retailers are also going to buy and sell infringing products, we will also stop shipping them”. Mr Larian then forwarded to Stuart and Gary Grant the cease and desist letter that had been sent to Mr Sivner (18:20 PDT, 01:20 BST on 24 May 2017).

[123]At 08:00 BST on 24 May 2017, midnight PDT, Gary Grant weighed in, repeating the complaint that The Entertainer had been victimised and “caught in the crossfire” given that “all major retailers” were listing Worldeez. He said that the correct course of action was to serve notice on the company placing the product on the market. Just over an hour later, however, the Grants capitulated. Stuart Grant sent an email to Mr Larian (copied to Mr Laughton and Gary Grant) stating that:

“In light of your legal action, not wanting to be caught in any cross fire and pending legal ruling we have put our worldeez orders on hold. Can you please confirm with this action you are prepared to re instate our lol orders? Isaac we really feel like a piggy in the middle and just want to be fair to both parties. And if the situation was against you I’m sure you would appreciate retailer support.”

[124]Both Mr Laughton and Mr Larian replied thanking Stuart Grant for his confirmation. Mr Laughton confirmed that he had reinstated The Entertainer’s orders. Mr Larian’s email (sent at 05:45 PDT) repeated his position that “We bring new innovation to the toy business and we will stop copycats and protect our IP Worldwide. We expect our friends and partners to respect our IP rights.”

[125]Mr Larian did not, however, leave things there. Later that day (i.e. 24 May) he emailed the Grants to tell them that he had seen Mr Sivner and Mr Lazarus at the toy fair, and that they had agreed that “their ball” was a “knock off” of LOL Surprise. He continued:

“They don’t want a law suit. So our lawyers are going to let them know how and what changes they must make. Which brings me to this. Why would Entertainer (or any retailer) support a blatant knock off? Whether MGA or any other toy company? … All retailers in the U.K. Have confirmed they aren’t going to buy this knock off. We will enforce our IP vigorously and cease any shipment of infringed products at custom. Please ask your buyers to respect original IP.”

Please ask your buyers to respect original IP.”

[126]Notwithstanding the capitulation of the Grants, Mr Larian expressed continued disappointment with The Entertainer in an email exchange with Mr Laughton the next day (25 May). Mr Laughton replied claiming that he was even more disappointed, and noted that “I am reviewing their orders and forecast and will allocate accordingly … They are greedy”. The clear inference from that email was that Mr Laughton was intending to reduce The Entertainer’s allocation of LOL. Mr Laughton begrudgingly accepted, in cross-examination, that he was thereby seeking to appease Mr Larian.

[127]On 31 May 2017 Gary Grant emailed Mr Larian to confirm The Entertainer’s position: “I have followed the email trail regarding [Worldeez] with interest. I am not sure I agree with your view that the products conflict but clearly you are unhappy and if there is enough crossover I have better things to chase and life is too short than to fall out with you. I also value the relationship we have with you, this has taken years to build and it’s not something we wish to put at risk … For the reasons above we have decided not to run the [Worldeez] ball which I understand is the only product in the range you are unhappy with. The concept clearly does not clash with LOL in-fact it’s more like Shopkins than anything else … for this reason we have decided only to stock the 5 and 10 pack [Worldeez] product, I am sure you will be happy with this decision? I hope you can see we have responded and respected to your concerns and look forward to another successful trading year with all your brands”.

[128]Stuart Grant accepted that, as this email made clear, the reason that The Entertainer decided not to stock the Worldeez globe was because of the objections of MGA, and that The Entertainer would have supported Cabo and stocked the Worldeez globe if it were not for MGA’s intervention. He maintained that this was an “independent” decision by The Entertainer. I do not accept that characterisation. The decision was one that was taken with great reluctance in the face of MGA’s threats to withdraw all supplies of LOL Surprise to The Entertainer.

[129]Mr Larian replied to Mr Grant with a lengthy email on 1 June 2017, which included the claim that Mr Sivner had agreed that Worldeez was “too similar” to LOL and that Cabo would change the design of the shrink wrap to make it “very different” to LOL before launching the product. In the same email Mr Larian asserted that The Entertainer could “go ahead and buy [Worldeez] … if you really don’t see a product conflict”, but noted, pointedly, that “you, as a retailer can buy anything you want, the same way a supplier can choose who to sell to and not”. It was put to him in cross-examination that this was a reminder of MGA’s threats to stop supplying LOL to The Entertainer. Mr Larian’s initial denial was followed by a comment which confirmed that this was indeed precisely the intention of that email:

“If you choose to go and buy a knock off, that’s your choice. I don’t like it, but you don’t need the original product. That’s my choice not to sell the original product to – whether it’s Entertainer or another retailer who buys the knock off.”

[130]MGA’s position remained unchanged through the course of 2017. On 7 September 2017 Gary Grant emailed Mr Laughton attaching a picture of Worldeez and asking “These are now [in] Toyshop’s that are selling LOL … can we buy now?” Mr Laughton replied, with careful opacity, “You can of course buy what you want Gary. It’s totally your choice.” Gary Grant complained:

“You know that’s not a straight answer … we support you in all ways and feel we are being penalised as we have bowed to pressure yet others are still receiving LOL stock and selling worldeez … what does this level playing field look like? We have only one weeks stock of LOL left … can you find us more stock as we stay exclusive to MGA?”

[131]Mr Laughton replied, again opaquely:

“As I have said it’s totally your decision Gary, it’s your business. I believe you have stock arriving.”

The inference to be drawn from that email exchange was, however very clear: as Mr Laughton eventually accepted when cross-examined on this point, he didn’t want to say explicitly that if The Entertainer bought the Worldeez globe then it would not be supplied with LOL, but that was what Mr Laughton wanted Gary Grant to understand.

[132]On 15 February 2018, Stuart Grant again emailed Mr Laughton seeking clarification of the products to which MGA took objection. Mr Laughton again refused to answer the question, claiming that “We have no issue with you stocking anything”. Gary Grant, in response the next day, yet again complained about the position being taken by MGA:

“we clearly need to avoid direct copies but we feel the net has been thrown very wide beyond what is reasonable … It seems to me that if MGA believe that a toy is a copy and infringes its IP or [copyright] it will take legal action and in the unlikely event that we have any of these potentially infringing products in stock we will withdraw them from sale without question. To date I am unsure that any such action has been taken so we must assume that all current known products are non-infringing and therefore acceptable to MGA. If by inference MGA are trying to restrict our right to stock non infringing products it does not feel like a partnership. We go the extra mile for all our partners and value genuine relationships so the subtle threats of non-supply feels unreasonable. If we stock lines where MGA have not taken legal action and cease and desist action has not been issued, we have to assume that is fine to sell, otherwise this leaves us not quite knowing where we stand … hence our request for guidance.”

[133]Mr Laughton forwarded that email exchange to Mr Larian saying “Don’t comment or stop Stuart [Grant] coming to showroom – let’s see how big their balls are. Asda and Argos confirmed they are not stocking.”

[134]A week later on 23 February 2018, in response to a request by Stuart Grant for more stock of LOL products, Mr Larian replied copying Mr Laughton:

“Let me look. WM has the exclusive Zuru SHIT surprise and we don’t want LOL to hurt those sales at WM. May be we can give some of their allocation to others in the UK.”

Mr Larian then asked Mr Laughton “Do you think he got the message?” Mr Laughton replied “Well I certainly did and im quite slow towards the end of the week.” Again, as the exchange between Mr Larian and Mr Laughton confirmed, Mr Larian’s email to Stuart Grant was a thinly-veiled threat to The Entertainer that their allocation of LOL would be reduced if they sold products to which Mr Larian took objection.

Toys R Us

[135]On 23 or 24 May 2017 Mr Laughton telephoned the buying director for TRU UK, Andrew Brocklehurst, asking whether TRU planned to stock Worldeez. Mr Laughton’s evidence was that he assumed that he would have said, “We may have a problem with regards patent with regards the balls”. (See the transcript extracted at §439 below, with further discussion of this section of Mr Laughton’s evidence.) Mr Laughton then emailed Mr Brocklehurst on 24 May asking “Any news Andy – I have to report back to Isaac”. Mr Brocklehurst replied, confirming that TRU “won’t be taking the item in its current format” and noting that Singleton was “looking to rework the concept in a way that is acceptable to MGA”.

[136]Mr Laughton then reported back to Mr Larian (still on 24 May): “I have reached out by telephone to all major retailers and although some were presented it in a different packaging they have all agreed that they will not be taking it in its current format. I believe from TRU that Singleton are looking at changing packaging to avoid issues with us”.

[137]Mr Laughton’s evidence was that this email reflected his conversations with TRU, Smyths, Argos and Tesco. Mr Larian responded “TRU buys and you cut their allocation big time”. He followed this up with a lengthy email to TRU management executives in the US and Canada, including Kevin Macnab (then the president of TRU International) attaching the Tiana unboxing video and saying:

“We have patent pending and full copyright and trade dress protection on the LOL concept. We are told that TRU (at least in the UK) is buying this knock off. We are taking legal action against this company in the UK. As we speak. We will stop shipping original LOL to any retailer who supports knock off. Please don’t buy knock off.”

Please don’t buy knock off.”

[138]That email was untruthful: Mr Larian sent it after he had just been told by Mr Laughton that TRU in the UK was not planning to stock Worldeez. Mr Larian forwarded that email to Mr Laughton commenting “See my email to TRU management and send to their U.K. MD. They buy this knock off and we will stop shipping. No exceptions.” Mr Laughton replied confirming “TRU not buying.”

[139]In Mr Larian’s third witness statement, filed in October 2023, he said that he could “now recall” that Kevin Macnab had called him around late May 2017, and had said that his “independent view” was that “Worldeez was a blatant knock-off of LOL. He told me that he had spoken to Dave Brandon (Chairman of TRU) and told him not to buy the knock-off. Kevin told me that TRU UK would not buy the product either.”

[140]I do not accept that account. The call with Mr Macnab is not mentioned anywhere in Mr Larian’s first or second witness statements, filed in January and February 2022, and there is no explanation in Mr Larian’s evidence of why, over a year later in October 2023, he “recalled” this conversation. MGA said that Mr Larian’s memory was “refreshed” by a privileged contemporaneous document. That document has, however, not been disclosed: MGA offered to waive privilege in the document, but on conditions that Cabo did not accept. Mr Larian’s third witness statement, moreover, made no reference to his memory having been refreshed by any document, but simply claimed that he could now “clearly recall” Mr Macnab’s call to him.

[141]As noted above, when Mr Spector attempted to make communication with Mr Macnab to discuss Mr Larian’s evidence on this point, Mr Macnab said that he had no recollection and did not wish to become involved. Since Mr Macnab has not given evidence on the point, I do not place decisive weight on that exchange of messages, but it adds to the general picture of an account given by Mr Larian which is self-serving and unsupported by the other evidence before the court.

[142]Mr Larian’s account of this conversation is also at odds with the contemporaneous evidence indicating that TRU was in fact interested in stocking Worldeez at that time. On 6 June 2017 Mr Brocklehurst emailed Mr Laughton asking to speak to him about Worldeez. A week later, Paul Mitchell, a consultant engaged by Singleton who was previously head of the toy buying division at TRU, emailed Mr Sivner to report on an email exchange with Mr Brocklehurst regarding Worldeez, in which Mr Brocklehurst had said that “the head of TRU US is going to discuss with Isaac to see if they can get an agreement”.

[143]On 5 July 2017 Mr Laughton emailed Mr Larian saying that “Had call from TRU merch manager who is asking what we are doing, they aren’t buying but want to know our stance before making a decision”. There is no evidence of any response to that by MGA, but the email indicated that TRU remained interested in stocking Worldeez, if MGA had lifted its objections to the product.

Smyths

[144]Mr Laughton also telephoned Smyths on either 23 or 24 May 2017, to find out whether they were stocking Worldeez. Mr Laughton accepted that this discussion – and his discussion with the other retailers he called then – would have been similar to his call with Mr Brocklehurst, and that he would have said that Mr Larian believed that the Worldeez globe infringed MGA’s patent. When asked whether he would have communicated Mr Larian’s policy of not supplying LOL Surprise to any retailers interested in stocking the Worldeez globe, Mr Laughton was initially evasive, but he eventually agreed that he would have done so.

[145]Given the terms of Mr Laughton’s email to Mr Larian on 24 May 2017 claiming that he had spoken to all the major retailers (§136 above), it is in my judgment highly likely that Mr Laughton did indeed make very clear to Smyths (as he and Mr Larian had made clear to The Entertainer and TRU) that if Smyths stocked the Worldeez globe it would not receive supplies of LOL Surprise. It is also in my judgment likely, on the basis of that email, that Smyths agreed not to stock the globe. Mr Smyth’s evidence, in his witness statement, was that he did not recall any threat or coercion by MGA in relation to the stocking of Worldeez by Smyths; but as already noted that evidence needs to be treated with caution given that Mr Smyth was not cross-examined. The conduct of Mr Larian and Mr Laughton at the time renders it highly improbable that they would have exempted Smyths from the very clear line that was being taken in relation to The Entertainer and TRU.

[146]Mr Cohen said in his first witness statement that Smyths cancelled the order it had placed for Worldeez, but (as set out above) he accepted in cross-examination that Smyths had not placed an order for Worldeez prior to MGA’s intervention; there was therefore nothing to cancel. It is, however, common ground thatSmythsdidnot go on to place any orders of the Worldeez globe; nor did it take the 5- or 10-pack Worldeez products.

[147]Mr Smyth’s explanation, in his witness statement, was that this was because of Singleton’s history as a close-out company, and because the Worldeez product was not of sufficiently good quality. That explanation is not consistent with the contemporaneous emails up to 23 May 2017 (§§98–100 above) indicating ongoing discussions on an order to be placed for Worldeez, which then evaporated at precisely the time that MGA intervened. The more likely explanation is that although Smyths did not immediately confirm an order following the meeting with Cabo, it was about to do so around 23 May 2017, but then decided not to proceed after MGA had made it clear that LOL Surprise would not be supplied to any retailer that stocked the Worldeez globe. B&M and

Other retailers

[148]On 24 May 2017 Mr Laughton emailed Ms White asking “On another note I assume you’re not going to be selling Worldeez?” Ms White replied “Where has that come from?” Mr Laughton responded “Saw the youtube video and we are protecting our L.O.L IP – all other retailers are not supporting but I thought perhaps because of your relationship with Marc [Sivner] that perhaps you may.” Ms White then asked whether she could buy LOL from MGA, saying that she was “still keen to buy but you said you couldn’t get stock”. Mr Laughton responded confirming that he did not have enough stock to meet his current orders. As discussed below, B&M went on to launch the Worldeez range in its stores in July 2017.

[149]Mr Laughton’s evidence was the retailers he had contacted by telephone on 23 or 24 May 2017 included Argos and Tesco, and that they had agreed not to take the Worldeez globe: see §§136–137 above. When cross-examined, Mr Laughton indicated that he had also telephoned Sainsbury’s and Asda. On 25 May Mr Sivner told the Cabo founders in a Whatsapp message that “Argos will take when you get a letter”, i.e. a letter confirming that Worldeez did not infringe MGA’s IP rights. In a later WhatsApp message on 18 August Mr Sivner reported that “Tesco love it” but would not take the globe “because of lol”. These reports from Mr Sivner were not entirely consistent with the evidence set out at §103 above, indicating that even prior to MGA’s intervention Tesco and Argos were not interested in stocking Worldeez.

[150]There is one contemporaneous message indicating that Hamleys declined to stock Worldeez “because of LOL”, but there is no evidence that MGA ever contacted Hamleys. In the circumstances I do not make any findings of threats by MGA to Hamleys.

AB Gee

[151]Cabo contends that MGA also coerced AB Gee, a toy wholesaler, into dropping Worldeez from its range. It relies on an email to Mr Laughton from Alan McKellar, the general manager of sales at MGA, in June 2018 stating that “AB Gee are well briefed and understand we can remove their distribution agreement at any time, last year they removed Worldeez from their range as per my request”. Mr Michaelson also said that he had been told in around July 2017 that AB Gee had been threatened by MGA and would not be buying any more Worldeez, and were taking the product off their website.

[152]Mr Hunter, the co-owner of AB Gee, was adamant that no-one from MGA ever reached any kind of agreement with him about AB Gee not stocking Worldeez. Indeed, his evidence was that AB Gee’s internal records showed that it had taken stocks of the Worldeez globe, 5-pack and 10-pack in July and August 2017, and that the product was sold by AB Gee until May 2018. He said that AB Gee had not ordered more stock of Worldeez simply because the product wasn’t selling; and that from its initial purchases in July 2017 through to December 2017 AB Gee was never out of stock of any of the Worldeez range. The product was then removed from AB Gee’s ongoing range at the end of 2017. He maintained that if the product had taken off AB Gee would have bought more. He was not able to comment on whether Worldeez was removed from AB Gee’s website, but did not believe that this would have made a difference to sales.

[153]As I have already noted, Mr Hunter was a straightforward witness and I accept his evidence. There is no direct contemporaneous evidence of a threat made to AB Gee, or that AB Gee responded by refraining from placing further orders for the product that would otherwise have been made. Nor has any evidence been adduced as to any changes made to AB Gee’s website. Mr Hunter’s evidence as to the reasons for AB Gee’s failure to re-order Worldeez after the summer of 2017 was entirely credible. I do not, therefore, make any finding of threats made by MGA to AB Gee. Worldeez repackaging and relaunch Immediate steps taken by Cabo

Worldeez repackaging and relaunch

[154]Following MGA’s intervention, as set out above, The Entertainer cancelled the initial order it had placed for the Worldeez globe. TRU and Smyths likewise did not place any orders for the launch of the globe. The various planned meet and greets with Tiana at the stores of all three retailers were cancelled.

[155]As an immediate measure, to try and launch the product in a form to which MGA would not take objection, Cabo repackaged some of the globes as blind bags. The Entertainer purchased a small number of those, along with some of the 5- and 10- pack products, and a video was later filmed with Tiana at The Entertainer’s Birmingham store to promote the Worldeez blind bags. Meanwhile Cabo redesigned the packaging of the Worldeez globe by replacing its blue plastic design with a white plastic design, and removing the image of the “Bella” figurine, as shown below.

[156]The new design was sent to the Chinese factory on 2 June 2017, and existing stock in the UK was rewrapped at a local factory. This was all done before the product was launched, such that no globes in the original blue wrapper were ever marketed. That did not, however, cause MGA to lift its objections to the Worldeez globe: as described above, even in September 2017 Mr Laughton was making clear to The Entertainer that LOL would not be supplied if The Entertainer stocked the Worldeez globe. None of The Entertainer, TRU and Smyths ultimately stocked the Worldeez globe, even in its repackaged form. Nor did TRU or Smyths ever order any of the other Worldeez range (i.e. the blind bags or 5- and 10-packs). The Entertainer ultimately only ordered very small quantities of Worldeez.

[157]The Second Claimant, The Licensing World (TLW), was incorporated on 30 May 2017, with the same shareholders as Cabo. An assignment agreement transferred to it Cabo’s goodwill and Worldeez trademarks for the sum of £1. This appears to have been done in an attempt to protect any Worldeez intellectual property rights in the event that Cabo was sued, as MGA had threatened to do.

Launch of Worldeez globe in B&M

[158]As an alternative to launching the globe in the main toy retailers, Mr Sivner contacted Mr Bobby Arora, owner of B&M, a large discount retailer chain which did not stock LOL Surprise (and for which, as set out above, Mr Laughton had expressly confirmed that no stock would be available). They discussed Cabo’s issues with MGA, and Mr Arora expressed interest in stocking Worldeez at B&M. The Worldeez globe then launched in B&M stores on 1 July 2017, with a wholesale price of £1.50 and retail price of £2.99. Its launch was accompanied by a Tiana video filmed in one of the B&M stores, which got around three million views on her channel.

[159]The launch did not escape MGA’s notice. On the day of the launch, Mr Laughton noticed an advert for Worldeez at B&M, and (apparently before speaking to Mr Larian about this) fired off an email to Mr Arora saying “I’m away at the moment but you might want to call me on Monday re your purchase of Worldeez. I’m afraid this is going to be a real problem”. He then forwarded the advert to Mr Larian, leading to the following exchange:

“LAUGHTON: What you want me to do? I know the owner of this group LARIAN: Nothing. Don’t sell them LOL LAUGHTON: Nothing Isaac ??? The other retailers will be asking how they can do this? … Are you saying legally all our retailers can buy this? LARIAN: It will die fast. I don’t want to waste legal fees. LAUGHTON: But did you see how many they have sold Isaac? What am I to tell Entertainer and Smyths and TRU when they all rung on Monday? LARIAN: will ask the lawyers. You should tell entertainer and Smyth if they buy knock off, they don’t need the real thing. We are truly oversold by at least 3 million pieces. Send a couple of samples here please. LAUGHTON: I will. It hope this at 2.99 won’t damage my own lol business”

LARIAN: Nothing. Don’t sell them LOL LAUGHTON: I will. It hope this at 2.99 won’t damage my own lol business”

LARIAN: Nothing. Don’t sell them LOL

[160]On a separate email thread Mr Laughton again asked Mr Larian to “Let me know Monday what to do re worldeez – really don’t want them in my market.”

[161]Initial sales of the Worldeez globe at B&M were reasonably good, and on a WhatsApp chat between the Cabo founders together with Mr and Mrs Sivner and Mr Lazarus, Mrs Sivner reported being told by “Hayley” that the globe was B&M’s top-selling toy over the weekend of 8–9 July 2017. It is not clear whether this was a reference to Ms White or Mrs Michaelson (especially given that there is no evidence that Mrs Sivner was in contact with Ms White); nor is it clear what the basis for that report was. Ms White’s evidence at the trial was that she did not “recall saying it was the number 1 toy, because I don’t think it was the number 1”. In these circumstances I do not consider that it can be inferred, from Mrs Sivner’s WhatsApp message alone, that Worldeez was indeed B&M’s top selling toy at that time. Success on that scale also appears unlikely given Mr Arora’s comments a little over a week later (set out below) as to the poor sales of the product.

[162]The contemporaneous materials indicated that by 13 July 2017 B&M was selling around 2,000 units of Worldeez per week. Ms White in her evidence described that as “very high” in terms of unit numbers, and said that the globe was a “good line”, but she noted that the sales were well below B&M’s expectations even when they later increased:

“I feel like we were disappointed in terms of we had been sold this story that – by Marc Sivner – this was going to be the biggest hottest new brand. You will sell 20,000 globes a week. So even though we are selling 3,000, 4,000 a week, it’s a good line, I think we were disappointed in the sales of it.”

[163]It is apparent, moreover, that the 5- and 10-packs did not ever sell well. On 4 July 2017 Ms White reported to Cabo that the sales of those were “quite poor” and on 19 July she described sales of the 5-packs as “dreadful” and said that sales of the 10-packs “still aren’t great”, such that she would only re-order them at a lower price.

[164]On 19 July 2017 Cabo emailed Mr Arora suggesting a further marketing video featuring Tiana in a helicopter. Mr Arora’s response was that:

“Sales have been poor so I believe this is 100% necessary. … Alternatively if sales don’t pick up, we have to have all stock collected and the account debited + handling costs.”

[165]The new Tiana video was then put online on 29 July 2017, and was immediately very successful, getting three million views on its first day; it ultimately went on to gain around 20 million views. It is common ground that this video was an impressive piece of marketing, and MGA acknowledged it as such, both at the time and in its evidence at the trial.

[166]Prior to the release of that video, sales of the globe in B&M had increased somewhat, to around 3,500 a week by 25 July 2017. Cabo remained disappointed, with Mr Sivner reporting “Only selling 3500 a week. Not fabulous”. Mr Michaelson and Mr Cohen’s reactions were to suggest that B&M’s figures were inaccurate, and that sales must have been better than reported. There is, however, no evidence of B&M suppressing or underreporting their sales figures. After the helicopter video was released, sales of the globe peaked at 4,500 units of the globe in the second week of August.

[167]An internal MGA email indicates that MGA was continuing to monitor the sales of Worldeez. On 10 August 2017 Deyanira Rubio emailed Mr Laughton forwarding one of the videos promoting Worldeez, noting “I am sharing this you tube video, of Worldeez, they said that is selling incredible in UK and Europe”. Mr Laughton replied “Only selling in a discounter in UK (B&M Bargains) we are the leading brand in UK so making it extremely hard for them to try and get real estate. As far as I’m aware it’s not in Europe”.

Decline in B&M sales after August 2017

[168]By the end of August Ms White reported to Cabo that sales at B&M had slowed, and she therefore cancelled a further order that had been placed for the 10-packs. The 5-pack had not been reordered after July. Again, the Cabo founders doubted the sales figures they were being given by B&M, with Mr Michaelson commenting in September that “There is no reason to think that B&M for example are selling less than they were originally. It is more than possible that the low numbers they gave in the beginning were genuine and the numbers could be building.” Again, however, there is no evidence that B&M’s reporting was inaccurate. Rather, the reality appears to have been simply that the product did not maintain its popularity beyond the mid-August peak.

[169]There is a dispute about when B&M stopped purchasing Worldeez altogether. Ms White’s recollection was that there were no further orders after the end of 2017. While there were invoices dated 19 February and 5 March 2018, Ms White believed that those would have related to orders placed at the end of 2017. Whether or not that is the case, there is no evidence of any purchases by B&M after the start of March 2018.

[170]On 20 February 2018, Mr Arora emailed Mr Laughton offering to “drop Worldeez in favour of LOL”. In a subsequent WhatsApp conversation on 28 February, Mr Arora again offered to drop Worldeez (as well as the Zuru 5 Surprise toy which Mr Larian also objected to). Given the evidence as to the fall in sales of Worldeez in B&M after August 2017, I consider that this offer was simply posturing, or as MGA put it a “negotiating bluff” by Mr Arora. Mr Arora was offering to give up a product which in reality either had already been discontinued by B&M, or was about to be discontinued, in the hope of using that ostensible sacrifice to obtain supplies of LOL Surprise.

[171]Mr Laughton responded on 1 March, “I cannot tell you to do that Bobby but you have to decide to take the brand or not. PS I have found 20000 lol pets and 20000 lol sisters for you – delivery first week in may … hope this makes you happy”. Mr Arora replied “We will drop all other skus. Need few weeks to sell down.” On 1 March 2018 Mr Arora sent an email to Ms White, copied to Mr Laughton, confirming the arrangement:

“Andrew sorting deals and lol. If deal Ok, drop 5 surprise and worldeez. Hopefully will sell out over Easter hols. And then push lol. Rrp. Andrew: pls send deal stock to Hayley today.”

In cross-examination, Mr Laughton accepted that Mr Arora had in these emails agreed with him to drop the Zuru 5 Surprise toy and Worldeez in order to get supplies of LOL Surprise.

Sales to other retailers

[172]Other than B&M, Worldeez stock was sold to AB Gee in both July and August 2017 for onward supply to independent toy retailers. As discussed above, AB Gee did not place any orders after August 2017. Small amounts of stock were sold to other customers (including ToyTown, The Works and WH Smith) and through Amazon.

Licensing and international distribution

[173]Cabo had hoped to secure licensing deals and international distribution of Worldeez. Discussions with potential licensees did not, however, prove fruitful. As for international distribution, prior to MGA’s intervention Stuart Grant had started discussions regarding possible international distribution with Spin Master, a Canadian-based global children’s entertainment company. An initial meeting was arranged to take place in London; but that meeting did not then go ahead after Spin Master learned of MGA’s objections to Worldeez. When Spin Master was approached again in 2018 it was not enthusiastic, citing the competition from other brands in the US market.

[174]In late May or early June 2017 there were also initial discussions with Funtastic, an Australian distributor. However, Funtastic later informed Mr Cohen that it considered that the product was too similar to LOL. Other international distributors were approached, but they indicated that Cabo’s price and cost model did not work for them (as discussed further below at §§582–584).

Nickelodeon advertising

[175]Cabo’s negotiations with Nickelodeon for a long-term advertising contract continued into August 2017, and some adverts for Worldeez were run by Nickelodeon during July 2017. On 3 December 2017 Mr Michaelson had a telephone call with Nickelodeon, in which he told Nickelodeon that Singleton was willing to cover £50,000 for advertising until the end of the year, but was no longer willing to act as Cabo’s guarantor for the longer-term contract that had been under negotiation. He also informed Nickelodeon that because of MGA’s intervention and the consequent lack of retailer support for Worldeez, Cabo would not be able to generate the revenue foreseen by the original proposed contract.

[176]Nickelodeon’s response was that given Cabo’s change of position and the demand for Nickelodeon airtime, it would have to reassess the situation and come back with a new proposal. It does not appear that any such proposal was subsequently made, and no contract was ultimately signed with Nickelodeon.

Demise of Cabo

[177]By September 2017 Singleton was not willing to commit to significant further spending for further stock or marketing. Mr Sivner and Mr Lazarus were trying to recoup Singleton’s loans from any funds received by Cabo, and were exploring options to extricate themselves from the business. On 20 September, without discussion with the Cabo founders, they had a meeting with the toy company Creata to discuss the possibility of Creata taking over production and international distribution of Worldeez, with an option to buy out the brand. When that offer was (some days later) communicated to the Cabo founders, they rejected it. Mr Cohen’s only explanation for this in his evidence was that he thought that they “wanted to cling on” to the product.

[178]There was, nevertheless, by then no consensus between the Cabo founders and Singleton on how the Cabo business was going to be funded and managed going forward. This was apparent from a September 2017 document with notes prepared by Mr Michaelson and Mr Cohen for a meeting with Mr Sivner, setting out a litany of complaints by the Cabo founders regarding Singleton’s participation in the business. The comments in that document made clear that the Cabo founders blamed Singleton for mismanaging the distribution of Worldeez, and felt that Mr Sivner and Mr Lazarus were being unrealistic in their expectations of profitability, were charging Cabo for sums not previously agreed, and were being too pessimistic about the future of the product.

[179]On 6 December 2017, Avrom Bishop, Singleton’s finance director, emailed Mr Cohen, informing him that Cabo’s balance stood at £700 and that any money coming in had been directed to Singleton. The Cabo founders nevertheless continued to explore options for the continuation of the Worldeez. In a WhatApp conversation with the Cabo founders, the Cabo licensing agent Russell Dever reported on discussions regarding a retailer rollout programme for 2018, including the launch of seasons two and three of the product. A new potential packaging design (with a tube rather than a globe) was discussed, and a sample presented at a meeting with Mr Sivner and Mr Lazarus in April 2018. Mr Singleton and Mr Lazarus were, however, by then no longer enthusiastic. Mr Lazarus said in his evidence that he had already started to lose interest by that point, and was focused on not losing further money. As he put it, “I was checked out of the project at this point”. He and Mr Sivner therefore decided not to invest further in Worldeez.

[180]On 14 May 2018 Mr Bishop emailed the Cabo founders to say that Singleton was unwilling to put any further money into Cabo, that it would exercise a lien on the stock and bank account, and that Mr Sivner and Mr Lazarus would not have any further involvement in the company. That effectively marked the end of the Cabo founders’ relationship with Singleton. What the Cabo founders did not know at the time (and did not discover until later in the year) was that on 9 May 2018 Mr Sivner and Mr Lazarus’ other company Sinco had agreed a merchandising and licensing agreement with MGA, which entitled Sinco to sell LOL branded watches and jewellery. That was no doubt a trigger for the decision by Mr Sivner and Mr Lazarus to cut ties with Cabo.

[181]On 12 July 2019, the Cabo founders agreed to pay Singleton £52,496.00 in full settlement of Singleton’s claimed outstanding debt to the company of £361,918.75, and Mr Sivner and Mr Lazarus sold their shares in Cabo to the Cabo founders for a total of £504. Mr Lazarus resigned as a Cabo director, and in his place Mrs Michaelson and Mr and Mrs Cohen were appointed as directors.

Procedural background

[182]Cabo’s claim was issued on 15 May 2020, and (as noted above) the case was originally set down for a four-week trial listed to commence on 27 June 2022. Disclosure was provided in October 2021, and factual and expert evidence was exchanged in early 2022. Unfortunately, MGA then discovered that there had been flaws in its disclosure process, such that a large number of potentially relevant documents had not been harvested. The trial was therefore adjourned by consent on 9 June 2022, and following a further hearing before Joanna Smith J on 20 July 2022 MGA was ordered to pay the costs of the adjourned trial on the indemnity basis, albeit that the failures on the part of MGA were not found to be deliberate: Cabo v MGA

[2022] EWHC 2024 (Pat)

[183]MGA then re-ran its disclosure exercise and provided additional disclosure during January and March 2023.Both parties made consequential amendments to their case, and there was a further round of factual and expert evidence.

[184]On 22 March 2024, MGA issued an application for sanctions against Cabo on the basis of Cabo’s own breaches of its disclosure obligations, which MGA and the court were not aware of in July 2022 when the court was considering the adjournment of the trial and the costs consequences of that adjournment. Cabo’s disclosure failings did not come to light until correspondence between the parties in April and June 2023, as a result of which Cabo produced a large number of further documents in August and November 2023. MGA contends that if Cabo’s own disclosure failings had been known at that time, they would in themselves have likely necessitated the adjournment of the trial, such that a different order as to the costs of adjournment would have been made. The relief sought therefore includes varying the indemnity costs order of Joanna Smith J.

[185]By consent, MGA’s sanctions application was adjourned to be heard after the trial (and judgment following the trial). The application nevertheless remains potentially relevant to the issues in the trial in one respect, namely the credibility of the Cabo witnesses. That point arose because the documents not initially disclosed by Cabo included a large number of WhatsApp chats between the Michaelson and Cohen couples, including a chat entitled “Top Secret” containing 583 pages of material, which appears to have been the main channel of communication between the Cabo founders during the period of time relevant to the claim. Cabo’s explanation for the omission, given in correspondence in February 2024, was that by the time the disclosure exercise was conducted, the Cabo founders had forgotten about the existence of this and other relevant WhatApp chats.

[186]MGA contended that this was not credible given that it was clear that the Cabo founders were continuing to use the “Top Secret” WhatsApp chat for their communications with each other at the time of the disclosure exercise, and the fact that they were also, while preparing for these proceedings, evidently searching WhatsApp chats that were not included in the original Cabo disclosure. At the trial, it was put to both Mr Michaelson and Mr Cohen that they had in fact deliberately decided not to disclose those chats because they were known to contain adverse material; and that this was a matter which went to their credibility. (Both Mrs Michaelson and Mrs Cohen said that their husbands did the disclosure exercise and that they were not involved in that; MGA did not take issue with that evidence.) Mr Michaelson and Mr Cohen both said that they had thought that the chat was a personal family discussion and not disclosable.

[187]Had Mr Michaelson and Mr Cohen carried out a thorough search for relevant material to disclose, that should have captured the “Top Secret” chat as well as other messages. There is no doubt, therefore, that the failure to disclose the “Top Secret” WhatsApp chat and other relevant WhatsApp communications was careless. I do not, however, consider that Mr Michaelson and Mr Cohen deliberately sought to avoid disclosing it in order to hide adverse material. By the time the disclosure order was made (in October 2021) over three years had passed since the demise of the Cabo business, and the Cabo founders had since then moved on to using their group WhatsApp chat for personal communications between their families. That appears to have led to their assumption that it did not contain relevant disclosure. While that indicates a lack of diligence on their part, I do not consider that it indicates dishonesty or otherwise undermines the credibility of Mr Michaelson and Mr Cohen’s evidence. Abuse of dominance claim Overview of the legal framework

Abuse of dominance claim

[188]Cabo’s claim was filed before IP completion day on 31 December 2020, relying on the provisions of both EU and UK competition law. Cabo therefore brings its abuse of dominance claim under both the Chapter II prohibition and Article 102 TFEU, although neither Cabo nor MGA identifies any material difference in the scope of these provisions.

[189]The Chapter II prohibition in s. 18 of the 1998 Act prohibits conduct which amounts to the abuse of a dominant position in a market, which may affect trade within the UK. Section 18(2) specifies that conduct may in particular constitute an abuse if it consists in:

“(a) directly or indirectly imposing unfair purchase or selling prices or other unfair trading conditions; (b) limiting production, markets or technical development to the prejudice of consumers; (c) applying dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage; (d) making the conclusion of contracts subject to acceptance by the other parties of supplementary obligations which, by their nature or according to commercial usage, have no connection with the subject of the contracts.”

[190]Article 102 TFEU sets out essentially the same prohibition, save that the requirement is for the conduct to affect trade between Member States.

[191]The basic legal framework is not in issue in the present case. The issues in dispute are(i) the definition of the relevant product market (it being agreed that the relevant geographic market is the UK);

(ii) whether MGA is dominant on that market; and

(iii) whether MGA’s conduct amounted to an abuse as a matter of law. It is not disputed that MGA’s conduct was capable of affecting trade within the UK for the purposes of the Chapter II prohibition, or between Member States for the purposes of Article 102 TFEU. The relevant market definition Legal principles

The relevant market definition

[192]The classic definition of a dominant position is set out in Case 27/76 United Brands v Commission EU:C:1978:22, §65, and is:

“a position of economic strength enjoyed by an undertaking which enables it to prevent effective competition being maintained on the relevant market by giving it the power to behave to an appreciable extent independently of its competitors, customers and ultimately of its consumers.”

[193]No single factor is determinative in the assessment of whether an undertaking occupies a dominant position on a relevant market. Rather, it is necessary to consider all relevant matters including the nature of the conduct alleged to constitute the abuse: United Brands, §66–68. The conventional starting point, however, is the definition of the relevant market. It is nevertheless important to bear in mind that defining the market in the context of an abuse of dominance claim is a means to an end, i.e. the determination of whether the relevant undertaking has or had market power amounting to dominance, rather than an end in itself: Socrates v Law Society

[2017] CAT 10

, §106. As the Commission’s 2024 notice on the definition of the relevant market for the purposes of Union competition law (the Market Definition Notice) puts it, at §§8 and 10, market definition is an “intermediate tool” to structure and facilitate the competitive assessment in appropriate cases, by enabling the calculation of market shares which may be used to assess an undertaking’s market power.

[194]For that purpose, the relevant product market has been defined as comprising all the products which customers regard as interchangeable with or substitutable for the products of the relevant undertaking, based on their characteristics, prices and intended use, and taking into consideration the conditions of competition and the structure of supply and demand on the market. That does not, however, require the inclusion of all products that might potentially be substitutable with the focal product. Rather, the focus is on the products that exert “effective and immediate” competitive constraints on the relevant undertaking, during the relevant timeframe. As discussed further below, competitive constraints from products falling outside the relevant market should nevertheless be taken into account in the more general competitive assessment: Market Definition Notice §§6, 12 and 17.

[195]The standard analytical structure for the assessment of the relevant product market is to start with the “focal product” under assessment, i.e. the product in relation to which the competition concern arises, and to add readily available substitute products in an iterative manner. Starting with the correct focal product is important. The Competition Appeal Tribunal (CAT) noted in Allergan v CMA

[2023] CAT 56

, §185(4) that:

“As a matter of good practice, the focal product ought to be defined as conservatively i.e. as narrowly as possible. … If one tests the focal product narrowly, then if that definition proves to be too narrow, no harm is done: the substitutes will be captured, not as focal products, but as substitutes for the focal product. On the other hand, if too wide a definition of the focal product is adopted, one runs the risk of ‘baking in’ an erroneous assumption, and thereby adopting an incorrect definition of the market by including within the market definition products that should not be so included.”

[196]As that comment indicates, the problem with the adoption of a market definition that is too wide is that it may not capture market power arising in a more narrowly-defined market. This point is also made in Niels et al. in Economics for Competition Lawyers (3rd ed, 2023), §3.35, noting that taking the smallest market “avoids overlooking pockets of market power”.

[197]The starting point must therefore be defined as narrowly as possible, so as to be able to test whether there is indeed a relevant narrow market on which the undertaking under investigation is dominant. If the starting point is confined to the product under assessment, appropriate substitutes can then be added to that in order to define the boundaries of the relevant market. If, however, the point of departure is a broader market which includes a range of products, that assumes that all of the products within that range are indeed substitutable. The standard process of market definition has no means of testing whether that starting assumption is correct.

[198]Put another way, the standard iterative process of market definition, which stops when no further competitive constraints are evident, is reliable only when the starting point is defined narrowly and products are then added. That process does not envisage the subtraction of products. So if the starting point is too broad and includes products that do not in fact exert “immediate and effective” competitive constraints on the product under assessment, then the standard iterative process leads to an unreliable conclusion.

[199]Once the focal product is identified, the primary means of identifying substitute products is an assessment of demand substitution by customers. In that regard, products are regarded as effective and immediate competitive constraints if customers would switch easily from the focal product to those alternative products, in response to a deterioration in the conditions of supply of the focal product relative to other undertakings, e.g. an increase in price or a deterioration in the quality or innovation of the product.

[200]The notional theoretical test of whether an alternative product exercises such a competitive constraint, such that it should be added to the defined market is the so-called “SSNIP test” or “hypothetical monopolist test”. That test posits a hypothetical monopolist supplying the focal product, and asks whether sufficient consumers would switch to an alternative product in response to the imposition by that hypothetical monopolist of a “small but significant non-transitory increase in price” above competitive levels (usually specified as being around 5–10%) so as to render that price increase unprofitable. If so, then the alternative product is deemed to fall within the relevant market, and the exercise is repeated with further alternative products, until the set of products is such that a SSNIP can be sustained profitably. If not, then the focal product is the relevant product market and the analysis stops there: see Market Definition Notice §§27–29; and Allergan §185(7).

[201]In practice, however, in most cases there will not be reliable market data on which a SSNIP test as such can be conducted. The Market Definition Notice also notes that the SSNIP test may be difficult to apply when undertakings compete on parameters other than price, such as quality or the level of innovation. While, therefore, the SSNIP test may serve as a useful conceptual framework, in most cases the definition of the market will need to rest on other evidence: Market Definition Notice §§30–31.

[202]In addition to the consideration of demand substitution, as discussed above, supply side substitution may also be relevant where the constraining effect of suppliers that are able to switch production is equivalent to that of demand substitution in terms of effectiveness and immediacy. Supply substitution may be relevant, in particular, where suppliers are able to switch production between products in the short term so as to supply customers who might otherwise switch to other substitutable products: Market Definition Notice §§32–33 and fn 58.

[203]It will readily be appreciated from the above points that the exercise of defining a relevant market is not a precise science. Rather, it is an exercise of judgment which requires consideration of all the available evidence.

[204]In relation to demand substitution, relevant evidence includes consideration of the product characteristics, prices, functionalities and intended use, as well as an assessment of the parameters that are most relevant for the choice of customers: Market Definition Notice §§48–50. Where available, evidence of past substitution can be informative, particularly where that is caused by factors such as an unexpected cost shock or unavailability of certain products in the market. However, evidence of customers shifting away from a product as a result in a change of preferences or consumption patterns is less informative for demand substitution: Market Definition Notice §§51–52.

[205]In relation to supply substitution, relevant evidence includes evidence of the ability and willingness of undertakings to switch production and supply in the short term, and the barriers and costs to doing so. Again, where available, any evidence of past supply substitution will be informative: Market Definition Notice §§60–61.

[206]Particular considerations arise where the relevant markets contain products that are significantly differentiated, such that the customer’s choice turns on specific attributes of the product such as design, brand image or technical specifications. One approach is to define a narrow market, while recognising the competitive constraints which may be exerted by other products along the continuum of differentiated products. An alternative approach may be to define a relatively broad relevant market, but to assess the competitive dynamics in specific market segments within that market: Market Definition Notice §§85–86 and fn 113. Both approaches ultimately reflect the same principle, which is that in a highly differentiated market it will generally be necessary to assess the closeness of competition between different products, rather than basing the assessment on a hard-edged market definition analysis: see Market Definition Notice §110 and Whish and Bailey, Competition Law (11th ed, 2024), pp. 33 and 38–39.

The parties’ submissions

[207]The parties’ submissions were based on the evidence of Mr Colley (for Cabo) and Mr Parker (for MGA) on this issue. The experts agreed on the overall framework of analysis, namely that the focal product was LOL Surprise, and that the relevant product market for the purposes of Cabo’s abuse claims was (conceptually) a wholesale market on which LOL Surprise was sold to retailers, since the abuse is said to have been committed by MGA at the wholesale level of the distribution chain. They also agreed that the retail demand on that market was essentially derived from the consumer demand on the downstream retail market, because retailers make decisions about which products to carry and stock based on the purchasing behaviour of consumers. The wholesale market should therefore be defined by reference to the extent to which consumers would regard other products as substitutable for LOL Surprise.

[208]The experts also agreed that although the SSNIP test would in principle be the right approach to adopt, the relevant data are not available in this case: there are simply no data which would make it possible to determine the extent of consumer switching in response to a “shock” in the supply of LOL, such as a cost shock or supply shortage. It is, therefore, necessary to consider other evidence to determine the products that should be regarded as substitutable for LOL Surprise, and therefore within the relevant product market.

[209]The dispute between the experts lay in their assessment of what further evidence was relevant in that regard, and the conclusions to be drawn from that evidence as to the definition of the product market. On those points their positions were unfortunately a very long way apart, and reflected a fundamental disagreement as to the approach to be taken to the available evidence, in respect of what was agreed to be a highly differentiated product market.

[210]Mr Colley proposed that the market should be defined as surprise collectible toys with a sophisticated unwrapping experience aimed at girls aged 6–9 years. He accepted that this was based on an essentially qualitative assessment. Starting with LOL Surprise as the focal product, he identified from the evidence of the industry experts and other evidence what he considered to be the three key features of LOL Surprise which drive demand for that product, namely that it is(i) a surprise collectible toy;

(ii) targeted at girls aged 6–9 years;

(iii) with a sophisticated unwrapping experience. A “sophisticated unwrapping experience” was defined on the basis of Ms Munt’s evidence as requiring more than a single layer of unwrapping, in other words something more than a simple blind bag or equivalent.

[211]On that basis, Mr Colley’s candidate market comprised(i) various different LOL Surprise products,

(ii) all of the Worldeez products, (ii) a list of collectible toys produced by the following major brands: Shopkins, Num Noms, Hatchimals Colleggtibles and Mash’ems/Fash’ems, and (iv) a few smaller brands: Fisher Price, Fizz N Surprise Mermaids, Gift’ems, My Mini Mixie and Twozies. Those products all fell within the playset dolls and collectibles class of the NPD playset dolls segment, although Mr Colley did not base his analysis specifically on the NPD classifications.

[212]Although Mr Colley recognised that in a highly differentiated market the line between products within and outside the defined market might be a somewhat arbitrary one, he said that it was necessary to draw a line somewhere, and maintained that his set of characteristics was the most coherent basis upon which to do so. His view was that since the surprise element was a fundamental feature of LOL Surprise, he did not consider that collectibles without that feature, and without some form of innovative packaging, to be close substitutes for LOL Surprise. Likewise, he did not consider that collectibles targeted at boys or a different age range of girls would be regarded as close substitutes for LOL Surprise. Nor did he consider that other dolls generally (outside the collectibles category) were substitutes for LOL Surprise, particularly in light of the differences in the play experiences between collectibles and other types of dolls, such as nurturing dolls and fashion dolls.

[213]Mr Parker’s view evolved during the course of the proceedings. He originally took the view that the market should be defined as encompassing at least the whole “dolls” supercategory in the NPD classification, and possibly even all toys, if supply-side substitution was taken into account. Even on the narrower of those two potential market definitions, the dolls supercategory would have included all four of the NPD segments of that supercategory: nurturing dolls, fashion dolls, playset dolls and large dolls. He subsequently revised that analysis and his final primary market definition was a market comprising the two NPD segments of fashion dolls and playset dolls. Even that market, however, extended far beyond Mr Colley’s market to include playset dolls with no surprise unwrapping element, such as Sylvanian Families, as well as fashion dolls such as Barbie and Bratz. As discussed below, Mr Parker’s market also included accessories for dolls, such as dolls’ dresses, and role play and dressing up accessories, such as glitter tattoos.

[214]Mr Parker arrived at that market definition by taking as his starting point the NPD product segment containing LOL Surprise, namely the playset dolls segment, and testing whether there was substitution between that segment and the fashion dolls segment, by looking at NPD revenue data for the two segments over the period 2017–2021. He concluded that the overall market size of the two market segments was broadly constant over that period, but that there were significant fluctuations in the size of the two segments. Crucially, however, he found a very strong negative correlation of -0.98 between the monthly annual rolling average of sales of the playset and fashion dolls segments over the period: as one segment grew, the other almost always shrank. His conclusion was that this was consistent with a market that included both segments, and inconsistent with any narrower market.

[215]In response to a question from the court during the opening submissions as to how one might analyse the spectrum of products which are (more strongly through to more weakly) substitutable for LOL Surprise, Mr Parker provided two successive reports (his eighth and ninth reports) which considered the individual brands in the playset and fashion dolls segments which lost monthly sales during periods in 2017 and 2018, during which period there was a significant increase in the LOL Surprise monthly sales. He said that this analysis showed the brands that were the strongest economic substitutes for LOL Surprise. For 2017 those were, in order, Shopkins, Monster High, My Little Pony, Trolls and Twozies. The results for 2018 were similar but included several further brands, including Num Noms and Barbie whose sales increased during 2017 but decreased during 2018. He said that this analysis would provide a useful guide if the court wanted to build up a market starting from LOL Surprise, although he maintained that the market was more appropriately defined as encompassing all products in the playset and fashion dolls segments.

[216]Each party robustly rejected the approach taken by the other party’s expert. I will address the criticisms of each of the two approaches, before setting out my conclusions on the approach which should in my judgment be adopted in the present case.

Mr Colley’s approach

[217]MGA advanced both general and specific criticisms of Mr Colley’s approach, which I will consider in turn.

[218]General points. The first general criticism was that Mr Colley’s analysis was not in any sense an application of the SSNIP test, since he had not considered the question of whether it would be profitable for a hypothetical monopolist in his postulated market, or indeed any market, to sustain a small but significant price increase.

[219]I agree that Mr Colley’s approach was not an application of the SSNIP test, as such. Rather, it was an assessment of substitutability based primarily on a qualitative assessment of product characteristics, in circumstances where (as the experts agreed) there were no data available to conduct a SSNIP analysis as such. There is, however, nothing inherently wrong with that approach – as explained above, in many or even most cases it will be necessary to look at alternative evidence of substitutability, rather than conducting a SSNIP analysis.

[220]MGA’s second general criticism was that Mr Colley did not iteratively assess whether products competed with the focal product. Rather, he simply populated his product market on the basis of his proposed set of key characteristics.

[221]Again, that is true but I do not consider that there is anything wrong with that approach. If one starts (as Mr Colley did) with the narrowest possible focal product, namely the specific product under investigation, a definition of a broader relevant product market does not necessarily require the addition of successive individual products. Rather, if there is an identifiable group of products which can be said to exercise an immediate and effective competitive constraint on the relevant focal product, that group as a whole can properly be added as to the relevant market. Nor is there anything wrong with defining the relevant market by reference to the characteristics of that group of products, if the available evidence indicates that the only products which exert an immediate and effective competitive constraint on the focal product are products with the identified characteristics of that group.

[222]The third general criticism was that Mr Colley had not in fact conducted an economic analysis of the disputed product market. Rather, his assessment was almost entirely based on an assessment of product characteristics drawn mainly from Ms Munt’s evidence.

[223]It is fair to say that Mr Colley’s assessment of product characteristics was based heavily on the evidence of Ms Munt, together with references to other factual material, such as the factual witness statements and contemporaneous documents. Mr Colley did, however, conduct some analysis of price trends and overall sales revenues in different NPD market segments, which he said provided some support for his analysis, although he pointed out the limitations of reliance on patterns of sales revenues (a point which I address further below). There is nothing unusual in an economic report on market definition addressing both qualitative and quantitative evidence, particularly where there is an interplay between the two, and commenting on the insights that could be drawn from each of those types of evidence.

[224]I do not, therefore, accept the general criticisms of Mr Colley’s approach. The important question is whether the results of his analysis stand up to scrutiny. That is the subject of MGA’s specific criticisms.

[225]Specific points. MGA’s first specific criticism was that Mr Colley’s boundary between a “sophisticated unwrapping experience” and what Ms Munt described as a “simple element of surprise” was unclear and arbitrary.

[226]There is in my judgment some force in that point. Ms Munt’s witness statements on this point were not entirely consistent, and when cross-examined her evidence came down to an assertion that there was a material difference between a “single” and a “double” element of surprise. So, for example, for collectible toys designed for the target group, a product in a single blind bag would be outside Mr Colley’s market, whereas a product in a blind bag within a blister pack would be within the market. Mr Colley, likewise, maintained that (in principle) a product inside a plain capsule would be outside his market, whereas the same product in a capsule with a layer of plastic shrink-wrap would be inside his market.

[227]Both Ms Munt and Mr Colley struggled to articulate a coherent reason why consumer demand for a surprise collectible toy, as a substitute for LOL Surprise, would turn on the sort of distinctions which they drew in their evidence. Ms Munt could not explain why, for example, consumer demand would have been materially different for a toy inside a capsule, compared with the same toy in a blind bag in the same capsule. Mr Colley eventually simply said that, in a highly differentiated product market, with a chain of substitution along a continuum, he had to draw the line somewhere.

[228]The point that it may be difficult to draw a clear line between products inside and outside the relevant market where the case concerns highly differentiated products is not controversial. As discussed above, that is why a simplistic analysis based on market definition and market shares is unreliable in a highly differentiated product market. While it may be necessary, as Mr Colley said, to draw the line somewhere, that should not lead to the arbitrary exclusion from the competitive assessment of products falling outside the defined market, which in some cases may exert a competitive constraint that is very similar to that of products within the market.

[229]In the present case, there was no evidence to support the suggestion that the distinction between a single and double element of surprise would have made a significant difference to the pattern of consumer demand, so as to produce a significant difference in the strength of the competitive constraint on LOL Surprise. By contrast, there was no dispute that having at least some element of surprise was a highly relevant feature for consumer demand.At most, MGA said that account should be taken of supply substitution, given that a non-surprise collectible could relatively easily be packaged with a surprise element, for example by putting it in a capsule or blind bag.

[230]On that latter point, it was apparent that minor changes to toy packaging could be implemented relatively quickly if necessary. Cabo itself changed its outer packaging after MGA objected to the original design of the Worldeez globe, including repackaging an initial batch of figurines in blind bags for the launch of the product in The Entertainer. There was, however, no evidence before the court which suggested that non-surprise collectibles exerted an effective and immediate competitive constraint on LOL Surprise arising from the fact that it might in theory have been possible to repackage and redesign them as surprise collectibles. Nor was there any evidence of that ever having occurred in fact. I do not, therefore, consider that Mr Colley was wrong to base his market definition on products with at least some element of surprise, even if (as set out above) I do not consider that a hard line may be drawn between products with more or less sophisticated unwrapping experiences.

[231]Secondly, MGA criticised Mr Colley’s selection of characteristics on the grounds that it ignored the doll play potential of LOL Surprise. MGA pointed, in that regard, to the evidence of both of the toy experts that one of the particularly appealing features of LOL Surprise was that it was a mini-doll which enabled a classic doll play pattern.

[232]There is no doubt that the play experience of LOL Surprise was an important characteristic of the product. It was not, however, clear how “doll play potential” could meaningfully be used to define products that exerted an “immediate and effective” competitive constraint on LOL Surprise, when in any event all of the products under consideration (on both Mr Colley and Mr Parker’s market definitions) fall within the dolls market. That is perhaps why, although it was common ground between the toy experts that the “doll” features of LOL Surprise were both appealing and popular, the evidence indicates that the unwrapping experience of LOL was a more important driver of consumer demand than the doll play experience. Ms Munt said that the surprise unboxing experience was “a key part of what girls love” about LOL, and Mr Harper said that a “key innovation in the LOL product was to make the packaging, the unwrapping of the product, a key part of the product experience”.

[233]Contemporaneous MGA internal documents likewise identified the “surprise and unwrapping” of the multiple layers of packaging as the elements most liked by girls buying into the brand, and listed “The extended and engaging consumer unboxing experience” as a feature that made LOL Surprise unique. MGA recorded its preference test findings as showing that “Unwrapping the surprise is the preferred play pattern over other toy elements, second is the doll itself”. Those conclusions appear to have been based on external research conducted by the market research company SSI, which found that 63% of girls ranked the unwrapping experience as their favourite element of LOL Surprise. Consistent with those findings, in evidence filed in support of MGA’s application for a US patent for LOL Surprise, MGA’s vice president of business affairs and planning (Sam Khare) stated that:

“The remarkable success achieved by the LOL Surprise product was driven by the originality of the unboxing play pattern experience that kids have come to enjoy on their own and in online videos combined with the excitement created by the blind bag toy experience.”

[234]Thirdly, MGA placed considerable emphasis on the source of LOL’s revenue growth, a point that was explored by Mr Colley in his analysis of toy sales revenues. Mr Colley noted that LOL Surprise revenues increased by £80m between 2017–18, and that around £36m of that growth appeared to have come from within brands within the playset dolls and collectibles NPD subclass. In cross-examination, Mr Colley accepted that it could be inferred that around half of that growth would have come from products within his candidate market – i.e. that the growth in revenues of LOL Surprise cannibalised sales that would otherwise have been made by those products. MGA contended that since, on that analysis, the majority of the sales of LOL Surprise came from outside Mr Colley’s market, that suggested that the market was wrongly specified.

[235]I do not accept that criticism. As Mr Colley noted, when cross-examined on this point, an analysis of revenue patterns across different categories of dolls might show a correlation between changes in different segments, but that does not establish causation, and is therefore no more than “weakly suggestive” of demand switching. Given that the dolls market is characterised by rapidly changing consumer preferences and trends, and highly differentiated products at different price points, the sources of the revenue growth of LOL Surprise cannot reliably be deduced from such an analysis. Indeed, even if the evidence had directly shown switching (which it did not), the Market Definition Notice makes clear that this would be less informative for demand substitution if it resulted from a change of consumer preferences: see §204 above.

[236]The decline in the overall toys market revenue during the same period is a further complicating factor. Mr Colley explained that the fact that the dolls segment revenue remained (by contrast) fairly constant could indicate that LOL Surprise increased the revenue of the dolls market compared with what would otherwise have occurred, rather than diverting revenue that would have gone to other dolls. There is, moreover, an asymmetry problem: that even if switching from certain categories of dolls to LOL Surprise could be identified, it does not follow that there would be a switch back from LOL Surprise to the same categories of dolls in response to a price rise for LOL. As Mr Colley noted, a finding of an asymmetric competitive constraint may arise, in particular, where a new superior product diverts sales from an existing product.

[237]Ultimately, the cross-examination of Mr Colley on this point reflected the difficulty of seeking to draw inferences on switching from general data on changes in revenue patterns, for which there could be many different explanations. That is why Mr Colley was (properly, in my judgment) cautious regarding the conclusions that could be drawn from this sort of analysis.

Mr Parker’s approach

[238]As set out above, unlike Mr Colley, Mr Parker placed very heavy reliance on an analysis of NPD revenue data. His central observation was that over the period 2017–2021, when the fashion dolls segment grew, the playset dolls segment shrank, and vice versa (see §214 above). He also noted that the combined size of the two market segments was broadly constant during that period. He contended that this demonstrated “a pattern of demand being substituted between Playset Dolls and Fashion Dolls”. On that basis he concluded that fashion dolls and playset dolls were in the same relevant product market during that period, and that there was not, during the relevant period, a narrower relevant market for substitutes for LOL Surprise. As explained at §215 above, Mr Parker’s alternative analysis of individual product substitutes for LOL Surprise likewise rested on considering the brands which lost sales at the period of LOL’s growth in revenue.

[239]There are several fundamental problems with that approach. The first is the point discussed above in relation to Mr Colley’s comments on sales revenues: that a correlation (even a very strong correlation) between revenue patterns does not establish causation and therefore demand substitution, given the numerous other factors that could have influenced those revenue changes. That problem undermines both Mr Parker’s primary correlation analysis, and his alternative analysis of individual brands.

[240]This issue was particularly evident on examination of Mr Parker’s alternative analysis of individual brands. That analysis produced lists of brands within the playset dolls category which Mr Parker considered were the “strongest economic substitutes” for LOL Surprise, on the sole basis that their sales fell during periods in 2017 and/or 2018 in which the sales of LOL Surprise increased. One obvious problem was that this analysis produced the incoherent result that some products would be regarded as close substitutes in one year but not the next, e.g. Num Noms and Barbie (on the list of close substitutes for 2018, but not for 2017). Both the 2017 and 2018 lists also notably excluded Hatchimals Colleggtibles, on the basis that during both periods the sales of this product increased rather than decreased. The toy experts agreed, however, that Hatchimals Colleggtibles was a surprise collectible product that offered a sophisticated unwrapping experience (featuring a small animal toy that “hatched” from an egg: see the first picture below). Its product characteristics therefore indicated that it was a close substitute for LOL Surprise – and at least as close a substitute as some of the other products in Mr Parker’s lists (e.g. Num Noms, which are scented food-related collectibles in the form of mini lip glosses contained within a pot designed to look like a yoghurt pot: see the second picture below).

[241]When cross-examined on these points, Mr Parker was forced to concede that his revenue correlation analysis did not work on an individual brand-by-brand basis (in particular, he said, because of the different lifecycle stages of individual brands). That does, however, call into question why he provided two reports to the court which offered precisely that analysis, contending that it allowed him to identify the products that were the strongest substitutes for LOL Surprise.

[242]Mr Parker nevertheless maintained that his revenue correlation analysis remained sound at the level of entire NPD segments, on the basis that an aggregation across products would “net out” factors affecting individual brands. But that fails to take account of the fact that even across entire NPD segments, demand will change as a result of changing consumer preferences, making such changes less informative for demand substitution. Mr Laughton explained that before the launch of LOL Surprise MGA’s business focused primarily on fashion dolls; after the launch of LOL Surprise, however, MGA’s sales of mini-dolls as a proportion of its business increased, and a number of competitors began to produce mini-dolls. By 2021, however, sales of mini-dolls had begun to decline, and Barbie was “back” and growing again as a brand. Mr Laughton concluded that “[s]uch changes in the nature of the competition happen all the time, as children’s interests alter and new trends develop”. Mr Parker’s analysis took no account of that point.

[243]The second fundamental problem is that Mr Parker’s primary correlation analysis took as its starting point the playset dolls NPD segment, and then sought to test whether the market should be widened further by adding in the fashion dolls segment. Mr Parker thereby assumed that if he could show substitutability in general between the two market segments, then it would inexorably follow that every product in the playset dolls segment was substitutable for LOL Surprise.

[244]That is a classic example of starting with too broad a focal set. What Mr Parker did not do was to start with the focal product, i.e. LOL Surprise, and consider which products in the playset dolls segment, or at least which types of products, were indeed substitutable for that focal product. In cross-examination, he asserted that he did not need to do so, because the logic of the SSNIP test was (in his view) that if the candidate market was tested and found to be too narrow, then it was not necessary to consider whether the candidate market was the correct starting point. That is, however, precisely the approach deprecated in the case-law and economic literature referred to at §§195–196, on the basis that it risks overlooking “pockets of market power”.

[245]As set out above, the correct approach is to start with the focal product, defined as narrowly as possible. Mr Parker’s primary analysis did not do that. His alternative brand-by-brand analysis did purport to do that, but (as set out above) he eventually accepted that his alternative analysis was unreliable. The result was that Mr Parker did not put forward any reliable analysis which adopted the correct starting point.

[246]Thirdly, Mr Parker’s analysis produced results which were inconsistent with any sensible assessment of competitive constraints by reference to product characteristics and functionality. The discussion at §240 above illustrates that point in relation to Mr Parker’s alternative analysis. The problem was equally evident for Mr Parker’s primary analysis, which resulted in a very broad market that included a whole range of products with wholly different characteristics and functionality to those of LOL Surprise. These included, for example: i) Accessories for dolls, such as a Barbie dress or car. Mr Harper sought to argue that those could be in some way competitive with LOL Surprise, but it was apparent that any competitive interaction was at best indirect. ii) Role play and dressing up accessories, such as glitter tattoos kits, and accessories sold for use with playset dolls, such as a “My Fairy Garden” growing kit. There was no evidence that these sorts of products could have exerted an “immediate and effective” competitive constraint on LOL Surprise; and Mr Harper accepted that neither of those products were strong substitutes for LOL Surprise.

[247]In fact, Mr Parker’s approach did not account for the product characteristics of LOL Surprise in any meaningful way at all. MGA’s answer was to rely on Mr Harper’s comments to the effect that children wanting a particular toy would not ask for a toy based on its product characteristics, such as a “small collectible doll with a sophisticated unwrapping experience”; and that trying to define a market based on those characteristics was therefore divorced from reality.

[248]Mr Harper was not, however, instructed as an expert on market definition and his comments did not reflect the purpose of market definition as a matter of competition law and economics. While the process of defining a product market by reference to demand substitutability will obviously look at consumer preferences and relevant drivers of demand, the definition of a market is a theoretical construct which is used to assist in the analysis of dominance. A market definition is therefore not intended to represent the actual decision-making processes of a consumer choosing to buy a particular product. Indeed, if the market in the present case were required to correspond to the way in which a small girl in the target age range would choose her toys, Mr Parker’s definition would also fail that test, since small girls do not habitually ask for toys by reference to particular NPD segments.

[249]Ultimately, MGA had no real answer to the problem that Mr Parker’s evidence on this issue did not meaningfully reflect any assessment of the types of qualitative evidence referred to in the Market Definition Notice that are consistently used to define markets, particularly in the absence of any data on which a proper SSNIP analysis could be conducted. His approach is a good illustration of the risks of conducting a statistical analysis without verifying that its results are grounded in the market realities and common sense. The result was an opinion that gave the impression of advocacy for MGA rather than more objective evidence to assist the court.

Conclusions on market definition

[250]As set out above, there were some problems with each of the experts’ evidence on the market definition. Overall, however, I consider that Mr Colley’s approach is to be preferred. Unlike Mr Parker, Mr Colley started (correctly) with the focal product, narrowly defined as the product under assessment, i.e. LOL Surprise. To that he added products based on an assessment of relevant product characteristics, i.e. characteristics that were meaningful for consumer choice. Mr Parker’s alternative approach, based on looking at negative revenue correlations between entire NPD segments, adopted a starting point that was unreliable for the reasons set out above.

[251]The evidence supports a market definition reflective of the characteristics identified by Mr Colley in respect of the target age range and the focus on surprise collectible toys. The main problem with Mr Colley’s market definition was that it sought to draw too bright a line between products with a “sophisticated” unwrapping experience and products with a less sophisticated unwrapping experience, such as products with a simple blind bag or other “single element of surprise”, as Ms Munt put it.

[252]The correct approach, as Mr Colley acknowledged in his cross-examination, is to recognise that the boundaries of the market definition in a highly differentiated product market will be difficult to draw, and should therefore not be regarded as hard-edged. The competitive analysis should therefore consider the extent of competition from products falling outside the defined boundary, and should if necessary test the effect of extending the market definition to some of those products, at least by way of sensitivity analysis.

[253]In the present case, given the difficulty in drawing a bright line between products with a single and a double element of surprise, it is appropriate to consider whether the extension of Mr Colley’s market to all collectible toys targeted at girls aged 6–9 years, with at least a single element of surprise (e.g. a blind bag), would make a material difference to the assessment of dominance, in particular by considering the impact on the market share calculations. I will refer to this market as the extended Colley market.

[254]In addition, the dominance assessment should consider whether there are other specific products falling further outside the relevant market which also act as competitive constraints in relation to LOL Surprise. MGA’s main argument in that regard was that LOL Surprise competed not only with collectible toys but also with fashion dolls such as Barbie. The dominance assessment should therefore consider the extent to which there is evidence of fashion dolls exerting a competitive constraint on LOL Surprise. Whether MGA was dominant on the relevant market Legal principles

Whether MGA was dominant on the relevant market

[255]It is well-established that market shares are a proxy for measuring market power. The Commission’s 2009 Guidance on its enforcement priorities in applying Article 102 TFEU (the Article 102 guidance) comments at §13:

“Market shares provide a useful first indication for the Commission of the market structure and of the relative importance of the various undertakings active on the market. However, the Commission will interpret market shares in the light of the relevant market conditions, and in particular of the dynamics of the market and of the extent to which products are differentiated. The trend of development of market shares over time may also be taken into account in volatile or bidding markets.”

[256]In broad terms, as summarised in Bellamy & Child, European Union Law of Competition (8th ed, 2018), §10-028 a market share of above 40% is typically regarded as a significant factor in evidencing dominance; and market shares significantly and consistently above 50% are likely to be strong indicators of dominance save in exceptional market conditions. Dominance is not likely if the undertaking’s market share is below 40%, unless there is specific evidence such as substantial disparities in market share and/or significant barriers to entry.

[257]Market shares are typically assessed over one year reference periods, assessed for the duration of the conduct under investigation. There is, however, no inflexible rule as to the reference period that will be relevant in a given case, and it may, in particular cases, be necessary to consider market shares over longer or shorter reference periods: Market Definition Notice, §113.

[258]Market shares cannot, however, be used simplistically in the assessment of dominance. While the defined market allows the identification of the closest competitive constraints on the undertaking under scrutiny, it will also be relevant to consider the extent of competition from products outside the defined market, while bearing in mind that such products will be more remote constraints: Market Definition Notice §17. That is particularly the case in a market where products are significantly differentiated.

[259]It is also always necessary to have regard to the features of the market, because the importance of the market share as evidence of a dominant position will vary from market to market, according to the structure of the market. It is therefore necessary to consider the market share in the context of the nature of the market: Socrates, §§120–1, citing Case 85/76 Hoffmann-La Roche EU:C:1979:36, §40. Factors typically taken into account in the assessment of dominance are barriers to expansion, entry and exit, the countervailing power (if any) of competitors, customers and consumers, and the stability of the market share of the allegedly dominant undertaking.

[260]The conduct of the undertaking alleged to be dominant may also be a relevant factor. As the European Court noted in United Brands, §68, in considering the competitive situation on the relevant market “it may be advisable to take account if need be of the facts put forward as acts amounting to abuses without necessarily having to acknowledge that they are abuses.” In

Case T-30/89

Hilti v Commission EU:T:1991:70, §19, the General Court thus observed that “Hilti’s commercial behaviour bears witness to its ability to act independently of, and without due regard to, either competitors or customers in the markets in question.” If there is evidence that the relevant undertaking has been able to impose terms that disregard the wishes of its consumers, that is the “hallmark of dominance”: Genzyme v OFT

[2004] CAT 4

, §255. Each case will, however, turn on its own facts, and the factors relevant in one case may not be the same as the factors that will be relevant in another case.

The parties’ submissions

[261]The following table sets out MGA’s percentage market shares on the retail market, month by month during 2017, calculated on the basis of the NPD data, on five alternative bases discussed during the trial. These were:(i) Mr Colley’s figures based on his defined market;

(ii) Mr Parker’s reconstruction of the market share figures for Mr Colley’s market, using a slightly different methodology;

(iii) Mr Colley’s sensitivity analysis of the extension of his market to all Shopkins toys in the playset dolls and collectibles class, and all Num Noms toys in MGA’s 2017 sales data (in both the playset dolls and collectibles and the playset doll accessories class);

(iv) Mr Parker’s figures based on the NPD market segment for playset dolls and accessories; and

(v) Mr Parker’s figures based on his preferred market definition of the NPD market segments for both fashion and playset dolls. 263. Table 2: 2017 market shares for Colley and Parker markets (%) Jan-17 Feb-17 Mar-17 Apr-17 May-17 Jun-17 Jul-17 Aug-17 Sep-17 Oct-17 Nov-17 Dec-17 (i) Colley mkt 28.68 37.05 33.13 36.26 34.31 39.52 56.28 62.95 63.84 60.94 71.57 58.43 (ii) Colley mkt recalc 28.06 33.94 38.93 35.49 34.27 43.73 54.09 63.29 72.37 66.91 65.30 66.80 (iii) Colley mkt + Shopk/NN 25.97 34.44 30.90 35.22 36.72 38.90 53.07 59.32 60.06 58.43 72.52 59.04 (iv) Playset dolls sgmt 9.66 11.37 11.88 12.70 14.62 17.51 18.84 24.69 25.94 26.69 25.46 24.66 (v) Fashion & playset dolls sgmt 6.78 7.99 7.98 8.55 10.06 11.53 11.83 15.26 15.56 15.15 15.21 15.03

[262]Detailed figures were not provided for the above markets for the period after December 2017. Mr Colley said that he had not conducted a systematic market definition exercise beyond 2017, but that his preliminary assessment of his defined market in 2018 suggested that MGA had around 80% of that market (produced by the significant increase in LOL sales in 2018). Mr Parker provided the following figures for MGA’s market shares in 2018–19 on the basis of the NPD playset dolls market segment, as well as the (narrower) playset dolls and collectibles class: 265. Table 3: Parker market share estimates for 2018–19 (%) 2018 2019 Playset dolls & accessories (segment) 54.9 55.1 Playset dolls & collectibles (class) 63 60

[263]As already explained, the relevant market in this case is the wholesale market. It was common ground that market shares on that market could be derived from the retail market shares, but the experts differed as to what adjustment that would require.

[264]Mr Colley said that the time lag of 2–4 months between placing a wholesale order and delivery to the retailer meant that, in order to assess MGA’s market share on the wholesale market during the main period of the alleged abuse, namely from May to December 2017, it is necessary to consider MGA’s retail share from July 2017 onwards. On that basis, and on Mr Colley’s market definition, MGA’s market share was well above 50%. Cabo’s submission was, therefore, that on a correct market definition and with an appropriate time lag MGA’s market shares were strongly indicative of dominance. Given that MGA’s market shares were even higher in 2018, Cabo said that this showed a position of market power which was sustained and therefore not ephemeral.

[265]Cabo also relied on further factors which in its submission supported the conclusion that MGA was dominant during the relevant period:(i) the fact that the key retailers regarded LOL Surprise as a “must-stock” product during the relevant period, and therefore lacked bargaining power in negotiations with MGA;

(ii) the barriers to entry and expansion in the toy market, which Cabo said are high;

(iii) MGA’s conduct, which Cabo said was probative of dominance.

[266]MGA’s starting point was that on the basis of Mr Parker’s market definition, MGA’s market shares were not at any point high enough to support a finding of dominance: its highest monthly market share was 26.69% for October 2017. In addition, MGA contended that in order to establish dominance, it is necessary to show a high market share over a sustained period. If MGA’s market share during 2017 were to be calculated using rolling averages for the previous 12-month period, that would give even lower market share figures.

[267]Mr Parker had not applied a wholesale to retail time lag in his market share calculations. At the trial, however, MGA did not seriously dispute the necessity to take account of a time lag, although it said that the better view was that the time lag should be closer to a month on the basis of the factual evidence. On Mr Parker’s market definition that did not make any difference to the conclusion.

[268]As to the other relevant factors, MGA said that barriers to entry were low, and that there was countervailing buyer power from strong retailers. Mr Parker’s view was that being a “must stock” product added nothing to the market share assessment. MGA’s conduct was, in MGA’s submission, simply an example of relative bargaining power: that LOL Surprise was simply more attractive than Worldeez, such that retailers preferred to stock LOL Surprise when given a choice.

[269]In relation to the court’s request, after the trial, for the parties to provide estimated market shares for the extended Colley market described at §253 above, the parties agreed that a robust calculation would need evidence from the industry experts to identify the additional collectible toys to include in that market. As a proxy, BRG (instructed by MGA) added into Mr Colley’s market all toys in brands identified by Ms Munt as “surprise collectible brands”, whether or not the specific toys had a surprise element or not, and also added all toys in the playset dolls and accessories subsegment with a name indicating a surprise element (i.e. products containing the words “mystery”, “surprise” or “blind” in their names).

[270]Cabo objected that these additions were too wide, because they included toys which did not have any surprise element, as well as toys falling under the playset doll accessories subclass which were therefore categorised as accessories rather than dolls. Cornerstone Research (instructed by Cabo) therefore provided an alternative calculation that was limited to the addition of toys in the playset dolls and collectibles class, with product names indicating a surprise element.

[271]The two sets of market share figures produced on these bases were as follows: 275. Table 4: 2017 market shares for extended Colley market (%) Jan-17 Feb-17 Mar-17 Apr-17 May-17 Jun-17 Jul-17 Aug-17 Sep-17 Oct-17 Nov-17 Dec-17 Ext Colley mkt (Cabo) 29.7 34.5 33.4 37.4 37.0 39.4 48.1 53.4 55.9 56.2 64.9 51.5 Ext Colley mkt (MGA) 20.52 23.08 22.33 26.41 28.51 29.78 37.55 41.98 45.86 47.77 56.77 43.58

Market shares

[272]For the reasons set out above, the markets which I consider most relevant for the assessment of market shares are(i) Mr Colley’s defined market, and

(ii) the extended Colley market, which should be considered in order to address the concerns about the boundary between products with a single element of surprise and products with multiple surprise elements. A precise analysis of the extended Colley market cannot be undertaken without further toy industry expert evidence, but both parties have provided indicative figures based on the inclusion of products with names indicating a surprise element.

[273]I consider that Cabo’s additional product set provides the closest proxy for the extended Colley market. MGA’s inclusion of all toys in brands identified as being “surprise collectible brands”, whether or not they had any surprise elements, is a significant departure from the characteristics identified by Mr Colley which I have found to be an appropriate starting point for the market definition. This materially influences the result: as Cabo pointed out, MGA’s calculation includes all products in very popular brands such as Shopkins whose bestselling product had no surprise element. There is also no compelling justification for including accessories, rather than limiting the additional products to toys in the playset dolls and collectibles class. While Mr Colley’s original sensitivity analysis (row (iii) of the table at §261 above) included for illustrative purposes all Shopkins toys in the playset dolls and collectibles class, and all Num Noms products including those in the playset doll accessories class, that does not mean that the alternative sensitivity analysis requested by the court, based on the extended Colley market defined in §253 above, should include products whose competitive influence on LOL is far more remote than the products falling at the boundary of or just outside of Mr Colley’s defined market.

[274]As for the relevant period for the assessment of market share figures, the focus of the abuse alleged by Cabo was the period around the end of May 2017. Cabo also relied on MGA’s continuing contacts with toy retailers regarding the sale of Worldeez in the subsequent months of 2017, in particular the contacts with TRU in June and July 2017, and with The Entertainer in September 2017. The question is then whether market shares should be assessed on the basis of the figures relating to that period, appropriately adjusted for the wholesale to retail time lag, or whether (as MGA contended) the calculation of market shares should use rolling averages for the previous 12-month period.

[275]That question must be answered by considering the features of the market in issue. The factual and expert evidence describes a market characterised by considerable volatility, in which there are rapid changes in trends and consumer preferences, leading to very short product lifecycles and continuous product innovation. Those characteristics manifest themselves in immediate shifts in patterns of demand at the retail and therefore also wholesale level. The very rapid and extraordinary success of LOL Surprise is a paradigm example of that.

[276]In a market exhibiting those characteristics, and in the context of an inquiry into whether MGA had market power in relation to its supply of LOL Surprise during May 2017 onwards, it would not be meaningful to base the market share assessment on a period before LOL Surprise had entered the market. Indeed, when cross-examined on this point, Mr Parker was not able to offer any convincing explanation as to why MGA’s market share during a period prior to the market entry of LOL Surprise would provide useful evidence of MGA’s market power in May 2017 onwards when LOL Surprise had launched in the UK with phenomenal success.

[277]I do not, therefore, consider that it is appropriate to use 12-month rolling average market share figures. Rather, the focus should be on the evidence of MGA’s market shares during the period from May 2017 to September 2017. In order to derive the wholesale market shares for that period from the retail market share figures, it is necessary to adjust for the time lag between placing orders and delivery of supplies (and as set out above, MGA ultimately did not dispute that as a matter of principle).

[278]The toy experts agreed that the typical time lag was around two months for domestic orders (i.e. the orders delivered by domestic distributors from warehouses in the UK), and at least four months for FOB orders (delivered directly from the manufacturers, typically in the Far East, without passing through an intermediary domestic distributor). That does not mean that every order placed for LOL Surprise was delivered in that timescale: MGA have identified some orders which suggested a shorter time lag. There has not, however, been any granular analysis of the order periods for LOL Surprise, and the toy experts did not suggest that any such analysis might yield results materially different from their agreed conclusion.

[279]I therefore accept the toy experts’ evidence as the best available evidence for what is inevitably a broad-brush adjustment to extrapolate from the retail to the wholesale market. The wholesale market share figures should therefore be assessed by applying a two to fourth-month time lag to the available retail market share figures. On that basis it is relevant to look at the retail market share figures from July 2017 to early 2018.

[280]MGA’s market share during that period, on Mr Colley’s market definition as well as the two sensitivity analyses produced by Mr Colley/Cornerstone Research, was well above 40%, and therefore in a range indicative of dominance. Indeed for all but the post-trial market share figures for the extended Colley market, the market shares were well above 50%, in a range indicating a presumption of dominance: i) Mr Colley’s market share figures produced a range of 56.28% to 58.43% for July to December 2017. ii) Mr Parker’s recalculation of the market shares for Mr Colley’s market produced a range of 54.09% to 66.80% for the same period. I note that there was a detailed debate between Mr Colley and Mr Parker about the methodology used by Mr Parker to derive these figures. Ultimately however the difference in the figures was limited, and Mr Parker’s recalculation showed precisely the same as Mr Colley’s figures, namely that MGA’s retail market share was above 50% from July 2017 onwards. iii) Mr Colley’s sensitivity analysis based on the inclusion in his market of Shopkins and Num Noms products, including those without any element of surprise, produced a range of 53.07% to 59.04% for that period. iv) Cabo’s version of the extended Colley market, using a product set which I consider to be the best available proxy, produced a range of 48.1% to 51.5% for that period. v) Even MGA’s version of the extended Colley market, which I consider to be overly broad, produced a range of 41.98% to 43.58% from August to December 2017, with the July 2017 figure given as 37.55%. That would indicate dominance using a time lag of three months, which would fall within the middle of the appropriate range (two to four months) identified by the toy experts. vi) There was no dispute that MGA’s market share was even higher during 2018, on any basis.

[281]I agree with Mr Parker’s comments that a “flash in the pan” spike in an undertaking’s market share should be regarded with some caution, and would not necessarily be a reliable indicator of market power. The case-law likewise cautions against placing reliance on large market shares which may turn out to be ephemeral (see e.g.

Case T-79/12

Cisco v Commission EU:T:2013:635, §69). In the present case, particularly when considered in the context of the characteristics of the toy market, MGA’s market share during 2017 was not ephemeral, nor is there any evidence that it was perceived as such by retailers: i) MGA’s market share was rising consistently throughout 2017, with a particularly dramatic rise in the retail market from around June 2017, translating to around April 2017 in the wholesale market. ii) That was no doubt because by spring 2017 the UK retailers would have been well aware of the phenomenal success of the product in the US, as well as the commercial success which it had already enjoyed in the UK. The retailers’ expectation, consistent with the market performance by then, was that the product would remain extremely popular, at least for the remainder of 2017. That was reflected in their orders in spring 2017 for substantial stocks for sale in summer/autumn 2017. iii) The evidence of MGA’s retail market shares during 2018 also shows that MGA’s wholesale market share continued to rise dramatically throughout the period of the alleged abuse.

[282]Finally, MGA’s market shares were very considerably higher than those of its closest competitors. Mr Colley provided the following diagram illustrating MGA’s and competitor market shares during 2017, based on his primary market definition. Again, as discussed above, the relevant period to consider is the period from July 2017 onwards.

Figure 1: Colley diagram of 2017 MGA and competitor market shares

[283]Mr Parker’s recalculation of the market shares on Mr Colley’s market similarly showed that the closest competitors to MGA during the period July to December 2017 were Flair Preziosi, the UK distributor of Shopkins, whose highest market share for the same period was 23.15% (July 2017), and Spin Master (the supplier of Hatchimals Colleggtibles), whose highest market share during that period was 25.76% (November 2017). Cabo’s analysis of the extended Colley market likewise showed Shopkins and Hatchimals Colleggtibles as the closest competitors, with Shopkins’ highest market share in that period being 14.0% (July 2017), and Hatchimals Colleggtibles’ highest market share being 16.1% (December 2017).

Competition from products outside the relevant market

[284]The next question is whether there is evidence of significant competitive constraints on LOL Surprise from products outside the markets used for the market share analysis. Leaving aside the submissions on the materiality of the distinction between products with a single and multiple layers of surprise, which is addressed with the extended Colley market analysis, Mr Parker contended that Barbie in particular competitively constrained LOL Surprise. The evidence on this point was, however, largely discredited at the trial. Mr Parker attempted to address this in his brand-by-brand revenue analysis in his eighth and ninth reports, which I have described above at §215. That listed Barbie as being a close substitute for LOL Surprise, at least during 2018. Mr Parker eventually conceded, however, that this analysis was unreliable (see §§240–241).

[285]Mr Parker also relied on the evidence of MGA’s toy expert Mr Harper to the effect that Barbie and LOL Surprise were competitors. Mr Harper’s assessment was, however, based on little more than the observation that the two products appeal to the same target audience, a point picked up in a 2018 internal MGA study. Ms Munt agreed that girls might own both brands and play with them together. She nevertheless took the view (in both her written and oral evidence) that collectibles such as LOL Surprise and fashion dolls such as Barbie have very different product features and play patterns, such that they should not be regarded as substitutable. Ms Munt’s reasoning was, in my judgment, more solidly based than that of Mr Harper.

[286]MGA also relied on internal MGA emails comparing sales of LOL Surprise with other brands, including Barbie, and a comment from Mr Larian “Let’s crush that bitch like we did in 2005 once and for all”. That reflects Mr Larian’s continuing rivalry with Mattel, but does not come close to showing that Barbie acted as an “immediate and effective” competitive constraint on LOL Surprise.

[287]MGA’s closing submissions eventually said little more than that Barbie and LOL Surprise “have a competitive interaction”. What MGA conspicuously did not say was that the closeness of the constraint was such that MGA could not be regarded as dominant in relation to the supply of LOL Surprise. This illustrates the importance of drawing a distinction between products that are close competitors of the focal product, and products for which there is some competitive interaction but which could not realistically be regarded as a sufficiently close substitute that they would constrain the market conduct of the supplier of the focal product. There was undoubtedly a degree of competitive interaction between LOL Surprise and Barbie, and indeed other products which were not in Mr Colley’s product market such as My Little Pony. That does not, however, mean that those products acted as effective competitive constraints on LOL Surprise: there was no evidence that they did so, or were likely to have done so by virtue of their characteristics.

Barriers to entry and expansion

[288]In assessing the relevance of barriers to entry to the assessment of MGA’s dominance in the present case, it is important to have regard to the relevant time period. The vigorously competitive nature of the toys market indicates that barriers to entry are not high, as such. It is common ground that the toys market is (as described above) characterised by continuous innovation and rapid product lifecycles. The question in the present case, however, is whether a new product could have entered the market (or expanded its existing market share) during the relevant time period so as to constrain the market power of MGA in relation the supply of LOL Surprise.

[289]It is clear that there was no realistic possibility of any serious competition to LOL Surprise emerging during the second half of 2017 in response to the success of LOL, bearing in mind the lead times for the development and launch of a new toy. Mr Larian said that the development period for a toy usually takes between 16–18 months. The toy experts’ evidence was that this normally takes around 18 months for large global companies and around 12 months for smaller companies. Those estimates are consistent with the time taken for the development and launch of Worldeez, from initial designs in early 2016 to the launch in the summer of 2017. For a new product to hit the shelves in late 2017, therefore, its development would have needed to start the previous year. There was no prospect of that happening overnight, to enable a new product to enter the market quickly enough to pose an effective competitive constraint to LOL Surprise during that period. Nor was there any plausible suggestion that LOL Surprise was in fact constrained by the potential for repackaging of existing products.

Countervailing buyer power

[290]MGA suggested, albeit somewhat half-heartedly, that retailers were able to exercise countervailing buyer power on the basis that they hold the power to space in their stores. I do not accept that submission: It is apparent that the key retailers held no such power in relation to MGA in the present case.

[291]It was clear from the toy expert evidence that the balance of bargaining power between manufacturers and toy retailers is fact sensitive and dependent on prevailing market trends, conditions of supply and demand, and the size of the manufacturers. In the present case LOL Surprise was, from spring 2017 onwards, a must-stock product, in the sense that the main toy retailers realistically had to stock it in order to avoid substantial diversion of sales and customers to their rivals. The case-law recognises this as a relevant factor when assessing dominance. In Socrates v Law Society the CAT commented at §123, in relation to the Law Society’s Conveyancing Quality Scheme, providing accreditation for solicitors engaged in residential conveyancing, that:

“By ‘must-have’ product we do not mean that the product was so essential that no residential conveyancing solicitors firm could operate outside the CQS: we use the term to describe the situation where for the majority of conveyancing solicitors there was little option but to seek CQS accreditation.”

[292]Similarly in its decision in Case AT.40134 AB InBev C(2019)3465 final, in relation to brands of beer, the Commission stated at §24 that:

“Some branded beer products are considered so important by consumers in a particular country in a particular point in time that a retailer or convenience store in that country considers it needs to put these particular products on the shop shelves to avoid a substantial loss of sales and/or customers.”

[293]The Commission then noted at §75 that:

“The three large retailers have stated that some of AB InBev’s products … are ‘Essential Products’. In other words, in their view they would lose significant sales and/or clients if they did not obtain them. This prevented these three large retailers from being able to strongly bargain on these products …”

[294]That is the sense in which the concept of a “must-stock” toy product is relevant in the present case. The “must-stock” label is in this sense something of a misnomer, because it is common ground that not every retailer will inevitably stock a “must-stock” product. Rather, what it signifies is a product with such consumer popularity that retailers are aware that if they do not stock it they are likely to lose significant revenue, as well as losing market share to other retailers. That puts them in a weak bargaining position in relation to supplies of that product. It should also be emphasised that “must-stock” status is not binary: within the range of very popular products that might be labelled “must-stock”, some will be even more popular than others, creating an even stronger imperative for retailers to stock them. Mr Harper therefore referred to a “spectrum of must-stock toys” at any given point in time, depending on how well particular products are selling. His view was that a toy retailer would have to take a meaningful percentage of the top sellers, even if it did not stock all of them.

[295]The consistent evidence was that LOL Surprise was, by May 2017, a “must-stock” product for UK toy retailers, and indeed at the top of the must-stock spectrum for toys. In other words, the demand for LOL Surprise was stronger than for any other comparable product. As set out at §77 above, it was well-known that the product had been an extraordinary success in the US market. The Forbes website reported in April 2017 that LOL surprise was the “hottest new craze in the collectible doll world” and had been ranked as the top selling doll since January. Since the US market is a good predictor of success in the UK, retailers in this country could and did expect LOL Surprise to go on to be successful here. Mr Harper said that LOL Surprise was “always destined for success” in the UK, given the social media interest in the US.

[296]That expectation proved well-founded: the product was an outstanding success in the UK market as soon as it launched in February 2017, selling out within weeks of its launch. By May 2017 it was the highest selling doll in the UK. Ms Munt’s evidence was that all retailers would have been “desperate to stock this red hot ‘must have’ range immediately”. Mr Harper agreed that LOL Surprise would have been considered by retailers as a must-stock item by May 2017, and that it was “probably at the top of the spectrum” because of the potential shown by its US sales.

[297]The toy experts’ assessment was consistent with the factual evidence: Stuart Grant said that by May 2017 LOL was “the number one brand in the industry”, and Mr Smyth said in his witness statement that he considered it to be a must-stock item when it was launched. The further problem for retailers was that the demand for LOL Surprise vastly outstripped supply, creating a scarcity situation. Stuart Grant explained:

“we were in a situation with LOL where … they couldn’t make enough. So it didn’t matter whether they gave us double our market share of allocation, I still needed four times our market share because the demand was just so high …”

[298]Stuart Grant agreed that these factors gave MGA greater bargaining power than the toy retailers, shifting the power dynamic that would otherwise have prevailed in a normal relationship where supply was matched with demand. That evidence was amply borne out by MGA’s conduct in the present case, as considered below.

[299]Mr Colley took account both of the “must-stock” status of LOL Surprise and the conditions of scarcity of supply in his assessment of MGA’s market power. Mr Parker disagreed, disputing that these factors were relevant indicators of market power. Mr Colley’s approach is, however, supported by the authorities and is sound as a matter of principle: the extent to which retailers regard a product as “must-stock”, and the conditions of supply and demand on the market, are factors which are relevant to the assessment of countervailing buyer power and, more generally, to the broader economic context which is a necessary part of the analysis of dominance.

MGA’s conduct

[300]The history of MGA’s threats (veiled and explicit) to the main UK toy retailers upon the launch of Worldeez is set out above. It is very clear that MGA was able to threaten to withhold supplies from retailers – in the case of The Entertainer, suspending sales worth around $2m – because it knew that the demand for LOL Surprise was such that any supplies withdrawn from one or more customers could be rapidly reallocated to rival retailers. MGA’s threats were, moreover, highly effective, as set out above and also summarised below at §§321–322.

[301]It is highly improbable that MGA would have felt able to behave in this way, but for its market power in the supply of LOL Surprise. Indeed it is quite apparent from (in particular) the exchange of emails with The Entertainer, between 23 and 31 May 2017, that The Entertainer’s decision not to stock the Worldeez globe was taken with great reluctance, following repeated objections from both Stuart and Gary Grant to MGA’s conduct. The Grants only capitulated because they were desperate to obtain supplies of LOL Surprise. MGA’s ability to compel The Entertainer, one of the leading toy retailers in the UK, to cease stocking a rival product, in the face of vehement opposition from The Entertainer, is (as Mr Colley noted) inherently demonstrative of its ability to act independently of competitors and customers. As the CAT commented in Genzyme (§259 above), the imposition by a supplier of terms that disregard the wishes of its customers is the “hallmark of dominance”.

[302]Mr Parker’s reports surprisingly took no account of The Entertainer’s capitulation (or the agreements of other retailers not to stock the Worldeez globe) in his assessment of dominance. That cannot have been because Mr Parker did not want to get drawn into commentary on the facts: on the contrary, he referred to and commented on selected parts of Mr Grant’s evidence regarding its negotiations with MGA. Notably, when cross-examined, Mr Parker agreed that evidence of insensitivity of an undertaking to the reactions of its customers was relevant to the assessment of dominance:

“Q. And therefore evidence of insensitivity is relevant to dominance, yes? A. I agree it is relevant, yes. Q. Now, that’s irrespective whether or not insensitivity is also relevant to the abuse in question. A. I understand that. Q. So MGA’s strongarming of retailers is evidence of MGA’s insensitivity to customer reactions and is therefore relevant to dominance. A. Correct.”

A. Correct.”

[303]Following the exchange above, however, Mr Parker went on to suggest that MGA’s conduct did not in fact necessarily indicate dominance; rather, it simply indicated that MGA/LOL was more important to retailers than Cabo/Worldeez. That was not an analysis previously offered in any of the (numerous) reports filed by Mr Parker in these proceedings. Nor do I accept it as a matter of principle. The fact that one toy product is more successful (and hence more important to retailers) than another product does not automatically give the supplier of the more successful product the ability to exclude the less successful product from the shelves of all of the main toy retailers by threatening to withhold supply. Absent dominance, that strategy would be unsuccessful, or would at least carry a high risk of failure, because retailers would be able to switch to other products. The fact that MGA was (in this case) well aware that it could profitably sustain that strategy, because retailers had no adequate substitute, is compelling evidence of its market power, which supports the other indications of dominance set out above.

[304]There was, moreover, no factual evidence of any similar strategy being adopted by a non-dominant toy supplier. Quite the contrary: Ms Munt said that during her 22 years of buying toys at Argos and Mothercare, she never had a supplier refuse, or threaten to refuse, to supply their toys to her on the basis of a stated belief that an item she had listed was a copy of one of their toys. Mr Mowbray also said that he was not aware of any other manufacturer refusing to supply their products because of a competitor product. The only evidence of any similar behaviour was a vague reference by Mr Grant in his cross-examination to “a situation exactly like that today with another manufacturer”. Mr Grant did not, however, provide any further details about this incident, so I do not consider that any weight can be placed on this comment.

Conclusions on dominance

[305]The evidence in the present case consistently and overwhelmingly supports a finding that MGA was dominant in relation to the supply of LOL Surprise. MGA’s market share was, during the relevant period for assessment, in a range indicative of dominance or even a presumption of dominance, both on Mr Colley’s market and by reference to a slightly wider market to include all collectible toys targeted at girls aged 6–9 years, with at least a single element of surprise. While barriers to entry in the market were not intrinsically high, the timescale for product development was such that there was no realistic possibility of serious competition to LOL Surprise emerging quickly enough to constrain MGA during the relevant period in 2017; nor was there evidence of a constraint arising from the potential for repackaging existing products. In addition, the “must-stock” nature of the product and the conditions of scarcity of supply of LOL Surprise reduced the potential for any countervailing buyer power, leading to a situation where MGA was able to act with complete disregard for the wishes of its major customers.The conclusion that MGA was dominant is, on the facts of this case, inescapable. Whether MGA’s conduct amounted to an abuse Legal principles

Whether MGA’s conduct amounted to an abuse

[306]Case 85/76 Hoffmann-La Roche v Commission EU:C:1979:36, §91 sets out what is now the established formulation of the concept of an abuse of a dominant position:

“The concept of abuse is an objective concept relating to the behaviour of an undertaking in a dominant position which is such as to influence the structure of a market where, as a result of the very presence of the undertaking in question, the degree of competition is weakened and which, through recourse to methods different from those which condition normal competition in products or services on the basis of the transactions of commercial operators, has the effect of hindering the maintenance of the degree of competition still existing in the market or the growth of that competition.”

[307]Specific examples of abuse are given in the Chapter II prohibition and Article 102 TFEU. The statutory examples are not, however, exhaustive and the courts have deprecated attempts to treat those examples as “pigeon-holes into which one must fit a case”: see e.g. Purple Parking v Heathrow Airport

[2011] EWHC 987 (Ch)

, §79. Rather, the case-law recognises that a range of practices may be regarded as abusive, where the effect is to impair competition on the market.

[308]In particular, a paradigm form of abuse is the exclusion of competitors by a dominant undertaking. In

Case C-307/18

Generics (UK) Ltd v CMA EU:C:2020:52 the CJEU referred at §§151–152 to conduct “intended to deprive parties demonstrated to be potential competitors of effective access to a market” as an example of an abuse of a dominant position, noting that even if the intention is to protect the commercial position of the dominant undertaking, that “does not justify resorting to practices that fall outside the scope of competition on the merits”.

[309]Where exclusionary abuse is alleged, it is in general necessary to demonstrate that the impugned conduct is capable of having exclusionary effects. In relation to certain types of conduct, however, an anticompetitive effect may be inferred. The Article 102 guidance gives at §22 various examples of conduct designed to exclude competitors, for which that is the case:

“If it appears that the conduct can only raise obstacles to competition and that it creates no efficiencies, its anti-competitive effect may be inferred. This could be the case, for instance, if the dominant undertaking prevents its customers from testing the products of competitors or provides financial incentives to its customers on condition that they do not test such products, or pays a distributor or a customer to delay the introduction of a competitor’s product.”

[310]In similar vein, the Commission’s draft guidelines on the application of Article 102 TFEU to abusive exclusionary conduct by dominant undertakings (draft Article 102 guidelines), published for consultation in August 2024, refer at §60(c) to “naked restrictions” in the following terms: “Naked restrictions: certain types of conduct by a dominant undertaking have no economic interest for that undertaking, other than that of restricting competition. These types of conduct are by their very nature capable of restricting competition. Only in very exceptional cases will a dominant undertaking be able to prove that in the specific circumstances of the case the conduct was not capable of having exclusionary effects. Examples of naked restrictions are:(i) payments by the dominant undertaking to customers that are conditional on the customers postponing or cancelling the launch of products that are based on products offered by the dominant undertaking’s competitors;

(ii) the dominant undertaking agreeing with its distributors that they will swap a competing product with its own under the threat of withdrawing discounts benefiting the distributors; or

(iii) the dominant undertaking actively dismantling the infrastructure used by a competitor. While it is in principle open to the dominant undertaking to seek to show that the naked restriction is justified on the basis of an objective justification, it is highly unlikely that such behaviour can be justified in this way”.

[311]In relation to the refusal to supply an existing customer which promotes the products of a competing brand, the court made clear in the United Brands case that an undertaking in a dominant position “cannot stop supplying a long standing customer who abides by regular commercial practice, if the orders placed by this customer are in no way out of the ordinary,” since such a refusal to sell would limit markets to the prejudice of consumers, and would risk eliminating a trading party from the relevant market (§§182–183). The court continued:

“189. Although it is true … that the fact that an undertaking is in a dominant position cannot disentitle it from protecting its own commercial interests if they are attacked, and that such an undertaking must be conceded the right to take such reasonable steps as it deems appropriate to protect its said interest, such behaviour cannot be countenanced if its actual purpose is to strengthen this dominant position and abuse it. 190. Even if the possibility of a counterattack is acceptable that attack must still be proportionate to the threat taking into account the economic strength of the undertakings confronting each other. 191. The sanction consisting of a refusal to supply by an undertaking in a dominant position was in excess of what might, if such a situation were to arise, reasonably be contemplated as a sanction for conduct similar to that for which UBC blamed Olesen. 192. In fact UBC could not be unaware of the fact that by acting in this way it would discourage its other ripener/distributors from supporting the advertising of other brand names and that the deterrent effect of the sanction imposed upon one of them would make its position of strength on the relevant market that much more effective. 193. Such a course of conduct amounts therefore to a serious interference with the independence of small and medium sized firms in their commercial relations with the undertaking in a dominant position and this independence implies the right to give preference to competitors’ goods.”

[312]A dominant undertaking is therefore entitled to compete on the merits to protect its commercial interests. It cannot, however, refuse to supply an existing customer whose conduct in promoting the products of a competitor is consistent with “regular commercial practice”. A similar point was made by the Commission in its decision in the Boosey & Hawkes case, [1987] OJ L286/36. After noting at §19 that it was well established that a refusal of supplies by a dominant producer to an established customer without objective justification may constitute an abuse, referring to the United Brands case, the Commission stated that:

“The injury to competition would be aggravated where (as is alleged here) the stated purpose of the action is indirectly to prevent the entry into the market of a potential competitor to the dominant producer. A dominant undertaking may always take reasonable steps to protect its commercial interests, but such measures must be fair and proportional to the threat. The fact that a customer of a dominant producer becomes associated with a competitor or a potential competitor of that manufacturer does not normally entitle the dominant producer to withdraw all supplies immediately or to take reprisals against that customer. There is no obligation placed on a dominant producer to subsidise competition to itself. In the case where a customer transfers its central activity to the promotion of a competing brand it may be that even a dominant producer is entitled to review its commercial relations with that customer and on giving adequate notice terminate any special relationship. However, the refusal of all supplies to GHH and RCH, and the other actions B&H has taken against them as part of its reaction to the perceived threat of BBI, would appear in the circumstances of the present case to go beyond the legitimate defence of B&H’s commercial interests.”

[313]Cabo also relied on the principle that an abuse of a dominant position may arise where a dominant undertaking brings litigation which cannot reasonably be considered as an attempt to establish its rights, and can therefore only serve to harass the other party, and where that litigation pursues a plan to eliminate competition:

Case T-111/96

ITT Promedia v Commission EU:T:1998:183, §30. In addition, Cabo referred to a line of decisions of the French Competition Authority on unjustified disparagement of a competitor’s products, including Decision No. 13-D-11 of 14 May 2013 Sanofi Aventis. For the reasons set out below, I do not consider it necessary apply either of these lines of authority in the present case.

The parties’ submissions

[314]Cabo’s pleaded case was that MGA’s conduct was an abuse of a dominant position in three respects:(i) MGA threatened to withdraw supply to its customers, in order to exclude Worldeez;

(ii) MGA’s threats to instigate legal proceedings were not a genuine assertion of MGA’s legal rights, but were solely aimed at stifling a competitor; and

(iii) MGA disparaged Cabo/Worldeez to toy retailers, in a way recognised by the French Competition Authority as constituting an abuse.

[315]At the hearing, and particularly in its closing submissions, Cabo maintained that each of the three pleaded elements of MGA’s conduct was abusive in its own right, but also said that MGA’s conduct amounted to an “overall anti-competitive campaign to exclude Worldeez from the market” which could be characterised as forming a single abuse or course of abusive conduct. Cabo’s contention was that MGA’s conduct was nakedly restrictive of competition, such that there was no requirement to go further and assess the effects of that conduct; but that in any event the anticompetitive effects were obvious.

[316]MGA’s response, as put in its closing submissions, was that(i) the threats to withdraw supply to customers who stocked Worldeez were a legitimate and proportionate response to a commercial attack, and that save for the cancellation of the initial orders placed by The Entertainer, MGA’s conduct did not cause the retailers not to stock Worldeez;

(ii) MGA’s threats of litigation did not amount to abuse on the basis of the criteria set out in the relevant case-law, in particular because there were reasonable grounds to support an allegation of passing off; and

(iii) the denigration or disparagement of a competitor is not a recognised category of abuse, and MGA’s conduct was in any event very different from the sort of conduct regarded as constituting an abuse in the French Sanofi Aventis decision.

The overall exclusionary campaign

[317]I consider that Cabo is correct to describe MGA’s conduct as, taken together, an overall campaign to exclude Worldeez from the market. While MGA in its submissions has sought to compartmentalise the different elements of the conduct referred to by Cabo as separate types of alleged abusive conduct, I do not think that such compartmentalisation is necessary or indeed appropriate. On the basis of the facts described above the threats of litigation and disparagement of Worldeez as a “knock off” should not be seen as freestanding aspects of MGA’s conduct, but as an integral part of MGA’s threats to withdraw supply of LOL Surprise. Taken as a whole, MGA’s conduct amounted to a strategy to ensure that the Worldeez globe was not sold in any of the major toy retailers in the UK.

[318]It is apparent that within minutes of receiving Mr Laughton’s first email forwarding the Tiana Worldeez unboxing video, Mr Larian embarked upon a campaign designed to exclude the Worldeez globe from the market, by making clear that he would withhold supplies of LOL Surprise from any retailers who stocked the globe. Those threats were accompanied by repeated claims that the Worldeez product was a “knock off” of LOL Surprise and infringed MGA’s intellectual property rights including (in particular) patent rights, and that MGA was going to take, or was in the course of taking, action to protect its rights.

[319]MGA thereby used its complaints of copying and its purported legal action against the suppliers of Worldeez as the purported justification for its threats of refusal to supply LOL Surprise, and as a means of reinforcing those threats. That can be seen in Mr Larian’s communications to the toy retailers which repeatedly made claims of infringement of MGA’s intellectual property rights, referring in particular to his patent claims, as the explanation for and basis of MGA’s refusal to supply policy. It was also evidently the reason why Mr Larian forwarded MGA’s cease and desist letter to the Grants. (I return to this point, specifically in the context of the patent infringement threats, at §§485–491 below.)

[320]The deterrent effect of MGA’s conduct was also reinforced by the fact that the retailers were well aware that MGA’s refusal to supply policy was being applied across the board, to any retailer that stocked the Worldeez globe, not least because they were told that expressly by MGA (see the emails at §§122, 125 and 137 above).

[321]As described above, MGA’s exclusionary strategy was highly effective: Gary and Stuart Grant at The Entertainer, who had been supportive of the Worldeez product from the outset and which had already placed an order, capitulated within a day and agreed to cancel their order for the globe. The Entertainer ultimately only stocked the 5- and 10-packs, and the blind bags which were sold there in place of the globe. TRU, which had held initial discussions with Cabo, did not go on to stock any of the Worldeez range. Smyths, for which the discussions had progressed to indications of initial order quantities, likewise did not ultimately stock any of the range. The Worldeez globe, which was the anchor product of the Worldeez range, was thus completely excluded from the stores of the three main toy retailers in the UK.

[322]It appears that Mr Laughton also procured agreements from (at least) Argos and Tesco that they would not stock the Worldeez globe. While it is unclear whether they would otherwise have stocked Worldeez, it is apparent that MGA sought to ensure that the globe was excluded not only from the main toy retailers but also from other more general retail outlets. Again, neither retailer went on to stock any of the Worldeez range. It is also evident from the email exchanges between Mr Larian and Mr Laughton that Mr Larian brooked no exception whatsoever to his policy: see for example Mr Larian’s email to TRU on 24 May 2017 stating that “We will stop shipping original LOL to any retailer who supports knock off”, and his subsequent instruction to Mr Laughton “They buy this knock off and we will stop shipping. No exceptions” (§§137–138 above).

[323]Notably, no threats were made to B&M at the time when the Worldeez globe was launched there, presumably because B&M (unlike the other retailers which MGA had contacted) was a discounter which was not, at the time, stocking LOL Surprise. MGA could not, therefore, exert any leverage over B&M by threatening to withhold LOL.

[324]MGA’s conduct was clearly motivated by a desire to eliminate the competitive threat posed by Worldeez. Mr Laughton’s comment to Mr Larian on 1 July 2017 “really don’t want them in my market” (§160 above) speaks for itself. Mr Laughton’s email to his colleague Ms Rubio, noting that Worldeez was only being sold in B&M, and commenting that as MGA was the leading brand in they UK, MGA was “making it extremely hard for [Worldeez] to try and get real estate” (§167) is also revealing: it shows that Mr Laughton was well aware that MGA was using its market power to try to exclude Worldeez from the market.

[325]I do not accept MGA’s submission that it is relevant to the analysis to consider how successful Worldeez would have been but for MGA’s conduct, assessing on a trader by trader basis what orders would have been placed in the counterfactual. That would be to conflate the analysis of liability under the Chapter II prohibition/Article 102 TFEU with the issue of quantum. For the purposes of considering whether MGA’s conduct was abusive, it is sufficient to show that its conduct was capable of hindering the maintenance or the development of competition in the relevant market.As set out at in the Commission’s Article 102 guidance and draft Article 102 guidelines, cited at §§309–310 above, an anticompetitive effect can be inferred where the dominant undertaking’s conduct can only raise obstacles to competition and creates no efficiencies, such as an inducement to a customer to delay the introduction of a competitor’s product.

[326]In the present case, MGA’s threats to withdraw supply, reinforced by the accompanying threats of litigation and claims that Worldeez was a “knock off”, were obvious and undeniable obstacles to competition, and precisely the sort of exclusionary conduct which has been consistently regarded, in the case-law set out above, as an abuse of a dominant position. MGA went far beyond simply attempting to delay the introduction of Worldeez products; rather, MGA completely prohibited its customers from stocking the Worldeez globe, if those customers wished to retain any supplies of LOL Surprise. It is clear that such conduct falls outside the scope of competition on the merits, and that its anticompetitive effect can be inferred. MGA notably did not seek to suggest that its conduct was not capable of restricting competition. That is not surprising: as set out above, the evidence makes clear that MGA’s conduct did indeed have exactly the exclusionary effect that MGA intended.

[327]It is therefore not necessary to consider whether MGA’s threats of litigation and disparagement of the Worldeez product might also be regarded as separate abusive conduct. The better analysis is that that conduct was combined with the threats of withdrawal of supply to constitute an overall strategy of excluding Worldeez from the market.

MGA’s “response to commercial attack” argument

[328]Ultimately, MGA’s only real defence to the abuse claim, in relation to its threats to withdraw supply, was its contention that MGA’s conduct was a legitimate and proportionate response to an attack on its commercial position. MGA had pleaded a passing off defence (which is addressed below). Its commercial attack argument did not, however, rely on a claim that Worldeez was passing off LOL Surprise. Rather, the way in which MGA put its case in its closing submissions was to say that even if there was no passing off as a matter of law, there was a commercial attack which involved “unfair or objectionable conduct, to which MGA legitimately and obviously wished to respond”, since:(i) MGA had expended time and effort in the creation of LOL Surprise;

(ii) the external appearance of the Worldeez globe “reasonably appeared in May 2017 to be a copy of LOL Surprise”;

(iii) it was likely that any sales made by Cabo would be because of the globe’s similarity to LOL Surprise. That was, in MGA’s submission, the sort of attack on its commercial interests which entitled it to take action.

[329]MGA’s argument therefore appeared in substance to be that it was objectively justified in threatening to withdraw supply, because it reasonably thought that the original packaging of the Worldeez globe was a copy of LOL Surprise and therefore objectionable in some (undefined) sense, even if could not bring claim in passing off. That argument is hopeless and I unhesitatingly reject it.

[330]First, it is well-established that it is entirely legitimate for a product to look like another trader’s product, so long as it is not passed off as a product of that other person (and provided, of course, that no other intellectual property rights are infringed – but MGA does not rely on any others). As Jacob J stated trenchantly in Hodgkinson & Corby v Wards Mobility

[1994] 1 WLR 1564

, at 1569–70:

“There is no tort of copying. There is no tort of taking a man’s market or customers. Neither the market nor the customers are the plaintiff’s to own. There is no tort of making use of another’s goodwill as such. There is no tort of competition. … At the heart of passing off lies deception or its likelihood, deception of the ultimate consumer in particular. … Never has the tort shown even a slight tendency to stray beyond cases of deception. Were it to do so it would enter the field of honest competition, declared unlawful for some reason other than deceptiveness. Why there should be any such reason I cannot imagine. It would serve only to stifle competition.”

[331]Absent passing off, therefore, there was nothing objectionable or unfair about the presence on the market of the Worldeez globe: it was normal and healthy competition on the merits. Cabo was entitled to market and promote the product, and the key retailers were entitled to stock it and advertise it to their customers. Doing so was (to use the language of United Brands)“regular commercial practice”: it was a legitimate and normal course of conduct for the retailers to stock the Worldeez range, if they chose to do so, alongside any of the other collectible toys in their stores.

[332]MGA was entitled to respond to that by competing on the merits, for example by offering a superior or more innovative product, or by implementing a better marketing campaign. What MGA was not entitled to do was to stifle legitimate competition by exercising its market power to cut off the main retail outlets for the globe. The fact that MGA had invested time and effort in the creation of LOL Surprise makes no difference to that analysis: the fact that an innovative product may, sooner or later, meet with competition from rivals is an entirely normal feature of a competitive market.

[333]Secondly, I do not accept that MGA’s conduct was a response to a genuine belief that the Worldeez globe was a copy of LOL Surprise even in the original Worldeez packaging design. Mr Larian’s initial barrage of emails, including instructions to Mr Laughton to cut supplies of LOL Surprise to The Entertainer, was sent before he had seen any physical sample of the Worldeez globe. Only several hours later, after multiple emails sent to The Entertainer making clear that its supplies of LOL Surprise would be cancelled, did Mr Larian’s ask Mr Laughton for a sample of the product “to see if it’s really a knock off”. It was then not until the next day that Mr Larian actually saw the globe, at his meeting with Mr Sivner and Mr Lazarus at the Las Vegas toy fair. By then MGA had already extracted an agreement from The Entertainer that it would not stock the globe.

[334]It is, moreover, apparent that while Mr Larian repeatedly protested that the Worldeez globe was a “knock off” of LOL Surprise, he was using that phrase in a loose and pejorative sense, on the basis of a cursory and impressionistic assessment that the external packaging of the globe looked somewhat similar to that of LOL Surprise. It did not reflect any considered analysis of what was objectionable about the globe, and certainly was not based on any assessment of how the globe might have infringed any specific intellectual property rights held by MGA. That is clear from the panoply of accusations made by Mr Larian before seeing the globe or receiving any advice from his lawyers, ranging from threats of patent and copyright infringement to references to MGA’s “trade mark and trade dress”, and vague complaints about “palm[ing] off of MGA’s good will”.

[335]The cease and desist letter sent from MGA’s in-house lawyers on 23 May 2017 (see §106 above) likewise referred in vague terms to a patent pending, without identifying which patent; contended that Worldeez (in general, not limited to the globe or its external packaging) was likely to mislead or deceive consumers in a whole range of respects; and demanded that Cabo not only refrain from offering the product in packaging that resembled the LOL Surprise packaging, but also that it refrain from other unspecified infringements of MGA’s intellectual property rights in relation to LOL Surprise. As set out above, Cabo’s request that MGA identify the patent or patent application upon which it relied met with no answer.

[336]Even when MGA did involve external solicitors, the letter from Mishcon de Reya sent on 26 May 2017 (§114 above) was even less specific, simply alleging that “your Worldzee [sic] product … infringes our client’s intellectual property rights in its L.O.L. Surprise! Products” and requiring the destruction of all existing stock. The objections in that letter were (again) not limited to the globe or its external packaging; and the letter did not identify any specific intellectual property right alleged to have been infringed. The letter should never have been sent. It was oppressive, entirely unreasonable, and provides no basis for an assertion that MGA was entitled to object to the Worldeez globe.

[337]Indeed, given that MGA involved its lawyers so quickly after it became aware of Worldeez, if there was any suggestion that MGA genuinely believed that it had any grounds to object to Worldeez (or at least the Worldeez globe), one would have expected to see some evidence to that effect. There is, however, none; and in fact the evidence indicates the contrary. In the email exchange between Mr Laughton and Mr Larian on the day of the launch of the Worldeez globe at B&M (§159 above), Mr Laughton asked Mr Larian directly whether retailers could legally buy the globe. Mr Larian replied saying that he would “ask the lawyers” (suggesting, surprisingly in light of the letters and emails that had been sent, many by him, that he did not know the answer to that question). A few days later, on 6 July 2017, Mr Larian told Mr Laughton “Retailers who buy knock off: we can’t tell them not to buy these. But, they are black listed … and will not be allowed in MGA showrooms if they buy ANY LOL knock off.” When asked about this email in cross-examination, Mr Larian avoided engaging with the questions put to him. The point of the email was, however, clear on its face: Mr Larian evidently understood by then that he had no legal basis for objecting to the sale of Worldeez, such that his only leverage was to withhold supplies of LOL Surprise.

[338]In similar vein, when MGA’s Amazon account manager asked Mr Laughton on 7 July 2017 whether she should initiate Amazon’s take-down procedure in relation to Worldeez, which would require confirmation that MGA had a good faith belief that the listings violated its rights, Mr Laughton replied “No you can’t do this. It’s up to them if they want to run it at the moment”. Mr Laughton accepted that he said this because he had been told by Mr Larian that MGA had no right to prevent retailers from taking Worldeez.

[339]Thirdly, MGA’s submissions in this regard glossed over the fact that the packaging of the Worldeez globe was rapidly changed by Cabo before the globe was launched, but it is clear that MGA did not lift its objections to the product. That belies the suggestion that MGA’s conduct was genuinely a response to a commercial attack.

[340]MGA’s attempted justification of its conduct as a response to a commercial attack therefore fails both as a matter of law and on the facts. Ultimately, if MGA had a genuine belief that the Worldeez globe interfered with its rights on any proper legal basis, its remedy was to bring proceedings on that basis. A dominant firm is not entitled to simply circumvent the proper legal procedures (with the inherent uncertainty of litigation) by taking retaliatory anti-competitive action against a product on the basis of a claimed belief that that product’s similarity to its own product was unfair or objectionable.

MGA’s passing off defence

[341]MGA has not ever brought passing off proceedings against Cabo. Nevertheless, MGA’s pleaded defence relied heavily on claims that its conduct was justified on the basis that Worldeez was passing off LOL Surprise, or at least that it had reasonable grounds to believe that this was the case. By the time of closing submissions, however, MGA’s case on passing off was advanced with considerable diffidence, limited to the point that MGA’s complaints of passing off were a genuine attempt to establish its rights and were not “manifestly unfounded”. On that basis MGA contended that the test for abusive litigation set out in ITT Promedia was not met.

[342]Given my conclusion above that MGA’s threats of litigation should not be regarded as freestanding abusive conduct, but as part of MGA’s threats to withdraw supply, it is not necessary to consider the application of the ITT Promedia test. MGA’s passing off defence, pursued only in that context, therefore does not fall to be determined. Since, however, both parties commented extensively on the passing off issue in their written closing submissions, I will address it for completeness.

[343]The fundamental principle underlying the law of passing off (as summarised by Arnold LJ in Lidl v Tesco

[2024] EWCA Civ 262

, §28, is that “no person may misrepresent their goods or services to be those of another person”. Both parties relied on the formulation of the test set out by Lord Oliver in Reckitt & Colman v Borden (the “Jif Lemon” case)

[1990] 1 WLR 491

, at 499. That requires the claimant to establish(i) goodwill or reputation attached to the goods or services which they supply in the mind of the purchasing public, by association with the identifying “get-up”;

(ii) a misrepresentation by the defendant to the public (whether or not intentional) leading or likely to lead the public to believe that goods or services offered by them are the goods or services of the claimant; and

(iii) damage or the likelihood thereof.

[344]As to the misrepresentation required, the most common form of passing off involves a misrepresentation as to trade origin. That is the misrepresentation alleged by MGA in the present case. That form of misrepresentation does not require proof that the defendant’s goods will actually be mistaken for the claimant’s goods; it is sufficient if customers are led to believe that the claimant is in some way responsible for the defendant’s goods, or that they share a common manufacturer: Glaxo Wellcome v Sandoz

[2019] EWHC 2545 (Ch)

, §163.

[345]Mere confusion or a likelihood of confusion is not the test, and it is not enough if members of the public are “merely caused to wonder” if there is a connection. Rather, passing off requires deception, or a likelihood of deception: Glaxo Wellcome v Sandoz, §159. Furthermore, it is not sufficient that careless consumers may be misled. The deception must be such that ordinary consumers who purchase with ordinary caution and who know what is fairly common to the trade are likely to be misled: Lidl v Tesco, §35.

[346]MGA asserted that there was a misrepresentation leading to, or likely to lead to, the public believing that Worldeez was either LOL Surprise or part of the same range as LOL Surprise. I do not accept that submission. Images of LOL Surprise next to the original design of the Worldeez globe are shown below. While they are both in spherical packaging with a blue background and an image of a doll/figurine, those are the extent of the similarities. The sizes of the products are, as shown below, markedly different (LOL Surprise being 10cm in diameter, compared with 6.5cm for Worldeez). The brand names are prominent and are completely different. The logos of the brands are also entirely dissimilar, as are the colours of the design other than the background colour.

[347]Moreover, to the extent that there was any similarity at all between the background colour of the packaging originally designed by Worldeez and the fact that the design incorporated an image of the “Bella” figurine, those features were both changed on the package redesign, which took place before any of the globes were marketed to the public. The redesign replaced the blue background with white, and removed the image of the “Bella” figurine (see §§155–156 above). The only remaining similarity was the spherical shape. But there is nothing distinctive about the spherical shape of the packaging of a collectible toy: numerous other toys used spherical or similar capsules, as shown in the following illustration provided by Cabo (the blue ball being a generic vending machine ball):

[348]That is presumably why MGA’s comments about the external appearance of the Worldeez globe were, in its closing submissions, confined to the position as it stood under the original packaging design in May 2017. MGA did not put forward any basis on which the redesigned packaging could have been regarded as a misrepresentation for the purposes of a passing off claim.

[349]It is also necessary to take into account the fact that the toy market is characterised by high levels of product awareness. Consumers know what brands they want, and go into stores to buy that particular brand (or rather, more often, children select the particular toy which their parent or other adult then buys). As Mr Harper put it in one of his reports:

“Girls do not go into a store saying, ‘I want a small doll and I want to unbox it’. They get excited about a brand. Certain brands excite them and they are going to buy that brand …”

[350]Ms Munt’s evidence was likewise that sales of collectible toys like LOL Surprise are strongly driven by branding, and that brand awareness by children is very high. Unsurprisingly, in that context, the toy experts essentially agreed that customer confusion between LOL Surprise and Worldeez was unlikely. Mr Harper said that although it was possible that consumer confusion could arise, there were many features that were very different between the brands, such that confusion was “probably unlikely”. Ms Munt’s evidence was that it was “highly unlikely” that customers might be confused, given the “vastly different” product sizes, brands, themes, packaging design/colours and imagery, and brand advertising (among other differences).

[351]MGA contended that the court should give weight to the fact that (it said) the Cabo founders had the intention to deceive in the presentation of the Worldeez globe, or at least knew of the risk and proceeded recklessly. I do not accept that submission. There is no doubt that the Cabo founders looked at the packaging of LOL Surprise and used that, among many other things, as inspiration for their packaging and marketing ideas; and that they were optimistic about the success of a new mystery surprise toy given the previous success of products such as LOL Surprise and Hatchimals Colleggtibles. The contemporaneous documents show, however, that they were well aware that they should not copy LOL Surprise and were endeavouring not to do so.

[352]There is certainly no evidence indicating that the design of the LOL Surprise packaging was copied in order to free-ride on LOL’s success. MGA referred to comments from the Cabo founders noting that the globe “will be packaged in a similar kind of way” to LOL Surprise, and discussing adopting some of the technical solutions used by LOL (e.g. the two layers of shrink wrap). That is, however, a long way from showing that Worldeez intentionally copied the design of the LOL Surprise packaging. Likewise, while Cabo apparently wanted the style of its television advertisement to be similar to that of the LOL advertisement, the content of the advertisements were entirely different, as Mr Harper accepted.

[353]MGA relied on factual evidence from third parties which it said was relevant to establishing misrepresentation. In fact, most of the factual evidence referred to by MGA went no higher than comments on similarity. Mr Lazarus said that he thought the design of the packaging looked similar to LOL Surprise, and there are comments to that effect from him in contemporaneous WhatsApp chats. Mr Hunter of AB Gee thought that the globe was “somewhat similar” to LOL Surprise. Mr Brocklehurst at TRU commented that the globe was “very close to LOL”. One of MGA’s distributors observed that the Worldeez globe was “very similar in the look and concept to LOL”; and Funtastic, a potential distributor for Worldeez, said that the products were “very similar”. MGA also referred to press reports referring to the similarity of the products. None of this suggests a risk that consumers would or might be misled: it merely shows that the two products were perceived as being similar, which is far short of what is required to establish passing off.

[354]The only evidence of confusion, or a risk of confusion, related to the trade origin of Worldeez was scant. There was some conflicting evidence about Mr Smyth’s reaction when he first saw the Worldeez product at the meeting described at §98 above. I do not consider any of that evidence to be reliable: Mr Smyth’s account was not tested by oral evidence at the trial; Mr Michaelson disputed Mr Smyth’s account, but I do not consider him to be a reliable witness; and Mr Lazarus was not at the meeting and gave only a second-hand account of what he had been told by Mr Sivner. MGA also relied on a WhatsApp exchange between the Cabo founders and Mr Sivner and Mr Lazarus on 24 May 2017, shortly after MGA’s initial reaction to Worldeez, in which Mr Lazarus said that he thought that the average person would be confused by the appearance of the globe, noting “Looks confusing to punter. Association with lol”. He said that he had been told that by Mr Beardall at Spin Master. Those were vague and general comments, in the course of a (rather heated) WhatsApp exchange in the wake of MGA’s initial threats, before the packaging of the globe was redesigned to remove the elements that might have been regarded as similar to the packaging of LOL Surprise.

[355]Finally, I note that there is no evidence that any actual consumers were deceived or misled by the appearance of Worldeez into thinking that the product was in some way associated with LOL Surprise.

[356]Considering all of the evidence set out above, in my judgment there was no misrepresentation capable of founding a claim in passing off. Having regard to the clear differences between the products, the evidence of the experts, and the minimal factual evidence of any potential confusion, I have formed the clear view that even in its original packaging the Worldeez globe was not likely to mislead consumers as to the origin of the product. That conclusion is even stronger in relation to the redesigned packaging, which removed the only real aspects of similarity which had previously existed.

Conclusion on abuse of dominance

[357]MGA’s conduct is, in my judgment, very clearly an abuse of its dominant position, contrary to the Chapter II prohibition and Article 102 TFEU, in the form of an exclusionary campaign based on threats to withdraw supply of LOL Surprise to existing customers, accompanied and reinforced by threats of litigation and disparaging claims that Worldeez was a “knock off” of LOL Surprise. I do not accept that MGA’s conduct was a response to a genuine belief that the Worldeez globe was a copy of LOL Surprise, but in any event even if that had been established that would not have justified MGA in implementing conduct which was designed to, and did in fact, eliminate Cabo from a substantial part of the market. Unlawful agreements claim Overview of the legal framework

[358]In light of my conclusion on abuse of dominance, it is not strictly necessary to consider the separate allegation of unlawful agreements. I will, however, address this issue since it was fully argued at the trial, and in case my conclusion above on abuse of dominance is wrong.

[359]As with the abuse of dominance claim, Cabo’s claim that MGA’s conduct amounted to unlawful agreements is brought under both the Chapter I prohibition and Article 101 TFEU; and again no material difference is identified in the scope of these provisions.

[360]The Chapter I prohibition is set out in s. 2 of the 1998 Act. Section 2(1) provides that agreements between undertakings, decisions by associations of undertakings or concerted practices which may affect trade within the UK, and which have as their object or effect the prevention, restriction or distortion of competition within the UK, are prohibited unless they are exempt under the relevant provisions of the Act.

[361]Section 2(2) specifies that ss. (1) applies in particular to agreements, decisions or practices which (among other things) “(b) limit or control production, markets, technical development or investment” and “(e) make the conclusion of contracts subject to acceptance by the other parties of supplementary obligations which, by their nature or according to commercial usage, have no connection with the subject of such contracts”.

[362]Section 9 provides for the exemption of certain agreements from the Chapter I prohibition, as follows:

“(1) An agreement is exempt from the Chapter I prohibition if it – (a) contributes to – (i) improving production or distribution, or (ii) promoting technical or economic progress, while allowing consumers a fair share of the resulting benefit; and (b) does not – (i) impose on the undertakings concerned restrictions which are not indispensable to the attainment of those objectives; or (ii) afford the undertakings concerned the possibility of eliminating competition in respect of a substantial part of the products in question.” (a) contributes to – (i) improving production or distribution, or (ii) promoting technical or economic progress, while allowing consumers a fair share of the resulting benefit; and (b) does not – (i) impose on the undertakings concerned restrictions which are not indispensable to the attainment of those objectives; or (ii) afford the undertakings concerned the possibility of eliminating competition in respect of a substantial part of the products in question.”

[363]Article 101(1) TFEU sets out essentially the same prohibition as in s. 2 of the 1998 Act, save that (as with Article 102 TFEU) the requirement is for the conduct to affect trade between Member States. Article 101(3) TFEU corresponds to the exemption in s. 9 of the 1998 Act.

[364]In the version of the 1998 Act in force at the relevant time, s. 10 provided for a “parallel exemption” from the Chapter I prohibition if the relevant agreement was exempted from Article 101(1) by virtue of an EU Regulation. Those Regulations included the VBER. The main relevant provisions of that Regulation (in the version in force at the relevant time were the following:

“Article 1 Definitions 1. For the purposes of this Regulation, the following definitions shall apply: (a) ‘vertical agreement’ means an agreement or concerted practice entered into between two or more undertakings each of which operates, for the purposes of the agreement or the concerted practice, at a different level of the production or distribution chain, and relating to the conditions under which the parties may purchase, sell or resell certain goods or services; (b) ‘vertical restraint’ means a restriction of competition in a vertical agreement falling within the scope of Article 101(1) of the Treaty; … (d) ‘non-compete obligation’ means any direct or indirect obligation causing the buyer not to manufacture, purchase, sell or resell goods or services which compete with the contract goods or services, or any direct or indirect obligation on the buyer to purchase from the supplier or from another undertaking designated by the supplier more than 80% of the buyer’s total purchases of the contract goods or services and their substitutes on the relevant market, calculated on the basis of the value or, where such is standard industry practice, the volume of its purchases in the preceding calendar year; … Article 2 Exemption 1. Pursuant to Article 101(3) of the Treaty and subject to the provisions of this Regulation, it is hereby declared that Article 101(1) of the Treaty shall not apply to vertical agreements. This exemption shall apply to the extent that such agreements contain vertical restraints. … Article 3 Market share threshold 1. The exemption provided for in Article 2 shall apply on condition that the market share held by the supplier does not exceed 30% of the relevant market on which it sells the contract goods or services and the market share held by the buyer does not exceed 30% of the relevant market on which it purchases the contract goods or services. … Article 5 Excluded restrictions 1. The exemption provided for in Article 2 shall not apply to the following obligations contained in vertical agreements: (a) any direct or indirect non-compete obligation, the duration of which is indefinite or exceeds five years; … (c) any direct or indirect obligation causing the members of a selective distribution system not to sell the brands of particular competing suppliers. For the purposes of point (a) of the first subparagraph, a non-compete obligation which is tacitly renewable beyond a period of five years shall be deemed to have been concluded for an indefinite duration. … Article 7 Application of the market share threshold For the purposes of applying the market share thresholds provided for in Article 3 the following rules shall apply: (a) the market share of the supplier shall be calculated on the basis of market sales value data and the market share of the buyer shall be calculated on the basis of market purchase value data. If market sales value or market purchase value data are not available, estimates based on other reliable market information, including market sales and purchase volumes, may be used to establish the market share of the undertaking concerned; (b) the market shares shall be calculated on the basis of data relating to the preceding calendar year; … (e) if a market share is initially not more than 30% but subsequently rises above 35%, the exemption provided for in Article 2 shall continue to apply for one calendar year following the year in which the level of 35% was first exceeded; …” (a) ‘vertical agreement’ means an agreement or concerted practice entered into between two or more undertakings each of which operates, for the purposes of the agreement or the concerted practice, at a different level of the production or distribution chain, and relating to the conditions under which the parties may purchase, sell or resell certain goods or services; (b) ‘vertical restraint’ means a restriction of competition in a vertical agreement falling within the scope of Article 101(1) of the Treaty; … (d) ‘non-compete obligation’ means any direct or indirect obligation causing the buyer not to manufacture, purchase, sell or resell goods or services which compete with the contract goods or services, or any direct or indirect obligation on the buyer to purchase from the supplier or from another undertaking designated by the supplier more than 80% of the buyer’s total purchases of the contract goods or services and their substitutes on the relevant market, calculated on the basis of the value or, where such is standard industry practice, the volume of its purchases in the preceding calendar year; … Article 2 Exemption 1. Pursuant to Article 101(3) of the Treaty and subject to the provisions of this Regulation, it is hereby declared that Article 101(1) of the Treaty shall not apply to vertical agreements. This exemption shall apply to the extent that such agreements contain vertical restraints. … Article 3 Market share threshold 1. The exemption provided for in Article 2 shall apply on condition that the market share held by the supplier does not exceed 30% of the relevant market on which it sells the contract goods or services and the market share held by the buyer does not exceed 30% of the relevant market on which it purchases the contract goods or services. … Article 5 Excluded restrictions 1. The exemption provided for in Article 2 shall not apply to the following obligations contained in vertical agreements: (a) any direct or indirect non-compete obligation, the duration of which is indefinite or exceeds five years; … (c) any direct or indirect obligation causing the members of a selective distribution system not to sell the brands of particular competing suppliers. For the purposes of point (a) of the first subparagraph, a non-compete obligation which is tacitly renewable beyond a period of five years shall be deemed to have been concluded for an indefinite duration. … Article 7 Application of the market share threshold For the purposes of applying the market share thresholds provided for in Article 3 the following rules shall apply: (a) the market share of the supplier shall be calculated on the basis of market sales value data and the market share of the buyer shall be calculated on the basis of market purchase value data. If market sales value or market purchase value data are not available, estimates based on other reliable market information, including market sales and purchase volumes, may be used to establish the market share of the undertaking concerned; (b) the market shares shall be calculated on the basis of data relating to the preceding calendar year; … (e) if a market share is initially not more than 30% but subsequently rises above 35%, the exemption provided for in Article 2 shall continue to apply for one calendar year following the year in which the level of 35% was first exceeded; …”

[365]As with the abuse of dominance claim, the basic legal framework is not in issue in the present case. The issues in dispute are(i) whether Cabo has established that anticompetitive agreements were indeed concluded with any of the toy traders;

(ii) whether any such agreements had as their object or effect a restriction of competition;

(iii) if so, whether they were exempted by virtue of the VBER, and

(iv) if not exempted under the VBER, whether they were exempt pursuant to s. 9 of the 1998 Act and Article 101(3) TFEU.

[366]The focus of the debate on both sides was the VBER issue. The other issues were addressed only briefly in the parties’ closing submissions. Agreements with the toy traders The parties’ submissions

Agreements with the toy traders

[367]The dispute between the parties on this point was a purely factual one. Cabo’s case was that MGA entered into agreements with at least The Entertainer, TRU, Smyths, B&M and AB Gee that they would not stock Worldeez (or at least the Worldeez globe). Cabo also says that it is likely that MGA also entered into agreements with other retailers such as Argos and Tesco, but does not seek specific findings of infringements in relation to those.

[368]MGA’s position changed repeatedly. Its original defence admitted at least some of the agreements. Its final pleaded position was that it did not admit any agreements with The Entertainer and TRU, and denied the agreements with Smyths, B&M and AB Gee. By the time of MGA’s closing submissions, MGA admitted a “temporary” agreement with The Entertainer on 24 May 2017, but maintained that Cabo had failed to prove agreements or concerted practices with any of the other traders.

[369]Neither Cabo nor MGA contended that anything turned on the distinction between agreements and concerted practices, for the purposes of this case. I will therefore refer simply to agreements in the discussion below.

Discussion and conclusions

[370]The relevant communications between MGA and The Entertainer are set out at §§117–134 above. It is now common ground that an agreement was reached between MGA and The Entertainer on 24 May 2017, pursuant to which The Entertainer agreed to put its orders of Worldeez “on hold” in return for MGA reinstating supplies of LOL Surprise. MGA said that this was temporary and quickly superseded by an “independent decision” taken by the Grants not to stock the Worldeez globe, based in particular on Stuart Grant’s evidence on this point. I have already rejected Stuart Grant’s characterisation of that decision as being “independent” (§128 above). It is clear that The Entertainer’s decision was not an independent one but was taken as a result of coercion from MGA. I find, therefore, that The Entertainer’s initial agreement to put Worldeez orders on hold was replaced, on 31 May 2017, with an agreement not to stock the Worldeez globe, as a condition of continued supplies of LOL Surprise. It is apparent from the exchanges of emails in September 2017 and February 2018 that MGA’s position remained unchanged throughout that period.

[371]The relevant communications between MGA and TRU are set out at §§135–143 above. It appears from those communications that TRU agreed not to stock (at least) the Worldeez globe, in return for continued supplies of LOL Surprise. I do not accept MGA’s contention that TRU had formed its own “independent view” that the Worldeez globe was a “knock-off” of LOL. That claim was based on Mr Larian’s evidence of a purported conversation with Kevin Macnab of TRU. I have already rejected Mr Larian’s account of that conversation (§§139–140 above). As with The Entertainer, I consider that TRU’s decision was not independent, but was taken as a result of coercion from MGA. TRU did not in the event stock any of the Worldeez range. The evidence indicates, however, that MGA’s concern was with the Worldeez globe rather than with the 5- and 10-pack, and that the discussion between Mr Laughton and TRU (and other retailers) focused on the packaging of the globe. While Mr Larian’s email to TRU on 24 May 2017 was not expressly confined to a complaint about the globe, that needs to be read in the context of Mr Laughton’s previous discussions with TRU. There is, moreover, no suggestion in any other evidence that TRU agreed to refrain from supplying the 5- and 10- packs of Worldeez. I do not therefore find an agreement going beyond the globe.

[372]The relevant communications between MGA and Smyths are set out at §§144–147 above. As I have already found, it is likely that Smyths agreed not to stock the globe, in return for continued supplies of LOL Surprise. As with TRU, while Smyths did not ultimately stock any of the Worldeez range, I do not consider that the evidence is sufficient to support the existence of an agreement going beyond the sales of the Worldeez globe.

[373]As described at §§158–168 above, the Worldeez range was sold in B&M stores from July 2017 through to early 2018. That included not only the globe but also the 5- and 10-pack products. §§170–171 above set out the subsequent agreement in March 2018 between MGA and B&M for B&M to start receiving supplies of LOL, ostensibly on the basis that B&M would drop both Worldeez and Zuru’s 5 Surprise toy from its range. By that time, however, it is clear from the evidence that Worldeez had either been or was likely to be discontinued by B&M.

[374]I have already found that the evidence does not establish an agreement between MGA and AB Gee (§§151–153 above). Anticompetitive object or effect The parties’ submissions

Anticompetitive object or effect

[375]Cabo submitted that the agreements were anticompetitive by object, on the basis that their entire purpose was to exclude Cabo from the market. In the alternative, Cabo argued that the agreements had an anticompetitive effect. MGA submitted that Cabo had failed to prove that the agreements were restrictive either by object or effect, on the basis that the agreements did not fall within the category of object restrictions, and had no appreciable effect on competition in circumstances where the agreements were vertical in nature, and concerned only a single competing product for which there were other routes to market. MGA also relied on its argument that in Worldeez would have failed in any event in the counterfactual case.

Discussion and conclusions

[376]There can be no doubt that MGA’s agreements with The Entertainer, TRU and Smyths did indeed have as their object a restriction of competition. It is therefore not necessary to go further and consider whether there was also a restriction of competition by effect.

[377]According to the established case-law, the “essential legal criterion” for determining whether an agreement involves a restriction of competition “by object” is whether the agreement reveals in itself a sufficient degree of harm to competition that it is not necessary to assess its effects: Allianz Hungária EU:C:2013:160, §34;

Case C-67/13

P Groupement des Cartes Bancaires v Commission EU:C:2014:2204, §§57–58; and

Case C-228/18

Budapest Bank EU:C:2020:265, §37.

[378]In determining that question, it is necessary to have regard to the content of the agreement, its objectives, and the relevant economic and legal context. That context includes the nature of the goods or services affected, as well as the conditions of the functioning and structure of the market: Allianz Hungária, §36; Cartes Bancaires, §53; Budapest Bank, §51.

[379]Many cases found to be restrictions of competition by object involve horizontal agreements. But it is also possible for a vertical agreement to be found to restrict competition by object. A notable example is Allianz Hungária, which concerned agreements between car insurance companies and car repair shops concerning the hourly charge for car repairs. The court commented there (at §43) that: “While vertical agreements are, by their nature, often less damaging to competition than horizontal agreements, they can, nevertheless, in some cases, also have a particularly significant restrictive potential. The court has thus already held on several occasions that a vertical agreement had as its object the restriction of competition”.

[380]Although the parties’ intention is not a necessary factor in determining whether an agreement is restrictive by object, it is a factor that may be taken into account: Budapest Bank, §53.

[381]In the present case the transparent purpose of the agreements with The Entertainer, TRU and Smyths was, as I have already found, to exclude Worldeez from the market (see §317 above). MGA was well aware that those three retailers were the most important toy retailers in the UK, and that a successful launch of a new toy on the market would be very unlikely if the product was not stocked by those retailers. MGA ensured, through its coercion of all three of those retailers, that none of them sold the Worldeez globe. As I have already found, it was able to do so through the significant market power which it wielded, which meant that the key retailers had little choice but to accede to MGA’s demands, albeit in the case of The Entertainer very reluctantly and with considerable unhappiness.

[382]The consequence of the agreements was that the main and most important distribution channels were unavailable to the anchor product of the Worldeez range, which in turn significantly reduced the likelihood of a successful launch and market recognition for the range as a whole. Cabo’s only option was to launch the globe in B&M, a discount retail chain which did not (at least at that time) have the market recognition of the leading toy retailers. MGA was, of course, well aware of this, as evidenced by Mr Laughton’s comment to a colleague in August 2017 that the Worldeez globe was “Only selling in a discounter in the UK (B&M Bargains)” and that MGA was “making it extremely hard for [Worldeez] to try and get real estate” (§167 above).

[383]While the agreements were vertical in their nature, as noted above the retailers were told by MGA that its policy regarding Worldeez was being applied to all other retailers who stocked LOL Surprise. Mr Larian also told The Entertainer in terms that all retailers in the UK had confirmed that they were not going to buy the Worldeez globe (§125 above). Whether or not the other retailers were told the same by MGA, it is clear that the retailers monitored the sales of Worldeez in other stores, and were aware of where it was and was not being sold (see the emails at §§122, 123, 130 and 159 above). Each of The Entertainer, TRU and Smyths would therefore have known that the others among them were stocking LOL Surprise and would therefore not be stocking the Worldeez globe. The agreements were thus cumulatively reinforcing at a horizontal level.

[384]As I have also found, MGA’s conduct was explicitly motivated by a desire to eliminate the competitive threat posed by Worldeez (§324 above). For the reasons already set out (§§333–337 above), I do not accept that MGA genuinely believed that the Worldeez globe was passing off LOL Surprise, such that it was trying to protect its rights in that regard.

[385]MGA referred to various decisions concerning single branding agreements (such as the beer supply agreements at issue in

Case C-234/89

Delimitis EU:C:1991:91) which were not found to constitute restrictions by object. Those decisions concerned very different types of agreements in very different markets, and provide very little guidance for the assessment of the agreements at issue in the present case. The same applies to the example of a customer in a purchase agreement requiring the supplier not to sell to a particular competing customer, at issue in Seafood v My Fish

[2017] EWHC 766 (Ch)

, also relied upon by MGA. Norris J found in the latter case that that there was no evidence that, having regard to the functioning and structure of the relevant market, competition in that market was significantly weakened (§48). In the present case, however, for the reasons set out above, the agreements were obviously and intentionally significantly restrictive in nature, by completely excluding the Worldeez globe, which was the anchor product of the range, from the main distribution channels in the UK.

[386]The agreement with B&M in March 2018 was rather different. As noted above, by that time Worldeez had either been or was about to be discontinued by B&M, and Cabo’s position was that the brand had failed by that point. The economic context of the market was therefore not such as to indicate a sufficient degree of harm to competition so as to find that agreement had as its object a restriction of competition. As to the effects of the agreement, the economic experts agreed that the agreement between B&M and MGA did not have any material effect on the supply of Worldeez. The B&M agreement was, therefore, not a restriction of competition within the meaning of the Chapter I prohibition. Exemption under the VBER The parties’ submissions

Exemption under the VBER

[387]MGA’s main argument in relation to the unlawful agreements claim was that any agreements that were concluded between MGA and the toy traders were exempted under the terms of the VBER. Cabo disputed the application of the VBER. The parties’ arguments raised three main issues as to the interpretation and scope of the VBER:(i) whether the agreements between MGA and the toy retailers were within the scope of the VBER;

(ii) whether MGA’s market share fell above the 30% threshold set out in Article 5; and

(iii) whether the agreements were excluded under Article 5.

[388]In light of my conclusion above regarding the B&M agreement, it is not necessary to consider the application of the VBER to that agreement. Rather, the focus is on the agreements between MGA and The Entertainer, TRU and Smyths.

Scope of the VBER

[389]Cabo’s primary case on the VBER was that the agreements between MGA and the toy retailers did not relate to the purchase, sale or resale of LOL Surprise or other products supplied by MGA, and were therefore not vertical agreements within the definition set out in Article 1(a) of the VBER. Rather, Cabo said, the agreements solely related to Worldeez and the retailers’ agreements not to purchase it. Cabo also contended that on a purposive construction the VBER should not apply to agreements such as those in issue in the present case, which were nakedly exclusionary and provided no benefit to consumers.

[390]I do not accept those submissions. Article 1(a) defines a vertical agreement covered by the VBER as an agreement between undertakings operating (for the purposes of the agreement) at a different level of the production or distribution chain, and relating to the conditions under which the parties may purchase, sell or resell certain goods or services. The terms of Article 5 make clear that the VBER in principle applies to vertical agreements which contain obligations not to stock competing products; if that were not the case, the provisions in Article 5 would be superfluous.

[391]In the present case, the agreements between MGA and the toy retailers plainly fell within the Article 1(a) definition of a vertical agreement: they were agreements setting out the conditions under which the retailers could purchase LOL Surprise from MGA, the conditions being that the retailers would not stock (at least) the Worldeez globe.

[392]Contrary to Cabo’s submissions, the agreements cannot sensibly be characterised as “freestanding agreement[s] not to purchase the goods of a competing supplier”. They were not simply agreements that the retailers would not stock Worldeez; rather, as is apparent from the description of MGA’s conduct at §§117–147 above, the entire purpose of the retailers’ agreements to MGA’s demands was to obtain supplies of LOL Surprise. That is demonstrated most clearly by the exchanges of messages between The Entertainer and MGA in 2017, but the same was also true of TRU and Smyths, considering the totality of the evidence as to those agreements.

[393]While there is no doubt that the VBER, as with any EU Regulation, should be construed purposively, that does not provide a licence to read in additional qualifications to the terms of the VBER. It is apparent from the recitals to the VBER that the intention was to limit the benefit of the exemption to vertical agreements which can be assumed to satisfy the conditions of Article 101(3) TFEU, and that the exemption should not apply to agreements containing restrictions which are likely to restrict competition and harm consumers (see e.g. recitals (5) and (10)). That purpose was, however, intended to be achieved through the conditions in the Regulation, as recital (12) makes clear:

“The market-share limitation, the non-exemption of certain vertical agreements and the conditions provided for in this Regulation normally ensure that the agreements to which the block exemption applies do not enable the participating undertakings to eliminate competition in respect of a substantial part of the products in question”

(emphasis added).

[394]The recitals recognise that in particular cases an agreement to which the exemption applies may have effects that are incompatible with Article 101(3) TFEU. In such a case the benefit of the Regulation may be withdrawn by the Commission or the competition authority of a Member State, pursuant to Article 29 of Regulation 1/2003: recitals (13) and (14). The foreseen remedy for an agreement exempted under the VBER which is nevertheless considered to be anticompetitive is therefore the withdrawal of the benefit of the VBER, not the implication of additional terms limiting the scope of the VBER.

[395]The agreements at issue in the present case therefore fell within the scope of the VBER as defined in Article 1(a).

Market share threshold

[396]Article 3 of the VBER sets out a market share threshold of 30% for each of the supplier and the buyer. It is not suggested that the market shares of any of the toy retailers exceeded 30%. Cabo contended, however, that MGA’s market share exceeded 30% of the market during the relevant period, assuming the relevant period to be 2017; or that MGA had not shown the contrary, if the relevant period under the VBER was in fact 2016.

[397]On my findings above in relation to the abuse of dominance claim, MGA’s market shares did indeed exceed 30% during the period of the agreements in issue, i.e. from May 2017 onwards. For the purposes of the VBER, however, the effect of Article 7(b) is that what is relevant is not the market shares during the period of the agreements, but rather the market shares during the “preceding calendar year”.

[398]Cabo relied on

Case T-419/03

Altstoff Recycling Austria v Commission EU:T:2011:201, §96, in which the General Court noted that (for the purposes of the application of an identical provision in Regulation 2790/1999, the predecessor to the VBER) in the case of “new markets”, the supplier’s market share can only be calculated from the time at which the market was created. That of course makes sense where, due to the creation of a new market, market shares cannot be calculated for the purposes of Article 3 of the VBER for the year preceding the conduct in issue. That does not mean, however, that Article 7(b) can simply be disapplied in every case where market shares fluctuate on a given market. On the contrary, Article 7(e) makes clear that even if one of the parties’ market shares rises above 35% (from an initial share of 30% or below), the exemption will apply for a calendar year following the year in which the level of 35% is first exceeded.

[399]In the present case, however, the market did exist in 2016 (i.e. the calendar year before the agreements commenced), and MGA was a supplier on that market (with Num Noms). It is therefore relevant to consider whether MGA’s market share was below 30% during that year. It is not disputed that MGA bears the burden of proof on this point, as the party claiming the benefit of the exemption: see by analogy Network Rail v Achilles

[2020] EWCA Civ 323

, §105.

[400]While the evidence as to MGA’s market share during 2016 is incomplete, there is enough before the court to make it clear that MGA’s market share very likely did not exceed 30% during 2016. Mr Colley did not calculate market shares during 2016, but Mr Parker provided calculations of market shares for June to December 2016, based on Mr Colley’s preferred market definition, which showed MGA’s market shares ranging from 7.91% to 14.56% during that period. There is nothing in any of the evidence before the court to suggest that MGA’s market share during the early part of 2016 was so significantly different as to change the conclusion that MGA’s market share was indeed well below 30% throughout 2016, assuming Mr Colley’s market definition. Nor do the market share comparisons set out at §§261 and 271 suggest that a different conclusion might be reached for the extended Colley market.

[401]On the basis of both Mr Colley’s market definition and the extended Colley market, MGA’s market share only exceeded 35% in early2017. Accordingly, on the basis of Article 7(e) of the VBER, the benefit of the exemption continued to apply throughout the period of the agreements at issue in the present case.

Excluded restrictions

[402]Cabo’s final argument on the scope of the VBER was that the agreements in the present case constituted indefinite non-compete obligations within the meaning of Article 5(1)(a), such that they were excluded from the exemption under the block exemption. The submission was that a “non-compete obligation” as defined in Article 1(d) encompasses any agreement not to sell competing products, including those limited to the competing products of a particular supplier.

[403]MGA disputed that interpretation. Its submission was that the definition of a “non-compete obligation” in the first half of Article 1(d) refers to a single-branding obligation, i.e. an obligation not to sell competing brands in general, rather than an obligation not to sell particular competing brands.

[404]MGA’s interpretation of Article 1(d) is, in my judgment, correct. The definition in Article 1(d) refers to a direct or indirect non-compete obligation alongside an obligation to purchase more than 80% of the buyer’s total purchases of the contract goods or services and their substitutes from the supplier (or undertaking designated by the supplier). That alignment indicates that the harm being addressed is a situation where, whether framed as a non-compete obligation or a minimum purchase obligation, competing goods or services can only make up less than 20% of the buyer’s total requirements. That is the interpretation set out in the Commission’s 2010 Guidelines on Vertical Restraints (the Vertical Restraints Guidelines)[2010] OJ C 130/1, which state at §66:

“Non-compete obligations are arrangements that result in the buyer purchasing from the supplier or from another undertaking designated by the supplier more than 80% of the buyer’s total purchases of the contract goods and services and their substitutes during the preceding calendar year … thereby preventing the buyer from purchasing competing goods or services or limiting such purchase to less than 20% of total purchases.”

[405]The Article 5(1)(a) exclusion of non-compete obligations, as defined by Article 1(d), must also be read in the context of the further exclusion in Article 5(1)(c) which relates to obligations not to sell particular competing brands when imposed on the members of a selective distribution system. If Article 5(1)(a) were to be interpreted as encompassing any type of non-compete obligation, including an obligation not to sell a particular competing brand, Article 5(1)(c) would be redundant. The wording of Article 5(1)(c) therefore strongly indicates that Article 5(1)(a) is not intended to extend to obligations not to sell specific competing brands. Rather, it is apparent that a distinction is being drawn between an obligation not to sell competing brands in general (Article 5(1)(a)) and an obligation not to sell particular competing brands, which is excluded when that obligation arises in the context of a selective distribution system (Article 5(1)(c)).

[406]The Commission’s explanation of Article 5(1)(c), at §69 of the Vertical Restraints Guidelines, reinforces that distinction:

“The Block Exemption Regulation covers the combination of selective distribution with a non-compete obligation, obliging the dealers not to resell competing brands in general. However, if the supplier prevents its appointed dealers, either directly or indirectly, from buying products for resale from specific competing suppliers, such an obligation cannot enjoy the benefit of the Block Exemption Regulation. The objective of the exclusion of such an obligation is to avoid a situation whereby a number of suppliers using the same selective distribution outlets prevent one specific competitor or certain specific competitors from using those outlets to distribute their products (foreclosure of a competing supplier which would be a form of collective boycott).”

[407]The only academic commentary addressing the scope of Article 5 of the VBER, to which I have been referred by the parties, is Wijckmans and Tutschaever, Vertical Agreeemnts in EU Competition Law (3rd ed, 2018). That likewise describes the obligation defined in the first half of Article 1(d) as a single branding obligation, and explains the relationship between that and the “80 percent rule” in the second half of Article 1(d) as follows (§§7.25 and 7.32, emphasis in original):

“The first type of non-compete obligation under Article 1(1)(d) of Regulation 330/2010 is a particular category of single branding. It covers those obligations which cause the buyer not to manufacture, purchase, sell or resell goods or services which compete with the contract products. … … single branding is when the buyer is restricted to manufacture or trade any competing products. If the buyer is required not to manufacture or trade certain competing products only, it will depend on what percentage of the total sales of the buyer that the sale of those competing products represents in order to determine whether there is a non-compete obligation in the sense of the 90 per cent rule or instead whether the clause escapes the scope of Article 1(1)(d) of Regulation 330/2010.”

[408]The agreements in the present case were not single branding agreements, requiring the toy retailers not to stock competing brands in general; nor did they require the retailers to take more than 80% of their stock of LOL Surprise and substitutable products from MGA. Rather, they only prohibited the retailers from stocking Worldeez (or at least the Worldeez globe). As such they could only have been excluded from the VBER under Article 5 if the retailers were members of a selective distribution system used by MGA, so as to engage Article 5(1)(c). That was not the case here.

[409]The Article 5(1)(a) exclusion is not, therefore, applicable to the agreements between MGA and the toy retailers. It is not necessary, in the circumstances, to address MGA’s further point that the duration of the agreements was not explored in the evidence at the trial, and cannot be treated as having been indefinite (as contended by Cabo) on the basis of mere assertion.

Conclusion on the VBER

[410]It follows that the agreements between MGA and the toy retailers were exempted under the VBER and were therefore not prohibited by the Chapter I prohibition or Article 101(1) TFEU.

Exemption under s. 9 / Article 101(3)

[411]Given my conclusion on the VBER, I do not need to address MGA’s alternative argument that even if the VBER did not apply, any agreements that toy traders would not stock Worldeez were exempt under s. 9 of the 1998 Act and Article 101(3) TFEU.

[412]For completeness, however, if it had been necessary to address this issue, MGA’s argument on this point was hopeless for at least two reasons. First and foremost, MGA’s argument rested on the proposition that the agreements (if there were any) contributed to improving the production or distribution of goods and the promotion of technical and economic progress, by preventing “free-riding” on its investment in LOL Surprise. That argument in turn rested on MGA’s repeated contentions that the Worldeez globe was a “knock off” of LOL Surprise. I have, however, already rejected that argument as a justification for MGA’s abuse of its dominant position. It fares no better in the context of the unlawful agreements claim. If the suggestion is that Worldeez was passing off LOL Surprise or that MGA considered that to be the case, I have found that MGA did not hold any such genuine belief, nor did the Worldeez globe in fact pass off LOL Surprise.

[413]Secondly, MGA claimed that the agreements did not impose restrictions that were not indispensable to the attainment of the objectives. That is belied, however, by the fact that MGA maintained its objection to the Worldeez globe even after the packaging was redesigned, despite the fact that MGA is now unable to put forward any basis on which the redesigned packaging could have been regarded as a misrepresentation of the trade origin of Worldeez.

[414]There is therefore no credible basis for an exemption under s. 9 and Article 101(3), even without considering the other criteria for the application of those provisions. Conclusion on the

Unlawful agreements claim

[415]I consider that MGA’s conduct involved agreements with The Entertainer, TRU and Smyths which were, in principle, restrictive of competition by object. Those agreements were, however, vertical agreements which were exempt under the VBER and therefore not prohibited by the Chapter I prohibition or Article 101(1) TFEU. Patent threats claim Overview of the legal framework

[416]Cabo’s claim for unjustified patent threats is brought under s. 70 of the 1977 Act. The basic legal framework is not in issue in the present case. It is common ground that the relevant version of s. 70 is the version which applied before it was amended on 1 October 2017. That provided:

“(1) Where a person (whether or not the proprietor of, or entitled to any right in, a patent) by circulars, advertisements or otherwise threatens another person with proceedings for any infringement of a patent, a person aggrieved by the threats (whether or not he is the person to whom the threats are made) may … bring proceedings in the court against the person making the threats, claiming any relief mentioned in subsection (3) below. (2) In any such proceedings the claimant or pursuer shall, subject to subsection (2A) below, be entitled to the relief claimed if he proves that the threats were so made and satisfies the court that he is a person aggrieved by them.”

[417]Section 70(2A) provided that a threat may be justified if the defendant proved that the acts in respect of which proceedings were threatened constitute or would (if done) constitute an infringement of a patent.

[418]The relief that could be claimed under s. 70(3) was a declaration that the threats were unjustifiable, an injunction against the continuance of the threats, and damages in respect of any loss sustained by the threats.

[419]If MGA’s conduct is found to constitute threats within the meaning of s. 70(1), MGA does not contend that those threats were justifiable under s. 70(2A). That is because although MGA filed an application on 6 April 2017 for a European patent entitled “Toy having multiple serial surprise reveals”, that application had not been published by May 2017, when the alleged threats were made. (The patent was granted on 19 June 2019, but subsequently revoked on 21 March 2023 following an opposition filed by Cabo.) MGA therefore accepts that Mr Larian’s claims in May 2017 to have a relevant patent for LOL Surprise were not accurate.

[420]MGA does, however, dispute(i) that its conduct constituted threats of patent infringement proceedings in the UK;

(ii) that Cabo was a “person aggrieved” by those threats; and

(iii) that Cabo can claim either damages or declaratory relief if s. 70(1) threats are established. This section addresses the first two of those questions; the issue of relief is addressed further below. Threats of patent infringement proceedings Legal principles

Threats of patent infringement proceedings

[421]A threat of patent infringement proceedings can be in writing or verbal. It must, however, be something which conveys to the recipient that the person making the threat(i) has patent rights against “another person”, which need not be the recipient of the communication; and

(ii) intends to enforce those rights by legal proceedings in the UK: see Best Buy v Worldwide Sales Corporation

[2011] EWCA Civ 618

, §20 (considering the equivalent provision for threats of trade mark infringement under s. 21 of the Trade Marks Act 1994); and The Noco Company v Shenzhen Carku Technology

[2023] EWCA Civ 1502

, §§21 and 23 (considering the version of the 1977 Act as amended from October 2017).

[422]The threat does not have to be an express threat, but may be veiled, covert, conditional or future: Generics (UK) v Warner-Lambert

[2015] EWHC 2548

, §693. Whether a communication amounts to such a threat is an objective test, and is determined by reference to what a reasonable person, with knowledge of all the relevant circumstances at the date of the communication, would have understood the communication to mean, in the context of the communication as a whole. Where there is a sequence of communications, the communications must be looked at as a whole: Best Buy, §18; The Noco Company, §§26–28. The parties’ submissions

[423]Cabo identified six specific emails which it said included threats of patent infringement: i) Mr Larian’s email to Stuart Grant at 07:33 PDT on 23 May 2017, “Please don’t buy knock off. We have a patent and will take action against the copycats” (§117 above). ii) Mr Larian’s email to Stuart Grant at 08:22 PDT on 23 May 2017, “We do have patent and copyright and we will take legal action and protect our IP” (§118 above). iii) Mr Larian’s email to Stuart Grant at 09:24 PDT on 23 May 2017, “We have a patent for this concept of layers of surprise in a ball. And this item has even copied the color of our balls as well as the shrink wrap in clear intent to palm off our good will and trade mark and trade dress. … We invest millions in creating original ideas and innovations and will vigorously protect them” (§120 above). iv) Mr Larian’s email to Stuart Grant at 18:15 PDT on 23 May 2017, “We will take action against Sinco Toys and any other parties involved to protect our IP” (§122 above). v) Mr Larian’s email to Stuart and Gary Grant at 05:45 PDT on 24 May 2017, “we will stop copycats and protect our IP Worldwide. We expect our friends and partners to respect our IP rights” (§124 above). vi) Mr Larian’s email to TRU on 24 May 2017, “We have patent pending and full copyright and trade dress protection on the LOL concept. We are told that TRU (at least in the UK) is buying this knock off. We are taking legal action against this company in the UK. As we speak” (§137 above).

[424]In addition, Cabo said that oral threats were made by Mr Laughton to each of TRU, Smyths, Argos, Tesco, Sainsbury’s and Asda on 23 or 24 May 2017 (§135–137 and 149 above). Cabo also contended that oral threats were made to AB Gee, but I have already rejected that on the facts (§§151–153 above) so do not need to say anything more about that.

[425]MGA said that when read as a whole, including with the cease and desist letter which was forwarded to The Entertainer on the evening of 23 May 2017 (PDT) the emails to The Entertainer would not have been understood as threatening patent infringement proceedings against either The Entertainer or Cabo, but would instead have understood the complaint to be one of passing off by Cabo. Likewise, MGA said that the email to TRU would have been understood as threatening passing off proceedings against Cabo. In respect of Mr Laughton’s telephone calls to TRU and other retailers, MGA said that it was improbable that these would have been understood as threatening legal proceedings of any kind.

Discussion

[426]MGA’s arguments differed across the different communications relied on by Cabo, and it is therefore necessary to consider these separately.

[427]Emails to The Entertainer. MGA emphasised that it did not at any point threaten legal proceedings against The Entertainer, and cited Mr Grant’s evidence to the effect that he did not consider Mr Larian’s emails to be threatening to sue The Entertainer. Rather, MGA said that its emails indicated that action would be taken against the maker of Worldeez.

[428]I agree that the emails to The Entertainer would have been understood as threats against Cabo rather than against The Entertainer. But that is irrelevant for the purposes of the patent threats claim, since the threat does not have to be made against the recipient of the communication; rather, it is sufficient that threats of patent infringement proceedings have been made against “another person”. MGA’s emails to The Entertainer explicitly threatened legal action against Cabo, as the maker of Worldeez, and also referred to Sinco which MGA thought was associated with Cabo.

[429]MGA’s attempt to characterise those threats as relating to passing off rather than threatening patent infringement is entirely unreal. MGA accepted (indeed positively asserted) that the emails to The Entertainer must be read as a whole. Read in isolation, the emails sent by Mr Larian at 18:15 PDT on 23 May 2017 and 05:45 PDT on 24 May 2017 did not specifically refer to the intellectual property rights said to be infringed, but merely cited “IP” or “IP rights” in general terms. The three preceding emails did, however, all expressly claim that MGA had patent rights and asserted that MGA would take action to protect its intellectual property rights. The natural and obvious interpretation of the emails, read as a whole by a reasonable person in the position of The Entertainer, was that MGA was intending to bring patent infringement proceedings against Cabo and potentially other entities involved in Cabo.

[430]The fact that Mr Larian expressed himself in hasty and intemperate terms, and referred in his emails to a miscellany of other supposed intellectual property rights such as copyright and trade mark rights, does not change the clear implications of Mr Larian’s communications. Indeed it is notable that the three emails which referred to specific intellectual property rights said to be held by MGA all referred expressly to patent rights. By contrast, none of the five emails sent to The Entertainer referred to passing off.

[431]MGA could therefore garner no support from the emails themselves for the contention that they should be understood as threatening a passing off action. That is why MGA’s main point on this was that the emails to The Entertainer should be read in the light of the cease and desist letter which was sent to Cabo on the evening of 23 May 2017 (PDT), and forwarded by Mr Larian around the same time to The Entertainer. MGA argued that the letter would have been understood as articulating a claim in passing off rather than a claim for patent infringement.

[432]While that letter (described at §§106–108 above) undoubtedly did include a claim which in substance was likely to have been understood as a complaint of passing off, it is notable that the first paragraph of that letter expressly claimed that “MGA currently has a patent pending for L.O.L. Surprise!”, and the second page of the letter demanded that Cabo “in general refrain from any infringement on L.O.L. Surprise! products or trade dress, or any other intellectual property of MGA”. Accordingly, far from dispelling any impression that MGA was intending to rely on its patent rights as against Cabo, that letter reinforced Mr Larian’s repeated references (in his emails) to his claimed patent rights in relation to LOL Surprise.

[433]Indeed, that was precisely the way that Cabo and its solicitors understood that letter. The response from Cabo’s solicitors sent to MGA on 26 May 2017 (see §114 above) noted the claims of “patent pending” and asked MGA to identify any patent or patent application on which it relied. Cabo went on to deny any patent infringement “for the simple reason that you do not hold any patent or patent application the territorial scope of which encompasses this jurisdiction”. With some prescience, the letter continued:

“… it has come to our client’s attention that you have also issued threats to UK retailers (including The Entertainer (Amersham) Limited) that dealings in the Worldeez product will constitute patent infringement. Our client perceives there to be a very real risk that orders for the Worldeez product will be cancelled, and that other orders will not be placed for that product, by such retailers directly as a result of your threats, in which case our client will suffer substantial financial loss directly attributable to your threats.”

[434]Cabo’s solicitors thus interpreted the cease and desist letter, and MGA’s communications with (at least) The Entertainer, in the way in which I consider those communications should objectively be interpreted, namely as making threats of patent infringement proceedings.

[435]Email to TRU. MGA’s email to TRU likewise clearly threated legal action against Cabo, and claimed that MGA had a “patent pending” on LOL Surprise. MGA’s suggestion that this email would have been understood, again contrary to its clear wording, as a threat of proceedings for passing off rather than patent infringement, is hopeless. The email did not refer to passing off or the conditions for passing off. MGA sought to infer a reference to passing off from Mr Larian’s complaint about TRU “buying this knock off”. But as I have already found, Mr Larian habitually used his “knock off” mantra in a loose and pejorative sense, alongside a miscellany of allegations of different intellectual property rights (see §334 above). Nothing in the context would have suggested to the recipient of the TRU email that, having referred explicitly to a pending patent, the statement that MGA was “taking legal action against this company in the UK. As we speak” meant a claim in passing off rather than a patent infringement claim.

[436]MGA also suggested that having seen the reference to “patent pending” in the email to TRU, a reasonable recipient would have taken advice which would have revealed that MGA could not immediately bring infringement proceedings. It relied in that regard on comments by Mann J in Nvidia v Hardware Labs

[2016] EWHC 3135 (Ch)

, §§21–22, that in an “appropriate case” a threats letter should be treated as it would be viewed with the benefit of legal advice. The situation contemplated there was, however, one in which a threats letter was “couched in legal technical terms”, as compared with a letter “addressed to a businessman about business matters”. The email to TRU, by contrast, fell into the latter category. It was not drafted in technical legal terms, but was sent by Mr Larian to the management of TRU, in what was clearly a business context, expressed in direct and forthright language, and in terms which were readily intelligible to the recipients. It is, as Cabo noted wholly unrealistic to suppose that TRU would have sought legal advice in order to interpret that email, and Mr Larian cannot have expected the recipients to do so.

[437]In any event, as MGA acknowledged, a threat of patent proceedings made prior to a patent grant can be construed as a threat to bring proceedings once the patent was granted, which is an actionable threat of patent infringement proceedings: Brain v Ingledew Brown

[1996] FSR 341

, pp. 347–8 (per Aldous LJ). Even if, therefore, legal advice had been taken, that would not have suggested that no infringement action was intended, but would rather simply have indicated that infringement proceedings were intended once the patent had been granted.

[438]The email to TRU was, in my judgment, clearly and obviously a threat of patent infringement proceedings.

[439]Mr Laughton’s telephone calls. On 24 May 2017 Mr Laughton reported by email to Mr Larian that he had spoken by telephone to “all major retailers” who “all agreed” that they would not take the Worldeez globe in its current format (§136). His oral evidence was that this referred to telephone calls with TRU, Smyths, Argos, Tesco and Sainsbury’s. He was asked about what he had said to TRU: “Q. And you also accept in your evidence you would have explained to them that Isaac Larian believed that Worldeez, for example, infringed MGA’s patent rights …? A. I probably didn’t go into too much detail. … I don’t remember what I said, but I assume I would have said, ‘We may have a problem with regards patent with regards the balls’.

[440]Later in the cross-examination, Mr Laughton was asked about his similar conversations with the other retailers:

“Q. … you fairly accept that you would have flagged your understanding from Isaac of MGA’s issues with the Worldeez globe at the time, which is it infringed MGA’s intellectual property rights? A. Yes. Q. So you would have said to them in the same way, ‘MGA’s view is this infringes the patent’, for example? A. We have some issues with it. Isaac believes it infringes the patent and we are just trying to solve it, yes.”

[441]MGA pointed out that these answers were given in response to leading questions, and that Mr Laughton was speculating as to what he had said seven years previously rather than recalling the precise words used. Mr Laughton was, however, asked about this twice, once in relation to TRU, and then again in relation to the other retailers, and gave essentially the same answer, in substance agreeing with the proposition that he had told the retailers that MGA believed that Worldeez infringed its patent. Mr Laughton was not shy of claiming a lack recollection when he wished to avoid answering a question, but notably did not do so on this point.

[442]Mr Laughton’s responses were, moreover, consistent with Mr Larian’s emails to The Entertainer and TRU (copied or forwarded to Mr Laughton), which as set out above had repeatedly asserted patent rights and threatened to bring legal proceedings. Mr Laughton had also himself complained to Mr Sivner in a text message exchange on 23 May 2017, either the same day as or the day before his telephone calls with the toy retailers, that Worldeez “infringes on our patent with LOL – we will protect so expect contact from MGA legal today” (§105 above). While MGA asserted that it was improbable that a retailer would have understood a layman such as Mr Laughton to be threatening legal proceedings of any kind, that is precisely what he did in his text message exchange with Mr Sivner.

[443]In that context it is, in my judgment, very likely that Mr Laughton in his telephone calls with retailers not only did indeed assert that MGA considered that Worldeez (or at least the Worldeez globe) infringed its patent. It is also likely that he made clear in those conversations (whether expressly or impliedly) that MGA intended to enforce those rights if necessary. Cabo is therefore right to characterise those calls as making oral threats of patent infringement proceedings. “Person aggrieved”

[444]In order to be a person aggrieved by a threat, the claimant must show that its commercial interests have been, or were likely to be, adversely affected by the threats in a real, as opposed to a fanciful or minimal, way: Warner-Lambert, §694. The aggrieved person does not, however, have to have been the recipient of the threats, or even the person against whom the threats are made: The Noco Company, §21.

[445]That condition is clearly met in the present case. The threats of proceedings were made precisely in order to interfere with Cabo’s commercial relationships with the toy retailers; and were likely to do so in fact. Indeed, I have found that The Entertainer cancelled its order for the globe as a result of MGA’s conduct; Smyths would have placed an order for Worldeez but for MGA’s conduct; and TRU was also interested in stocking Worldeez, but for MGA’s objections. Conclusion on the

Patent threats claim

[446]MGA’s emails and telephone calls listed at §423–424 are therefore, in my judgment, rightly characterised by Cabo as unjustified patent threats within the meaning of s. 70 of the 1977 Act. Causation and quantum Preliminary comments

Causation and quantum

[447]It follows from my conclusions on liability that it is necessary to determine the loss suffered by Cabo as a result of MGA’s exclusionary campaign, in breach of the Chapter II prohibition and Article 102 TFEU, and MGA’s unjustified threats of patent infringement proceedings.

[448]As indicated at the start of this judgment, the parties’ positions on causation and quantum were poles apart. Cabo contended that, but for MGA’s conduct, Worldeez would have been a successful and profitable product, both in the UK and internationally, and that Cabo would have gone on to generate additional revenue from licensing deals. Cabo’s initial claim for lost UK and international profits, and lost licensing revenues, was for over £170m. That had reduced to between £51–65m by the end of the trial. Cabo relied on quantum modelling by Mr Colley, together with the factual and toy expert evidence.

[449]MGA submitted that Cabo could not show that Worldeez would have been profitable (whether in the UK or internationally) or would have generated any licensing revenues in the absence of MGA’s conduct, and that the claim for damages should therefore fail. MGA put its submissions on essentially two bases. The first was that Cabo could not show on the balance of probabilities that it would have traded profitably, on the basis of the factual and toy expert evidence as to the quality of the product, extent of initial retailer support, the dynamics of the toy market and the sales volumes, revenue volumes and capital required to break even during 2017. Secondly, MGA relied on Mr Parker’s quantum models to submit that Cabo would have enjoyed very limited profitability on even generous assumptions as to its success, and would have suffered losses on more realistic assumptions. To the extent necessary, MGA relied on Mr Davies’ approach to valuation issues.

[450]The parties approached the issues of causation and quantum in very different ways, both as a matter of legal principle and as a matter of evidence. It is therefore necessary to start by considering a number of issues of legal principle, before addressing the issues between the parties on the evidence. Legal principles Actionable damage and causation of loss

Legal principles

[451]There was considerable debate, in both the opening and the closing submissions, as to the approach to be taken to the causation of loss. By the time of the oral closing submissions, however, the position was essentially common ground.

[452]Cabo’s cause of action under the Chapter II prohibition and Article 102 TFEU is treated as a breach of statutory duty. In order to complete the cause of action for that breach, a claimant must prove that some actionable harm or damage was caused by that breach. That is sometimes expressed as the damage which forms the “gist” of the tort action. Recovery is not limited to the “gist damage”, but the claimant must satisfy the test for causation in relation to (at least) the “gist damage” before the court can consider the quantification of the claimant’s actual loss: Royal Mail v DAF Trucks

[2023] CAT 6

, §§167–168 and §173.

[453]Causation of loss or damage must be established on a balance of probabilities, and must be assessed by reference to the type of harm alleged by the claimant. A claimant cannot rely on a particular type of actionable damage as completing its cause of action, if that has not been pleaded and addressed in the evidence before the court: see the CAT in Royal Mail v DAF, §§172–173.

[454]At some points in its earlier submissions, Cabo appeared to argue that once its cause of action had been established by showing some actionable damage, it was not necessary to prove causation on a balance of probabilities for any further head of loss claimed. The authorities consistently establish, however, that causation of any head of loss must be established on a balance of probabilities: see the comments of Toulson J and then Toulson LJ in Fyffes Group v Templeman

[2000] 2 Lloyd’s Rep 543

and Parabola Investment v Browallia

[2010] EWCA Civ 486

[2011] QB 477

, considered in more detail at §§459–460below, as well as the Court of Appeal in Royal Mail v DAF, considered at §462 below. By the time of the oral closing submissions, this was common ground.

[455]On the same basis, even where a tort does not require actionable damage to be shown to complete the cause of action, if damages are claimed it will still be necessary to establish causation in relation to each head of loss. That is the case for the patent threats claims in these proceedings. An action under s. 70 of the 1977 Act is a statutory tort of strict liability, which does not require proof of actual damage in order to be actionable: Global Flood Defences v Van den Noort

[2015] EWHC 2548 (Pat)

, §34. An injunction or declaration can therefore be claimed irrespective of proof of loss (and in this case Cabo does claim a declaration in addition to damages, which is addressed separately below). Where damages are claimed, as they are here, they are assessed on the usual tortious basis, which requires proof of causation of the loss on a balance of probabilities.

Quantification of the loss

[456]Once causation of a head of loss is established, damages are assessed on the compensatory basis. Cabo is therefore entitled to be placed in the position in which it would have been had the breach not been committed: Sainsbury’s v Visa

[2020] UKSC 24

, §194. That requires an assessment of Cabo’s performance in the counterfactual world in which its market entry was not impeded by MGA’s intervention.

[457]That assessment is necessarily uncertain: it is impossible to predict with precision how an undertaking would have performed on the market in a hypothetical scenario that is very different from what happened in fact. The basic common law approach to the assessment of damages in such a case requires the court to do the best that it can on the available evidence, notwithstanding forensic difficulties that may arise. This is often labelled the “broad axe” principle: see Mastercard v Merricks

[2020] UKSC 51

[2021] 3 All ER 285

, §§49–51 (referring to the well-known dictum of Lord Shaw in Watson, Laidlaw v Pott 1914 SC (HL) at 29–30). The basic principle is that the court should avoid “artificial demands for precision” but should instead achieve practical justice on the basis of material which may well be limited: Green LJ in London & SE Railway v Gutmann

[2022] EWCA Civ 1077

, §59. That may well require broad assumptions, estimates, generalisations and “informed guesswork” on the basis of the evidence before the court.

[458]For the same reason, where quantification of the loss turns on a hypothetical exercise, the court does not have to establish a particular quantum of loss on a balance of probabilities. That would demand artificial precision in a context where precise measurement is impossible. Rather, the court must make a reasonable assessment based on the evidence before it, taking all relevant and material factors into account.

[459]Toulson J explained that point in Fyffes Group v Templeman, p. 667, as follows:

“[T]here is an important distinction between identification and measurement of the wrong (injury) done to the claimant – or, to use legal terms, between causation of loss and quantification of loss. The distinction is easy to see in theory, but drawing the boundary line can be difficult in practice because questions of causation and quantum can be closely entwined. Causation of a head of loss must be established on a balance of probabilities. Quantification of a head of loss may involve an assessment of all shades of risks and possibilities. … As I have said, causation of a head of loss has to be established on the balance of probabilities.”

[460]He made similar comments (by then as Toulson LJ) in Parabola, §§22–24:

“22. … Some claims for consequential loss are capable of being established with precision (for example, expenses incurred prior to the date of trial). Other forms of consequential loss are not capable of similarly precise calculation because they involve the attempted measurement of things which would or might have happened (or might not have happened) but for the defendant’s wrongful conduct, as distinct from things which have happened. In such a situation the law does not require a claimant to prove the impossible, nor does it apply the balance of probability test to the measurement of the loss. 23. The claimant has first to establish an actionable head of loss. … [In the present case] the judge found that, but for Mr Bomford’s fraud, on a balance of probability Tangent would have traded profitably at stage 1, and would have traded more profitably with a larger fund at stage 2. The next task is to quantify the loss. Where that involves a hypothetical exercise, the court does not apply the same balance of probability approach as it would to the proof of past facts. Rather, it estimates the loss by making the best attempt it can to evaluate the chances, great or small (unless those chances amount to no more than remote speculation), taking all significant factors into account … 24. The appellants’ submission, for example, that ‘the case that a specific amount of profits would have been earned in stage 1 was unproven’ is therefore misdirected. It is true that by the nature of things the judge could not find as a fact that the amount of lost profits at stage 1 was more likely than not to have been the specific figure which he awarded, but that is not to the point. The judge had to make a reasonable assessment and different judges might come to different assessments without being unreasonable.”

[461]In Vasiliou v Hajigeorgiou

[2010] EWCA Civ 1475

, §25, Patten LJ cited the above passage of the Parabola judgment, commenting that this does not mean importing the loss of chance doctrine into the quantification of loss:

“Where the quantification of loss depends upon an assessment of events which did not happen the judge is left to assess the chances of the alternative scenario he is presented with. This has nothing to do with loss of chance as such. It is simply the judge making a realistic and reasoned assessment of a variety of circumstances in order to determine what the level of loss has been.”

[462]The “broad axe” approach was applied more recently in Royal Mail, both by the Competition Appeal Tribunal (at §§173–174) and the Court of Appeal. As Sir Julian Flaux C observed in the Court of Appeal

[2024] EWCA Civ 181

, §145, “once a Court has established loss on a balance of probabilities, the claimant is entitled to be compensated and the Court will do its best to quantify the compensation on the available evidence. … it is in the context of difficulties in the quantification of loss that the principle of the broad axe is deployed by the Courts”.

[463]It may be appropriate to err on the side of generosity to a claimant in the calculation of loss, where evidence is unavailable as a result of the defendant’s conduct. The classic example of this is the case of Armory v Delamirie (1722) 1 Str 505, where a jewel of unknown value was removed from its setting by the defendant, which was accordingly required to pay damages on the basis of a presumption of the highest value jewel which the setting would have accommodated. Likewise, the court may draw adverse inferences from a failure to adduce evidence, as the CAT did in the Royal Mail v DAF Trucks case in relation to DAF’s failure to provide evidence as to how the cartel operated and the benefits it derived from participation: see the judgment of the Court of Appeal in that case at §§143–144, upholding the approach of the Tribunal.

[464]That does not, however, mean that in any case where the defendant’s wrongdoing requires damages to be calculated by assessing a counterfactual scenario, even where there is voluminous evidence on both sides, the court is required to resolve all uncertainties in favour of the claimant. That would be wholly inconsistent with the approach set out in the cases above. Rather, the court will assess the totality of the material before it, giving such weight as is appropriate to the evidence on each side, depending on the court’s appraisal of the strength of that evidence. Where that analysis requires assumptions or inferences to be made, the court will make those assessments on the basis of a consideration of what is reasonable and realistic. There is no principle that requires that approach to be overridden by a default presumption in favour of the claimant, simply because the exercise being carried out is the assessment of a hypothetical counterfactual scenario which did not occur as a result of the defendant’s conduct.

[465]Similar comments were made in Porton Capital Technology v 3M UK Holdings

[2011] EWHC 2895 (Comm)

, concerning a claim for lost profits for breach of contract. Hamblen J observed at §§244–245 that:

“This is not a case concerning the value of goods which the defendant has refused to produce or of the suppression of evidence, as in Armory v Delamirie. … It is a claim for lost profits for breach of contract. There is factual and expert evidence before the court relating to that claim. There is documentation before the court relevant to the claim. The evidential playing field is a level one. Whilst it is correct that the claim involves a degree of conjecture, that is the case in relation to very many contractual damages claims and in all such cases it can be said that it is the defendant’s breach of contract which has made that conjecture necessary. As a matter of authority there is no requirement to apply the principle of Armory v Delamirie to a case such as the present, and as a matter of principle I consider that there is good reason not to do so and that the application of the principle should not be extended further than is necessary. Even if that be wrong … any presumption would only arise in a case of doubt and in arriving at the findings set out below I have not found there to be sufficient doubt to give rise to any presumption that might otherwise be applicable.”

[466]As in the Porton Capital case, there was in this trial an abundance of factual and expert evidence on both sides. While there were some aspects on which the evidence was not entirely complete by the end of the trial, that was not because it was inherently unavailable, but was because of the problems discussed at §§42–54 above, namely the divergence of the experts’ positions and the evolving nature of both parties’ positions during the trial. Where necessary, further submissions and evidence were provided by the parties following the trial. I do not consider there to be remaining material doubts in the assessment of the evidence, still less doubts of such a nature that might raise the application of the Armory v Delamirie principle.

The approach to claims for lost profits

[467]There was some debate as to the extent to which a claim for lost profits arising from the defendant’s conduct requires proof, on a balance of probabilities, that the claimant would in the counterfactual case have traded profitably. MGA’s consistent position was that it was indeed necessary for Cabo to show on a balance of probabilities that it would have traded profitably in the counterfactual case. Cabo’s position on this point was not entirely clear. Cabo’s opening submissions and written closing submissions contended that it was not necessary to establish on a balance of probabilities that Cabo would have traded profitably. Ms Kreisberger in her oral closing submissions, however, accepted that it was necessary to prove causation of any head of loss on a balance of probabilities. She also accepted that if a claim was made for lost profits caused by the defendant’s conduct, it would be necessary to prove on the balance of probabilities that there would have been profits in the counterfactual scenario; and she reiterated that Cabo’s case was that it would indeed have been highly profitable in the counterfactual case.

[468]To the extent that any doubt remains as to Cabo’s position on this point, the approach adopted by MGA and in Cabo’s oral closing submissions is, in my judgment, clearly correct. The contrary approach put forward in Cabo’s written closing submissions on this point relied on the judgments of the Court of Appeal in in Parabola and Vasiliou. Those cases do not, however, support the position advanced by Cabo. Both cases were claims for lost profits on the basis that, but for the conduct of the defendants, the relevant businesses (in Parabola a stock trading company, in Vasiliou a restaurant) would have traded profitably. In both cases the approach taken by the trial judges was to consider first whether, on a balance of probabilities, the businesses would have traded profitably, before going on to make a “reasonable assessment” of the profits which would have been achieved. In both cases that approach was upheld by the Court of Appeal.

[469]At §23 of Vasiliou, Patten LJ commented that Mr Vasiliou’s competence and the restaurant’s prospects of success were not matters that went to causation, but were at most relevant to the assessment of how profitable the restaurant would have been. Those comments do not, however, suggest that in a claim for lost profits which would have been made but for the defendant’s conduct, it is possible to avoid considering whether on a balance of probabilities the business would have been profitable at all. Quite the contrary: Patten LJ approved the approach taken by the trial judges in two successive claims in the dispute which led to the appeal, both of whom had made findings on a balance of probabilities that the restaurant would have been successful and profitable, before assessing what the lost profits would have been. Patten LJ also cited and relied upon the passage from the judgment of Toulson LJ in Parabola above, in which Toulson LJ had approved the same approach taken by Flaux J in that case.

[470]Nothing in Patten LJ’s comments at §23 suggested that it was possible to bypass consideration of whether Mr Vasiliou had established on a balance of probabilities that the business would have been a profitable one. Rather, those comments seem to have been made in the context of an explanation of the distinction between the Allied Maples loss of chance doctrine and the approach which had been adopted by the trial judges, rejecting the appellant’s submission that the quantification of profitability should have been subject to a discount to account for the possibility of failure. That can be seen in Patten LJ’s observations at §24:

“Judge Levy, in the passages I have quoted from his judgment, found as a fact that Zorbas would have been a successful restaurant and therefore assessed its lost profits on that basis. His analysis of the variable factors I have outlined which formed the agreed components of that calculation involved taking into account the time needed to establish a reputation and other everyday contingencies but did not involve a more general discount of the kind described in Allied Maples to take account of the statistical probability of failure. That was excluded by his finding that the restaurant would have been a success.”

[471]Patten LJ returned to these points at §44:

“As explained earlier, the issue of how successful the restaurant would have been was not an issue of causation. It was relevant only to quantum. Judge Dight and Judge Levy were satisfied that the restaurant would have been profitable and calculated the damages accordingly. One can express this in terms of them assessing the chances of success at 100% but either way there is no room for a further discount. The calculation of profits which they made was not determined as the best level of profits reasonably obtainable. It was the amount which on their findings he would have earned.”

[472]Patten LJ’s point was therefore that having concluded on the balance of probabilities that the restaurant would have been successful and profitable, the correct approach was then to make a reasonable assessment of how successful the restaurant would have been, taking account of the restaurant’s chances and everyday contingencies arising under that scenario, rather than assuming the “best level of profits reasonably obtainable”. Put another way, having determined whether on a balance of probabilities the restaurant would have been successful and profitable, the quantification exercise then required assessment of how successful and how profitable it would have been.

[473]As Nugee J commented in Wellesley Partners v Withers

[2014] EWHC 556 (Ch)

, §188(5)–(6), the important point arising from these cases is that where the claim is for a loss of profits arising from a lost opportunity to trade, the court must first decide on a balance of probabilities whether the claimant would have traded profitably. Nugee J continued at §188(7):

“… it is clear from Parabola and Vasiliou v Hajigeorgiou that if the court finds that trading would have been profitable, it then makes the best attempt it can to quantify the loss of profits taking into account all the various contingencies which affect this: see Parabola, para 23. This neither requires any particular matter to be proved on the balance of probabilities (see Parabola, para 24) nor has anything to do with the loss of a chance as such (see Vasiliou v Hajigeorgiou, para 25). The assessment of the loss will itself include an evaluation of all the chances, great or small, involved in the trading (see Parabola, para 23). Once the judge has assessed the profits in this way, any further discount is therefore inappropriate (see Vasiliou v Hajigeorgiou, para 28).”

[474]The analysis of Nugee J in Wellesley was cited at length by Bryan J at in Assetco v Grant Thornton

[2019] EWHC 150 (Comm)

, §412. At §§435–440 Bryan J also cited the judgment of Males J in SCF Tankers v Privalov

[2016] EWHC 2163 (Comm)

[2018] 1 WLR 5623

, applying essentially the same approach at §55:

“The true position is that in principle damages can be awarded for loss of profits even if a claimant might have made a loss. The approach which the court will adopt is to ask whether the claimant has proved to a sufficient standard (which may be the balance of probabilities, or sometimes merely that there was a real and substantial chance as in loss of a chance cases) that its trading would have been profitable. If so, the court will make the best assessment of the damages that it can, applying if necessary a discount to reflect whatever uncertainty exists, while recognising that a party seeking to show what might have happened is not required to perform an impossible task with unrealistic precision.”

[475]Bryan J then continued at §441 that SCF Tankers was a case like Parabola and Vasiliou where the claim is for the loss of profits arising from a lost opportunity to trade generally. “In such a case the court first decides if the claimant would have traded successfully – in SCF Tankers v Privalov whether a profit would have been made by the defendants.” The quantification exercise should then, he considered, be approached in the way described by Nugee J in Wellesley.

[476]The consistent approach taken in all of these cases is therefore that where the claim is for the loss of profits in a business resulting from a lost opportunity to trade, the first question for the court is whether the claimant has established on a balance of probabilities that it would have traded profitably, but for the matters relied upon as establishing the defendant’s tortious liability. If the answer to that question is “yes”, then the court will assess what that profit would have been, making the best attempt it can to establish that on the evidence (see the similar summary at §83 of BSV v Bittylicious

[2024] CAT 48

Conclusions on the overarching approach

[477]In light of the principles set out above, the issues to determine are as follows: i) Whether the patent threats were causative of any loss suffered by Cabo (relevant if my conclusion on abuse of dominance is wrong). ii) The identification of the heads of loss pleaded by Cabo. iii) Whether Cabo would have traded profitably in the counterfactual case. iv) The assessment of the parties’ respective quantum models, to the extent relevant.

Causative effect of MGA’s conduct

[478]It is common ground that if there are several causes of loss, it is not necessary to determine weight as between those causes. The classic statement is that of Devlin J in Heskell v Continental Express

[1950] 1 All ER 1033

, 1047:

“Where the wrong is a tort, it is clearly settled that the wrongdoer cannot excuse himself by pointing to another cause. It is enough that the tort should be a cause and it is unnecessary to evaluate competing causes and ascertain which of them is dominant.”

[479]If my conclusion on abuse of dominance is correct, it is not disputed that, if Cabo can establish any loss flowing from MGA’s conduct, the conduct alleged to be an abuse of a dominant position was at least “a” cause of that loss. In that event, therefore, no threshold question of the causative effect of MGA’s conduct arises. If however, contrary to the conclusions above, the patent threats claim were to be the only basis on which Cabo could establish liability, then a threshold question of causative effect would arise, because it is disputed whether the patent threats on their own were “a” cause of any loss suffered by Cabo.

[480]Cabo’s position was that even if the only unlawful element of MGA’s conduct was the threats of patent infringement, the entirety of Cabo’s damage could be claimed as flowing from that conduct, since the patent threats were at least “a” cause of Cabo’s loss. Its submission was that either the patent threats on their own would have caused the retailers to drop the Worldeez product, or those threats at least made a material contribution to the retailers’ decision not to stock Worldeez.

[481]MGA’s response was that any actionable patent infringement threats were neither a sufficient nor effective cause of any loss which Cabo suffered, because Cabo would have been in precisely the same position had those threats not been made. Its position was in essence that in so far as Cabo did suffer any loss, that was caused by MGA’s refusal to supply LOL Surprise to any retailers that supplied the Worldeez globe. In the context of that policy, MGA said that any threats of patent infringement made no material difference to the retailers’ decisions and were therefore not causative in fact or law of any loss suffered by Cabo. The logic of MGA’s argument was therefore that Cabo’s damages claim could only flow from the facts underlying the competition law arguments, such that if MGA’s conduct did not as a matter of law infringe the Chapter I/II prohibitions or Article 101/102 TFEU, then no damages claim could be brought.

[482]Both Cabo and MGA relied on the judgment of the Supreme Court in Financial Conduct Authority v Arch Insurance

[2021] UKSC 1

[2021] AC 649

concerning insurance claims for business losses as a result of the Covid-19 pandemic. The judgment of Lords Hamblen and Leggatt (with whom Lord Reed agreed) considered extensively the analysis of causation in circumstances of concurrent causes of a loss. Their starting point was the established proposition that to establish causation of loss it is not necessary to show that the relevant causative act is the sole cause of that loss; rather it is sufficient that the act is one of a combination of causes that are of equal or almost equal efficacy in bringing about the loss. §§172–173 of the judgment thus cited Heskell, at 1048, setting out the principle that “if a breach of contract is one of two causes, both co-operating and both of equal efficacy … it is sufficient to carry judgment for damages”, and ENE Kos 1 v Petroleo Brasileiro (No. 2)

[2012] 2 AC 164

, §74, where Lord Clarke stated that “where there are two effective causes, neither of which is excluded but only one of which is insured, the insurers are liable”.

[483]Lords Hamblen and Leggatt continued to note, at §175, that in these cases it could not be said that either cause that was characterised as a proximate cause “on its own rendered the loss inevitable in the ordinary course of events. In each case it was the combination of the two causes which together made the loss inevitable. Neither would have caused the loss without the other.”

[484]The difficulty in the Arch case was the conclusion that the relevant clauses in the insurance contracts covered only the effects of cases of Covid-19 occurring within the specified radius of the insured premises. If the relevant question was whether the Government would have acted in the same way if there had been no cases of Covid-19 within the specified radius, the answer would have been that it probably would have acted in the same way (§179). That problem led to a discussion of the situations where causation may nevertheless be established where a series of events combines to produce a particular result, but where none of the individual events is either necessary or sufficient to bring about the result itself (§§183–185).

[485]That specific problem does not, however, arise in the present case. This is not a case of damage caused by the combination of elements which were not individually either necessary or sufficient to cause the loss. Rather, this case is a classic situation where, if Cabo did suffer loss as a result of MGA’s conduct, that loss was caused by a combination of several elements of conduct, including the patent threats, each of which was necessary to produce the result.

[486]That is because, as I have found (§§317–319 above), MGA’s threats of patent infringement proceedings were bound up with its threats to withhold supplies of LOL Surprise from retailers that stocked the Worldeez globe. From the outset, MGA justified its refusal to supply policy on the basis of its objections that the Worldeez globe infringed its intellectual property rights, with the claims of patent rights prominent in those objections. That is why, although MGA’s communications with retailers did not always distinguish between the globe and the 5- and 10-packs, its real objection was to the globe rather than the other products in the Worldeez range, and MGA did not take issue with The Entertainer stocking the Worldeez blind bags or the 5- and 10-packs.

[487]It is also clear from the extensive exchanges between MGA and The Entertainer that the latter clearly understood that MGA’s objections, and its consequent refusal to supply policy, turned on its intellectual property claims. The Entertainer noticeably initially sought to push back on Mr Larian’s claims that Worldeez was a “knock off” of LOL Surprise, pointing out that the product was completely different to LOL. The Grants capitulated, however, after Mr Larian’s statement on the evening of 23 May 2017 (PDT) that MGA would be taking action against Sinco and others involved in Worldeez, followed minutes later by forwarding to the Grants the case and desist letter that had been sent to Cabo.

[488]Stuart Grant’s subsequent email explicitly stated that he had put Worldeez orders on hold because of Mr Larian’s legal action (§123 above). Stuart Grant reiterated that point in his oral evidence:

“… by him providing that cease and desist letter, it tells us that he is actually not threatening, he is actually moving forward with some kind of legal response to the other side, at which point we are saying, okay, we are just going to step back and we will leave you to figure out your differences between you. In the meantime, we will put the product on hold because we don’t want to sell something that’s illegal or that’s infringing anyone’s rights … … the defining moment in that cycle, as it were, was seeing that Isaac had gone from emails to us stating lots and lots of different things around the Worldeez product to actually issuing a legal document or giving us sight of a legal document that he was or is issuing and therefore, okay, this is actually serious. He is not just threatening, this is actually serious. There is some infringement here, so we will put our orders on hold.”

[489]It is apparent from The Entertainer’s subsequent correspondence with MGA, and in particular Stuart Grant’s 15 February 2018 email (§132 above), that even during the year after the events forming the focus of these proceedings, The Entertainer maintained the position that it would refrain from selling products which MGA genuinely believed to infringe its intellectual property rights, but considered that it was free to sell products for which MGA had not taken legal action.

[490]The communications between MGA and other retailers were not as extensively documented as those with The Entertainer, but it is clear from my findings above regarding the threats made to and responses of TRU and Smyths, in particular, that MGA’s patent infringement threats were (as with The Entertainer) central to its objections to the sales of the globe, and its demands that the retailers should not stock the globe if they wished to maintain supplies of LOL.

[491]The present case is, therefore, not one in which it is possible to separate the patent infringement threats from the refusal to supply policy, so as to say that the former had no causative effect on the retailers’ decisions. Rather, the evidence is that the refusal to supply threats were always made in conjunction with the intellectual property objections, and in particular the patent infringement threats, and were intended to be (and were) understood in that context. The patent infringement threats were therefore an inextricable part of the effect of MGA’s refusal to supply policy.

[492]This case therefore falls into the category of cases where, if loss can be established, that loss was the result of two effective causes which combined to produce the outcome relied upon as causing loss to the claimant. On that basis, if anything had turned on the point, I would have rejected MGA’s causation argument and would have found that the patent infringement threats were causative of any loss suffered by Cabo. Whether Cabo did in fact suffer loss as a result of MGA’s conduct is the next question.

Actionable damage and causation: Cabo’s heads of loss

[493]It follows from the discussion above that the starting point in considering causation of loss is to determine what heads of loss have been pleaded by Cabo.

[494]Cabo’s primary pleaded head of loss was for lost profits in respect of lost sales to toy traders (including but not limited to the UK launch retailers), and lost sales and licensing revenues more generally, extending to lost profits in other jurisdictions. The question for the first stage of the assessment of damages (whether for the abuse of a dominant position or patent threats under s. 70 of the 1977 Act) is therefore whether Cabo has established, on the balance of probabilities, that but for MGA’s conduct it would have traded profitably. The amount of the lost profits is then a matter for the quantification stage of the assessment.

[495]Cabo’s secondary pleaded head of loss was loss of value to the business that it would have been able to build, in so far as that loss was not already captured in the assessment of lost profits. As discussed below, the parties’ quantification models ultimately did not assess business value separately from lost profits, but considered both together under the framework of the DTM. No separate causation question therefore arises in this regard. In principle, however, I agree with MGA that if Cabo had advanced a separate and independent claim for loss of value to the business, it would have had to establish on a balance of probabilities that there would have been a valuable business in the counterfactual scenario.

[496]During the course of the trial, Cabo relied on various other types of loss which, it argued, were actionable heads of loss. These arguments were mainly advanced in support of a premise which was not pursued by the end of the trial and which I have rejected above, namely that establishing causation of a single head of loss was sufficient for the court to proceed to the quantification of damages, without needing to show causation on the balance of probabilities for any other head of loss. In any event, none of the alternative heads of loss stood up to scrutiny.

[497]First, Cabo claimed that it had lost revenue from orders that were cancelled by the launch retailers. That was, however, not Cabo’s pleaded case: the pleaded case was one of lost profits arising in respect of lost sales, including sales to the launch retailers. Consistent with that case, Mr Colley and Mr Parker modelled lost profits generally (with their estimates under different assumptions forming the basis of the cashflow projections in the DTM), and did not separately quantify lost revenues from cancelled orders. Indeed Mr Colley specifically confirmed in the second Colley/Parker joint expert statement that he had not considered this. Mr Parker addressed the matter only very briefly, noting that while there might have been “potential losses” from reductions in sales to the launch retailers, Cabo did in fact sell all of its stock, so any initial reduction in sales ultimately resulted in no net volume reduction. Indeed, as discussed further below, Cabo’s sales of Worldeez to B&M (as well as to several other smaller retailers) gave Cabo a larger margin than it would have made from supplying Worldeez to The Entertainer.

[498]Secondly, Cabo relied on additional expenses incurred as a result of MGA’s intervention, arising from the cost of rewrapping the globes in a different colour packaging, and repackaging some of the stock as blind bags for sale in The Entertainer. Again, however, no such expenditure is pleaded as a head of loss. Had it been pleaded, there would have been a further problem with this claimed loss, which is that the expenses of producing the Worldeez products were defrayed by Singleton, rather than Cabo, with Cabo incurring a liability in debt to Singleton. That liability was then written off under the July 2019 settlement agreement between Cabo and Singleton. In so far as necessary, MGA sought permission to amend its defence to rely on the settlement agreement in support of its submission that any loss suffered by Cabo had been avoided, and Cabo consented to that amendment. Ultimately, however, the issue fell away because Cabo did not pursue the repackaging point as a separate head of loss in its oral closing submissions.

[499]Finally and for completeness, there was some suggestion by Cabo during the trial that it would be sufficient (for the purpose of completing its competition law cause of action and/or establishing causation more generally) for it to show that it would have been less loss-making in the counterfactual case than it was in fact. Again, however, no such claim was pleaded or ever particularised during the trial, and Ms Kreisberger confirmed in her closing submissions that Cabo’s claim was not brought on this basis.

[500]Cabo’s damages claim therefore stands or falls with its claim that it would have traded profitably but for MGA’s conduct. That is therefore the next question to address. Whether Cabo would have traded profitably in the counterfactual case

The parties’ submissions

[501]It is important to note at the outset that the evidence of the economic experts (Mr Colley and Mr Parker) was not directed at the question of whether, on a balance of probabilities, Cabo would in fact have been profitable in the counterfactual scenario. Rather, what they did was to model the potential profits that Cabo would have obtained if it had enjoyed various levels of success, based on comparator products. The question of whether Worldeez would have been as successful as the various different comparators selected was, they recognised, ultimately not a matter falling within their expertise. Mr Davies’ valuation evidence likewise did not assume any particular level of success, but simply commented on the appropriate approach to the valuation aspects of the modelling.

[502]The evidence as to Cabo’s profitability in the counterfactual scenario therefore came from the factual witnesses and the toy industry experts. Cabo’s submission was that Worldeez would have been highly profitable, but for MGA’s conduct. It contended that Worldeez was an attractive and appealing toy, with a good brand name, an innovative unboxing experience, well-executed figurines and a strong marketing campaign. Cabo submitted that Worldeez was comparable to toys such as Hatchimals Colleggtibles and 5 Surprise Mini Brands, and was also comparable to Shopkins in many respects (and indeed Cabo claimed that Worldeez was potentially superior to Shopkins in some respects).

[503]Cabo relied on the successful initial launch of Worldeez in B&M, the evidence of the witnesses of fact as to the appeal of Worldeez, and the evidence of retailer support prior to MGA’s intervention. In addition to the factual evidence, Cabo relied on Ms Munt’s view that Worldeez would have been successful in stores. On that basis, Cabo contended that Worldeez would have enjoyed commercial success upon launch in the UK, followed by international sales and licensing deals.

[504]MGA submitted that Worldeez was not a particularly innovative product and had various design and manufacturing defects. More importantly, however, MGA said that however good a product Worldeez was, Cabo lacked the operational expertise necessary to turn it into a commercially successful and profitable business. In particular, MGA said that Cabo had done no consumer research, had no formal business plan or financial projections, was paying too much for the product, lacked experience in inventory management, and did not have sufficient working capital support from Singleton. Mr Harper’s opinion was that these problems meant that Worldeez would not have been commercially successful in the UK whatever level of retail distribution was achieved. MGA also relied on a breakeven analysis showing that on a range of assumptions as to the appropriate level of costs, Cabo would have required huge sales volumes merely to break even in 2017, which would not realistically have been achievable.

[505]On that basis, MGA submitted that there was no realistic likelihood of international and licensing revenues. In any event, however, MGA submitted that the international distribution routes would not have worked for Cabo given its cost model and the lack of sufficient working capital. Finally, although (as explained above) the economic experts did not opine on the likely commercial success of the Worldeez product, MGA said that its submission as to the likely unprofitability of Cabo was supported by Mr Parker’s quantum scenarios.

[506]It is appropriate to consider in turn the various factors relied upon by the parties. One preliminary overarching point should, however, be made. For the reasons set out at §§464–466 above, I do not accept Cabo’s submission that because the effect of MGA’s conduct was that Worldeez’ market access was significantly hindered, such that there are no hard data on what an “unhindered performance” would look like, the application of the principle in Armory v Delamirie means that Worldeez should be assumed to have “the highest level of success”, as Ms Kreisberger put it in her closing submissions. The fact that the assessment of causation of loss in this case is based on an assessment of counterfactual profitability, given the nature of the infringement, does not mean that every disputed point must be assessed in Cabo’s favour. Nor (unlike the situation in Armory v Delamirie) is this a case where nothing is known about profitability in the counterfactual case. On the contrary, there is in this case very extensive factual and expert evidence from which conclusions may be drawn as to the likely counterfactual success of Worldeez. My assessment of that evidence is set out in the discussion below.

Product quality

[507]It is common ground that Cabo did not carry out any formal consumer research before designing and launching Worldeez. That is not, however, fatal to its chances of success. The more important question is the quality of the product which Cabo ultimately produced. As to that, Cabo relied on the evidence of both witnesses of fact and the expert evidence.

[508]Starting with the factual evidence, it is clear that some of the toy retailers and other industry players were very positive when they were first presented with Worldeez (see in particular Start Grant’s comments at §94). It is, however, necessary to put the evidence of the initial reactions to Worldeez in context. Many of the individuals who first saw and commented on Worldeez were friends of Mr Sivner and Mr Lazarus, and wished to encourage Cabo as a new entrant into the toy manufacturing business. That was undoubtedly the case for the Grants at The Entertainer, and Mr Garnham. They were therefore predisposed to be supportive of the product. As Mr Garnham said, he liked to encourage entrepreneurs in the industry, and described his approach as being one of “extreme positivity” in this context. Similarly Mr Grant said that Mr Sivner and Mr Lazarus were close friends “who I have always chosen to support with any product range that they wanted to launch”.

[509]There is less reliable evidence as to the views of TRU and Smyths regarding the product quality of Worldeez. Mr Smyth’s witness statement stated that he did not consider Worldeez to be of a quality that would appeal to him as a retailer. As discussed at §147 above, however, while Smyths did not place an immediate order for Worldeez, the contemporaneous emails indicate that it was about to do so prior to MGA’s intervention. There is no evidence from TRU itself as to its reaction to the product, and for the reasons discussed at §§139–142 I do not consider Mr Larian’s account of a conversation with Mr Macnab to be reliable. As with Smyths, it is apparent that TRU was interested in stocking Worldeez, even if there is nothing suggesting that it was as enthusiastic as The Entertainer.

[510]The factual evidence therefore indicates that the main toy retailers perceived Worldeez to be good enough to stock, or at least consider stocking, when it launched. That evidence does not, however, address the likely appeal of Worldeez compared to other comparable products on the market. On that point the best evidence before the court was that of the toy experts, who considered in detail the features of Worldeez and the extent to which those features were likely to enhance or reduce the appeal of Worldeez.

[511]The toy experts agreed that Worldeez was an appealing toy, which tapped into the then-unfolding craze for unboxing collectibles, and which had a good brand name. They also agreed that Worldeez had well-executed figurines with good tactile appeal, appealing packaging and an extended unwrapping experience which included the use of a key to unlock the chamber of the globe which contained the figurines. The toy was lightly educational, and appealed to both genders although somewhat more to girls than boys.

[512]While Mr Harper accepted that the packaging (including the key to unlock the lower compartment of the globe) was an innovative element of the globe, which would appeal to children, he considered that the figurines themselves were not particularly innovative – and as noted above, the experts agreed that having an innovative product is critical to success in this industry. Most obviously, the Worldeez figurines were very similar to Shopkins. Ms Munt noted the similarities between Worldeez and Shopkins, and agreed that these made Worldeez less innovative than other collectibles in the market.

[513]It is apparent from the internal contemporaneous correspondence that Cabo’s designer, Mr Olivier, repeatedly warned the Cabo founders about the similarity with Shopkins, during the design process in 2016 and in March 2017. Of particular note was an email in September 2016 from Mr Olivier to Mr Cohen, expressing his concerns about the design direction:

“I can’t stop thinking about how this change of direction as bring Worldeez close to Shopkins, they are all unique and the concept is far from what they have implemented in the market, but at the ‘end of the day’ we have collectible toys that are characters within similar dimensions and with the same material … I am really concerned about how this will be translated in terms of intellectual property, but most important, I have no doubts about the potential of the concept, a potential that Shopkins as your competitors will see and soon or later a similar concept will be implemented by moose toys. … I really am liking this direction, but concerned about the implications that it may have considering the strong competitor that we have ahead.”

[514]Mr Michaelson’s response was revealing:

“I think we expected this? I think the important thing would be to get it in store without anybody seeing it at trade shows etc and feeding back to Moose. That is why I think the best approach is to make sure Marc gets it in to Smyths and the Entertainer … get big sales and then follow up with worldwide distribution.”

[515]When cross-examined about this, Mr Michaelson gave an evasive answer. His intention was however, transparent on the face of his email: he was hoping to “get big sales” on Worldeez before Moose Toys, the suppliers of Shopkins, objected. What he did not appear to have considered was whether the similarity with Shopkins would reduce the innovative quality and hence dampen the consumer appeal of Worldeez from the outset.

[516]The world theme itself was also not unique. Moshi Monsters had previously launched a range of products in 2010 called “Worldies”, with figurines similar to the Worldeez figurines that were later developed by Cabo. The trial bundle included the following picture as an example:

[517]The Shopkins brand also included world themed products. In addition, the Gift’ems toy (which was launched in August 2016) featured a range of surprise collectible doll figurines which related to different cities from around the world. As with Shopkins, the Cabo founders were aware of the similarities between Worldeez and Gift’ems during the design process: on 15 August 2016 Mr Sivner wrote a dismayed WhatsApp message:

“Gutted boys please read it’s the same concept as worldezz”

. Mr Michaelson sought to brush this off, saying that the Gift’ems doll was just a “city doll” with minimal reference to the country represented. Mr Sivner was not, however, persuaded, replying “The concept is the same mark so will the retailers buy both??”. His comments were, as it turned out, prescient: when Worldeez was launched, Tesco was apparently uninterested because it had previously taken Gift’ems, which had sold badly (see §103 above). The world theme collectibles already offered by Shopkins and Gift’ems were also noted by one of the toy companies approached by Cabo to discuss international distribution (see §639 below).

[518]Aside from product innovation, the other factor considered in some detail by the experts was the play value of the product. Ms Munt considered that the Worldeez toys offered considerable potential for imaginary play, referring to the key unlocking element of the packaging and the characters for each featured country. She accepted, however, that Worldeez did not reach the same level of sophistication as LOL Surprise, which was a larger mini-doll with far greater play value. Mr Harper’s view was likewise that Worldeez had more limited play value than LOL Surprise, given the numerous additional features of the LOL Surprise dolls: their limbs and head could be moved, they could be dressed and undressed and fed with water from the bottle, the dolls could either cry, spit or pee, and some dolls also changed colour when placed in water.

[519]Mr Harper agreed that the unwrapping experience of Worldeez, including in particular the key unlocking element, was engaging, and when put to him in cross-examination he agreed that this was better than the packaging of Shopkins. But he thought that Worldeez had more limited play appeal than Shopkins. That was, in particular, Shopkins featured accessories such as a shopping trolley which formed a central part of the play experience, as reflected in Shopkins’ marketing. By contrast, while the Worldeez range included a scooter, a car and a bus, these were figurines rather than accessories. The Worldeez play value was therefore largely focused on the unwrapping of the product and the collection of the range of figurines. His ultimate opinion was that Worldeez would never come close to matching Shopkins’ success, both for product/marketing and commercial reasons. There was no contrary evidence from Cabo on that point (as discussed below, Ms Munt’s evidence on Cabo’s likely commercial success was very limited).

[520]Mr Harper’s written evidence also compared Worldeez to various other collectible brands on the market around the same time as Worldeez. These included, in particular: i) Mash’ems/Fash’ems, which are small, squishy collectible toys packaged in plastic spherical capsules. The toys are themed around known brands such as My Little Pony (pictured below) and Disney’s Frozen. The characters were larger than Worldeez, and Mr Harper considered that they would offer more play value than Worldeez. He expected this brand to be more successful than Worldeez. ii) Num Noms, described and pictured at §240 above. Mr Harper considered this to be a superior product because of its various play features, and said that Worldeez would “not come close” to matching its success. iii) Hatchimals Colleggtibles, also described and pictured at §240 above. Mr Harper said that the collectible animal characters were “cute”, and that the brand would have benefited from the success of the original Hatchimals product (a much larger plush product priced at £60). He would therefore have expected this to be more successful than Worldeez. When cross-examined on the point, his evidence was that Worldeez was “comparable” as a product to Hatchimals Colleggtibles, but he remained of the view that he would have expected Hatchimals to do better than Worldeez, emphasising Spin Master’s experience of running a business. iv) Gift’ems, described at §517 above. Perhaps unsurprisingly given the design of that product, Mr Harper considered that this product was, among the collectibles that he was considering, the closest to Worldeez due to the world/travel theming, the surprise unboxing and the fact that both products featured dolls from different countries (Worldeez) or cities (Gift’ems). He therefore considered that Worldeez and Gift’ems would have had similar appeal to the target audience.

[521]Ms Munt did not compare Worldeez to Mash’ems/Fash’ems or Num Noms, and said very little about Hatchimals other than noting that Worldeez was more comparable to Hatchimals than LOL in terms of sophistication. She agreed that Gift’ems was a similar world-themed collectible, albeit that it was based around a mini-doll rather than products from the countries represented, and she noted that Gift’ems had particularly sophisticated packaging which showcased the doll.

[522]Overall, therefore, the evidence as to the quality of Worldeez indicated that Worldeez was an appealing and well-executed collectible toy, but that its main innovative element was its packaging. The figurines themselves were very similar to the Shopkins concept, and the world them had been used already in multiple previous collectible toys. The weight of the evidence indicated that Worldeez had more limited play value and was likely to be less successful than LOL, Shopkins, Mash’ems/Fash’ems and Num Noms, and that Hatchimals, while a comparable product, was more successful than Worldeez would have been. The most comparable product was perhaps Gift’ems. Notably, that was not a particularly successful product, ranking 37th among brands of playset dolls and accessories in 2017 based on the available NPD data.

[523]There was some debate as to various product defects which emerged with Worldeez. There was a known issue with the perforation of the packaging, which was hard to open. This was noted in various of the WhatsApp chats between the Cabo founders, including a comment on 8 June 2017 from Mr Michaelson that “[w]e could do with getting Tiana a Stanley knife”. This was not, however, a problem unique to Worldeez; indeed there were videos of LOL Surprise being unwrapped, where the unboxers found it difficult to remove the wrapping. Cabo did not, moreover, receive any customer complaints about the ease of unwrapping. I do not consider this to be a major issue with the appeal of the product.

[524]A more significant issue was the instances of globes containing two identical figurines, which the Cabo founders were aware of prior to the launch of Worldeez. Mr Michaelson sent an email to Cabo’s marketing agent on 17 April 2017 ahead of the filming of the first Tiana video, saying that Cabo was due to receive a shipment of globes from China, but “One major issue is that the 2 mystery items in each globe are going to be identical!” This issue did not appear to have been resolved (or at least fully resolved) by the time Worldeez was launched in July, because Cabo did receive subsequent complaints from customers about this. While it is entirely normal that purchasers of collectible toys will end up with some duplicates in their collections (as Mr Mowbray noted in his evidence), the problem in this case appears to have been that the instructions to the Chinese factory did not make sufficiently clear (at least initially) that each globe was to contain different items.

[525]There were also customer complaintsregarding globes containing no figurines, and broken globes. It appears that Mr Cohen discussed problems with the globe opening mechanism with the Chinese factory which manufactured the products, but this could not be improved for the initial batch of globes which had by then already been finished.

[526]Theevidenceof customer complaints itself should not be overstated: the overall number of customer complaints was small. The main issue, however, is that the materials before the court, including internal correspondence, indicates that the Worldeez toys were rushed to the market without sufficient quality control in advance, leading to problems being addressed in the course of the production process after initial batches of the globes had already been manufactured.

Marketing campaign

[527]Cabo placed great emphasis on its marketing campaign for Worldeez. In particular, it had a very successful online marketing campaign through the use of videos by Tiana, with a particularly popular helicopter video (see §158 above), as well as other videos which also attracted millions of views. In addition to the sponsored videos, there was also an unsponsored video in which Tiana selected Worldeez as her favourite collectible. Mr Larian candidly accepted that Cabo’s online marketing was impressive.

[528]In addition to the Tiana videos, Worldeez produced an advertisement for use as a “pre-roll” on YouTube (i.e. shown before a user’s selected video), and a TV advertisement was produced by Diaframma to be shown on Nickelodeon. Cabo’s channel on Popjam attracted a large number of followers (apparently more than MGA’s Num Noms Popjam channel). There was also a wide range of further marketing materials including “mummy blogger” campaigns, marketing in children’s magazines, and coverage in the trade press.

[529]Cabo’s marketing of Worldeez therefore covered a variety of media sources and there is no doubt that it was an impactful marketing campaign, at least initially. What is less clear is whether Cabo would, in the counterfactual case, have been able to invest sufficiently to maintain the momentum required for the product to achieve not merely initial consumer excitement, but ongoing commercial success translating to overall profitability.

[530]Cabo’s total initial marketing budget, including both online and television marketing, was around £275,000. The Nickelodeon component of that, in the contract under negotiation in early 2017, was an agreement for airtime at a cost of £50,000 plus a profit share, and the cost of Diaframma producing the advert was £19,000. In the end, as explained above, the contract with Nickelodeon was not finalised and Cabo only paid the £50,000 minimum guarantee.

[531]There was something of a debate as to whether the television slots offered by Nickelodeon (and in fact taken by Cabo) were particularly good slots. Mr Michaelson claimed that they were not “bad slots”; Mr Cohen said that they were “very good” slots. That evidence was, however, vague and subjective. A more objective measure of the quality of the slots could be ascertained by considering the “rating points” (referred to as TVRs) given to each of the slots by Nickelodeon. Mr Harper explained that TVRs are a measure of the number of consumers in the target audience who will see the advert, expressed as a percentage of the total audience, multiplied by the number of times the advert was viewed by them. A rating of 100 TVRs would therefore enable the advert to reach approximately 50% of the target audience twice (50% reach x 2 = 100 TVRs). On that basis, Mr Harper calculated that an advertising burst of 150–200 TVRs (which he said would normally be spread across several weeks) would be needed to reach the majority of the target audience on several occasions. The first burst of advertising for Worldeez on Nickelodeon in July to September 2017, however, only provided 111.5 TVRs, and the subsequent advertising in October and November 2017 provided 24.8 and 36 TVRs respectively. Mr Harper’s assessment was that this was a relatively low weight television campaign.

[532]Ms Munt accepted that the number of TVRs was “quite low” but considered that it was also necessary to consider the quality of the advert (which she thought was very high) and the timing of the television slots. Quality of an advert is obviously a relevant factor in the impact of a marketing campaign. The point about timing of slots, however, missed the point which was that the timing of an advert is one of the factors used to calculate that advert’s TVRs, rather than a factor which might counterbalance a low TVR rating.

[533]The weight to be attributed to Cabo’s television marketing as opposed to online marketing was also disputed. Cabo submitted that its relatively low television budget was sufficient in the context of its focus on an online campaign. It relied on Ms Munt’s opinion that by 2017 the focus of marketing was starting to shift from television to online marketing. She considered that overall Cabo ran a high weight marketing campaign, based mainly on its online marketing which she considered to have been impressive and effective.

[534]Cabo’s submission was not, however, consistent with the contemporaneous advice which it had received from its marketing agents Weird Lime. In April 2017 they had said that the typical spend by major brands for television marketing alone (i.e. excluding online influencers) to launch their collectibles was between £300,000 and £3m depending on the brand. Subsequently in early May 2017, when the Nickelodeon contract was being negotiated, Weird Lime advised Cabo that “TVR is low … for a launch campaign” and that Cabo should increase the funding for the launch weeks by “at least £50,000” given the direct competition from other collectibles such as Shopkins and Hatchimals Colleggtibles. Mr Michaelson forwarded that to Mr Sivner and Mr Lazarus to ask if they were happy to increase the spend. The response from Mr Sivner was that Cabo should instead ask for “more tv for less money”. Whether or not Cabo made that request, it is clear that Cabo did not ever increase the Nickelodeon budget.

[535]As to the overall marketing budget, there was broad agreement between Ms Munt and Mr Harper that a toy supplier could have delivered a solid UK launch campaign for £200–300,000. The question was what budget would have been required to maintain substantial ongoing sales of the levels sought by Cabo. Mr Harper’s evidence was that Cabo would have needed to increase its marketing budget to achieve greater commercial success than it did. He would have expected a total budget of at least £500-600,000 if Worldeez was going to achieve the sales of £2–3m which Cabo had projected for its first six months. Ms Munt considered that that level of marketing budget would have been exceptional. She also pointed out that the quality of the marketing campaign was as important as the overall spend; and it was common ground that the marketing of different brands of collectible toys varied widely depending on whether their focus was on television or online marketing (and Mr Harper acknowledged that his expertise did not extend to the latter). Cabo also relied on the evidence of Mr Mowbray, to the effect that Zuru’s marketing spend was typically around 6–7% of revenue.

[536]Taking as a whole, while generalisations are difficult, the evidence indicates that Cabo’s television marketing budget was low for the launch of a successful new collectible toy. While that may have been to some extent compensated by Cabo’s impactful initial online marketing campaign, Worldeez was competing in a crowded market, at a time when (as Mr Harper confirmed) television was still the main advertising medium for children, albeit that the shift from television to online marketing had by then begun. The evidence indicates that Cabo’s overall marketing budget of around £275,000 was reasonable for a solid launch campaign. The problem was, however, that Cabo needed more than that: it needed very high levels of sales from the outset merely to break even during 2017 (see further below), and even higher sales to achieve the revenues that it had projected for that period. Mr Harper was, in my judgment, right to take the view that this would have required a far higher overall marketing budget than Cabo had available to it.

Retailer support

[537]Cabo’s submissions and evidence painted a picture of enthusiastic support for Worldeez by toy retailers, prior to MGA’s intervention. Once Worldeez was launched in B&M, Cabo contended that its initial sales were very high, providing an indication of how strongly it would have performed but for MGA’s conduct. As with other aspects of Cabo’s case, I consider that to be an over-optimistic characterisation.

[538]It is common ground that Stuart Grant at The Entertainer was very supportive of the launch of Worldeez. He was a friend of Mr Sivner, was keen to launch the product before other retailers, placed an immediate order for the product, and capitulated to MGA’s threats only after considerable resistance. It is also apparent that The Entertainer remained keen to stock Worldeez in September 2017 (see Gary Grant’s email to MGA described at §130 above). Stuart Grant confirmed in his cross-examination that he had been extremely supportive of Worldeez and considered that there was a significant opportunity for the toy.

[539]There was, however, no evidence of comparable support from other retailers. As discussed above, both TRU and Smyths were interested in stocking Worldeez, but unlike The Entertainer had not placed firm orders for the product prior to MGA’s intervention at the end of May. Tesco and Argos had not shown interest in taking Worldeez, even before MGA intervened.

[540]One of the problems with obtaining retailer support appears to have been that while Cabo had planned to launch Worldeez in May, the lead time required to get the product into stores had been badly planned. Argos was a particular example of that. Even if Argos had been interested, the timing of the Worldeez launch was in any event not early enough for the product to be included in the Argos autumn/winter 2017 catalogue. This seems to have been a problem with other retailers too. In a frustrated WhatsApp exchange between Mr Michaelson and Mr Cohen in mid-April 2017, Mr Cohen commented on a discussion with Mr Sivner and Mr Lazarus: “For example when we said the dates, all you can hear is Alex saying ‘dreaming. Living in cloud cuckoo land. … Then saying the retailers need time to list. Hello … we’ve been telling them this for fuckin months”.

[541]More generally, Mr Michaelson noted in September 2017 that “distribution has been badly managed. Even with the MGA issue. Grocers should have been approached many months ago.” Mr Lazarus accepted this in his evidence, acknowledging that the approach to retailers should have been planned more strategically. In particular, he said that Cabo had underestimated how hard it would be to get into supermarkets, and should have considered more carefully the timing of when supermarkets should be approached.

[542]Cabo was, however, able to secure distribution of Worldeez in B&M, through Mr Sivner’s contact with B&M’s owner Mr Arora. While Cabo claimed that it remained “strangled” by MGA, Mr Arora was willing to take the product and was evidently not deterred by MGA’s objection to it – most likely because, as noted above, MGA was not at the time supplying LOL to B&M and could not therefore threaten B&M with refusal to supply, as it had done with other retailers. It is common ground that Worldeez did, initially, enjoy good sales at B&M. The dispute is as to whether those sales were of a level indicating that Worldeez would in the counterfactual case have enjoyed durable commercial success, such as to render Cabo profitable.

[543]As to that, Cabo’s submissions as to the strength of the initial sales at B&M were, ultimately, not consistent with the weight of the contemporaneous documentary evidence. It is apparent that notwithstanding good sales on launch of Worldeez, the volumes of the globe sold remained far below B&M and Cabo’s expectations, leading Mr Arora to state on 19 July 2017 that if sales didn’t pick up he would have to have all the stock collected by Cabo (§164 above).

[544]Even when sales had increased to 3,500 a week at the end of July, Mr Sivner described that as “not fabulous”, and the Cabo founders were sufficiently disappointed to question whether the figures they had been given by B&M were accurate (§166 above). At their peak, sales were 4,500 a week around the start of August. Ms Munt described that as “really strong” sales, but that figure remained considerably below the 20,000 a week which (according to Ms White) Cabo had predicted. It is, moreover, clear that by the end of August sales had dropped off. As set out above, the 5- and 10-packs did not ever sell well. It is also notable that although The Entertainer did initially stock the 5- and 10-packs of Worldeez, and some of the blind-bagged product in place of the globe, its orders were ultimately relatively small. It did not re-order any of the 5- or 10-packs after July 2017, and placed its last order for blind bags in September 2017.

[545]At best that evidence indicates that the Worldeez globe was successful in B&M for around six weeks, but did not maintain consumer momentum thereafter; and the 5- and 10-packs were never successful. That is not indicative of a product with sufficiently strong appeal to go on to achieve a significant degree of commercial success. Rather, it is a typical pattern of a product which launches well on the back of an attractive marketing campaign but which ultimately does not maintain strong sales. Mr Harper gave several examples of innovative toys which had failed despite strong retailer support and good marketing campaigns, and Ms Munt accepted that more toys fail than succeed. Their comments were consistent with the evidence of Mr Garnham (“I’ve seen unbelievable product that my gut says ‘That’s going to fly’, and … it has gone into the bargain bin within weeks”) and Mr Grant (“there is plenty of product that gets mass distribution in every single retailer in the world and doesn’t sell”).

Business plan/financial projections

[546]Cabo launched Worldeez with no formal business plan or financial projections. The only financial projection was one put together for the purposes of the negotiations with Nickelodeon in March 2017. Mr Cohen claimed that the figures in that projection were conservative. In a WhatsApp message between the Cabo founders in April 2018, however, Mr Michaelson described the projection as “very ambitious”, and Mr Lazarus in his evidence said that the projection was “trying to big up the company” to pitch the deal to Nickelodeon. He said that the figures were “all over the place” and not based on any forensic science; rather “they were largely based on dream and hope from both Singleton and the Cabo team”.

[547]The evidence of Mr Lazarus was, on this point, entirely consistent with the other material before the court. I will address in turn the main elements of the financial projection, namely sales volumes, retail and wholesale prices, cost of goods, and the profit margin figures.

[548]Sales volumes. The toy experts agreed that forecasting demand was critical for suppliers to meet demand and to be profitable. Mr Cohen accepted that the likely level of sales was not a matter within the experience of the Cabo founders. The sales volume projections were therefore initially prepared by Singleton (likely Mr Lazarus), but those figures were not based on any market research or analysis of comparators. They seem, rather, to have been figures plucked from the air. Mr Cohen then significantly increased the 2019 projected sales figures without discussing this with Singleton, and with no analytical basis other than his view that he “would have expected growth”.

[549]No-one appears to have considered whether these projections were realistic, given the market conditions at the time. In fact, even the projections for 2017 (1.2m units of the globe in the seven months from June to December) would have made Worldeez the best-selling doll (by sales volumes) in the UK, ahead of both Shopkins and LOL Surprise. That was wildly implausible – there was nothing in any of the evidence to support the suggestion that Worldeez would have outperformed both of those phenomenally successful products. Mr Cohen’s projections of over 2.7m globes in 2018 and almost 3.4m globes in 2019 were even more far-fetched.

[550]Retail price. The projection assumed a retail price for the globe of £3.49. In reality, that was rapidly reduced to £2.99, with Mr Sivner explaining in a WhatsApp message on 30 March 2017 that £3.49 was above the market average for a 2-pack collectible, and that the price point needed to be a maximum of £2.99. The toy experts agreed that this was the right price. There was no evidence of any realistic prospect that the price could thereafter have been increased; indeed a 2018 email discussion with a potential distributor indicated that it was unlikely to be possible to raise the price after the £2.99 had been established on launch in 2017.

[551]Wholesale price. The projection assumed a wholesale price of £1.75. Mr Cohen, when asked about this, said that Cabo had discussed a lower price of £1.50, but had retained the figure of £1.75 in the marketing deck on the basis that “God loves a tryer. May as well try it at £1.75, and it would give us a bit of negotiation, if necessary, down to the £1.50.” The problem with a wholesale price of £1.75 was that it would have given a retail margin of only 30% by reference to the ex VAT retail price of £2.49. That wholesale price was only paid by one retailer who took a very small amount of Worldeez stock (Bargainmax). ToyTown paid £1.65, but again only took a small quantity of stock.

[552]No larger retailer agreed a wholesale price which gave a margin of less than 40%. B&M agreed a wholesale price of £1.50, which provided a margin of 40%. The Entertainer negotiated a £1.20 wholesale price, giving a retail margin of 52%. Mr Michaelson, who was responsible for attending sales meetings with the retailers, initially contended that this gave Cabo a “good margin”. When pressed on the point, however, he was unwilling to explain further, saying that he would leave it to Mr Cohen to respond. The reality was that he was well aware that the wholesale price negotiated with The Entertainer was unviable. He had commented in May and September 2017 that Cabo was “making nothing” from its sales to The Entertainer, and that “with the margin Entertainer were offered we would have made no money from them”.

[553]Cabo tried to get around that difficulty by describing the £1.20 price for The Entertainer as an “introductory rate” which would have been increased if Worldeez had performed well. But there is no evidence that this was what The Entertainer agreed. On the contrary, Mr Grant said that the margin of 52% was The Entertainer’s “absolute minimum” margin requirement. He was not challenged on this point, and Ms Munt confirmed that The Entertainer was well-known for expecting a minimum 50% margin.

[554]Cabo also contended that Smyths and TRU would have purchased the globe at £1.75, based on the fact that those were the wholesale prices presented to them on the marketing deck. The difficulty with that submission is that (as set out at §§96–102 above) while Cabo have expected orders to follow the meetings with Smyths and TRU, neither of those retailers did in fact place an order for Worldeez, so there was no agreement by either of them to any price. Nor is there any evidence of either Smyths or TRU confirming, even provisionally, that a wholesale price of £1.75 would be acceptable to them. As to the expert evidence on this point, the toy experts agreed that small suppliers tend to offer retail margins of 45% or more (which would have implied a wholesale price of £1.37 or less). Ms Munt’s evidence was that if a retailer did accept a lower headline margin, it would expect additional preferential trading terms such as discounts, rebates and promotional support.

[555]Ultimately, therefore, there was no evidence supporting a general wholesale price of £1.75. Rather the evidence indicated that the realistic wholesale price for the larger retailers was in the range of £1.20 to £1.50. The average wholesale price would therefore have fallen somewhere between those figures. A mid-point of around £1.35 is in my judgment a suitable figure to adopt.

[556]Cost of goods. The projection gave figures for the cost of goods as £0.75 for the globe, £1.60 for the 5-pack and £2.75 for the 10-pack, but MGA contended that the costs were in fact £0.82, £1.78 and £3.19. Whichever figures were accurate, there was a consistent concern that Cabo’s costs were too high when compared with the selling price of the product. The contemporaneous documents making this point included a message from Graham Mottram, who was assisting Cabo with sales in September 2017, suggesting that Cabo should have been paying $0.50 for the globe. That was also Mr Mowbray’s evidence.As discussed further below, this presented a significant problem for Cabo’s attempts to secure international distribution of Worldeez.

[557]Profit margin. The projection forecast a gross profit margin for Cabo of over 55% for every year covered by the forecast (2017 to 2019), and operating profit margins in the same years of 40–43%. Those figures were unrealistic. A wholesale price of £1.35 for the globe would have given Cabo a gross profit margin of 44%, assuming cost of goods as being £0.75 (and obviously a lower margin if the cost of goods was higher than that). As for the operating profit, Mr Harper’s evidence was that successful global toy companies would make an operating profit of in the range of 5–20% of net sales, with the best audited profit performance of the biggest toy companies being Hasbro in 2017 with profits of 16% of annual sales. He considered that for UK-only companies the profits were generally in the 5–10% range, and that this reflected the profit range of the smaller toy companies he had worked with. He was not aware of any individual brand exceeding an operating profit of around 24%. Ms Munt said that in her experience the operating profits of smaller toy companies ranged between 10–15%. Whatever the differences between the experts as to the precise ranges, it was clear from their evidence that Cabo’s projected profit margins were unachievably high. There is, moreover, no evidence that these forecasts were ever revised, or kept under review through (for example) regular P&L accounts or cash flow reports.

[558]The overall picture is of a set of financial projections which were implausible, indeed on some points wildly implausible, and which were never reviewed even when it became clear that the various elements of those projections were inaccurate and/or unachievable. That is indicative of a business with no robust financial control mechanisms – or indeed any clear structure for making financial decisions. None of the Cabo founders were paying serious attention to the accounts or P&L figures; rather, they appear to have left this to Singleton. But Singleton, likewise, was not exercising any real financial management; it was simply providing Cabo with the cash flow required during 2017. I agree with MGA’s submission that this was a serious problem for the viability of the business. As Mr Harper said, “any toy business that does not have a firm grip on its finances is unlikely to succeed”.

Inventory management

[559]The toy experts agreed that inventory control was a further factor that was critical for toy suppliers to meet consumer demand and to be profitable. As with the financial management of the business, the Cabo founders left inventory management to Singleton. But Singleton had no real experience of that and did not establish any systematic inventory control mechanisms. There appears to have been no advance inventory planning: no retailer forecasts were obtained at any stage, and Cabo produced no forecasts of its own. That meant that it was unable to plan the supply of stock, leading for example to an expensive decision to airfreight stock to The Entertainer for the intended launch of Worldeez in May 2017.

[560]While Mr Cohen was responsible for liaising with the Chinese factory producing Worldeez, he did not have access to the stock records, which were held by Singleton. He accepted in his oral evidence that no stock management system was ever shared with him. The lack of oversight of stock movements was also apparent from the WhatsApp discussions within Cabo in July and August 2017 questioning the sales figures they had been given by B&M (§166 above), as well as messages around the same time noting problems with stock deliveries to B&M and other retailers.

Working capital

[561]Beyond the initial capital contributions totalling £100,000 to establish Cabo (§82 above), the Cabo founders agreed that they would work free of charge “until the project was off the ground”, in return for Singleton covering the costs of the Worldeez business by way of loans. There was no formal agreement as to when those loans were going to be repaid. Instead, Singleton controlled Cabo’s bank account and was able to (and did) withdraw funds from that, generated by sales revenues, to repay its loans.

[562]The loan schedule attached to the July 2019 settlement agreement between Cabo and Singleton showed that Singleton started to withdraw significant funds from Cabo from August 2017 onwards. That deprived Cabo of working capital. This point was raised by Cabo’s licensing agent, Mr Dever, in January 2018, in the discussion referred to at §179 above. After a comment that “you must start to pick up revenue into Cabo or you may as well close it down”, he said that the loan agreement:

“… needs to be renegotiated. Repayment of loan in first position denies Cabo of working capital. Not all of that money needs to go back to Singletons right now. Cabo needs some revenue to survive until such time as the loan can be paid back in full given that you end up with a board and shareholders in a locked position. … If I could fund you out of this I would but it’s a Faustian deal whereby you have signed away your souls for a working capital loan.”

[563]Cabo’s ability to develop the business was, moreover, dependent upon the willingness of Mr Sivner and Mr Lazarus to provide further loan funds. That willingness was, however, likewise limited. While Cabo contended that Singleton could afford to support Worldeez in substantial sums, the evidence does not show that Mr Sivner and Mr Lazarus were willing to do so. On 10 August 2017 in a discussion about US distribution of Worldeez, Mr Sivner noted that “if we have to fund are selfs in us it will [mean] over 1 mill $. Can not afford”. Notably, that comment came at a time when sales of the globe were at their peak, following the Tiana helicopter video, and Mr Siver could therefore have been expected to be relatively optimistic about the product.

[564]The issue came to a head by September 2017, when it is clear that there was considerable frustration by the Cabo founders that Singleton was not willing to invest more in the continuation of the brand and was expecting to recoup its initial outlay more quickly than was feasible. The document prepared by the Cabo founders for a meeting with Mr Sivner (referred to at §178 above) reflected the tensions between the Cabo founders and Singleton, with comments that “£150–£300k is really not a lot to spend for IP like this” and “To expect break even in 2 months is ludicrous”.

[565]Cabo suggested that the attitude of Mr Sivner and Mr Lazarus, and their willingness to support Worldeez, had been tainted by MGA’s conduct. While there is no doubt that Mr Sivner and Mr Lazarus were concerned by MGA’s threats (see, for example, Mr Lazarus’ comments cited at §115 above), the September 2017 document indicated that there had been problems in the relationship between the Cabo founders and Singleton from the start of the project, before MGA intervened. The document noted, for example, that Singleton’s attitude had “Been a disrespect from the start – underestimated our abilities at not being able to achieve what we said we were going to – i.e. getting product ready for May”. The same document commented that the Cabo founders had “felt throughout” that Mr Lazarus was “not enthusiastic about the brand and only comes to life when there’s a problem”, and queried whether Mr Lazarus wanted to exit the business.

[566]Mr Lazarus in his evidence agreed that while he had agreed to be involved with Worldeez, he did not have the same enthusiasm for the project as the Cabo founders. He said that if he had genuinely had “money-is-no-object” faith in the Worldeez project, he would have been willing to support a multimillion pound spend on marketing from the outset, but he did not do so because he was not going to risk taking more than a “punt” on the business, and was “acutely aware of the potential for failure”. For those reasons, while he said that he would probably have funded production if there had been a guaranteed return on the investment, for example with a large purchase order placed by a retailer, he would not have been willing to invest large sums of money purely in the hope that sales would materialise. He explained the point further in his oral evidence:

“We are running a business at the same time. So our money is used in our business. … We took a punt with a certain amount of money in our heads and then we let it roll until we felt comfortable. I don’t have millions at my disposal, otherwise I’ll strangle my core business.”

[567]Those comments were consistent with the contemporaneous evidence set out above. There is no contrary evidence suggesting that either Mr Sivner or Mr Lazarus were willing to use Singleton’s funds to support a very large speculative investment in Worldeez, for example to finance international expansion. Mr Chacksfield suggested that there would in fact have been little risk to Singleton, because there would have been larger orders underpinning the cashflow requirement, with the effect that Singleton’s capital investment would have resulted in “free money”. That was, however, a matter of assertion: the evidence did not suggest that any working capital requirement would have been risk-free. I am therefore not persuaded that Singleton would, in the counterfactual case, have been prepared to commit cashflow support in significantly larger sums than it did in fact (the peak balance of Singleton’s loan to Cabo being less than £600,000).

[568]Nor, importantly, does the evidence indicate that Singleton was willing to commit funds to Cabo for anything beyond a relatively short initial start-up period. It is apparent from the loan schedule that Singleton was requiring repayment of substantial sums less than two months after Worldeez launched at B&M. The expectationor at least hope of Mr Sivner and Mr Lazarus therefore seems to have been that the business would quickly be profitable and therefore self-funding, such that they could rapidly recover their loans to the company, hence the comment in the September 2017 document, cited above, that Singleton expected “break even in 2 months”. As Mr Dever said, it was unfeasible for Cabo’s business to progress on that basis.

Toy expert evidence on commercial success

[569]Ms Munt considered that there was a high chance (which she put at a 60% chance) that Worldeez would have been a “success”. What she meant by that was explored at the end of her cross-examination. She confirmed that she did not mean a 60% chance that Worldeez would have been profitable, and she candidly accepted that she did not have sufficient experience of the financial and operational side of running a toy business to opine on Cabo’s financial and operational competence. What she meant was “creating a buzz” or “noise” with customers wanting the product and children talking about it. In response to further questions from the court, she said this:

“Q. You are talking about consumer excitement, but you are not saying profitability? A. I’m not necessarily saying profitability, but I am saying that … that noise … Q. All right. But if something is successful in your definition, you would include something that wasn’t profitable but had a lot of buzz on social media and sales? A. Yes. Q. So when you say 60% chance of success, you mean 60% chance of making a lot of noise and having sales even if not necessarily profitable? A. Correct.”

A. I’m not necessarily saying profitability, but I am saying that … that noise … A. Correct.”

A. I’m not necessarily saying profitability, but I am saying that … that noise …

[570]I consider that Ms Munt’s evidence on this point was vague and speculative. She did not have any concrete underpinning for her opinion on Cabo’s likely success. Nor was she able to provide any evidence on whether Cabo would in the counterfactual case have been able to turn a product which created a “buzz” in the marketplace (if it did indeed achieve that) into a profitable product.

[571]Mr Harper did, however, give evidence on that point. He explained that even with an innovative product and an effective marketing strategy, success of a new product in the UK is determined by “achieving strong enough financials to deliver a profit, accurate forecasting/inventory management and good retail distribution”, and that strong UK results could then in turn be used to drive international and licensing revenue. His evidence was that Worldeez would not ultimately have been successful in the UK, whatever level of retail distribution it might have achieved.

[572]That view was in part based on his assessment of the quality of the Worldeez product, but also (and importantly) took into account his assessment of Cabo’s marketing strategy and budget, and its overall operational capabilities. Put in percentage terms, Mr Harper estimated that the chances of Cabo securing sufficient sales in its first year in the UK market to secure year two retailer listings were well below 10% and probably as low as 1–3%, and said that this would have been likely to result in a “significant loss”, depending on the inventory commitment. He was not challenged on this latter point, nor was it suggested that he lacked the knowledge and/or experience to make this assessment.

Breakeven analysis

[573]In addition to the qualitative evidence set out above, MGA relied in its closing submissions on a breakeven analysis showing the revenues and sales volumes which Cabo would have had to achieve to break even in 2017, calculated by reference to MGA’s and Cabo’s alternative cost stacks, with (for each of those two alternative cost stacks) three different assumptions of the wholesale price of the Worldeez globe. For MGA’s cost stack the assumptions were £1.20 (the price paid by The Entertainer), £1.50 (the price paid by B&M) and £1.35 (the mid-point of those two prices). For Cabo’s cost stack the assumptions were £1.20, £1.35 and £1.52. The last of those figures represented Cabo’s position in closing submissions as to the average wholesale price which it said that it would have achieved, based on the blended average of the wholesale prices which were in fact paid by The Entertainer and Smyths, and the £1.75 wholesale price figure which it claimed that Smyths and TRU would have paid. The calculations then showed how, in MGA’s submission, the Worldeez globe would have had to perform against other collectible toys in order to break even during 2017 (i.e. assuming £0 profits), comparing both sales volumes and sales revenues.

[574]Because this analysis had not been produced by MGA until its written closing submissions it could not fully be addressed during the trial, and the parties’ oral closing submissions on the analysis were therefore supplemented by the post-trial written submissions filed on 17 February 2025. Those submissions confirmed that there was no dispute as to the accuracy of MGA’s calculations as such. Nor did Cabo take issue with the relevance of considering whether Worldeez could have reached a breakeven point by the end of 2017. I agree that this is a relevant and indeed important question to consider, in the assessment of Cabo’s likely profitability, given that there is no evidence suggesting that Singleton would have continued to fund Cabo if it remained unprofitable in 2018, nor any evidence of realistic alternative sources of funding in that event. Indeed, as set out above, it appears from the evidence discussed above that Singleton expected breakeven within an even shorter period.

[575]There was, however, a significant dispute about the reliability of MGA’s comparisons with other products. In particular, for both sales volumes and sales revenues MGA had used NPD data, which only covered 75% of the retail market. In addition, the data set used for the sales revenue comparison only covered the period January to November 2017. While MGA contended that these limitations were effectively offset by the fact that the Worldeez sales related to only a seven month period, a reliable comparison would need to take into account the strong seasonality of toy sales, and in particular the weighting towards Christmas, as well as the fact that other brands in the comparator data sets were also not on sale for the full year 2017. Cabo also disputed the reliability of comparisons with products at very different price points to the Worldeez products.

[576]These were points on which the economic experts might have been expected to opine, but since the breakeven analysis had not been produced until the closing submissions there had (obviously) not been any opportunity to explore this in the expert evidence. While MGA said that its analysis had been prepared with “expert input” (presumably Mr Parker or BRG), the precise nature of that assistance was not disclosed. I do not, therefore, consider that I can place any significant weight on the part of the breakeven analysis which ranked Worldeez against other toy products.

[577]The remainder of the breakeven analysis is, however, informative, and is as follows. 582. Table 5: Volumes and working capital required to break even in 2017 Cost stack/wholesale prices Breakeven volumes (all Worldeez) Globe breakeven volumes Working capital required (£) MGA cost stack/£1.20 2,678,431 2,299,050 3,910,679 MGA cost stack/£1.35 1,313,330 1,127,306 2,076,039 MGA cost stack/£1.50 869,948 746,726 1,480,152 Cabo cost stack/£1.20 917,026 787,136 1,273,711 Cabo cost stack/£1.35 686,227 589,027 1,035,955 Cabo cost stack/£1.52 554,009 475,427 938,162

[578]When considering the competing cost stacks, it should be noted that by the end of the trial there remained significant disputes between the parties on numerous points which led to the very different breakeven figures set out above. The disputed issues included not only the wholesale prices and cost of goods, which I have addressed above, but also a dispute as to the payments that would have been due to Nickelodeon had a TV advertising contract been agreed on the basis discussed by Cabo and Nickelodeon. The latter point had a particularly significant effect on the cost figures, because MGA’s contention was that Cabo would have agreed a profit-share contract which provided for Nickelodeon to receive 6.5% of retail revenue less Cabo’s costs of manufacture and distribution, whereas Cabo contended that the agreement would have been for Nickelodeon to receive 6.5% of wholesale revenue less costs. In addition, there were various other disputes on smaller points such as whether Singleton was entitled to a management charge, and the costs of product test, samples and depreciation. It was not, however, possible to assess the impact of any individual specific disputed cost items on the breakeven analysis, because MGA’s calculations were simply provided on the basis of a binary choice between its cost stack and Cabo’s cost stack.

[579]Fortunately, however, that does not matter, because the figures set out above show that even on assumptions favourable to Cabo, Worldeez would have had to be a highly successful product in order to break even during 2017. The evidence does not, however, indicate that Worldeez could have achieved success of that magnitude. In particular: i) On Cabo’s cost stack and an average wholesale price for the globe of £1.52 (which I consider to be too high for the reasons set out at §§551–555 above), Cabo would have needed to sell over 475,000 globes during 2017. That translates to an average of over 15,000 globes a week throughout June to December 2017. As described above, even in the immediate aftermath of a hugely successful marketing video with Tiana, resulting in peak sales at B&M during the August summer holiday period, the best that Cabo achieved was sales of around 4,500 a week at B&M. There is no evidence supporting the suggestion that even if Cabo had been on sale in the three main specialist toy retailers, it could have achieved (on average) over treble those sales for the entirety of June to December 2017. ii) At a more realistic average wholesale price of £1.35 (see §555 above), and again using Cabo’s cost stack, Cabo would have had to sell over 589,000 globes in 2017, or on average over 19,000 a week, which would have been an even more unattainable figure. iii) On any of the assumptions in the table above, Cabo would have needed very considerable capital (i.e. cashflow) funding from Singleton just to break even during 2017. Notwithstanding Mr Sivner’s apparent enthusiasm for Worldeez, and the family ties between Mr Sivner and Mr Michaelson, it is very improbable (particularly considering the evidence set out at §§563–568 above) that Singleton would have committed sums of a million pounds or more just to break even in 2017, let alone the larger capital sums that would have been required to tip the business into profitability.

International sales

[580]If Worldeez would not, in the counterfactual case, have been profitable in the UK, then (as was common ground) there would not have been a realistic prospect of international sales. In any event, however, it was clear from the evidence that Cabo’s hopes of penetrating the international market were highly optimistic, given the competition from established brands such as Shopkins. While Cabo had some discussions with Spin Master regarding US distribution, in June 2018 Spin Master said that Worldeez would be a “tough sell”, not primarily because of MGA but “Just on trends etc Shopkins starting to lose steam”. The international toy company Jay@Play likewise told Mr Sivner in August 2017 that the market for world-themed collectibles in the US was “very crowded with Shopkins leading the way with their World Vacation w little dolls and collectibles … Also Gift Ems by Jakks is all about all of the countries and it is marked down at Walmart”. In a transcription of a call between Mr Michaelson and Mr Cohen and an international toy agent, Tom Hodgkins, who had been discussing Worldeez with various international toy distributors, Mr Hodgkins made similar comments:

“… up there you have a healthy MGA and a healthy Shopkins so, there’s not a lot of room at retail for us to wedge ourselves in. That’s the other big problem that we’ve got overall, is just the landscape right now, we’re coming at the end of what’s been a couple years’ worth of new girls’ collectible items being launched.”

[581]A further problem was the question of how Cabo would distribute its products internationally. On that point, there was significant debate during the trial and in the post-hearing submissions as to whether international sales would have been made through the use of distributors, which would have borne the inventory risk, or through local sales representatives, for which Cabo would have had to bear the inventory risk, with a consequent increase in its requirements for working capital.

[582]Cabo’s position was that a distribution model would have been feasible. The evidence at the trial did not support that position. Rather, the consistent picture which emerged was that an international distribution model would have been unfeasible for Cabo, because Cabo’s cost prices were too high for it to be able to offer a suitable wholesale price to distributors.

[583]In the September 2017 internal document referenced at §178 above, Mr Michaelson and Mr Cohen said that “We were shocked that in last meeting Marc S said he spoke to ‘all’ distributors who have said the model doesn’t work”. The comments in the same document indicated that Mr Michaelson and Mr Cohen did not agree with that assessment, but the contemporaneous documents bear out Mr Sivner’s reported statement, with numerous examples of Cabo being told that its prices were simply too high. These included a comment from Jay@Play that “The prices do not work” for the US market, and a message from Mr Mottram reporting on discussions with various international distributors, saying that “there was a hard stop when we went over pricing … it will not work as is … pricing is an issue (as we spoke when I did the USA model also) in particular the globe … respectfully we need to re-look at costs and sell prices if we wish to generate sales”. Cabo was likewise told in March 2018 that the “costing simply [doesn’t] work for the Spanish market”.

[584]When asked what pricing would work, Goliath Games, a big international distributor, said that the FOB price should be under $1. Mr Mottram similarly suggested that Cabo should be paying $0.50 for the globe and selling it for $1 to distributors. Mr Cohen suggested that this would be possible, but Mr Mottram replied “Johnny – have to ask respectfully – how can you sell Globe to distributors at 1.00$ if your cost is 97 cents?” Mr Cohen suggested in his oral evidence that higher prices to distributors could have been negotiated if Worldeez had been more successful in the UK, giving it more clout. That was, however, pure speculation: Mr Cohen accepted that he had no evidential basis for that claim, and he did not offer any reason why prices that he was told categorically did not work for distributors might have become viable in the counterfactual case.

[585]Cabo could undoubtedly have reduced its costs slightly by placing orders for larger quantities. But a WhatsApp discussion on this point in August 2017 between the Cabo founders confirmed that even with volume discounts the factory costs would only come down by 5–8 cents, which would still have put the cost price at significantly above what Cabo had been told would be viable. Mr Michaelson then asked whether “the distributors are being unrealistic? Is it same story with rest of the world?” Mr Cohen replied, “All the dist are saying the same”. Mr Michaelson then responded “If they are all saying the same we could bite bullet and hope for good licensing deals”. That comment was revealing: it indicated that already, by that stage, Cabo was aware that it was unlikely to make any significant profits from international sales using a distributor approach.

[586]Mr Harper addressed this point in his evidence, and considered that Cabo’s profit margins would have made it difficult for Cabo to secure international distribution deals even if it had met its UK sales projections. At the trial he referred in particular to the US market of which he had recent experience. On the basis of that experience, he thought that Cabo would have had difficulty in securing US distribution due to the generally lower toy prices in the US than in Europe, and the consequent more limited margins available for Worldeez. While he suggested that it might be easier to secure distribution agreements for Australia and European markets, he noted that securing European distribution was administratively heavy, and South America had limited sales potential. His evidence was, therefore, broadly consistent with the factual evidence at the trial.

[587]Ms Munt, for her part, did not consider whether Cabo’s margins were sufficient for it to expand overseas. She did, however, consider that a sales representative international distribution model would generally have been more appropriate for small to mid-sized suppliers. The problem with that model in practice, however, was the increased cash flow requirement, which in Mr Harper’s view meant that sales representative arrangements would have been unfeasible for Worldeez.

[588]At the outset of the trial the economic experts accepted Mr Harper’s position on the viability of sales representative arrangements, and therefore did not include calculations for a sales representative model in the initial version of the DTM. After the trial, however, Mr Colley provided estimates for Cabo’s potential international business under a sales representative model, which were incorporated into the final version of the DTM. His evidence on this point had, obviously, not been explored at the trial. In any event, that evidence indicated that international sales under the sales representative model would have required additional working capital of over £5m for the US alone (and more if other countries were included). There was no evidence suggesting that this would have been feasible for Cabo; quite the contrary, the contemporaneous comment from Mr Sivner at §563 above indicates that this would not have been remotely affordable.

[589]Cabo’s answer to this was to suggest that cashflow requirements to sell into the US on a sales representative model could have been offset by revenues generated from international sales outside the US on a distributor model, as well as UK sales and licensing revenues. There was no evidence as to how that might have worked from a cashflow perspective (and it was not modelled even in the final version of the DTM). Moreover, that analysis obviously presupposed that it would have been feasible for Cabo to sell to non-US markets under distribution arrangements, which the factual evidence set out above indicated was unlikely. It also presupposed cashflow from UK sales, which was also unlikely given the analysis above.

[590]I do not, therefore, consider it at all likely that Cabo would have been able to generate international sales in the counterfactual case.

Conclusions on whether Cabo would have traded profitably

[591]Having regard to all of the factors set out above, my conclusion is, unequivocally, that in the counterfactual case Cabo would not have traded profitably. There is no doubt that Worldeez was in principle an appealing product which had an initially successful marketing campaign (particularly online). But it was competing in a crowded market for collectibles, and needed to be more than that to succeed. Unfortunately, it is apparent that the Cabo founders and Singleton were naïve and inexperienced, and lacked the operational capabilities which would have been required to provide a better chance of commercial success. The business was launched with hopelessly unrealistic financial projections, which were never updated; there was no real inventory control or stock management system; the cost of goods was too high for the product to be profitable without extraordinarily high sales volumes; and Singleton was not willing to provide the working capital that would have been required to support a stronger UK sales campaign, let alone international distribution. It is also apparent that, ultimately, the product was simply not as popular as Cabo had hoped and predicted. Despite the promising initial sales of the globe in B&M, the 5- and 10-packs never sold particularly well, and the sales of the globe quickly dropped off after the August 2017 peak. That is indicative of a product which did not have the enduring consumer appeal required to maintain sales momentum after the initial marketing push.

[592]In my judgment, therefore, Cabo could not in those circumstances have achieved the domestic sales volumes it needed to break even in 2017; there was no evidence that Singleton was prepared to continue to invest in the product if it remained unprofitable beyond that point; and there would therefore have been no realistic prospect of international sales.

[593]Cabo’s claim for damages therefore fails. It follows that it is not necessary to consider the quantification of Cabo’s loss. It is, however, appropriate to address the quantum models relied on by the parties (Mr Colley’s models for Cabo, and Mr Parker’s models for MGA), not only because there was considerable debate about these both throughout the hearing and in the post-hearing submissions, but also because an assessment of those models supports the conclusion that I have reached as to the likely profitability of Worldeez in the counterfactual case. The parties’ quantum models Preliminary comment on lost profits v valuation approaches

The parties’ quantum models

[594]Before considering the specific quantum models of Mr Colley and Mr Parker, it is necessary to address a preliminary dispute that arose as to the basis on which that modelling was conducted.

[595]It was common ground between the economic and valuation experts that 1 January 2018 should be taken to be the date on which Cabo exited the market. That was also consistent with Cabo’s position on the evidence, namely that by early 2018 it was clear that the brand had failed, and that no major retailer was going to stock it. The parties disagreed, however, as to how the date of demise should be reflected in the modelling of Cabo’s losses.

[596]Cabo’s position was that notwithstanding the amendment of its pleaded case to include a claim for loss of value to the business (as described at §43 above), its claim for loss comprised a single head of loss in the form of lost profits. This was not always easy to understand, as Mr Colley’s explanation of what he was doing shifted somewhat over time.

[597]Mr Colley’s ostensible approach was to model Cabo’s lost profits for two time periods:(i) the first five years up to June 2022, which was his assumed product life of Worldeez; and

(ii) a “perpetuity value” for the business, expressed as a capital sum, to reflect the profits that Cabo would have earned thereafter, whether on Worldeez or other products. Mr Colley initially described the perpetuity valuation element of his modelling as reflecting the ongoing ability of the business to generate cashflow, or (looked at a different way) the value which the business would have generated had it been purchased in July 2022. He said that this valuation was based on standard valuation techniques. In subsequent evidence, however (and in particular his oral evidence at the trial), Mr Colley said that although for convenience his modelling applied different approaches to different time periods, his overall approach was to model lost profits and he was not conducting a “business valuation” as such for any period of time.

[598]MGA took a different approach. MGA said that the proper approach was for the court to award damages for lost profits up to the assumed date of demise, i.e. 1 January 2018, and then to assess the hypothetical capital value of the business at that date, the latter of which would reflect the risks and uncertainties of future trading. On that basis, and having been given permission to call a separate valuation expert as a result of the amendment to Cabo’s pleaded case, MGA’s evidence on quantum was divided across its two economic experts as follows: i) Mr Parker’s instructions were specifically limited to the modelling of lost profits. He did so on two bases: first, across the full lifetime of Worldeez, which he assumed to be three years rather than five; and secondly up to the assumed date of demise of Cabo. Mr Parker did not respond to Mr Colley’s evidence on the perpetuity value of the business. ii) Mr Davies was instructed to consider only the “valuation assumptions” in Mr Colley’s evidence. He did not put forward any positive valuation himself, but simply addressed what he considered to be the “valuation aspects” of Mr Colley’s evidence, both in relation to lost profits in the period following the date of demise, and Mr Colley’s evidence regarding the perpetuity value of the business from summer 2022 onwards.

[599]MGA’s closing submissions addressed extensively the issue of whether the correct approach should be a valuation approach after the assumed date of demise. In that regard MGA relied in particular on the judgments of the Court of Appeal in UYB v British Railways Board

[2000] EWCA Civ 265

, Crehan v Inntrepreneur

[2004] EWCA Civ 637

, and Salford City Council v Torkington

[2004] EWCA Civ 1646

. On the basis of those authorities, MGA submitted that there was a “consistent approach” taken by the courts in cases such as the present, that the court will not award damages for lost profits in the period after a business has been lost, but should instead take a capital value of the hypothetical business at that point, based on the value which the business would have fetched in the open market.

[600]Cabo’s closing submissions maintained the position that it was not appropriate to postulate a “valuation” of its business at the date of demise, on the grounds that Cabo did not intend to sell its business then. Ultimately, however, Ms Kreisberger said in her oral closing submissions that she considered the difference between the parties on this point to be one of labelling rather than a dispute of substance.

[601]The authorities relied upon by MGA (and several others) were discussed extensively in the March 2022 judgment of Joanna Smith J, addressing the same submission made by MGA at that hearing. I do not need to repeat that discussion. It is sufficient to say that the judge concluded that the authorities did not establish the rule for which MGA contended. Rather, she considered that every case must be carefully considered by reference to its own specific facts, and having regard to the factual and expert evidence, so as to ensure full compensation to the claimant: see in particular §29 and §§62–63 of the judgment.

[602]MGA’s submission at the trial therefore repeated an argument which had already been advanced and had failed before Joanna Smith J. Even leaving aside the question of whether it was open to MGA to reopen the point, I agree with the conclusions of Joanna Smith J. The authorities cited by MGA reflect particular approaches adopted in those cases, and do not fetter the court in the assessment of any other case. It is not, in my judgment, appropriate or even possible to be prescriptive about the way in which the court will assess a claim for losses arising from the failure of a business. As Joanna Smith J said, each case will turn on its own facts, and the court will need to consider those facts carefully together with any expert evidence relied on by the parties.

[603]In the present case, the experts ultimately agreed on the quantification framework set out in the DTM, described at §§58–60 above, which incorporated aspects of both profitability and valuation modelling techniques. When asked in closing submissions what difference it made whether the analysis using the series of steps defined in the DTM was described as a purely lost profits assessment (Cabo’s position) or as a valuation exercise after the date of demise (MGA’s position), Ms MacLeod’s answer was that the assessment of loss by reference to the period after the date of demise had to be carried out through a “valuation lens”, and by using “valuation principles”. She said that this was required because a valuation approach was the only legitimate basis for exercising judgment on the points of detail under the DTM for the period after the date of demise.

[604]I do not accept that submission. If (contrary to my conclusions above) it had been established that Cabo would have been profitable in the counterfactual scenario, it would then have been necessary to make an assessment of Cabo’s loss in a manner which took account of the inevitably speculative and uncertain nature of the exercise. As Toulson J put it in Fyffes Group v Templeman, that exercise may involve an assessment of “all shades of risks and probabilities”. It may be that in some cases the question of whether a profitability or valuation approach is taken will make a difference to that assessment. In the present case, however, the overall framework of the DTM does not change according to whether the analysis is described as a profitability assessment or a valuation assessment after the date of demise – whatever label is used, the DTM requires the court to consider the appropriate counterfactual scenarios used for the inputs of the model, and to consider whether to apply further discounts to the figures produced by those scenarios to take account of the uncertainty that Cabo would in fact have achieved the success predicted by those scenarios.

[605]The court’s exercise of judgment on those points will turn on the factual and expert evidence, and there was significant disagreement between the parties as to the inputs to the DTM. Those disagreements ultimately turned, however, on each party’s assessment of the evidence before the court, rather than on the question of whether a “valuation approach” was being taken to that evidence.

Mr Colley’s quantum models

[606]Mr Colley’s expert reports calculated the potential profitability of Worldeez, using two alternative models. The first model was a Bertrand simulation model, with various different assumptions as to the success of Worldeez within that model. The alternative model was a simpler model of profitability using the financial projections prepared by Cabo for Nickelodeon.

[607]As already noted at §501 above, Mr Colley’s evidence was not that Cabo would in fact have achieved the success modelled in any of the permutations of his simulation model, or the Cabo projection model. Rather, his models sought to provide an indication of the profits that Cabo could have obtained if it had achieved the level of commercial success assumed in the variants of his simulation model or the Cabo projection model. Whether or not Cabo would have achieved those levels of commercial success was, Mr Colley accepted, a matter for the court based on the factual evidence and the evidence of the toy experts.

[608]There is no dispute that a Bertrand simulation model can be a useful model to quantify damages arising from a competition law infringement. The use of simulation models (including Bertrand models) is discussed in the Commission Staff Working Document “Practical Guide: Quantifying harm in actions for damages based on breaches of Article 101 or 102 of the Treaty on the Functioning of the European Union”, C(2013) 3440. In particular, the Practical Guide notes that: i) The Bertrand model of competition describes a market with a relatively small number of firms (and high barriers to entry) that compete on price and not output quantity. Under this model, prices are set simultaneously by firms based on beliefs about the prices their competitors will charge. Prices increase with the degree of product differentiation (§98). ii) Bertrand oligopolies in markets with differentiated products will normally lead to prices and volumes somewhere between perfect competition and monopoly levels, with the exact outcome depending on the number of firms in the market and barriers to entry, the degree of differentiation, and other characteristics of the market such as demand characteristics and the capacities and cost structure of producers (§99). iii) Simulation models can be used to simulate the sales volume and market share a foreclosed competitor would have attained if an exclusionary infringement had not taken place. It should be constructed to replicate the most significant factors influencing supply (in particular the competitive interaction between firms and their cost structures) and demand conditions (§§100 and 103). iv) Although such models rely on a simplification of reality, they may provide useful insights regarding the likely damages. Their reliability depends, however, on the right assumptions being made, particularly as to the likely conditions of competition and consumer demand in the non-infringement scenario (§104).

[609]Mr Colley started from the premise that a Bertrand simulation model was a reasonable approach to the modelling of damages, bearing in mind the inherent difficulty of estimating Worldeez’ performance in the counterfactual scenario. His chosen model was a differentiated product Bertrand model, which assumed that equilibrium prices could be significantly above marginal cost, because of the quality differences between products.

[610]In order to populate that model, Mr Colley looked at five comparator brands in the market in 2017–2018: LOL Surprise, Shopkins, Hatchimals Colleggtibles, Num Noms, and Mash’ems/Fash’ems. He used the actual world retail prices and market shares (by both revenues and volumes) in those years to derive a “quality index” for each of those brands. Mr Colley described the quality index as reflecting the “consumer appeal of a brand” based on factors such as the features of the products and age of the brand. It was, however, common ground that the quality index also inevitably reflected other non-price factors that affected sales volumes, such as marketing and supply chain issues. As Mr Colley accepted in his oral evidence, “quality” in this context did not mean purely product design quality, but was rather a measure of the commercial success of the products in the market, in terms of sales volumes.

[611]Mr Colley then defined hypothetical quality indices for Worldeez, based on his calculations of the quality index for the comparator brands, weighting those comparators in different ways. The simulation model then predicted how Worldeez would have performed in the market if it had achieved the assumed quality level, taking into account its actual world prices, marginal costs and the simulated competition from the other brands if Worldeez had been present on the market. In short, therefore, the models predicted what Worldeez’ revenues might have been if Worldeez had achieved the same commercial success as the brands in the comparator sets chosen, with the various different weightings he selected.

[612]Mr Colley’s primary calculations were made using several different comparator sets: i) Base Case 1 specified a Worldeez quality index calculated as the average of the five comparator brands. ii) Base Case 2 assumed that Worldeez’ market performance would have been more similar to that of LOL Surprise and Hatchimals Colleggtibles than the other three comparator brands. This model therefore calculated the Worldeez quality index by doubling the weight given to the LOL Surprise and Hatchimals Colleggtibles quality indices, compared to the weight given to the quality indices of the other three comparators. iii) The two base case assumptions were then tested using various sensitivity analyses, one of which gave Worldeez the same quality index as the Hatchimals Colleggtibles brand.

[613]Mr Colley used his simulation model to predict Worldeez’ counterfactual sales volumes and revenues in 2017 and 2018. He then extrapolated from those predictions to estimate volumes and revenues in subsequent years, as well as licensing revenues and international sales, based on various assumptions set out in his evidence. His profitability calculations assumed that the cost of goods would be as in the Nickelodeon projections.

[614]The alternative profitability model based on the Cabo projections was far simpler. It was not a simulation model but simply calculated lost profits taking the Cabo projections as a starting point, extending that to further years, further retailers (Argos, Amazon and Claire’s), and adding in licensing revenues and international sales, on the same assumptions for those as used for his simulation models.

[615]When cross-examined on the relevance and reliability of his profitability models, Mr Colley emphasised that the purpose of his evidence was simply to model different versions of success in the counterfactual, and that it was not for him to opine on the chances of Worldeez achieving those levels of success. He accepted that his simulation modelling was only relevant if the court concluded that Worldeez would have achieved sales volumes comparable to those of the comparators used his model. Since the Cabo projections model assumed even higher sales volumes than considered in the simulation modelling, Mr Colley accepted that if the court considered that Cabo would not have achieved the commercial success assumed in in the simulation models, that would also undermine the utility of the Cabo projection model:

“… it’s got to be reasonable that if your main concern looking at the simulation results is that you don’t think that Worldeez was going to be of sufficient quality to … compete with those comparators that were achieving that certain level of sales, and the Cabo forecasts imply an even higher level of sales, it would be difficult how you could sort of square the one with the other.”

[616]In its opening submissions Cabo relied on all three of Base Cases 1 and 2 and the Cabo projections model. The version of the DTM originally provided to the court for the trial therefore included, in the input cashflow assumptions, a midpoint of the two Base Cases, the Cabo projections model, and an average of the Base Case midpoint and the Cabo projections model. By the time of the closing submissions, however, Cabo had abandoned Base Case 2. As for the Cabo projections model, the most that was said was that it was a “further data point to which weight should be attached”, on the basis that it provided some insight into the evolution of the business which Cabo and Singleton were expecting at the time. Notably, however, the inputs for the final version of the DTM did not include either Base Case 2 or the Cabo projections model: in reality, therefore, both were abandoned by Cabo as providing any useful quantum model for the court.

[617]Instead, Cabo’s closing submissions relied on the following three versions of the simulation model, which were the versions included in the final version of the DTM: i) Base Case 1as described above. ii) A scenario giving Worldeez the same quality index as Hatchimals Colleggtibles. iii) An average of Base Case 1 and the Hatchimals scenario.

[618]Assuming a five-year lifecycle for Worldeez, and using Cabo’s costs stack, Cabo provided the following estimates for its losses (in millions of pounds), on two alternative bases. The first was that Cabo would have been successful at securing both international distribution for Worldeez and licensing deals (in both the UK and overseas), and did not therefore make any adjustment for the probability of those outcomes. The second was that the loss calculation should be subject to probability adjustments to reflect an 80% probability of international distribution and a 65% probability of licensing deals. The calculations also included future revenue streams reflecting future unknown product, as well as interest. The figures are shown in the following table, with brackets indicating the probability-adjusted figures. 624. Table 6: Cabo calculations of losses (£m) UK International Total (inc unknown products + interest) Base Case 1 6.4 (5.6) 40.3 (27.8) 71.8 (53.3) Hatchimals 8.3 (7.3) 52.3 (36.5) 90.3 (67.4) Average of BC1/Hatchimals 7.4 (6.5) 46.3 (32.2) 81.0 (60.4)

[619]On all of these scenarios, while UK sales were calculated as being profitable overall, the revenues from international sales and international licensing made up the vast majority of the loss figures.

Assessment of Mr Colley’s models

[620]I do not need to address MGA’s submissions on the Base Case 2 simulation model or the Cabo projections model, given the position reached on these by the end of the trial. The relevant question is the reliability of the Base Case 1 and Hatchimals simulations, as presented in Cabo’s closing submissions. As to that, MGA criticised both the overarching approach of the model, as well as its main input assumptions.

[621]Overarching approach. MGA said that Mr Colley’s simulation approach was based on a hypothesis of market dynamics which did not reflect the reality of the market for collectible toys. I have some sympathy for that submission. Ultimately, any estimate of the profits which Cabo might have made, in a counterfactual scenario which did not occur in fact, is likely to be highly speculative and based on very broad-brush estimates and assumptions. The court’s analysis in such a case is a paradigm example of the wielding of the broad judicial axe. In those circumstances the construction of a sophisticated simulation model carries the risk of turning out to be an exercise of spurious precision. Whether or not its results are robust will depend upon the extent to which the model corresponds to the reality of the market. The more sophisticated the model, the more sensitive it is to the fallibility of its assumptions.

[622]The Bertrand model used by Mr Colley in this case was a model of static price competition in a market with a relatively small number of firms competing on price. That is likely to be an unreliable predictor of market dynamics in a volatile market with a large number of competitors, with significant competition on non-price factors (in this case product differentiation).

[623]An example of the problem of using this sort of model was the fact that the output prices for Worldeez predicted by the model, based on the simulation of the price competition that would occur under the model, were far higher than Cabo was able to charge in fact. For example, the weighted average retail price for the globe predicted by Base Case 1 was over £4 in 2017 and around £3.50 in 2018, whereas the globe was in fact sold for £2.99, and the evidence supported this as the right price for the product both in the actual and the counterfactual case (see §550 above). Mr Colley, when cross-examined on this point, confirmed that he was aware of the actual selling price of the globe, but that the model “comes up with another equilibrium” as a mathematical result. That answer did not, however, grapple with the question of how the court could usefully use a mathematical model which resulted inoutcomes that were on any basis (even in the counterfactual scenario) wholly unrealistic.

[624]In Cabo’s closing submissions, Cabo proposed that this problem could be addressed by overriding or constraining the output retail price in the model for 2017 (only), so as to correspond to the actual Worldeez launch prices, while retaining a higher retail price for subsequent years. The final version of the DTM reflected that price constraint for each version of Mr Colley’s model. This proposal was not, however, addressed in any of Mr Colley’s evidence prior to the trial. His only mention of the point was a very brief comment, on the afternoon of the second day of his cross-examination, that it would be possible to constrain the prices in the model. That was far too late for this to be explored with him and indeed with Mr Parker.

[625]Cabo’s proposed solution was, moreover, internally contradictory. On the one hand, Cabo said that Mr Colley’s model predicted a slightly lower retail price in 2018 than in 2017 because of the hypothesised increased competition from LOL Surprise, which would have required other brands to lower their prices to compete. On the other hand, Cabo’s price-constraint solution implied that Cabo would have increased the price of Worldeez from 2017 to 2018, with Cabo justifying that by saying that in the light of early success it was not unrealistic to assume that Worldeez would have raised its price. Ultimately, Cabo’s solution simply highlighted the artificiality of the model on which it relied.

[626]The selected comparators. Turning to Mr Colley’s input assumptions, it was common ground that since Mr Colley was not himself opining on the prospects of success, all his model could do was to predict profitability if Cabo had enjoyed the levels of success identified as the starting assumptions in the various versions of the model. That in turn depended on the whether the comparators chosen by Mr Colley to calculate his “quality index” for Worldeez were reliable predictors of the commercial success that Worldeez would have achieved in the counterfactual case. I agree with MGA’s submission that the comparators chosen by Mr Colley were in fact wholly unrealistic.

[627]Mr Colley’s Base Case 1 specified as the comparator products five brands (LOL Surprise, Shopkins, Hatchimals Colleggtibles, Num Noms and Mash’ems/Fash’ems) which were by any measure extremely successful toy products – indeed, in the case of both LOL Surprise and Shopkins, phenomenally successful. All five were, moreover, brands marketed by major international toy manufacturers, with a track record of producing successful toys in the UK. Base Case 1 assumed that Worldeez would have achieved commercial success equivalent to the average of the success of those five brands.

[628]The evidence did not, however, suggest that Worldeez could realistically have achieved that level of success. Mr Harper’s view was that Worldeez would not have done as well commercially as Hatchimals Colleggtibles, for the reasons outlined at §520 above and discussed further below. He considered that Num Noms and Mash’ems/Fash’ems were also both better products than Worldeez (see also §520 above). As noted above, Ms Munt said very little about these products. As for LOL Surprise and Shopkins, there was no serious suggestion, at least by the time of the closing submissions, that Worldeez would have achieved anything approaching the commercial success of those products (whatever might be said about some aspects of the comparability of Worldeez with Shopkins, discussed above). Particularly when considered in the light of the extensive evidence as to the operational failings of Cabo, the selection of the Base Case 1 comparator set was hopelessly unrealistic as a measure of Cabo’s likely profitability.

[629]That left the comparison with Hatchimals Colleggtibles alone, or alongside Base Case 1 to produce an average figure. Prior to Cabo’s closing submissions, this comparison featured only briefly in an Annex to one of Mr Colley’s earlier reports as a sensitivity analysis, and was described as such in the first joint economic expert report. In Cabo’s closing submissions, however, it emerged as one of the preferred scenarios alongside Base Case 1. I do not consider that to be entirely satisfactory: unsurprisingly, since this comparison had never previously been relied on as one of Cabo’s primary profitability models, it was not addressed in any detail by Mr Parker, and was not specifically explored in Mr Colley’s cross-examination.

[630]In any event, the Hatchimals comparison was no more realistic than Base Case 1. Hatchimals Colleggtibles was one of a small set of extremely successful toy products in 2017, and was even more successful in 2018. While there was no dispute that, as a product, Hatchimals might be regarded as comparable to Worldeez, Mr Harper’s considered view (in both his written and oral evidence) was that he would have not expected Worldeez to perform as well as Hatchimals Colleggtibles, both because that collectible product was launched following the success of the original (larger and more expensive) Hatchimals product, and because of the fact that it was produced by an experienced international toy company, Spin Master. There was no evidence to contradict this: Ms Munt did not consider whether Worldeez would have enjoyed a similar level of commercial success to that of Hatchimals Colleggtibles.

[631]In fact, because of Hatchimals’ continued success after 2017 the Hatchimals comparison in the simulation model produced an even higher estimated level of sales and therefore revenues across the lifecycle of Worldeez than Base Case 1. If the level of success implied by Base Case 1 would have been unachievable for Worldeez, then the same would have been true a fortiori in relation to the Hatchimals benchmark. This alternative benchmark therefore simply substituted an unrealistically high success model with an even more unrealistically high success model.

[632]Cabo sought to characterise Mr Colley’s comparators as being “conservative”, suggesting that they placed Worldeez (in the counterfactual case) at the bottom end of the successful brands at the time. Cabo’s comparison was presented as a bar chart in its written closing submissions, with an amended version in its post-trial submissions. That comparison, however, was not explored in the expert evidence and suffered from similar problems to those arising with MGA’s comparisons with other products in the breakeven analysis (see §§575–576) above, namely that Cabo’s comparison was based on sales during the entirety of 2017, in circumstances where Worldeez did not launch until June in that year. Even if Worldeez had been as successful as (say) Hatchimals Colleggtibles in that year, a comparison based on sales only in that calendar year would present Worldeez as having lower revenue figures simply because of the shorter sales period. Cabo’s comparisons were also based on sales revenues alone, which did not reflect sales volumes given the quite different price points of the various products being compared. In addition, while Mr Colley’s simulation predicted profitability across a five-year lifespan, Cabo’s bar chart was limited to comparative sales in 2017, thereby masking the fact that Hatchimals went on to be more successful in 2018 product than in 2017. I do not, therefore, think that this comparison provided any meaningful indicator of the reliability of Mr Colley’s comparators.

[633]In my judgment, the comparators used to populate Mr Colley’s simulation model could not provide a reliable indication of Cabo’s likely profitability in the counterfactual case. While Mr Colley said, at the trial, that it would be a very easy matter to re-run the model with a different selection of comparator products (“It is a flick of a switch”), his evidence did not consider any other comparisons, despite his detailed assessment of other similar products for the purposes of his market definition. That was indicative of a lack of objectivity in his approach. Given that Mr Colley was not purporting to give evidence on the level of commercial success which Cabo would have enjoyed in the counterfactual case, it would have been helpful if Mr Colley had considered a broader spread of potential comparators, rather than confining his analysis to a small set of extremely successful products.

[634]Cabo initially suggested that if the court considered that the comparisons used by Mr Colley assumed a level of commercial success that Cabo would not in practice have achieved, that could be corrected by use of the probability adjustment feature of the DTM, i.e. by selecting one of Mr Colley’s models as the starting input, but then reducing the cashflows implied by that model by selecting a probability of less than 100%. While I agree that this would be a possible approach in theory, the mechanism is (as noted above) not a true probability adjustment, but simply a means of reducing the cashflow by a specified percentage. That would be a paradigm example of spurious precision: if the starting point is inherently unreliable and improbable, any reduction of that by the court is a venture even further into the realms of speculation.

[635]A far better approach, if available to the court, is to select as the starting point the model which most closely approximates the likely success of Worldeez, based on the evidence before the court, and to consider what that model indicates regarding the profitability of Worldeez in the counterfactual case. Notably, by the time of oral closing submissions, that point was essentially common ground, with Mr Kuppen agreeing that if the court was able to form a view (based on one of the models) as to the likely counterfactual trajectory of Cabo, there should then be little need for any probability adjustment.

[636]Licensing and international sales. MGA’s other main criticism of Mr Colley’s model was that it assumed that Worldeez would have been able to secure licensing deals, and also assumed very high levels of international sales. As shown above, the international revenues made up the vast majority of the profits predicted by all three versions of the model put forward by Cabo in its closing submissions. Given my findings above as to the suitability of Mr Colley’s model itself and the comparators used to populate it, the issue of licensing and international sales based on that model does not arise. For completeness, however, I agree with MGA that Mr Colley’s assumptions regarding licensing and international distribution were unrealistic and flawed.

[637]For licensing, Mr Colley based his calculations on an estimate produced by MGA in 2017 of its prospective sales of licensed products for LOL Surprise, relative to its LOL toy sales revenue, and a royalty rate of 18% on those sales. Both of those assumptions were wildly optimistic. There is no basis whatsoever to assume that Worldeez would have obtained licensing sales on the level of LOL Surprise, an extraordinarily successful toy marketed by an established and experienced toy brand. Nor was there any basis to assume that, even if Worldeez had concluded licensing deals, it would have obtained a royalty rate of 18%. Ms Munt suggested typical royalty rates of 10–12%, based on indicative rates in a third party licensing handbook, the Brand Licensing Handbook. She accepted, however, in cross-examination that she had not been involved in the negotiation of royalty rates. Mr Harper did have direct experience in negotiating licensing agreements (albeit not in the toy market), and was able to give evidence of his knowledge of royalty rates drawing on his extensive experience in the toy industry. His evidence was thatwithout substantial TV programming support for a brand, its royalty rate on licensing was likely to be around 5%. The one licence deal which Cabo was offered, in March 2018, was a magazine licence which offered a royalty rate of 5–7%, depending on sales. Neither the factual nor the expert evidence therefore supported the royalty rate of 18% used by Mr Colley.

[638]Mr Colley also accepted in cross-examination that his estimates as to LOL’s licensing revenue (from which he derived his predicted licensing revenues for Worldeez) were themselves very significantly overstated. While he had estimated that MGA had made £6.7m in UK licensing revenues for LOL Surprise in 2017, in fact MGA’s internal reporting indicated that it had generated only $2.4m global licensing revenue in that year. Mr Colley’s methodology did not, therefore, provide a reliable basis for an estimation of Worldeez’ prospective licensing revenues, even if Worldeez could realistically have been expected to generate licensing revenues on a scale comparable to those of LOL Surprise (which it plainly could not).

[639]As for international sales, for the reasons discussed above I do not accept that if MGA had not intervened Cabo would have been successful in selling Worldeez outside the UK, the essential problem being the feasibility of any international distribution arrangement given Cabo’s cost model and the working capital requirements for international sales. These were not matters considered by Mr Colley, who simply assumed that international sales in Europe, North and South America, and Australia would have been likely once Worldeez had “gained traction” in the domestic market.

[640]It is therefore apparent from the evidence that Mr Colley’s estimates of licensing and international sales revenues were unrealistic, even leaving aside the problems with the comparators used for Mr Colley’s simulation model. This adds to a picture of a model which could not ever have provided a reasonable and objective estimation of Cabo’s counterfactual losses.

Mr Parker’s quantum models

[641]Mr Parker adopted a different (and simpler) approach to that of Mr Colley. Rather than constructing a simulation model, he calculated what Cabo’s profitability would have been if it had achieved the same volume of sales as a selected comparator product or basket of products, using NPD sales data. As with Mr Colley’s models, Mr Parker’s evidence was not that his models were reliable proxies for what Cabo’s likely success in the counterfactual scenario. Rather, he sought to assess what Cabo’s profitability would have been if Cabo had achieved the sales volumes or revenues set out in his models.

[642]Mr Parker constructed two models illustrating alternative levels of commercial success. Each model calculated profitability over an assumed lifecycle of three years for Worldeez: i) Mr Parker’s “significant success” model assessed the likely lost profits if Worldeez had achieved the same sales volumes in its first three years as My Little Pony Fash’ems during the years 2016–2018. This product (pictured at §520.i) above) was the top selling toy in the Mash’ems/ Fash’ems range during that period, and was included in both Mr Colley’s and Mr Parker’s proposed market definitions. It was comparable in price to the Worldeez globe, with an average retail price of £2.84 in both 2016 and 2017, increasing to around £3.00 in 2018. The NPD data showed sales (by volume) for My Little Pony Fash’ems of 479,572 in 2016, 498,748 in 2017 and 269,428 in 2018. Those sales were taken as indicating significantly successful sales volumes in the first three years of marketing Worldeez. ii) Mr Parker’s “moderate success” model was based on a basket of comparators made up of the best performing product of every brand newly launched between July 2018 and July 2019 in the playset dolls and collectibles class (using that period because of limited data availability for the period before then). Mr Parker used average month by month revenues for that basket of products, over three years, as a proxy for the revenues which Cabo might have achieved in its first three years. Mr Parker placed more weight on this model than his significant success model, contending that this was the better way of considering the likely success of Cabo in the counterfactual case.

[643]Mr Parker’s models did not include any licensing revenues, on the basis that he considered licensing to be unlikely given the limited UK revenues that his scenarios predicted, and Cabo’s lack of experience with licensing. He did, however, include a “broad brush” estimate of international sales during the second year of sales onwards, assuming that these would generate 10 times the UK sales, and that Cabo would have obtained a 15% margin on its sales to international distributors.

[644]Both models used cost assumptions which were based on the costs estimated in the Cabo projections. What Mr Parker did not consider was whether those costs assumptions would be realistic if Cabo’s sales levels were (as in his scenarios) far smaller than those predicted in the Cabo projections.

[645]In its closing submissions, and in the final updated version of the DTM provided after the trial, MGA updated Mr Parker’s calculations in several ways. In particular: i) MGA updated the costs assumptions for Mr Parker’s models in light of the questions put to Mr Parker in cross-examination. It also (again in light of the cross-examination of Mr Parker) uplifted the sales figures in both models by 33% to account for the fact that the NPD data covered only 75% of the market. ii) In respect of the significant success model, MGA produced a further sensitivity analysis based on the sales figures for the entirety of the Mash’ems/Fash’ems range (i.e. including all of the Mash’ems and Fash’ems toys, rather than only using the sales generated by the My Little Pony Fash’ems product). iii) In respect of the moderate success model, MGA produced a further calculation which assumed that Worldeez would have been twice as successful as the average product in the basket of comparator products.

[646]MGA’s updates of Mr Parker’s calculations produced the following figures for UK and international profits over an assumed three-year lifecycle of Worldeez. 653. Table 7: MGA calculations of losses (£) January 2018 UK only UK – 3 years International – last 2 years Total Significant success -83,451 -243,168 1,210,066 966,899 Significant success/all Mash’ems & Fash’ems 15,700 629,072 3,280,646 3,909,718 Moderate success -166,652 -550,399 -12,272 -562,671 Moderate success/2x average comparator sales -98,874 -329,676 121,174 -208,520

[647]Mr Parker’s models assumed an average wholesale price for the globe of £1.33 in 2017 and £1.37 in 2018 onwards, based on a weighted average of the prices paid by the retailers that did stock Worldeez. The figures set out in the table above are based on that price assumption.

[648]In its closing submissions, however, MGA contended for the first time (contrary to Mr Parker’s position) that the correct average wholesale price to assume in the quantum analysis was in fact not Mr Parker’s figures of £1.33 and £1.37, but was the price paid by The Entertainer of £1.20. Further profitability calculations were therefore provided by MGA using that price base, and the final version of the DTM included this as an option. Those price assumptions inevitably led to far lower profitability figures (or even greater loss figures) than the figures based on MGA’s cost stack set out in the table above.

[649]As Cabo pointed out, this was a new argument which it had not been able to explore with Mr Parker and other witnesses. Nor was it supported by the evidence in any event (particularly given that Cabo’s main retail outlet B&M had in fact agreed a higher wholesale price of £1.50). As set out at §554 above, a reasonable average wholesale price to assume for the counterfactual scenario would be around £1.35. Mr Parker’s weighted average figures are very close to that, and it is not necessary to comment further on MGA’s alternative profit figures using a wholesale price of £1.20.

Assessment of Mr Parker’s significant success model

[650]Cabo advanced three main criticisms of Mr Parker’s significant success model, by reference to the comparator chosen and the cost and revenue inputs to the model.

[651]Product comparison.Cabo’s first objection was that My Little Pony Fash’ems was a poor indicator for the performance of Worldeez in a “significant success” scenario, on the basis that the product was first launched some years previously, in 2013; each series had a smaller number of figurines to collect than the Worldeez launch series; and the packaging was simpler than Worldeez.

[652]I do not accept those objections. In a model which attempts to predict profitability in a counterfactual scenario using a real-world comparator, the question is not whether the comparator product shares precisely the same qualitative features as the product under investigation (which in a highly differentiated product market will inevitably be difficult to replicate). Rather, the question is whether the comparator is a reasonable proxy for the performance of the product under investigation, in the counterfactual scenario being investigated. That will undoubtedly involve consideration of the features of the product, but will also take into account other factors relevant to the comparison of likely market performance, such as the price point at which it was sold and the experience and reputation of the supplier.

[653]In that regard, My Little Pony Fash’ems had qualitative features which made it in at least some relevant respects comparable to Worldeez: it was one of the main collectibles included in Mr Colley’s market definition, and as such it was considered by Cabo (and its toy expert Ms Munt) to have a “sophisticated unwrapping experience”. It was also sold at a price point very similar to that of Worldeez. Although each series had only six or seven characters to collect, which was a much smaller number of characters than the figurines in the Worldeez series, numerous series had been released since the initial launch of the product (and further series were released in the years 2016–18). There was, moreover, no evidence that the limited number of characters in each series diminished the playability or likely commercial success of the product compared with that of Worldeez. On the contrary, Mr Harper considered Mash’ems/Fash’ems to be a better and more playable product than Worldeez.

[654]The main reason for caution in respect of this comparator is that My Little Pony Fash’ems was not merely a significantly successful product, but was in fact a very successful product in its price range. The NPD sales data showed that in 2016, for similar priced collectibles, My Little Pony Fash’ems was only outsold by the Shopkins 2-pack and Num Noms. In 2017 there were only three more successful collectible products at a similar price point: the Shopkins 2-pack, Trolls blind bags and Hatchimals Colleggtibles single pack. The product was, moreover, marketed under the banner of an established and very successful brand, produced by an experienced international toy company.

[655]Given the findings I have already made as to Worldeez’ product quality, marketing and retailer support, and Cabo’s operational abilities, I consider that My Little Pony Fash’ems represented a level of success that was greater than Worldeez would have been likely to achieve in practice. That conclusion is supported by consideration of the sales volumes of My Little Pony Fash’ems during the years used for the comparison. The NPD data for 2016, uplifted to account for its 75% market coverage (see below), indicated My Little Pony Fash’ems unit sales volumes of 639,429, or on average almost 12,300 a week. In similar vein to the discussion at §579.i) above, that is far greater than Cabo could have been expected to achieve, given the known peak sales of 4,500 per week at B&M.

[656]Use of a single product. Cabo’s second criticism was that the My Little Pony Fash’ems product was only a single product in the Mash’ems and Fash’ems range, and was therefore not a reliable benchmark for the success of the entire Worldeez brand. That criticism, in my judgment, failed to appreciate the fact that, as MGA pointed out, Mash’ems/Fash’ems split their products by theme, e.g. My Little Pony Fash’ems, Disney Frozen Fash’ems, Peppa Pig Mash’ems, Paw Patrol Mash’ems, etc, with each theme having a completely different character range which did not necessarily interact with the other themes. It was therefore reasonable for this scenario to use My Little Pony Fash’ems rather than the whole of the Mash’ems/Fash’ems range, particularly given that My Little Pony Fash’ems was by far the best-selling Mash’ems/Fash’ems product at the time, with almost double the sales of the next-best selling product (Peppa Pig Mash’ems). Again, the real concern with the comparator is that it is likely to represent an overstatement of Worldeez’ counterfactual success.

[657]It follows that while MGA’s closing submissions provided a profitability analysis based on the sales volumes of the entirety of the Mash’ems/Fash’ems range, by way of a sensitivity analysis, Worldeez could a fortiori not realistically have expected to achieve that level of sales. It is therefore not necessary to comment further on that analysis.

[658]Cost and revenue assumptions. Cabo’s third criticism was that Mr Parker’s model contained serious errors in its assumptions of costs and revenues. In particular (among other things), Mr Parker had adopted the costs figures from the Cabo projections without giving any consideration to whether those figures would have been reasonable in a scenario which assumed far smaller sales figures than the predictions of the Cabo projections.

[659]That was, in my judgment, a fair criticism. As an example of the problems with Mr Parker’s approach to costs, his year three model (based on the My Little Pony Fash’ems 2018 sales figures set out above) assumed that domestic revenues would have reduced by almost 50%. But his marketing and product development costs figures simply lifted the Cabo projections, which had assumed a significant increase in revenues, and on that basis a significant increase in the marketing and development costs. As a result, in Mr Parker’s model the assumed marketing and product development costs in year 3 exceeded the assumed domestic revenues. When asked about this in his cross-examination, Mr Parker did not even attempt to justify what he had done:

“Q. So in essence, in the actual outcome, management does what one might call the sensible thing and cut its losses, whilst in your scenario you assume revenues fall off the cliff, we know it is the end of the product life and we keep on spending on product development at 6 times the rate that the toy experts say we should be doing and spend essentially 100% of revenues on marketing. A. I agree that doesn’t seem very sensible. Unfortunately … those are the numbers that are in the Cabo projections and they don’t come with revenue shares attached. Q. I am not talking about the Cabo projections. I am talking about your scenario. A. Agreed, yes. So I don’t make any adjustments.”

[660]While it was to Mr Parker’s credit that he gave frank answers to these questions, the fact that he had seen fit to base his analysis on costs figures which related to a very different revenue projection, without any reflection on whether those costs figures were realistic in the light of his reduced revenue projections, indicates a lack of objectivity in this part of his evidence. It was also apparent that Mr Parker had based his annual revenue figures on the NPD data for My Little Pony Fash’ems (see §642.i) above) without uplifting those figures to account for the fact that the NPD data covered only 75% of the market.

[661]These and other errors were corrected in MGA’s amendments to Mr Parker’s model in its closing submissions, producing the updated profitability calculations set out in Table 7 above. It is, however, necessary to consider the effect of other disputes on the cost stack which remained by the end of the trial (see §578 above). The following table shows the effect of using(i) MGA’s cost stack with Mr Parker’s wholesale prices (£1.33 and £1.37) – as above in Table 7;

(ii) Cabo’s cost stack with Cabo’s proposed wholesale price (£1.52); and

(iii) a combined calculation requested by the court, after the trial, using Mr Parker’s prices but Cabo’s cost stack. 669. Table 8: Loss calculation for significant success model, comparing MGA and Cabo cost stacks (£) Cost/price model January 2018 UK total International Total MGA cost stack/Parker prices -83,451 -243,168 1,210,066 966,899 Cabo cost stack/Cabo prices 84,578 464,658 1,882,979 2,347,636 Cabo cost stack/Parker prices -8,554 114,753 1,882,979 1,997,731

[662]For the reasons discussed above, Mr Parker’s wholesale price assumptions were reasonable and are to be preferred to Cabo’s assumed wholesale price of £1.52. The figures to consider are therefore the first and third rows of Table 8 above. Those figures show that on either Cabo’s or MGA’s cost stack, using Mr Parker’s wholesale price assumptions, even if Worldeez had enjoyed sales of the level of My Little Pony Fash’ems, it would have remained loss-making by January 2018. On Cabo’s cost stack it would have been minimally profitable if it had remained in the UK market in 2018 and beyond, but that would have been unlikely given Singleton’s lack of appetite to support a loss-making business. On MGA’s cost stack it would have been entirely loss-making in the domestic market.

[663]Worldeez’ profitability as indicated by this model. Having addressed Cabo’s criticisms of Mr Parker’s significant success model, and corrected the model in respect of the errors in the cost and revenue assumptions, the final question is what therefore the extent to which Mr Parker’s model of significant success reflected the success which Cabo could reasonably have achieved for Worldeez in the UK and internationally, given that (like Mr Colley) Mr Parker was not qualified to opine on the likely commercial success of Worldeez as a product, on the basis of its product quality taken together with the other factors which are relevant to an assessment of likely commercial success.

[664]The figures set out above therefore overstate the revenues that Cabo could realistically expect to have obtained in the UK, indicating that Cabo would in the counterfactual case have been even more loss-making in the domestic market than is set out in the first and third rows of Table 8 above, rendering it even less likely that Singleton would have been prepared to continue to invest beyond the end of 2017. On either Cabo or MGA’s cost stacks, therefore, Mr Parker’s significant success model indicates that Cabo would have remained loss-making and would have exited the market during 2018.

[665]Mr Parker was therefore right, in my judgment, to disregard UK licensing revenues, on the basis that licensing would have been very unlikely on the levels of sales which his model predicted. Any significant profitability (albeit on a far lower scale than predicted by any of Mr Colley’s scenarios) would therefore have rested on international expansion, and the ability of Cabo to fund that.

[666]As noted above, it was common ground that international revenues would likewise not have materialised if Worldeez was not profitable domestically. In any event, as discussed above, international distribution was not likely to have been viable for Worldeez given Cabo’s cost model and the competition from other similar products.

Assessment of Mr Parker’s moderate success model

[667]As set out above, Mr Parker’s moderate success model estimated Worldeez’ revenues by reference to the average revenues of the best performing product of every brand in the NPD playset dolls and collectibles class, over a period of three years, specifically focusing on products that were newly launched between July 2018 and July 2019. Mr Parker used NPD data for the top 50 products of this category, and excluded any product which did not achieve sales of at least £1000 in at least three months. That gave him a comparator set of 27 products. The average monthly revenue of those products for their first three years in the market was as set out in the following diagram. Mr Parker used those revenue figures as a proxy for Worldeez’ revenues during its assumed lifecycle of three years. Figure 2: Parker moderate success model: average monthly revenue (£)

[668]As set out in Table 7 above, the moderate success model indicated that Worldeez would have been significantly loss-making, including on international sales. The further calculation provided with MGA’s closing submissions assumed that Worldeez would have been twice as successful as the average product in Mr Parker’s basket of comparators, and on that basis suggested (limited) profitability on international sales, but overall losses over three years when domestic losses were taken into account.

[669]Other than the various points on Mr Parker’s costs and revenue assumptions which were corrected in the updated figures provided with MGA’s closing submissions, Cabo’s main criticism of this model was that by basing itself on an average of the performance of the top products in all newly-launched brands post-July 2018, it took no account of what was known about Worldeez’ performance. Given that the vast majority of the products in the comparator set generated less revenue in year 1 than Cabo’s actual performance, in a situation where Cabo’s market access was impeded by MGA’s conduct, a straight average across all of the comparator products would not (Cabo submitted) provide a reasonable picture of what Cabo might have expected had it been able to pursue its original launch plans unimpeded by MGA.

[670]Mr Parker maintained that his moderate success model was more reliable than his significant success model, because “that is the expected outcome of a product you don’t know anything about”. But that misses the point that the commercial success of Worldeez was not a complete unknown. In addition to the extensive evidence discussed above regarding the quality of Worldeez as a product and the operational abilities of Cabo, the evidence in the trial included detailed sales figures for all the retail outlets where Worldeez was sold.

[671]Accordingly, while the moderate success model provides some insight as to the average trajectory of a typical product in the playset toys and collectibles class which enjoyed some degree of initial commercial success, it cannot be regarded as a useful proxy for the revenues of a product for which there is considerable real world post-launch information, including data showing revenues in a competitively constrained situation to have been better than that of most of the products in the comparator set.

[672]MGA’s attempt to address that issue, by providing a further analysis which assumed that Worldeez would have been twice as successful as the average from the comparator set, did not really resolve the problem. As Mr Kuppen put it in his closing submissions on this point, taking an average that is meaningless and multiplying it by two “may give a bigger number but it doesn’t solve the problem with this scenario”. I do not, therefore, consider Mr Parker’s moderate success model to be informative as a model of Worldeez’ expected performance in the counterfactual scenario. Notably, while MGA’s written closing submissions relied on both Mr Parker’s significant success and moderate success models, MGA did not in its oral closing submissions suggest that any significant weight could be given to the moderate success model.

Conclusions on the quantum models

[673]The different approaches of the parties to the modelling of profitability highlighted some of the limitations of the use of statistical modelling for the purposes of estimation of counterfactual profitability. Bearing in mind the inherently speculative nature of the exercise, a good counterfactual model is one that is based on reliable and unbiased data, with assumptions that are as close to reality as possible. As Oxera noted in its 2009 report for the European Commission “Quantifying antitrust damages: Towards non-binding guidance for courts”, §3.1.1:

“A model will only be as good as the quality of the input data used to populate it. … A sophisticated model based on unreliable or biased data is likely to be less robust than a simpler model based on better data. A critical question for a court to ask when reviewing a model is therefore whether the data used is of sufficient quality and reliability, and whether a simpler model relying on less, or more easily available, data could be used. There are many methods and models to describe the impact that an infringement of competition law might have. The robustness of each of these models partly depends on the validity of the assumptions that the model uses in the situation to which it is being applied.”

[674]By the end of the trial, Cabo only really relied on Mr Colley’s simulation model. That was a sophisticated model which ultimately did not closely reflect the nature of the market being investigated. It therefore predicted a mathematical equilibrium which simply did not correspond to the reality of the market, based on the known facts. That problem was compounded by the fact that the model was populated only with data biased heavily towards extremely successful comparator products. That displayed a lack of objectivity and provided the court with no assistance if the premise of those comparators (that Worldeez would have enjoyed levels of success comparable to the most successful similar products in the market) was rejected, as I have done. Mr Colley said that it would be easy to re-run his simulation with alternative comparators, but he notably did not carry out that exercise, despite having had a period of over two and a half years between his first expert report (in February 2022) and the trial.

[675]I do not, therefore, consider that the simulation model provided any useful and reliable indication of Cabo’s profitability in the counterfactual scenario. Nor do I consider that it would be useful to ask for the simulation to be re-run using different comparators, given the inherent problems with the application of that model to this particular market.

Figure 2: Parker moderate success model: average monthly revenue (£)

[676]Mr Parker’s comparator models were simpler and, in my judgment, in principle more appropriate for the quantification exercise in this case. The moderate success model provided some insight into the sales trajectory of an average collectible product during the relevant time period, but did not take any account of the features of Cabo and what was known about its level of success while it was on the market. Ultimately therefore it did not provide a useful indication of Cabo’s counterfactual profitability.

[677]That left Mr Parker’s significant success model using My Little Pony Fash’ems as a benchmark. Even if Worldeez had in its first three years achieved sales comparable to those of My Little Pony Fash’ems in the years 2016–18, on a reasonable assumption as to the wholesale prices at which Worldeez would have been sold, Mr Parker’s model showed that Worldeez would have been (on either version of the cost stack) either entirely loss making domestically or only marginally profitable in later years, if it had remained in the market despite losses in 2017. Moreover, given my earlier conclusions as to Worldeez’ product quality, marketing and retailer support, and Cabo’s operational abilities, and what is known of Cabo’s initial sales in 2017, I consider that this comparator overstated the likely success of Worldeez. Accordingly, in my judgment, Cabo would on any basis have been loss-making in the domestic market and would therefore not have generated either licensing revenues or international sales revenues.

[678]The significant success model therefore supports my conclusion reached on the basis of the other evidence examined in detail above, that Worldeez would not have been profitable in the counterfactual scenario. The basic reason for that is clear from the evidence before the court: Cabo’s financial model simply did not allow it to break even without huge sales volumes, which it was never realistically going to be able to achieve without a better and more innovative product, stronger retailer support and (crucially) better operational capabilities than it had, even with the support of Singleton.

[679]In those circumstances it is neither necessary nor appropriate to consider the further issues raised by the parties concerning the quantification of Cabo’s claim, based on the remaining options under the DTM.

Declaratory relief

[680]The final issue is the claim for declaratory relief on the patent threats claim, which Cabo seeks even if its primary damages claim fails. MGA says that there is no need for a declaration in circumstances where the threats occurred some time ago, Cabo has ceased trading, and MGA has given undertakings not to repeat the acts complained of as threats.

[681]I do not accept MGA’s submissions on this point. As Laddie J observed in Dimplex v De’Longhi

[1996] FSR 622

, p.627, s. 70(2) of the 1977 Act reflects a general policy that prima facie where the tort of patent threats has been made out, then the claimant is entitled to relief unless there are good reasons for deciding otherwise. In the present case, notwithstanding its undertakings not to repeat the conduct, MGA vehemently disputed the existence of any actionable threats throughout the proceedings, contending that the communications in question did not constitute relevant threats and that Cabo was not a person aggrieved. Given my conclusion that MGA’s conduct was, contrary to MGA’s protestations, unjustified patent threats within the meaning of s. 70, it is appropriate to remove any doubt on the issue by making a declaration to that effect.

Conclusions

[682]For the reasons set out above, my conclusions are as follows: i) MGA’s conduct was an abuse of its dominant position, contrary to the Chapter II prohibition and Article 102 TFEU, in the form of an exclusionary campaign based on threats to withdraw supply of LOL Surprise to existing customers, accompanied and reinforced by threats of litigation and disparaging claims that Worldeez was a “knock off” of LOL Surprise. ii) MGA’s conduct involved agreements with The Entertainer, TRU and Smyths which were in principle restrictive of competition by object; but those agreements were exempt under the VBER and therefore not prohibited by the Chapter I prohibition or Article 101(1) TFEU. iii) MGA made unjustified threats of patent infringement proceedings, within the meaning of s. 70 of the 1977 Act, to The Entertainer, TRU, Smyths and other retailers. If necessary, I would have found those threats to be causative of any loss suffered by Cabo as a result of MGA’s conduct. iv) Cabo’s damages claim ultimately turned on a claim that Cabo would have traded profitably but for MGA’s conduct. Having regard to all the evidence before the court, including in particular, the evidence of Worldeez’ quality as a product, the strength of Cabo’s marketing campaign, the extent of retailer support for Worldeez, Cabo’s financial projections and operational abilities, the working capital available to Cabo, the toy expert evidence on Cabo’s likely commercial success, and the analysis of the sales required to break even in 2017, I consider that Cabo would not have traded profitably in the counterfactual case. v) It is therefore not strictly necessary to consider the quantification of Cabo’s loss. Nevertheless, an assessment of the quantum models of Mr Colley and Mr Parker, for Cabo and MGA respectively, supports the conclusion that Cabo would not have been profitable in the counterfactual case. vi) Cabo’s claim for damages therefore fails. It is, however, appropriate to make a declaration that MGA’s conduct constituted unjustified threats of patent infringement proceedings.

order

8,000 x 2 Pk 3,000 x 5 Pk 5,000 x 10 Pk All above to be delivered in mixed FSDU Full requirements and delivery dates to be confirmed.” All above to be delivered in mixed FSDU Full requirements and delivery dates to be confirmed.”